Home

ISOCAM Interactive Analysis User`s Manual Version 5.0

1. 2 10 04 04 04 04 04 04 04 04 04 04 04 04 04 04 04 04 04 04 04 04 04 04 04 04 04 04 04 04 04 04 04 04 04 04 04 04 10 10 10 NN O1 O1 O1 C1 Aa a N 10 512 512 512 512 512 512 512 512 512 512 512 512 512 512 512 512 512 512 512 512 512 512 512 512 512 512 512 512 512 512 512 512 512 512 512 512 512 512 512 512 512 1 1 31 20 20 20 20 20 20 19 20 20 20 20 20 19 20 19 3 3 4 51 20 20 20 19 20 20 20 20 20 19 20 20 20 20 19 2 4 107 12 ra 161 161 161 161 161 161 161 161 161 161 161 161 161 161 161 161 161 244 282 319 356 386 349 311 274 303 341 378 415 445 407 370 333 333 KK kK kK kK kK K 161 161 161 161 161 161 161 161 161 161 161 161 161 161 161 161 161 161 161 161 161 244 244 282 319 356 386 349 311 274 303 341 378 415 445 408 370 333 333 333 333 333 55 55 56 56 55 55 55 55 55 55 55 55 55 55 55 55 55 967 983 000 016 995 979 962 946 925 941 958 974 953 937 920 904 904 Kk kk kK kK k 55 55 55 56 56 55 55 55 55 55 55 55 55 55 55 55 55 55 55 55 55 966
2. 2009 RANGE Y SCALE a Next frame Previous frame Pixel value Eubec1 9 106 4 0345 Figure 14 9 x3d window 134 CHAPTER 14 IMAGE ANALYSIS AND DISPLAY number If the IMAGE has had no dark correction or is badly dark corrected a regular saw tooth pattern will be easily discernible in the lines away from any source signal If the CUBE has good dark correction then the plot should just reveal background noise e Stabilisation CAM s CCD detectors can be slow to stabilise when the intensity of the incident light changes This is most evident when a strong source comes in to a pixel s FOV In x3d click on the button temporal cut and then on any bright pixel in the image window If the pixel contains a glitch then you may see that it takes several read outs to recover If the pixel contains a source then it is very likely that you will see a slow response or transient behaviour in the pixel to the source detection If the PDS has been calibrated without transient fitting with s90 m90 then click on the button mask to look at pixels flagged in the MASK The detected unstable pixels and dead pixels column 24 in the LW detector will now be marked in the plot window calibration has been performed well then only those pixels which exhibit transient behaviour will be masked Other methods of stabilisation which involve transient fitting i e fs inv and vision
3. This section introduces the CIA user to advanced concepts in slicing CVF data 19 5 1 Up and down CVF observation Since the LW CVF is split into two segments one complete LW CVF scan must be performed in two observations This means that a complete up and down LW CVF scan is comprised of four observations Consequently spdtoscd will output four SSCDs from the CISP data product CIA spdtoscd cisp10402702 fits raw sscd dir OLP1 IA SVAL 7 O B 3 CIA nowrite CIA cleaned sscds sscd clean raw sscd Out of 345 SCDs 5 are rejected due to mode 2 are rejected due to csh flag 8 are rejected due to qla flag In total 337 are accepted 16 Jun 1998 19 19 37 00 SSCD CLEAN v 2 0 lt Splitting SSCD into 4 segments I gt CIA print cleaned sscds 55 104027020001_98061619193759 55 104027020002_98061619194660 55 104027020003_98061619195839 55 104027020004_98061619201089 These four SSCDs may be concatenated into SSCD with sscd concatene Below the last three SSCDs are concatenated to the first SSCD 216 CHAPTER 19 ADVANCED SLICING liatia um 4 056 12 17 340 mre Figure 19 1 Spectrum from an up and down LW CVF scan The overlap between LW CVF segments at 9 5 jum can clearly be seen CIA sscd_concatene cleaned_sscds 0 cleaned sscds 1 CIA sscd concatene cleaned sscds 0 cleaned sscds 2 CIA sscd concatene cleaned sscds 0 cleaned sscds 3 A CVF PDS contai
4. s 343 Patched ASTROLIB and IDL routines in CIA 345 CONTENTS I Upgrading old CIA structures L1 Upgrading 2 0 5 L2 Upgrading 1 05 J Reporting problems and suggestions J 1 Problems with CIA software J 1 1 Template for a Software Problem Report 7 2 Comments on this document K Technical reports xiii 347 347 347 351 351 351 352 353 CONTENTS List of Figures 2 1 typical ciainfo display for UNIX 5 2 2 typical ciainfo display for VMS 6 2 3 typical cia help display 7 2 4 typical cia_html display for all routines 8 2 5 typical cia html display for graphic routines 9 2 6 typical cia html display for one routine 9 3 1 A raster IMAGE before calibration 27 3 2 A raster IMAGE after dark correction deglitching and stabilization 29 3 3 The calibrated raster 1 30 4 1 tviso display of EXPOSURE from staring observation 33 5 1 tviso display of the MOSAIC image from a solar system object observation 39 6 1 x3d display of a calibrated BS PDS CUBE 44 6 2 tviso display of beam switch 5 1 45 Tek
5. 15 5 2 PDS sca ye hehe qe wot ey Bl mu Rr 15 53 CVF PDS ak mox mue es Re BOR Mens 15 5 4 Paster PDS Ros EA GI Botte 15 5 5 BS PDS iae stu acd ey Ae todo an dade Se 15 5 6 CAL G PDS substructure CONTENTS CONTENTS 15 5 7 INFO PDS substructure 15 5 8 CCIM s 155 9 ASTR 16 Data structure manipulation 16 1 CIA data structure interface routines eh 16 1 1 structureinit 16 1 2 structure extract 16 1 8 structure put 16 1 4 structureget 16 1 5 structure write 16 1 6 structureread 16 1 7 structurelist 16 1 8 structuredel 16 1 9 structureinfo 16 1 10structurefind 16 1 11structureelem 16 2 An example of SAD manipulation 16 3 Saving and restoring PDSs 16 4 Manipulating MASK 16 4 1 Extracting the MASK from CIA data structures 16 4 2 Modifying the MASK 16 5 CDS data extraction 16 6 Manipulating CIA data structure history 16 6 1 Extracting the history 16 6 2 Replacing the history 17 Importing ISO data products to CIA 17 1 Importing FITS to CIA data structures 17 1 1 Assigning working directories 17 1 2 SADs from AAR aa2sad 17 1 3 SCDs from SPD spdtoscd 17 1 4 SCDs from ERD erdtoscd 17 1 5
6. and For the second separator you can use or but commas can t be used For example OUTPUTS imt block intarr 300 n the IMT block CLEAN CUBE Data output cube CUBE INPUT with undefined values instead of identified glitches Name The name in the header can be on the same line as the keyword NAME or on a single line after This is compulsory Calling sequence The calling sequence must be on the line after the keyword CALLING SEQUENCE The calling sequence can be on a single line or on two consecutive lines This is compulsory D 4 Automatic inclusion of new processing algorithms in CIA In order for easy addition of new algorithms into CIA the core CIA calibration routines have been designed to automatically call new low level processing routines In order for this automatic mechanism to work the name and calling sequence of new routines must conform to a standard As described below this is dependent on the type of processing the new routine will perform In all cases additional keywords can be used by the new routine keyword inheritance ensures that keywords are passed by the core calibration routine down to the new low level processing routine raster scan Raster MOSAIC routines should have the calling sequence raster method raster pds For example let us say that I have a new routine which implements a raster MOSAIC creation algorithm or method called
7. polari polari polar2 polar2 polar3 polar3 w2 w3 roll pds polar angle raster raster rmsraster raster rmsraster raster rmsraster dwi dw2 dw3 Stokes i sigma i stokes q sigma q stokes u sigma u polar rate rmspolar rate polar angle polar angle north rmspolar angle The last three lines of the above command contains the required output For further details on the meaning of these parameters see the online help on comp stokes Part II CIA Basic Guide Introduction The purpose of Part CIA Basic Guide is to allow you to view reduced images from the Auto Analysis processing of your ISOCAM data and make a first attempt at doing your own analysis No previous knowledge of Cam Interactive Analysis is needed to use the CIA Basic Guide e Chapter 9 introduces you to your data products and how they are managed in CIA e Chapter 10 introduces you to a CIA session and shows you how to display your Auto Analysis Results data e Chapter 11 overviews the CIA analysis process e Chapter 12 describes the first step in the analysis process data slicing e Chapter 13 introduces you to the data calibration routines of CIA e Chapter 14 describes CIA s image analysis and display routines 59 60 Chapter 9 The data products and CIA data structures In this chapter we will look at how the data are stored in the data products as FITS files how these data products are
8. 104 CHAPTER 13 DATA CALIBRATION Figure 13 1 x cia window 13 83 DATA CALIBRATION WITH X_CIA 105 e Subtract the user s own dark current frame an IDL 32 x 32 array called my dark in the following e Use the spatial deglitch method e Correct transient effects using the SAP model fitting e build flat field by selecting a few frames of the raster if no assumption on the extension or intensity of sources can be done Then the sequence of commands would be the following 1 2 10 11 12 13 14 Start a CIA session Start x cia with the following CIA x cia indark my dark Choose AOT Type default is raster scan Data Load SSCD Data AOT Info Dark User Input Deglitch Spatial Transient SAP Model Fitting Flat field Manual Process Selected Use all available functions of View and Tools menu to explore calibrated data Tools Hardcopy to create hardcopies Data Save IDL File If result is worth saving CIA Quit 13 3 3 3 Processing several data calibration methods in a single session The PDS contains two fields in which calibrated data are stored CUBE computed IMAGEs and IMAGE reduced per raster position per CVF wheel position or per filter The PDS field MASK which has the same size as CUBE is flagged to 1 for all pixels containing a bad value glitch not stabilised or dead pixel or for slewing frames during w
9. 13 3 3 5 Example 3 data calibration for false beam switching Let s assume that a 2x1 raster was performed the first position corresponding to the source position ON source and the second one to the background position OFF source With such an observing mode which could be called false beam switching it is possible to remove simultaneously the dark current and the background In that case data just need to be deglitched before this operation Transient correction can first be considered as optional and will be detailed in Example 4 Hence the sequence of commands would be the following 1 Start a CIA session 108 CHAPTER 13 DATA CALIBRATION 2 Start x cia simply type x_cia the CIA command line 3 Choose AOT Type default is raster scan 4 Data Load SSCD 5 Data AOT Info 6 Process None 7 Deglitch Spatial 8 Process Selected 9 Data Save IDL File my file xdr 10 CIA Quit If the user answers Yes to the question Average ISODATA CUBE into ISODATA IMAGE after step 10 the removing of dark and background can be made in a CIA session with the following IDL commands CIA restore my file xdr verb CIA source isodata image 0 CIA background isodata image 1 CIA image source background CIA tviso image If the user answers No to the question Average ISODATA CUBE into ISODATA IMAGE after step 9 the removing of
10. in the form MMMYY eg APROO The default will usually be the latest version 2 3 2 Using the Online Help ciainfo cia html and cia help The CIA help system uses the old style IDL help This is invoked by entering widget olh on the CIA command line The alias ciainfo will also invoke widget olh provided of course that the alias list is compiled CIA r alias 2 8 GETTING STARTED 5 The organisation of the headers within ciainfo is slightly different on VMS and UNIX platforms In the VMS version the routines are organised into two groups CIA USER and CIA HELP The former contains just those routines of most interest to the user while the latter contains the full suite of routines see Figure 2 2 In UNIX you have also CIA USER and CIA PRGM but additionally the routines are split into several groups see Figure 2 1 The titles of each should be self explanatory In addition to ciainfo there is a dedicated help invoked by typing cia_help on the command line see Figure 2 3 Note that this is an IDL widget based program so use of the command line is suspended under VMS during its operation In the cia_help opening widget the CIA routines are organised into eight supergroups These are listed in the top half of the widget and accompanied by a brief description in the bottom half of the widget To get deeper into the help choose a supergroup you are interested in and click on the appropriate button A supergroup widget will then a
11. mask dead CSCD143006010110 97092611592803 16 4 2 Modifying the MASK Following on from Section 16 4 1 we will now create and modify our own MASK 1 Firstly create an empty array that will hold our MASK let s call this our mask We will make it the same size as original mask CIA our mask bytarr 32 32 19 2 Suppose that the first column of all our CAM IMAGES is blind we can set the first column pixels to the appropriate value with put mask CIA our mask 0 put mask blind our mask 0 3 Now look at the first element in our mask This should be a blind pixel CIA print our mask 0 16 Since it is a blind pixel it has a value of 16 4 Our MASK can be added to a CIA structure with scd put The keyword put decides how out MASK is inserted into the structure For example to combine the MASK bits in our MASK with the CIA structure s existing MASK bits i e an OR operation CIA scd put mask mask CSCD143006010110 97092611592803 put put Actually this is the default behaviour of scd put To replace the existing MASK with the input MASK then put should be set to set Setting put to clear will cause the existing MASK bits to be set to zero i e cleared where the input MASK bits are one Alternatively we can create a MASK and put it directly into a structure merging it with the structure s MASK The structure could be a PDS or ordinary structure extracted from a CIA str
12. x cia when running on VMS is not able to load SSCDs or data structure files if they are not in the current directory 2 5 Acknowledging CIA in publications CIA is a joint development by the ESA Astrophysics Division and the ISOCAM Consortium The ISOCAM Consortium is led by the ISOCAM PI C Cesarsky Contributing ISOCAM Consor tium institutes are Service d Astrophysique SAp Saclay France and Institut d Astrophysique Spatiale LAS Orsay France and Infrared Processing and Analysis Center IPAC Pasadena U S A When publishing ISOCAM Data reduced with this analysis package please mention this in the acknowledgment the following way The ISOCAM data presented in this paper was analysed using CIA a joint development by the ESA Astrophysics Division and the ISOCAM Consortium The ISOCAM Consortium is led by the ISOCAM PI C Cesarsky If you want to cite CIA in your bibliography please refer to Design and Implementation of CIA the ISOCAM Interactive Analysis System Ott S et al 1997 in ASP Conf Ser Vol 125 Astronomical Data Analysis Software and Systems ADASS VI ed G Hunt amp H E Payne San Francisco ASP 34 2 0 REPORTING PROBLEMS WITH CIA 17 2 6 Reporting problems with CIA Considering the complexity and size of CIA it is not unlikely that you may find bugs in some routines If you do encounter what you suspect is a bug please check it first with your local CIA expert If s he can t he
13. 1w6 CSCD143006010001_96082815175532 C8CD143006010002 96082815175578 16 2 AN EXAMPLE OF SAD MANIPULATION 193 etc CIA print noscds 41 16 2 An example of SAD manipulation In Section 10 2 we used sad display to create SADs from the AAR data products At that point you were probably only interested in a quick look at your data now we will delve a little deeper into the data and take a look at the SADs themselves Most likely you have exited the session where you first used sad display to look at your AAR data products To regenerate the SADs in a CIA session use sad display as described in Section 10 2 or if the SADs are on disk use ssad_read Upon exiting sad display all the SADs created will remain in memory Try the following command CIA print ssad list CSSA000014300601_96091713360439 CSSA000014300601_96091713360451 ssad_list lists all the SSADs in memory Two such SSADs exist in our example one cata logues the set of SADs that hold the EXPOSUREs from the CCIM and CMAP data products and the other the SADs that hold the set of MOSAICs from the CMOS data products The former SADs are known as origin SADs and the latter as future SADs To see the full set of future SADs we can use the CIA routine ssad_elem Given the name of an SSAD it will list all of its SADs CIA print ssad_elem CSSA000014300601_96091713360451 CSAD000014300601 96091713370548 C5AD000014300601_96091713370625 CSAD00001430
14. Center coler Print Scale x 1 2 Scale x 2 Quit Figure 14 14 xv raster window 144 CHAPTER 14 IMAGE ANALYSIS AND DISPLAY 14 4 12 Cube animation with xmovie An animated movie of all the images in an IDL cube can be displayed with xmovie The cube can be taken directly from a CIA data structure for example the CUBE from a PDS CIA xmovie cvf pds cube One note of warning since xmovie needs a lot of memory to perform the animation a large cube may crash it Don t worry your data should not be affected 14 5 Simple image display 14 5 1 tviso The simplest form of image display is with tviso The purpose of this routine is to display a single image and intensity scale in an IDL window All images are rebinned to 320 x 320 for visualisation Its calling sequence is simple tviso image 14 5 2 Cube display with show_frame All the frames in an IDL cube can be displayed in a single window with the routine show_frame see Figure 14 15 Each frame can be enlarged by clicking on it If there are too many frames to fit in the window you can scroll through the cube by clicking on the buttons previous or next Clicking on postscript produces a postscript file of the window The button colour invokes xloadct and the button clone clones the show_frame widget As an example since the EXPOSUREs from a CVF PDS is essentially an IDL cube a call to show_frame could be CIA gt show_frame cvf_pds image 14 6 Image
15. Remove Ghosts Flat field None Flat field User Input Flat field Cal G Flatfield Auto Flatfield Manual Process None Process Default Process Selected Tools xv_temp Tools xv raster 299 300 CHAPTER 23 X_CIA REFERENCE GUIDE Tools isocont ISOCAM axis Tools isocont RA DEC axis Tools x_isocont Tools hardcopy gif Tools hardcopy ps Additional corrections of secondary order are Remove Dark Residuals Smooth Undefined Values FOV Distortion linear and calibration methods Spectral Deconvolution Some other buttons concern the type of data to be calibrated i e AOT Each time a new AOT is selected a new data set must be loaded The last buttons concern the projection method for building raster maps as well as spectrum output units 23 2 2 Short description of commands A list of all available commands together with a short description is given below CIA ISOCAM Display a beautiful image to say Welcome to the ISOCAM world at CEA Saclay CIA Log File Display the log file of all commands and all warning messages occurring during the x_cia session By default the log file is named x_cia_log txt and is created in the current directory A user named file can be created by using the command CIA gt x_cia logfile my_file txt CIA Help Display a help file CIA Info Display some copyright information about the Cam Interactive Analysis CIA packa
16. W ALL PIXELS A W no new selec lt BUILDER DATA lt gt VIEW gt BACKGROUND XLOaICT PRINT Dera PRINT VIEW PRINT BACKGROUND CANCEL DONE pixel 25 17 Frame 3 value 10 8073 cted zone of frame 3 mean 9 34446 Figure 20 5 bkg_builder s main window 255 256 CHAPTER 20 ADVANCED DATA CALIBRATION is involved Depending on the type of calibration data certain parameters are given precedence and others are completely ignored Usually all this is handled automatically by CIA When get sscdraster get sscdstruct and get sscdcvf build a PDS from an SSCD they choose the optimal calibration record from the CDS for your observation data These records are placed in the substructure CALG Sec tion 15 5 6 However if you have your own CDS and an SPD SCD then likewise you can use find best or find best psf to extract the most suitable record for you data 20 12 1 best find best is used for all calibration data types except data see Section 20 12 2 Suppose that we have a an SCD and an OFLT CDS in memory CIA spd scd CSCDLW3FOV6V4P00_96052803472412 CIA oflat cds CCGLWOFLT_96030223551347 CIA best oflat find best spd scd oflat cds best oflat is a full record from the CDS CIA help oflat cds str CIA OFLT oflat cds image 0 Note that find best finds the flat field with the closest wavelength to that of the observation d
17. e The IDL version does not work well with irregularly gridded data as can occur in the pixel history if a glitched pixel is recorded by the complex MASK Conse quently it is recommended to only use the simple MASK with the IDL version See Section 2 3 4 for configuring the MASK e The Fortran and C versions can handle irregularly gridded data In this case both MASK settings are fine though the complex MASK is recommended as it correctly indicates that glitched pixels should be ignored routine called corr transient fouks PDS side effects Unstable pixels are flagged in MASK and CUBE is modified reference Coulais A and Abergel A 2000 method s90 method Pixels which have not achieved 90 of their stable values are masked This is the default stabilization method routine called corr_transient_s90 PDS side effects Unstable pixels are flagged in MASK and CUBE remains unmodi fied reference ISOCAM Handbook Chapter Data processing methods Section Find the first stabilized pixel at each position without using a model method m90 method This method applies a model to find the first stabilized pixel at each STATE Pixels which have not achieved 90 of their stable values are masked routine called corr_transient_m90 PDS side effects Unstable pixels are flagged in MASK and CUBE remains unmodi fied reference ISOCAM Handbook Chapter Data processing methods Section Find the first stabilized pixel
18. seq entwhl mode fltrwhl pfov tint gain size m raster n raster 0 HOLE IDLE LW2 6 0 25 20 1 1 1 1 1 HOLE IDLE LW2 6 0 2 10 1 1 1 1 2 HOLE IDLE LW2 6 0 2 10 2 2 1 1 3 HOLE OBS LW2 6 0 2 10 2 3 1 1 4 HOLE OBS LW3 6 0 2 10 2 106 1 1 20 18 HOW TO SAVE SPOILED OBSERVATIONS 271 33 HOLE OBS LW3 6 0 2 10 2 14 27 2 34 HOLE OBS LW3 6 0 2 10 2 14 26 2 35 HOLE OBS LW3 6 0 2 10 2 1 0 0 36 HOLE OBS LW3 6 0 2 10 2 14 25 2 37 HOLE OBS LW3 6 0 2 10 2 14 24 2 The number of SCD increased from 398 to 399 and the data for the three pointings with the raster point IDs 26 2 25 2 and 24 2 have been nearly completely recovered and the observation can be processed the usual way target not acquired If the ISO satellite did not acquire the target but had a stable pointing then you can try to recover the data using the no_qla_flag option of get_sscdstruct CIA gt struct get_sscdstruct sscd no_qla_flag 1 This will declare all data as good so care has to be taken to exclude slews by manually setting PDS MASK bad QLA flag Either as a result of a failure to uplink commands for ISOCAM or due to a faulty programming of the command sequence the QLA flag normally indicating non stabilized data remains bad You can try to recover the data by manually discarding bad SCDs see Section 12 2 1 and consequently using the no_qla_flag option of get_sscdstruct CIA spdtoscd cispxxxxxxxx fits sscd nomode CIA gt sscd_info sscd CIA scd
19. 14 DEC DEC J2000 of the centre of the final MOSAIC that is constructed from the EX POSUREs in the SCD Type double Unit decimal degrees 15 ROLL As for SCD see Section 15 2 2 15 2 4 Science Analysed Data SAD The SAD may contain an EXPOSURE or a MOSAIC In CIA programming terminology the set of EXPOSUREs used to create the MOSAIC is referred to as the origin and the actual MOSAIC itself as the future This gives rise to two flavours of SAD origin SAD and future SAD Two array substructures exist in the SAD data structure CCIM and CMAP Don t confuse these substructures with the data products of the same name they have been named CCIM and CMAP for historical reasons We use these substructures for 1 CCIM Holds an averaged EXPOSURE in detector coordinates It may be used to hold your CIA processed data or AAR from the CCIM data product file 2 CMAP In the origin SAD it holds an EXPOSURE calibrated in astronomical coordinates In the future SAD it holds a MOSAIC It may be used to hold your CIA processed data or AAR from the CMAP data product file or the CMOS data product file Because the size of the MOSAIC may vary IDL pointers are used to handle the CMAP data structure The routines of Chapter 16 make this transparent to the user so you need only think of this structure in terms of fields that you may manipulate The CMAP substructure is of dimensions 32 x 32 for holding an EXPOSURE dynamic when used to h
20. 266 CHAPTER 20 ADVANCED DATA CALIBRATION 20 16 Faint source data reduction with PRETI Several methods for ISOCAM faint source data reduction exist triple beam method D sert et al 1999 Lari method Lari et al 2001 This method is coded in IDL and available as add on package to CIA For more infor matation please contact Carlo Lari lari ira bo cnr it Metcalfe method Blommaert et al 2002 This method is coded in IDL and completely based on CIA routines For more information please contact the ISO helpdesk helpdeskGiso vilspa esa es PRETI method Starck et al 1999 This method is based on the multiresolution package MR 2 The corresponding C executable for Sun Solaris is provide is with CIA together with its IDL interface re duce faint source which performs the following data reduction steps 1 Slice the CISP file and create a raster_pds 2 Dark correction Iv Calling the PRETI C executable which corrects glitches and performs a median flat fielding 4 Optional Transient correction 5 Conversion to milli Jansky 6 Averaging the CUBE into EXPOSURES T Projection of the IMAGET TES into the MOSAIC Calling syntax is CIA reduce faint source cisp_file fits raster 20 17 Error handling in CIA In this section we discuss the error calculated by CIA during the calibration and data reduction stages We use the standard definitions N 1 1 Mean amp N 2 Tj 20 2
21. 9 LW OBS 10 1 5 2 10 2 512 21 295 437 50 547 10 LW OBS LW10 1 5 2 10 2 512 26 295 463 50 518 gt 42 CHAPTER 6 BEAM SWITCH OBSERVATION 11 LW OBS LWi0 1 5 2 10 2 512 22 295 462 50 517 12 LW OBS LWi0 1 5 2 10 2 512 26 295 417 50 501 13 LW OBS LWi0 1 5 2 10 2 512 21 295 417 50 501 14 LW OBS LWi0 1 5 2 10 2 512 26 295 463 50 518 15 LW OBS LWi0 1 5 2 10 2 512 22 295 463 50 517 16 LW OBS LWi0 1 5 2 10 2 512 26 295 488 50 488 17 LW OBS LWi0 1 5 2 10 2 512 21 295 488 50 488 18 LW OBS LWi0 1 5 2 10 2 512 27 295 463 50 518 19 LW OBS LWi0 1 5 2 10 2 512 21 295 463 50 518 20 LW OBS LWi0 1 5 2 10 2 512 26 295 508 50 534 21 LW OBS LWi0 1 5 2 10 2 512 2 295 508 50 534 22 LW IDLE LWi0 1 5 2 10 2 512 25 295 463 50 519 23 LW OBS LW2 6 0 25 20 1 512 1 295 462 50 518 Get an overview of the SCDs CIA gt sscd_info sscd deg There are quite a few SCDs here Some of these we will discard For each source and reference pointing there are a pair of SCDs The first SCD contains on target data and the second SCD contains data from an intermediate step when ISO is slewing This SCD is considered to contain invalid data though in reality the data may be quite usable Section 19 4 1 describes how such data may be incorporated into the data reduction We normally can not use sscd_clean on a beam switch observation see Section 19 4 1 for a work around However it is a simple manner of looking at the above list and deleting the
22. ASNUMBER Its meaning depends on the type of on board processing as indicated by the field CAL OBC see Section 15 2 2 20 which is performed Type integer e CAL OBC 0 No on board processing ASNUMBER will also be 0 e CAL OBC 1 IMAGEs computed on board and accumulated before transmission to ground ASNUMBER indicates the number of on board accumulations value description 0 no accumulation i e EOI and RESET FRAMES in telemetry 1 two IMAGEs accumulated 2 three IMAGEs accumulated 15 16 IMAGES accumulated e CAL OBC 2 Only sampled IMAGEs in telemetry ASNUMBER indicates the sampling frequency E g ASNUMBER 4 implies that 1 in 4 IMAGES are transmit ted to ground CRPIX1 RA reference pixel of IMAGE in FITS convention Type float CRPIX2 DEC reference pixel of IMAGE in FITS convention Type float RA RA J2000 of CRPIX1 Type double Unit decimal degrees DEC DEC J2000 of CRPIX2 Type double Unit decimal degrees ROLL The spacecraft roll see Appendix E Type double Unit decimal degrees ENTWHL Entrance wheel position see the ISOCAM Handbook for possible values of ENTWHL Type string e g HOLE SELWHL Selection wheel position see the ISOCAM Handbook for possible values of ENTWHL Type string e g FABRY MIRROR LW 3Usually on board processing only takes place when TINT 0 28s or for CAM parallel data By FITS convention CRPIX1 CRPIX2 are the coordinates of the cen
23. Eubec1 8 106 134 00 Figure 3 1 raster IMAGE before calibration Dark current causes the alternately dark and bright horizontal lines The bright pixels at the bottom left of the IMAGE are due to a glitch this is clear from the sharp spike in the history of the currently selected pixel raster pds cube l1 9 106 Displayed by x3d 28 CHAPTER 3 RASTER OBSERVATION CAMOI CIA corr dark raster pds CIA deglitch raster pds CIA stabilize raster pds After some time and many messages printed to the screen you will have a fully calibrated raster PDS Now take a look at the cube of IMAGEs again CIA x3d raster pds Cube 1 9 106 4 03450 You will see that the IMAGE displayed by x3d Figure 3 2 has been greatly transformed The dark current bright and dark lines have gone and so has the glitch In fact a previously obscured source is now visible in the top half of the IMAGE You may notice also that the data values of the pixel have decreased This is because before dark correction all the IMAGES are normalised to ADUs gain second We can complete the calibration be reducing the IMAGEs to EXPOSUREs and flat fielding the EXPOSUREs CIA reduce raster pds CIA corr flat raster pds You can use x3d to take a look at the EXPOSUREs CIA x3d raster pds image Finally we can create the raster MOSAIC All the EXPOSUREs will be projected on to the raster field of view CIA raster scan raster
24. In the end of the display routine you have a button beginning by SEE ALSO it is a link to another related routines process routines graphics routines io routines misc routines astro routines az routines AZ List of Routines AA2SAD ABSCAL AD2XY ADD ARRAY ADSTRING Figure 2 4 typical cia html display for all routines 2 8 GETTING STARTED 9 rocess routines graphics routines io routines misc routines astro routines az routines GRAPHICS List of Routines ISOCONT PS COLOR SHOW FRAME SHOW SCD Figure 2 5 typical cia html display for graphic routines process routines graphics routines io routines misc routines astro routines az routines Previous Routine Next Routine GRAPHICS X3D NAME X3D PURPOSE X3D is a widget program for cube analysis Several operations can be done by using the mouse and press buttons Figure 2 6 typical cia html display for one routine 2 3 3 Displaying ISOCAM Auto Analysis products Part I is a quick start guide to analysing CAM data products It should get you up and running If want to have a look at your ISOCAM images asap then you can display your Auto Analysis data products data which has been delivered after standard automatic processing by CIA sad display windows 1 For more details see Section 10 2 10 CHAPTER 2 ABOUT CIA 2 3 4 Customizing your CIA session First time users and anyone not interested should skip th
25. Martin Heemskerk Univ of Amsterdam Mih seh Kong IPAC Francois Lacombe DESPA Wayne Landsman NASA IDL Astronomy Users Library David Landriu CEA Olivier Laurent CEA Patricia Le Coupanec DESPA Jing Li IPAC Leo Metcalfe ESA responsible for flight calibration Marc Antoine Miville Desch nes IAS Koryo Okumura ESA IAS Stephan Ott ESA responsible for IA system Michel P rault IAS responsible for ground calibration Andy Pollock ESA Daniel Rouan DESPA Pilar Roman ESA Michael Rupen NRAO Jaqui Sam Lone ESA CEA Marc Sauvage CEA responsible for the French ISO Data Centre Ralf Siebenmorgen ESA Jean Luc Starck CEA Richard Tuffs MPIK Dan Tran CEA MPIA Dave Van Buren IPAC Laurent Vigroux CEA responsible for IA algorithms Florence Vivares IAS Herve Wozniak Observatoire de Marseille Th y Vo ESA li iv Contents 1 About the CIA User s Manual 1 1 1 Organization of the CIA Users Manual 1 1 2 What you need to begin 1 13 Reporting comments on the CIA Users 1 2 2 About CIA 3 2 1 History and Purpose of CIA 3 2 2 System requirements Lus c4 i ped Re RORU RCACROR RE RO e 4 2 9 Getting started oi Douce emo eb io ole e eg RC as Pp dad 4 2 3 1 How to Start CLIA i loe RR heck on EUR RUP us 4 2 3 2 Using the
26. Multiresolution Median Transform The Multiresolution Median Transform works on the principle that in temporal space glitches are small scale structures and source signal will always be large scale structure Consequently this method may fail for glitches of very long duration In general though it is very robust and handles non stabilized data well It is especially good for observations where many IMAGEs are accrued This method is the default routine called mrid_deglitch PDS side effects Glitches removed from IMAGEs in CUBE i e CUBE is modified Glitched pixels also flagged in MASK reference ISOCAM Handbook Chapter Data processing methods Section Deglitching using the Multiresolution Median Transform MMT 5 method sky method sky cube deglitching Faders and dippers can be rejected from rasters with redundant observation Flags MASK of a PDS The input data cube and image should be flat fielded The algorithm is a The mosaic is created and back projected into the image again b The image is rebinned into a cube c Standard deglitching is performed on the difference between the original and the back projected cube routine called deglitch_sky PDS side effects Glitched pixels are flagged in MASK A new RASTER is created 20 2 CORE CALIBRATION 235 6 method ksig method Second order deglitching on stabilised data Remaining glitches glitch tails and residuals from transients can be rejec
27. OTF BYTE Array 348 DU FLOAT Array 348 DV FLOAT Array 348 UTK LONG Array 348 BOOTTIME LONG Array 348 TEMPERATURE FLOAT Array 10 348 IMAGE FLOAT Array 32 32 16 IMAGE_UNIT STRING i RMS FLOAT Array 32 32 16 NPIX FLOAT Array 32 32 16 CCIM STRUCT gt Anonymous 1 INFO STRUCT gt Anonymous Array 1 DARK FLOAT Array 32 32 FLAT FLOAT Array 32 32 CALG STRUCT gt Anonymous Array 1 HISTORY STRING 70 SSCD NAME STRING CSSC143006010002_98060117274484 SAD_NAME STRING CSAD143006010204 98060119435439 Before proceeding it is recommended to save this structure using IDL s SAVE with xdr set for portability CIA save filename lw6_raster dat lw6 raster xdr 13 1 3 general PDS This PDS is for general use and currently accommodates data from the following observations e Staring AOT 1 e Tracking AOT 1 98 CHAPTER 13 DATA CALIBRATION e Polarization AOT 5 Again like other PDSs the general PDS is created with get_sscdstruct from an SSCD containing sliced SPD SCDs For example CIA gt staring_pds get_sscdstruct staring_sscd where it is assumed that staring_sscd is an IDL variable holding the name of the sliced SSCD In structure it is the simplest of all PDSs see Section 15 5 2 for more on the architecture of the general PDS 13 1 4 BS PDS The BS PDS is designed to handle data from a beam switch observation AOT 3 Once you have
28. Spatial 9 Transient SAP Model Fitting 10 Process Selected 13 3 DATA CALIBRATION WITH X_CIA 107 11 Data Save IDL File my file xdr 12 Data Load IDL File my file xdr 13 Process None 14 Flat field Auto 15 Process Selected 16 Use all available functions of View and Tools menu to explore calibrated data 17 Data Reload Original Data 18 Flat field Manual 19 Process Selected 20 Use all available functions of View and Tools menu to explore calibrated data 21 Data Reload Original Data 22 Flat field User Input 23 Process Selected 24 Use all available functions of View and Tools menu to explore calibrated data 25 Tools Hardcopy to create hardcopies 26 27 CIA Quit Note that once data have been corrected from dark current glitches and transient effects it is necessary to save them in a IDL file step 13 in order to be able to reload this IDL file step 14 just after Indeed the Data Reload Original Data commands steps 19 and 23 permit to retrieve data in the same state since the last loading i e already dark glitch and transient corrected data in that case Then it make sense to try a new flat field correction steps 17 21 and 25 Note that if Data Save IDL File step 13 is omitted then data would be retrieved in the same state after the Data Load SSCD command i e raw data and dark glitches and transients effect would be visible in final images
29. The different methods can be selected by setting the keyword method In gen eral all the methods involve a weighted averaging of EXPOSURE pixels where the weight is the square root of the EXPOSURE pixel weight That is for each EXPOSURE pixel IMAGE i j k the weight is taken as NPIX j k In all cases the impact of raster MOSAIC creation on the PDS is the same The field RAS TER is filled with the computed raster MOSAIC The fields NPIXRASTER is filled with the number of IMAGE pixels from which each pixel in RASTER has been derived The field RMS is filled with the RMS of these IMAGE pixels Note the raster MOSAIC size is determined when the raster PDS is created from an SSCD with get sscdraster The size is computed to be optimum for the raster step size and PFOV However the user may increase the computed size by a specified magnification factor For example to have a MOSAIC which is twice as large in both dimensions as the computed size CIA raster pds get sscdraster sscd magnify 2 244 CHAPTER 20 ADVANCED DATA CALIBRATION Furthermore the size of each MOSAIC pixel is determined when the raster PDS is created from an SSCD with get sscdraster The size is computed to be optimum for the raster step size and PFOV This might result in rectangular e g not square pixels Setting the square option ensures the creation of square pixels CIA raster pds get sscdraster sscd square Normally the orientation of the raster
30. Then my routine would have the calling sequence raster xxx raster pds and would take a raster PDS as input corr dark Dark correction routines should have the calling sequence darkmethod dark dark error ack ack where dark is the output DARK image and dark error is the error on dark 326 APPENDIX D GUIDELINES FOR WRITING CIA ROUTINES corr flat Flat correction routines should have the calling sequence flat flat_method input error error ack ack where flat is the FLAT image input is the user input to corr flat and error is the error on flat deglitch Deglitching routines should have the calling sequence deglitch method cube cube out mask out nsigma nsigma where cube is all the IMAGEs from one pointing of CAM cube_out is the deglitched cube maskout flags glitched pixels in cube and nsigma is the sigma deglitching threshold stabilize Transient correction routines should have the calling sequence corr transient method cube mask in mask out where cube is all the IMAGEs from one pointing of CAM mask in is the MASK corre sponding to cube and mask ut flags unstable pixels in cube Appendix E ISOCAM astrometry angles and coordinates With the roll angle of ISO constrained by the sun position CAM does not produce images conveniently oriented to the standard astronomical convention This appendix should help you to understand how your IMAGEs EXPOSUREs and MOSAICs are oriented and and how to go about changing
31. cleaned sscd sscd clean spd_sscd sscd_elem returns a list of SCDs from an SSCD In the example the list is stored in the string array scds CIA scds sscd elem sscd scd del deletes SCDs from memory In the example delete the first SCD named in the string array scds CIA scd del scds 0 get sscdraster creates a raster PDS from an SPD SSCD containing raster or micro scan ob servation data CAMOI CIA raster pds get sscdraster spd sscd magnify magnify get sscdstruct creates a general PDS from an SPD SSCD containing staring observation data CAMOI CIA staring pds get sscdstruct spd_sscd get sscdbs creates a BS PDS from an SPD SSCD containing beam switch observation data CAMO3 CIA bs_pds get sscdbs spd_sscd get sscdcvf creates a CVF PDS from an SPD SSCD containing CVF data CIA cvf pds get sscdcvf spd sscd B 2 Data calibration All routines listed here that have the keyword method have a selection of different algorithms that can be applied to the data For an exhaustive list of methods see Chapter 20 or the on line help In general the variable pds scd refers to an SCD an SSCD or any flavour of PDS The variable pds refers to any flavour of PDS An appropriate variable name will be used in examples of routines which operate on a PDS of a particular flavour corr dark Remove dark current from IMAGEs in CUBE using the method dark model CIA corr dark pds s
32. convert wheel back LW FILTER LW2 CIA pfov_position convert wheel back PFOV LW 3 0 3 As described in Section 15 3 2 all calibration data reside in the DATA substructure To extract DATA from our PSF CDS CIA data cds get data psf cds 4 Inside this substructure you will find the cube DATA IMAGE and the arrays DATA PFOV and DATA FCVF Respectively these hold a variety of PSFs with their corresponding lens wheel positions and PFOVs Using the IDL command WHERE we can pick out the PSFs that we require CIA nn where DATA FCVF eq 1w2 position AND CIA DATA PFOV eq pfov position CIA psf data nn image 0 Now the variable psf will hold a cube of PSF images corresponding to the CAM configu ration of filter wheel LW2 and a PFOV 3 You could preview these with x3d CIA x3d psf 16 6 Manipulating CIA data structure history The processing applied within CIA to CIA data structures also known as the history is recorded in the CIA data structure field PROCESS The following sections will show you how to extract and replace this field 198 CHAPTER 16 DATA STRUCTURE MANIPULATION 16 6 1 Extracting the history The history information stored in a CIA structure can be accessed using the structure_get routine As an example suppose we have an SCD in memory Let s assign its name to the variable scd just for convenience CIA scd CSCDMKN297L6P100 9704051
33. cvf_pds str Structure lt ad39580 gt 47 tags length 9171368 refs 1 CVF_NAME STRING gt LW CVF2 8 78200 0 110000 ASTR STRUCT gt ASTR STRUC Array 1 ENTWHL STRING HOLE SELWHL STRING LW large Mirror CVF_INCR INT 4 13 1 CREATING A PDS FROM AN SSCD 99 WAVELENGTH START FLOAT 8 78200 WAVELENGTH END FLOAT 16 5200 RESPONSE STRUCT gt Anonymous Array 76 SPEC NAME STRING CCGLWSPEC_98040514330660 AOT STRING STARING TARGET STRING NGC1342 OBSERVER STRING ANON TDTOSN LONG 5805004 CHANNEL STRING LW PFOV FLOAT 6 00000 TINT FLOAT 2 10007 GAIN FLOAT 1 00000 FLTRWHL STRING LW CVF2 WAVELENGTH FLOAT 12 6510 NSCD LONG 76 NBR FRAME INT 1630 FROM INT Array 76 TO INT Array 76 TAB FRAME INT Array 76 ADU_SEC_COEFF FLOAT Array 76 TABFLATCOEF FLOAT Array 76 CUBE FLOAT Array 32 32 1630 CUBE_UNIT STRING ADU MASK BYTE Array 32 32 1630 OTF BYTE Array 1630 DU FLOAT Array 1630 DV FLOAT Array 1630 UTK LONG Array 1630 BOOTTIME LONG Array 1630 TEMPERATURE FLOAT Array 10 1630 IMAGE FLOAT Array 32 32 76 IMAGE UNIT STRING dd RMS FLOAT Array 32 32 76 NPIX FLOAT Array 32 32 76 CCIM STRUCT gt Anonymous 1 INFO STRUCT gt Anonymous Array 1 DARK FLOAT Array 32 32 FLAT FLOAT Array 32 32 CALG STRUCT gt Anonymous 1 HISTORY STRING Array 70 SSCD NAME STRING CSSC203056040001_96091918462427 SAD_NAME STRING CSAD2030
34. flat_thresh 10 nplanes 60 CIA gt flat_smooth_window 4 CIA gt act set_act make_map CIA gt slice_pipe The results of this first flat field determination are displayed on figure 21 2 At that stage the improvement from CIA is not yet striking Now let s try the DivSky method It is set with 5 parameters tdt flat_thresh nplanes size_filter and divsky Of these only two are new Let us explicit them e divsky this is just a boolean keyword i e true false set to indicate which method is to be used if it is not present on the command line as divsky then the Sliding Mean Flat Field method is used which is not what we want 282 CHAPTER 21 USING SLICE WITHIN CIA Table 21 3 Our choice of parameters for the Perturbed Single Flat Field method Filter tdt flat Snooth window flat EE E EE An LW2 65801627 35 0000 10 0000 25 0000 6 00000 Figure 21 2 The resulting maps for the Perturbed Single Flat Field determination Note that the map orientation has changed as SLICE always produces maps with North up and East left Imprints of the individual raster pointings are still visible 21 5 A WORKED EXAMPLE 283 Table 21 4 Our choice of parameters for the DivSky method tat size filter flat thresh uplanes LW3 65801627 15 10 60 LW2 65801627 15 10 30 e size filter this is an integer whose meaning is rather similar to the flat_smooth_window of the previous method Here also t
35. http www iso vilspa esa es IDLE See OP MODE IIPH Instrument Instantaneous Pointing History Contains instantaneous pointing information for the prime instrument during an AOT For most users with CAM data CAM is always prime IRPH Instrument Reference Pointing History Contains reference pointing information for the prime instrument during an AOT For most users with CAM data CAM is always prime IMAGE Refers to an image computed from the EOI and RESET FRAMES In theory IMAGE EOI RESET For the CAM LW detector this is indeed true but for the SW detector it is slightly more complicated and CCD lines have to be interleaved see MOSAIC micro scan observation A particular CONFIGURATION of CAM AOT 1 This is a spe cial case of a raster observation with a small pointing step size so that EXPOSUREs greatly overlap MOSAIC Refers to the final image from an AOT It is computed from the EXPOSUREs which are in turn computed from IMAGEs which are computed from the RESET and EOI FRAMEs See glossary entries for an explanation of each of these terms M direction M direction and N direction are the axes along which CAM performs a raster N direction See M direction observation data structures A broad term that refers to all the data products you need to make up uncalibrated CAM IMAGEs In other words it excludes calibration data The observation data structure is any structure holding such data see SCD SSCD SAD SSAD ob
36. in sec 21 7 we present a list of frequently asked questions as well as frequently encountered problems l Taken from Sauvage M 2001 An Introduction to SLICE inside CIA 273 274 CHAPTER 21 USING SLICE WITHIN CIA A detailed description of the algorithms in SLICE can be found in M A Miville Desch nes paper 2000 A amp A 146 519 while the package has it own user s manual available at in the doc directory of the SLICE delivery An important note for SLICE V1 2 Although optical distortion has been computed for all ISOCAM configurations and can be corrected in CIA SLICE V1 2 incorporates an older correction scheme As a result only the most widely used optical configurations can be corrected for distortion in SLICE This limitation is known to M A Miville Desch nes and should be uplifted in subsequent releases of SLICE 21 3 Organization of data in SLICE In CIA data are either stored in hidden structures the SCD SSCD SAD SSAD that you access through pointers or in standard IDL structures the so called PDS that you manipulate directly In SLICE this is different data are stored in IDL variables i e you can manipulate them directly but these also belong to common blocks that are accessed by SLICE routines As a result their names are defined once and for all and you cannot use them for other purposes This is why SLICE is not loaded by default when you start CIA Although you can do it in SLICE we assume here
37. may flag pixels in the MASK that are to be discarded in future analysis and modify other unstable pixels in the CUBE The flagged or masked pixels can of course still be seen by clicking on mask and the temporal behaviour of the modified pixels of the CUBE can be compared with that of the unmodified CUBE Glitches Looking at any single 2D CAM image it may be easy to confuse a glitch with a source However if you look at the temporal behaviour of a pixel in the IMAGEs accrued by CAM in a STATE i e DATA in an SCD or in the CUBE from a PDS then glitches are usually easily distinguishable from sources A CAM observation may be made of many STATEs each STATE containing many IMAGEs of the same sky position In principle when CAM is pointing at a source then source signal will be present in all the IMAGEs of that particular STATE but most glitches are of such a transient nature that their duration will be quite short though intense appearing in one or possibly a few IMAGEs Therefore in most cases sources will appear in temporal space as relatively low intensity long duration signal and glitches will appear as spiky intense events Click on temporal cut Scroll through the images of the CUBE until you find an IMAGE containing a pixel of high intensity Click on the pixel and a plot of its history will appear in the plot window The shape of the plot should reveal whether the selected pixel of the current IMAGE was hit by a cosmic ray or is a
38. mosaic will contain holes in the example below 39 readouts for the three pointings with the raster point IDs 26 2 25 2 and 24 2 are merged in one SCD SCD 35 270 CHAPTER 20 ADVANCED DATA CALIBRATION CIA spdtoscd cisp23300257 fits sscd nowrite There are 5732 records in the SPD file Determining cuts please wait There are 398 SCDs Slicing 1 2 3 4 5 6 7 8 9 10 11 CIA sscd info sscd pol 398 SCDs in the SSCD CSSC233002570101_02022614445300 seq entwhl mode fltrwhl pfov tint gain size m raster n raster 0 HOLE IDLE LW2 6 0 25 20 1 1 1 1 1 HOLE IDLE LW2 6 0 2 10 1 1 1 1 2 HOLE IDLE LW2 6 0 2 10 2 2 1 1 3 HOLE OBS LW2 6 0 2 10 2 3 1 1 4 HOLE OBS LW3 6 0 2 10 2 106 1 1 33 HOLE OBS LW3 6 0 2 10 2 14 27 2 34 HOLE OBS LW3 6 0 2 10 2 1 26 2 35 HOLE OBS LW3 6 0 2 10 2 39 0 0 36 HOLE OBS LW3 6 0 2 10 2 3 24 2 37 HOLE OBS LW3 6 0 2 10 2 14 23 2 To avoid this the ERD file cier fits or the SPD file cisp fits can be patched by the routine repair _rpid CIA repair_rpid cisp23300257 fits Reading cisp23300257 fits 38 RPIDs out of 39 could be recovered Writing cisp23300257 fits CIA gt spdtoscd cisp23300257 fits sscd nowrite There are 5732 records in the SPD file Determining cuts please wait There are 399 SCDs Slicing 1 2 3 4 5 6 7 8 9 10 11 CIA sscd info sscd pol 399 SCDs in the SSCD CSSC233002570101_02022615223101
39. referential is indirect in the drawings e Rotation of a positive angle rotates in the direct trigonometrical direction or counter clockwise i e unlike the IDL rot function Taken from Sauvage M 1996 Angles and ISOCAM LW v3 0 327 328 APPENDIX E ISOCAM ASTROMETRY ANGLES AND COORDINATES N V Y Column 24 7 Figure E 1 Schematic of an ISOCAM LW image Y and Z are the satellite axes Column 24 is indicated to help you visualize the array o called the roll angle is the position angle of the Z axis i e the angle between the celestial North and the Z axis counted positively eastward e We assume that you have already used CIA and are quite familiar with its structure organization and basic routines E 1 1 Definition of the roll angle for CAM LW The roll angle for ISOCAM s LW detector is defined in Figure E 1 Note that while this diagram is correct the orientation of your display can change the direction in which the axes point In IDL the orientation of a displayed image depends on the system variable ORDER Refer to Section E 2 3 for different axis orientations and the effects of ORDER E 1 2 Rasters along the satellite axes These rasters are called for simplicity s sake Y axis rasters and an example is given in Fig ure E 2 The sky is scanned along the satellite Y and Z axis The raster axes are called M and N and the referential M N is direct In that referential ISO always starts its ras
40. s Help or only the help for high level routines User s Help is displayed defining whether the alias file is run inclusion of your routines in IDL s search path Some users have reported problems running CIA on 24 bit X Windows displays One suggested solution is to force IDL to use an 8 bit PseudoColor To do this uncomment the following line to your CIA startup file device pseudo color 8 or to force IDL to use true color 24 bit visuals device true color 24 device decompose O run cia_vers graphics xloadct_5_4 pro and comment the line cia vers graphics xloadct 3 6 pro For more on this topic look for information on the procedure DEVICE in the IDL online help Other solutions to this problem that does not involve the CIA startup file is suggested in Section 2 4 As already stated all of the above can be configured in a CIA user startup file The following sections describe how you can create your own CIA startup file on both the UNIX and VMS platforms 2 8 GETTING STARTED 13 2 3 4 1 VMS Normally your system administrator should have set up the global symbol CIA_ENV this should point to the generic command file CIA_VMS_ENV COM It should be set to something like sh sym CIA_ENV CIA ENV SAPIO1 DKC200 CIA_VMS_ENV COM This command file sets up the environment for a CIA session It must be executed before CIA is started For convenience you can include a line in your login com
41. see Section 15 5 8 e INFO See Section 15 5 7 e Standard PDS fields of Section 15 5 1 182 CHAPTER 15 CIA DATA STRUCTURE HIGH LEVEL ARCHITECTURE 15 5 3 CVF PDS The CVF PDS is used to hold CVF data and so has specific fields for holding data that is pertinent to a CVF observation All the fields and substructures are listed below e CAL G see Section 15 5 6 e Standard PDS fields of Section 15 5 1 e CVF PDS specific fields These fields are listed in the table below along with a brief description and where appropriate a reference to the SSCD or SCD field where the data originates or otherwise the type of the data in the field subfield description reference type CVF NAME CVF SAD name string SSPEC NAME filter amp CVF spectral data CDS name string ENTWHL entrance wheel Section 15 2 2 9 SELWHL selection wheel Section 15 2 2 10 CVF INCR CVF increment step Section 15 2 3 12 WAVELENGTH START CVF start wavelength Section 15 2 3 12 WAVELENGTH END CVF end wavelength Section 15 2 3 12 RESPONSE WAVELENG wavelength float RESPONSE TRANS transmission at WAVELENG float RESPONSEL SENSITIV ADU gain sec pixel to mJy pixel factor float RESPONSE FCVF index wheel number integer RESPONSE TRANSMIS strange transmission fltarr nscd 15 5 4 raster PDS The raster PDS is a raster observation dedicated data structure so as with the CVF PDS it has fields specifically relating to a raster observation These fields
42. separated by blanks e g entering display product gt will list you all routines which contain the words display and product in their headers The IDL help functionality available with widget olh or ciainfo works also for the ASTROLIB We have now the command cia html which provides help in HTML format e additional capabilities for creation of rasters by get_sscdraster The keywords north and camera can overrule the default behaviour mosaic north oriented for north axis rasters and camera oriented for spacecraft axis rasters thereby giving the user the choice to determine the orientation of the mosaic F 2 NEW AND IMPROVED FUNCTIONALITY 339 The square option ensures the creation square mosaic pixels For certain raster steps rectangular pixels are produced as default which leads to elliptical looking sources Using the nocheck keyword also rasters with missing or repeated images positions can be treated e New display tool xcube You might like to use this instead of x3d xcube has many features The input can be 2 cubes so it easy to compare them e g before and after transient correction You can select the data to be displayed cube image mask by clicking Range in the cube the borders of the scd appears in the time plot of the cube and you can zoom on one scd one block etc You can also type the range e g from frame 400 to 500 You can select the scale of th
43. telemetry drops Short telemetry drops affecting only a few readouts will normally go un noticed For longer telemetry drops were the data for one or more SCDs is missing the following actions can be taken e raster observations Calling get sscdraster with the option nocheck will permit you to continue with the data reduction CIA raster pds get sscdraster sscd nocheck Alternatively sscd clean will fake empty SCDs so that get sscdraster is tricked into believing the raster is complete CIA cleaned sscd sscdclean sscd CIA raster pds get sscdraster cleaned sscd nocheck e CVF observations CIA data reduction is not affected by telemetry drops e Beam Switch observations With most beam switch observations consisting only of one on and one off position the observation is quite likely lost If there are more on off positions use get sscdstruct on the cleaned dataset e g a dataset containing no spurious SCDs any more and combine the results of on off positions manually e Polarization observations It will depend on the redundancy of the observation whether the observation is lost If there are sufficient observations with all polarizors use get sscdstruct on the cleaned dataset and combine the results manually bad raster point IDs Some observations are affected by an invalid 0 0 raster point ID ssed_clean will remove these SCDs so you can progress with your data reduction However the final
44. toggle button a new window appears you can select a region and get the statistics of this region e Color mode multiple choice menu selects the color table Reset button resets the luminosity and contrast of the selected color table you can also click or drag the mouse inside the main image left to right increases the luminosity top to down increases the contrast y e Overplot mode in this mode a button labelled skyview appears valid only when the input is a raster it overdraw on the main image the limits of each sky pointing 14 4 8 5 Title This panel contains a text widget with the target name of the image and some information on the configuration this is valid only for PDS input The user can type his or her own title it will be used as title in all the plots created by ximage e g File Plot image or Raster plot 14 4 8 6 Intensity and coordinates This panel contains a text widget only readable with the intensity and coordinates with its astrometry 14 4 8 7 Main Image The clicking behaviour depends upon the mouse mode e Datamode dragging the mouse on the image the zoom in thumbnail2 the intensity and the coordinates are those of the mouse position and not the cursor dragging the mouse outside the image the zoom in thumbnail2 the intensity and the coordinates are those of the cursor eft click moves the cursor to the mouse position middle click mo
45. 1 2 1 First we make a copy of the uncalibrated raster PDS CIA original lw6 raster lw6 raster 2 We apply a basic calibration treatment to the raster PDS by calling calib raster with the keyword all set see the calib raster in the on line help for the details of all CIA calib raster lw6 raster all Then save the calibrated MOSAIC so we can later compare with different calibration results CIA 1w6 raster all 1w6 raster raster 3 We can look at the results of our calibration by loading the cube of IMAGEs and the MASK from our raster PDS into x3d CIA x3d lw6 raster cube lw6_raster mask Clicking on the button mask indicates the pixels that have been masked as unstable By looking at the temporal history of the pixels you can immediately see if good stabilisation and deglitching have been achieved A vertical profile of some background pixels will tell you if a good dark correction has been applied Another way of judging the quality of the calibration is to look at the signal to noise ratio of some of the EXPOSUREs CIA xsnr lw6 raster image 0 Or simply display the MOSAIC with tviso CIA tviso lw6 raster all 13 4 CALIBRATING A PDS THE OLD WAY 111 4 Try another calibration using deglitch method MM stabilisation method s90 and the 5 CAL G flat CIA l1w6 raster original lw6 raster CIA calib raster lw6 raster dark deglitch mm stab s90 CIA flat calg
46. 204 eA ed ee EA 323 D 2 323 D 3 How to write a header 324 D 4 Automatic inclusion of new processing algorithms in CIA 325 ISOCAM astrometry angles and coordinates 327 Definitions ae AV EP uu eae gt 327 E 1 1 Definition of the roll angle for CAM LW 328 E 1 2 Rasters along the satellite axes 328 E 1 3 Rasters referenced to the celestial North axis 330 E 2 Trouble shooting astrometry in CIA structures 332 E 2 1 Incorrect astrometry in beam switch data 332 E 2 2 Astrometry 1 332 E 2 3 Roll image orientation and ORDER 332 E 3 Using FITS in CIA new problems a 332 E 3 1 FITS convention IDL s astrolib 332 E 3 2 From CIA structures to 1 336 What is new in CIA 5 0 337 F 1 New and improved algorithms a a 337 F 2 New and improved 338 E Bue Xess gena ptus s een dP OR dog p e d Lat sue te E ce do t 340 Warning messages in CIA 343 G1 Error messages Rode d Robo Rede a 343 G 2 Warning messages 343 Information messages
47. 6 0 5 04 1 512 1 180 498 18 849 3 LW OBS LW2 3 0 5 04 1 512 46 180 498 18 849 4 LW OBS LW2 3 0 5 04 1 512 24 180 490 18 871 5 LW OBS LW2 3 0 5 04 1 512 24 180 482 18 893 6 LW OBS LW2 3 0 5 04 1 512 24 180 473 18 915 7 LW OBS LW2 3 0 5 04 1 512 23 180 450 18 908 8 LW OBS LW2 3 0 5 04 1 512 24 180 458 18 886 etc If you look at the output of sscd info you will see more than the expected 32 SCDs These others correspond to STATEs where CAM is busy doing other things other than 3 2 DATA ANALYSIS 25 observing Our next step will be to discard those states sscd clean will do this for us It will perform another important task at the moment there is only one SSCD cataloging all the SCDs clearly it would be neater if we could divide all these SCDs into four distinct groups where each group contains all the data from a single CONFIGURATION So let s run sscd clean on our SSCD CIA cleaned sscd sscd clean sscd Out of 48 SCDs 12 are rejected due to mode 9 are rejected due to csh flag 13 are rejected due to qla flag In total 32 are accepted 27 Feb 2002 15 01 09 00 SSCD CLEAN v 2 7 lt Splitting SSCD into 4 segments I gt The variable cleaned sscd is an array containing the 4 names of the SSCDs cataloging our reorganised SCDs Because the SSCD name is derived from your computer s system clock when you reproduce this example your SSCD names will differ CIA print cleaned sscd 55 026005060001_02022715010900 C88C
48. 8 2 6 Photometry Before performing astrometry we should convert the pixel values to mJy in both the EXPO SUREs and the MOSAIC image CIA conv flux pds_polari image CIA raster scan pds polari Now perform photometry on each reduced EXPOSURE use xphot or any other program of your choice CIA xphot pds_polari image 0 You can also perform photometry on the raster MOSAIC The results should be consistent with flux measurements on the EXPOSUREs 8 3 Calculate Stoke parameters At this stage you need to have a calibrated PDS for each of the polarizors and photometric measurements for the source as seen through each polarizor To calculate the Stokes parameters we first need to calculate the polar weights with get polar weights CIA fltrwhli convert wheel back FLTRWHL pds polari channel CIA pds_polar1 fltrwh1l CIA get polar weight fltrwhli pds polari pfov 1 w2 w3 9 1 dw2 dw3 not exact matching for pfov I take the first complete set of data for polarizors for the pfov 360 CIA print wi w2 w3 1 dw2 dw3 0 986200 0 993700 1 00000 0 00100000 0 00100000 0 00000 56 CHAPTER 8 POLARIZATION OBSERVATION CAM05 DEDICATED CAM99 Now we supply the MOSAIC images for each of the three polarizors along with their asso ciated RMS images and the polar weights to the routine comp stokes note that you may also supply flux measurements comp stokes pds pds pds pds pds pds wi
49. 967 983 000 016 995 979 962 946 925 941 958 974 953 937 920 904 904 904 904 904 dec 81 The SCDs numbered from 2 to 17 all have FLTRWHL equal to LW3 and MODE equal to OBS This means that these SCDs make up the first CONFIGURATION of this observa tion The SCDs numbered from 21 to 36 also have FLTRWHL equal to LW6 and MODE 82 CHAPTER 12 DATA SLICING equal to OBS So these SCDs must correspond to the second configuration T he remaining SCDs represent STATEs of CAM between CONFIGURATIONS e g SCD number 16 and 17 or CAM acquiring calibration data e g SCD number 38 and 39 At the moment we are only interested in the LW6 CONFIGURATION so there are many SCDs we need to discard There are two ways of achieving this i sscd clean will automatically break up an observation into its constituent CONFIGURATIONSs one SSCD per CONFIGURATION ii scd find can be used to search for SCDs we wish to discard and in doing so create an SSCD for the CONFIGURATION we desire a sscd clean is invoked simply with CIA cleaned sscds sscd clean spd sscd Out of 41 SCDs 7 are rejected due to mode 6 are rejected due to csh flag 8 are rejected due to qla flag In total 33 are accepted 1 Jun 1998 17 27 36 00 SSCD CLEAN v 2 0 lt Splitting SSCD into 3 segments I gt sscd clean will have split spd sscd into three SSCDs of course the number of SSCDs returned will vary from observation to observation just as the num
50. ADVANCED SLICING variable new_method Andy gap mode F1 W1 POS yes no no F1 W2 POS yes no no F1 W3 POS yes no no F1 WA4 POS yes no no F1 W5 POS yes no no F1 W6 POS yes no no F2 0PER FLAG yes yes no F2 AOT OBS yes yes no F2 AOT AOT yes yes no F2 AOT OPM yes yes no F2 AOT CNF yes yes no F2 IM ORIG yes yes yes F2 INT TIME yes no no F2 IM PROC yes no yes F2_ADC_GAI yes no no F2_ADC_OFF yes no no GPSCRPID yes yes no WMOTION no yes no GPSCRPID is the variable containing Instrument Time Key ITK for each record of the file and is therefore not really a slicing variable WMOTION is non zero variable if one wheel is moving Table 19 2 Slicing variables used by CIA s automatic slicers 19 6 ADVANCED SLICING WITH X_SLICER 227 19 6 6 3 Selecting and Merging SCDs The x_handle_slice window allows you to select SCDs one by one To select or unselect the SCDs just click on the radio button on the SCD The x handle slice window allows you also to merge SCDs Merging two SCDs is simply to glue them together in order to have only one two rules must be followed e you can ONLY MERGE SELECTED SCDs e you can ONLY MERGE CONTINUOUS SCDs i e no hole in the data If you do not follow these rule it is very likely that the slicer will crash Merging SCDs is very interesting if you work on data that have suffered from telemetry drop outs for example if you have false M and N RASTER values see above If this is the case
51. CAM OP MODE Type string MODE may have the following values value CAM mode description IDLE idle OBS performing an astronomical observation DARK obtaining a dark frame FLAT obtaining an internal flat field image CLEAN detector is being cleaned FLTRWHL Filter wheel used in the observation Type string e g LW6 GAIN Detector electronic gain Possible values 1 2 4 Type integer TINT Integration time of each FRAME Type float Unit seconds PFOV Pixel field of view Type float Unit arcseconds WAVELENGTH Wavelength of the current filter Type float Unit microns See the ISOCAM Observer s Manual for possible values of FUTRWHL 15 2 OBSERVATION DATA STRUCTURES 165 15 2 2 Science CAM data SCD Following the definition of an SCD at the beginning of Section 15 2 we will look at where the FRAMEs IMAGEs and CAM parameters are stored All of these data are stored in the fields of the SCD Those fields unique to the SCD are listed here the rest being listed in Section 15 2 1 1 10 N_RASTER The position of the STATE in a raster in the N direction see Appendix E The first position corresponds to N_RASTER 1 For non raster SCDs N_RASTER is always one Type integer M_RASTER The position of the STATE in a raster in the M direction see Appendix E The first position corresponds to _ 1 For non raster SCDs N_RASTER is always one Type integer
52. EXPOSUREs EXPOSURE units jitter offsets in spacecraft x axis jitter offsets in spacecraft y axis Uniform Time Key long term transient compensation factor CAM boot time CAM temperature data DARK used for data calibration FLAT used for data calibration info on PDS creation etc 181 reference or type Section 15 2 1 3 Section 15 2 1 3 Section 15 2 1 5 Section 15 2 1 1 string e g raster Section 15 2 1 1 Section 15 2 3 1 Section 15 2 1 8 Section 15 2 1 10 Section 15 2 1 14 integer fltarr nscd Section 15 2 1 13 Section 15 2 1 12 Section 15 2 1 11 intarr nscd intarr nscd Section 15 2 2 18 bytarr nbr frames fltarr 32 32 nscd fltarr 32 32 nscd Section 15 2 2 12 string fltarr 32 32 nscd string Section 15 2 2 19 Section 15 2 2 19 Section 15 2 2 19 intarr nscd lonarr nbr frame fltarr 10 nbr frame fltarr 32 32 fltarr 32 32 strarr 70 OTF is not used by CIA Information stored in OTF is propagated into MASK Note that the data in OTF is derived from the OTF and qla flag see Section 19 6 7 This means that OTF 0 is bad and OTF 1 is good gt TABFLATCOEFF is filled by rel cal see Chapter 20 15 5 2 general PDS The general PDS is used for holding data from all observations other than a raster or CVF observation Its structure is a basic version of the raster PDS and CVF PDS It has the following substructures and fields e CAL G see Section 15 5 6 e CCIM
53. HANDLE LONG 70 3 Feb 2000 14 05 23 00 IA make array V 1 0 lt First argument array handle is not an handle E gt HANDLE CREATE Invalid handle identifier 72 etc It is recommended that you do not use sscd del in this manner When CIA is processing a CIA data structure for example using sscd info on an SSCD it is not recommended to interrupt the processing by typing Ctrl C This can cause a corruption of the data structure in a similar way to the sscd del problem detailed above There are problems running IDL on display that are 24 bit If you can force your display into 8 bit mode then it is recommended to do so For Solaris systems with the M64 Graphics Accelerator this can be done with the Solaris command m64config usr sbin m64config depth 8 Now log out of CDE or whichever windows manager you run When you log back in the display should be set to 8 bits To confirm the display configuration usr sbin m64config prconf Other suggestions are startx depth 8 or to change the colors via etc XF86 config You might also want to try adding the following line to your user init pro see Section 2 3 4 for details of customizing CIA This command forces IDL to use an 8 bit PseudoColor 16 CHAPTER 2 ABOUT CIA device pseudo 8 decomposed 0 There are some incompatibilities between different implementations of X Windows This can cause some widgets to be incompletely rendered on screen This sit
54. ISOCAM Observer s Manual Siebenmorgen R et al 1999 ISOCAM Handbook ESA Document SAI 99 57 DC Siebenmorgen R 1999 Polarisation Observations with ISOCAM ESA Document Tran D and Gastaud R 2000 Report on ISOCAM CVF faint source photometry in prepa ration Bibliography Abergel A et al 2000 Transient Behaviour of LW Channel of ISOCAM Experimental As tronomy vol 10 page 353 368 Biviano A et al 2000 The ISOCAM LW Detector Dark Current Behaviour Experimental Astonomy vol 10 page 255 277 Biviano A et al 1998 ISOCAM Flat field Calibration Report Technical report ESA 1998 http www iso vilspa esa es users expl_lib CAM Biviano A et al 1998 ISOCAM CVF Calibration Report Technical report ESA 1998 http www iso vilspa esa es users expl_lib CAM Blommaert J 2000 ISOCAM Photometry Report Experimental Astonomy vol 10 page 241 254 Boulade O and Galais P 2000 The ISOCAM Detectors An Overview Experimental Aston omy vol 10 page 227 239 Claret A et al 1998 Glitch Effects in ISOCAM Detectors Experimental Astonomy vol 10 page 305 318 Coulais A and Abergel A 2000 Transient correction of the LW ISOCAM data for low contrasted illumination A amp AS vol 141 page 533 544 D sert F X et al 1999 A classical approach to faint extragalactic source extraction from ISOCAM deep surveys Application to the Hubble Deep Field A amp A vol 342 page 363 377 Dzitko
55. LW7 1 5 2 10 2 18 3 1 9 HOLE OBS LW7 1 5 2 10 2 18 4 1 10 HOLE OBS LW7 1 5 2 10 2 18 5 1 11 HOLE OBS LW7 1 5 2 10 2 18 6 1 12 HOLE OBS LW7 1 5 2 10 2 10 7 1 13 HOLE IDLE LW7 1 5 2 10 2 3 7 1 14 HOLE IDLE LW7 1 5 2 10 2 1 7 1 15 HOLE OBS LW7 1 5 2 10 1 1 7 1 16 HOLE OBS LW8 1 5 2 10 1 3 7 1 17 HOLE OBS LW8 1 5 2 10 1 18 8 1 18 HOLE OBS LW8 1 5 2 10 1 18 9 1 19 HOLE OBS LW8 1 5 2 10 1 19 10 1 20 HOLE OBS LW8 1 5 2 10 1 18 11 1 21 HOLE OBS LW8 1 5 2 10 1 18 12 I 22 HOLE OBS LW8 1 5 2 10 1 2 13 1 23 HOLE IDLE LW8 1 5 2 10 1 13 1 24 HOLE IDLE LW8 1 5 2 10 1 1 13 1 25 HOLE OBS LW8 1 5 2 10 2 13 1 26 HOLE OBS LW6 1 5 2 10 2 11 13 1 27 HOLE OBS LW6 1 5 2 10 2 18 14 1 28 HOLE OBS LW6 1 5 2 10 2 18 15 1 29 HOLE OBS LW6 1 5 2 10 2 18 16 1 30 HOLE OBS LW6 1 5 2 10 2 18 17 1 31 HOLE OBS LW6 1 5 2 10 2 11 18 1 32 HOLE IDLE LW6 1 5 2 10 2 18 1 33 HOLE IDLE LW6 1 5 2 10 2 1 18 1 34 HOLE OBS LW6 1 5 2 10 1 18 1 35 HOLE OBS LW 1 5 2 10 1 1 18 1 36 HOLE OBS LW9 1 5 2 10 1 18 19 1 37 HOLE OBS LW9 1 5 2 10 1 18 20 1 38 HOLE OBS LW9 1 5 2 10 1 19 21 1 39 HOLE OBS LW9 1 5 2 10 1 18 22 1 40 HOLE OBS LW9 1 5 2 10 1 18 23 1 41 HOLE OBS LW9 1 5 2 10 1 1 24 1 42 HOLE IDLE LW9 1 5 2 10 1 8 24 1 43 HOLE OBS LW2 6 0 25 20 1 1 24 1 5 2 DATA ANALYSIS 3T Depending on the apparent velocity of the target the number of SCDs STATEs for each configuration can vary significantly 4 Remove unwanted SCDs with sscd clean CIA cleaned sscd sscd clean sscd
56. MOSAIC is determined when the raster PDS is created from an SSCD with get sscdraster The mosaic is north oriented for north axis rasters and camera oriented for spacecraft axis rasters However the user may override this CIA raster pds get sscdraster sscd north CIA raster pds get sscdraster sscd camera 1 method project method EXPOSUREs are corrected for FOV distortion and then projected onto the raster MOSAIC FOV according to their individual coordinates A C coded program projection performs the actual projection This is probably the best method for raster MOSAIC creation It does a good job of taking into account all the different raster types e g North axis micro scan It also works even if the actual coordinates of the EXPOSUREs do not match their intended positions in the raster It is the default method For more on projection and coadding images with different astrometry see Sec tion 20 15 called routine projette and the executable projection 2 method idlproj method Same method as method project the difference being that an IDL coded projection routine is used called routine raster scan idlproj 3 method camera method By this method the EXPOSUREs are rebinned to 256 x 256 images and then coadded according to their position in the raster The EXPOSUREs are always kept aligned to ISOCAMs axes even if the raster is a NORTH axis raster this means that the axes of t
57. Online Help ciainfo cia html and cia help 4 2 3 3 Displaying ISOCAM Auto Analysis products 9 2 3 4 Customizing your CIA session 10 E MESIOCUcU MENTEM 14 2 5 Acknowledging CIA in publications 16 2 6 Reporting problems with 17 I Quick Start Guide 19 3 Raster observation 23 3 1 Description of the observation 2e 23 39 2 Data analysis Bu Re cee es BUR UE dus 23 4 Staring observation CAMOI1 31 4 1 Description of the observation 31 4 2 Data analysise 6 ee ae ee eee ee pues dem pede tope 31 5 Solar System Object observation CAMOI 35 5 1 Description of the observation 2e 35 5 2 Data analysis ew e glee dw edm EOD Role le p D d a 35 6 Beam switch observation CAMO3 41 6 1 Description of the observation 2e 41 0 2 Dataanalysis c2 4 94 a de te OS WW heut IDE ESROEO ERE es 41 7 CVF observation CAMOA AT 7 1 Description of the observation 4T 7 2 Data analyslS i uod RR Rogo Rhodo mw x SOR BOR EUR EORR Roos 4T vi CONTENTS 8 Polarization observation CAMO5 dedicated CAM99 51 8 1 Description of the observation 51 8 2 Data Analysis 4 2 eie dot d uem Rob ou uem Dono ere 51 8 2 1 Overview of ca
58. Out of 44 SCDs 11 are rejected due to mode 7 are rejected due to csh flag 20 are rejected due to qla flag In total 22 are accepted 26 Feb 2002 17 07 45 00 SSCD CLEAN v 2 7 lt Splitting SSCD into 4 segments I gt We are left with four SSCDs corresponding to the four configurations Let s concern ourselves with the first clean SSCD Using sscd info compare the charac teristics of its component SCDs with those of the whole observation As expected the first SSCD returned by sscd clean contains all the data from the first CONFIGURATION an observation using the LW7 filter Also it can be seen that 5 pointings were performed to track the movement of the comet CIA sscd info cleaned sscd 0 parameter channel ENTWHL mode CIA fltrwhl pfov tint gain size m_raster n raster 5 SCDs in the SSCD 55 812002020001_99121614325400 seq channel ENTWHL mode fltrwhl pfov tint gain size m raster n raster 0 LW HOLE OBS LW7 1 5 2 10 2 18 3 1 1 LW HOLE OBS LW7 1 5 2 10 2 18 4 1 2 LW HOLE OBS LW7 1 5 2 10 2 18 5 1 3 LW HOLE OBS LW7 1 5 2 10 2 19 6 1 4 LW HOLE OBS LW7 1 5 2 10 2 9 7 1 5 Now we must place the contents of the SSCD into a PDS For an SSO observation we use a general PDS This is created with get_sscdstruct CIA gt sso_pds get_sscdstruct cleaned_sscd 0 6 Now we can proceed with the calibration We will perform the standard calibration steps on the cube i
59. Section 15 2 2 M_RASTER As for SCD see Section 15 2 2 ASNUMBER As for SCD see Section 15 2 2 CAL As for SCD see Section 15 2 2 15 2 5 Set of SADs SSAD The SSAD is a catalogue of a set of SADs either origin SADs or future SADs The SSAD has the same relationship to SADs that SSCDs have with SCDs It also has similar structure to the SSCD In addition to the standard fields of Section 15 2 1 it has the following fields 1 2 o N QC 10 NSAD The number of of SADs belonging to this SSAD Type integer ORIGIN If a future SSAD returns the name of the origin SSAD which catalogues the origin SADs Type string FUTURE If an origin SSAD returns the name of the future SSAD which catalogues the future SADs Type string RASTER COLUMNS As for SSCD RASTER LINES As for SSCD RASTER ORIENTATION As for SSCD RASTER ROTATION for SSCD RA RA J2000 of the centre of the final MOSAIC that is constructed from the EXPO SUREs in the SAD Type double Unit decimal degrees DEC DEC J2000 of the centre of the final MOSAIC that is constructed from the EX POSUREs in the SAD Type double Unit decimal degrees ROLL As for SCD see Section 15 2 2 15 3 CALIBRATION DATA STRUCTURE CDS 171 15 3 Calibration Data Structure CDS One generic structure is used to contain all of the different types of calibration data This is the Calibration Data Structure It contains a dy
60. You don t want to overwrite the old SSCD with the modified SSCD d Restore the original SSCD and slice another CONFIGURATION if there is one 12 2 2 Slicing a raster observation AOT 1 An example procedure which illustrates the use of both automatic slicers follows begin the procedure at Step 5 if you are only interested in slicing SPD data products and skip that step if you are only interested in slicing ERD data products 1 Start a CIA session 2 Assign useful directory paths to IDL variables In VMS CIA product dir DKA200 MDELANEY PRODUCTS CIA scd dir DKA200 MDELANEY SCDS CIA sad dir DKA200 MDELANEY SADS and in UNIX CIA product dir home mdelaney products CIA scd dir home mdelaney scds CIA sad dir home mdelaney sads 3 Convert the ERD data products on disk to ERD SCDs in CIA Use the CIA routine erdtoscd to perform this conversion and automatically save the ERD SCDs to disk CIA erdtoscd cier 4300601 fits erd sscd dir product dir CIA scd dat scd dir ack ack 12 2 AUTOMATIC DATA SLICING 79 The ERD SCDs are both in memory and saved on disk This is useful if you wish to end your CIA session with sscd read you can easily recover the ERD SCDs from disk in a later session Let s take a look at the contents of memory as follows CIA gt print sscd_list C83C143006010001 96082811465989 CIA print erd_sscd C88C14300
61. a micro scan refers to more than the special case of where the step size is not an integer number of pixels 19 6 ADVANCED SLICING WITH X_SLICER 225 e M RASTER and N_RASTER can be set to zero by a break of telemetry on a raster position for one or two frames Check these variables carefully because even if M RASTER and N_RASTER are false the images will be correct The x slicer will propose to break the SCD corresponding to this raster position into three parts You will have to use the Merge buttons to rebuild a whole SCD 19 6 5 5 On target flag and sequence number The On Target Flag OTF variable appears in two places while using x_slicer First you can select it as a slicing variable This is no longer recommended but that possibility has been left available Second you are asked in the x handle slice window if you wish to use the Enhanced OTF These are two different things It will be explained in a whole section below The Sequence number can sometimes be useful when more than one observation is contained in a file as it can help to discriminate between them The Sequence number STN in the SCDs are recomputed according to their place in the SSCD It is likely that you will get different values in the STN field than the one displayed by x slicer 19 6 5 6 Automatic slicers slicing variables As you probably know by now there are two other slicers in CIA These are the automatic slicers spdtoscd and erdtoscd See Section
62. an implicit agreement to its distribution to all CIA sites e It should be clear to all parties involved with whom responsibility for the maintenance of the routine lies e Routines should conform to the requirements outlined in Section D 2 Users can also supply routines to be distributed with CIA as contributed routines found in the directory cia vers contrib These are not supported by the CIA team Bug reports should be made directly to the author D 2 Basic requirements Please ensure that your routine conforms to the following requirements e All files must contain a header see Section D 3 below e A routine should display its calling parameters if it is called without any parameters e High and medium level routines should use the CIA error reporting conventions CIA is full of examples Normally this will be routines in the User category e Mask handling and the setting of undefined values should follow the appropriate conven tions 323 324 APPENDIX D GUIDELINES FOR WRITING CIA ROUTINES D 3 How to write a header In order for automatic formatting programs to work as well as for purposes of standardisation these rules must be followed in all headers e A header begins with and ends with e There is only one header per file regardless of the number of routines contained therein It may be anywhere in the file but most programmers prefer to find it at the top e The keywords are NAME PU
63. analysis 14 4 1 Extracting images from cubes with xselect frame This is a simple widget based program to interactively allow the user to select a single image from a given cube see Figure 14 5 It is invoked as CIA frame xselect_frame cvf_pds image frame index frame would contain the selected image upon quitting and frame index is the index of the selected image in the cube 14 4 2 Extracting images from cubes with xsubcube xsubcube is similar to xselect frame though it allows for a subcube of images to be extracted from a cube rather than just a single image to be extracted It is a widget based program with several parameters that may be interactively selected e The top slider specifies the number of frames per packet where a packet refers to an image in the final output subcube In other words it specifies the number of images in the input cube to average to produce an image in the output subcube 126 CHAPTER 14 IMAGE ANALYSIS AND DISPLAY e The next two sliders allow for choosing of the first and last image in the input cube to make up the output subcube e The last slider scrolls each image or the packet of the output cube e The button Area Extraction calls another widget which allows for the selection of a subim age from the images of the input cube e The buttons reset and cancel do the obvious apply applies the selection and quits A typical application may be to select a subcube from a CUBE or IMAGE o
64. calib struct automatically smooths column 24 in the and CUBE However the weight of every pixel i e the values in NPIX in column 24 will always remain as zero Therefore column 24 will appear as a dead column in the MOSAIC However it will no longer be column 24 but column NX RASTER 8 Dead pixels appearing in the raster MOSAIC are optionally smoothed by calib raster 1 dead method Where dead pixels appear in the raster MOSAIC i e where NPIXRASTER is zero a median smoothing method is applied Only the dead pixels are smoothed into their neighbors no other pixels are affected called routine im smooth PDS side effects Dead pixels in RASTER are smoothed 252 CHAPTER 20 ADVANCED DATA CALIBRATION SELECTION INSIDE OUTSIDE uNDo DISPLAY pixel 23 19 value 0 992318 mean 3 0 988307 FLATFIELD MEAN HEDIRN CAL G PROFILES DISPLAY Click on frame number to select unselect it Flatfield 4 015 O14 O13 E12 O10 11 O7 O68 05 4 Oo 01 02 03 3 mean 1 00059 3 t sigma 0 013355 I MEDIAN method se 33min 0 902911 I frames 4 11 12 3 1 09131 W MEDIAN IS NOT SI M ALL PIXELS ARE 5 Ee 3 00000 M no new selection 4 mean 0 998907 4 sigma 0 011900 I MEAN method sele 4 i min 0 905289 I framets 4 11 12 4 i max 1 06555 7 no new selection 7 Ca M gt XLORDCT TEST FLAT CANCEL DONE Figure 20 4 flat_builder
65. can choose a particular deglitching method by setting the keyword method to a particular string value The possible values for method and the associated deglitching methods and the associated routines which deglitch calls are described in the following subsections Technical details of some of these methods can be found in the references given below 1 method spat method Spatial and temporal deglitch This method attempts to use both temporal and spatial information of the CUBE to perform deglitching It finds the difference between successive IMAGEs the CUBE we will refer to this as the difference image Then it applies a thresholding criterion to each pixel of the difference image based on the median of the standard deviation of the difference image and the median of the mean of the difference image A pixel failing this criterion corresponds to a glitch in the IMAGE It is considered to be a good method when the CUBE is not stabilized However it may be poor at detecting long duration or close successive glitches routine called deglitch spat PDS side effects Glitches removed from IMAGEs in CUBE i e CUBE is modified Glitched pixels also flagged in MASK reference ISOCAM Handbook Chapter Data processing methods Section Spatial and temporal deglitch 2 method temp method Temporal deglitch By this method pixels failing a combined median and stan dard deviation test in temporal space are considered g
66. choice of parameters for the DivSky method 283 Common variables used in CIA and the CIA User s Manual when identifying data products o ce X due de Lo aa X Rn mc m ee Se 322 xix LIST OF TABLES Chapter 1 About the CIA User s Manual 1 1 Organization of the CIA User s Manual The CIA User s Manual is split into four parts Part I Quick Start Guide Part II CIA Basic Guide Part III Data Management and Part IV Advanced Use of CIA To further understand this organization read the introduction that can be found at the beginning of each of these parts To get maximum benefit from the CIA User s Manual novice users of CIA may find it helpful to read each chapter sequentially beginning obviously with the CIA Basic Guide while more advanced users may wish to jump in at the sections of interest to them In any case it is hoped that adequate cross references exist to make the CIA User s Manual useful wherever you choose to begin reading 1 2 What you need to begin Check that you have the following before continuing 1 Some knowledge of IDL 2 Data products Either retrieved from ISO s Post Mission Archive IDA at Villafranca a CD ROM issued to you by ESA see Chapter C or an archive at your site Additional documentation which you may need as you progress with CIA data analysis is described here 1 CIA documentation CIA documentation is distributed with CIA Documentation is generally in p
67. cvf_scds 76 6 The CAM parameter of most interest in a CVF observation is of course the wavelength of the CVF position in each STATE in fact this is the only parameter we expect to see changing We can use sscd info to specifically list this information per SCD i e per STATE CIA sscd info cvf sscd param wavelength mode 19 Sep 1996 19 31 37 00 PARCHECK V 1 0 SUCCESS S gt 77 SCDs in the SSCD CSSC203056040001_96091918462427 seq wavelength mode 0 8 78200 OBS 00300 OBS 11300 OBS 22300 OBS 33300 5 44200 OBS oP oO 86 CHAPTER 12 DATA SLICING 70 16 0500 OBS 71 16 1400 OBS 72 16 2400 OBS 73 16 3300 OBS 74 16 4200 OBS 75 16 5200 OBS The sequence number of each STATE and wavelength of the CVF in microns is listed along with the OP MODE 7 The data are now sliced We can store our results with sscd write CIA sscd write cvf sscd dir scd dir 12 3 Data slicing with x slicer This section introduces you to the CIA s x slicer This is a widget based program which allows you to interactively select your own slicing criteria The advantage of this way of preparing the data is that you get to see exactly how your observation was performed including all intermediate states and you can create an SSCD so it contains the states that you choose Generally you will use x slicer to create an SSCD for each configuration of an AOT and then create a PDS from the SSCD x slic
68. cvfi2 fits cvf1i6 fits cvf2 fits cvfb fits cvf9 fits sad write fits Expert users of CIA may use SADs to hold their calibrated data To write the SAD CMAP to a FITS file CIA sad write fits CSAD000014300601 98072013530646 cmap Note that the FITS files created by the routines described above can be easily loaded into IDL with the ASTROLIB routine READFITS 18 3 Export for archiving For archiving purposes we want to save the entire contents of a CIA data structure to an extended FITS file struct2fits writes an entire CIA data structure to a FITS file The data is stored in extensions CIA struct2fits raster pds name raster_archive fits fits2struct recovers the output of struct2fits It initializes the appropriate PDS and then fills it as best it can It is very useful for upgrading the architecture of an obsolete PDS 18 3 EXPORT FOR ARCHIVING 207 CIA fits2struct raster_archive fits hdr raster pds recovered CIA help raster pds recovered str Structure lt 123a0f0 gt 54 tags length 1399800 refs 1 RASTERCOL INT 4 RASTERLINE INT 2 M STEPCOL FLOAT 84 0000 N_STEPLINE FLOAT 84 0000 etc Optionally setting the keyword orig will cause fits2struct to recreate the original ar chitecture of the PDS This means that an archived PDS with obsolete architecture will be recovered into a later version of CIA without being upgraded 208 CHAPTER 18 EXPORT OF CIA DATA STRUCTURES Part IV
69. different filters from modem calculation you can use it as input to corr colour The SED data points must be first written to a text file in a simple two column format with wavelength microns in the first column and flux density Jy in the second This name of this text is supplied as input to corr colour CIA more sed dat 2 38014 440 23920 38163 429 91971 38311 421 62091 38460 419 00839 38610 424 57556 38759 436 57462 etc CIA corr colour 1 10 sed sed dat 20 14 A note on the infamous column 24 At this stage you should be very familiar with the LW detector s dead column 24 or by IDL convention column 23 By default most CIA processing will attempt to interpolate column 24 so that it is no longer appears in the reduced data Not all observers will desire this To avoid any filling of column 24 use the keyword dead col The routines that accept this keyword are e By default get sscdstruct get sscdbs get sscdraster and get sscdcvf will fill col umn 24 in the output PDS field CUBE with the average of its neighboring columns If dead_col is set then column 24 is set to zero 20 15 ADVANCED PROJECTION 259 e corr dark will ignore column 24 if dead col is set otherwise it will call corr col24 to fill it after performing dark correction on CUBE e corr flat will ignore column 24 if dead_col is set otherwise it will call corr_col24 to fill it after performing flat correction on I
70. disk and give them the correct official name 10 1 1 Copying on VMS READ CDROM is a DCL script distributed with CIA It will automatically copy rename the CD ROM data products to disk space Before using it ensure that your CD ROM drive is not already mounted and then insert your CD ROM into the drive To execute READ CDROM change to the directory in which it resides and call it with W40 OREAD CD COM Initially the program prompts you for an existing destination directory It then mounts the CD ROM drive and reads the file datalist txt directly form your CD You ll be asked now where you want your files copied to Name of target directory lt gt IS0W40 DKA200 SOTT OLP1 IA STEPHAN TMP READ CDROM contains some lines which have to be edited for the particular system you are working on e g the CD ROM device name must be inserted These are indicated at the beginning of the file It is expected that your CIA administrator will have taken care of this 67 68 CHAPTER 10 FIRST LOOK AT THE DATA Temporary mount to read contents of CD ROM 4MOUNT I WRITELOCK volume is write locked AMOUNT I CDROM IS0 10058016 10058016 1 of 1 mounted on _ISOW40 DKA600 ACOPY S COPIED ISOW40 DKA600 000000 DATALIST TXT 1 copied to OLP1 IA STEPHAN TMP DLIST TMP 1 14 blocks It then dismounts the drive and prompts for the data you wish to copy you can choose to copy data by instrument Type a single letter ls c and
71. for drizzling via the keyword shrink e Improved treatment of solar system objects by the new routine project sso CIA project sso struct eph eph e Upgrade of flux conversion calibration conv flux uses improved flux conversion sen sitivity factors and is now time dependent Additionally the calibration file ccglwsens contains now the column senserr to hold the uncertainty on the conversion factors 337 338 APPENDIX F WHAT IS NEW IN CIA 5 0 e New deglitching routines sky cube deglitching sky and two pass sigma deglitching ksig Sky cube deglitching can do miracles rejecting faders and dippers from rasters with redundant pointings ksig deglitching is a second order deglitching on stabilized data to remove 1 Glitches 2 Tails of glitches 3 Residual tails of up downward transients You can call these methods by CIA deglitch raster method sky CIA deglitch data method ksig e The median filtering for the treatment of faint source observations was improved in cluding a new algorithm to best window size The calling syntax is CIA stabilize data method med F 2 New and improved functionality e With xcorr astro you can correct for the astrometric shift due to ISOCAM s wheel jitter e Several routines can exist now in multiple copies xdisp ximage xphot cvf display show frame xcube e Improved help help functionality cia help can now search for multiple items
72. is selected then the Z axis is an index if s is selected the Z axis is time counted in seconds The Plot Window is updated accordingly go the the next frame go the the previous frame 14 4 CUBE ANALYSIS 131 Lower we have three buttons which are a multiple exclusive choice button only on the three can be active They act on the Plot Window Temporal Cut the default choice the Z axis of the plot is time Horizontal Cut the Z axis of the plot is now X axis e Vertical Cut the Z axis of the plot is now Y axis Lower we have three buttons for the mask They are not active if the input is a naked cube Bad Pixels toggle button can be on off The default is off and the bad pixels are not visualized e Interpolated values are used in the plot and dark pixels in the image Symbols toggle button can be on off If the button is on the following symbols are used masked pixel struct mask eq 1 undefined value 0 32768 NaN Inf Inf rejected frame struct rejected eq 1 off target frame struct on target eq 0 used for deglitching enabled when xcube is called with the option block First efine the region with the mouse left button then click the middle button to label the region as bad or right button for good The mask is updated The following button is Image which enables to choose the source of the image It is a menu bar with the choices depending upon the input
73. is the number of above and below 0 that we use as a threshold in the examination of the residual maps The range of acceptable values depend quite strongly on the noise used either global or local see below e local f set the current error map will be used to examine the residual maps instead of a median noise level In principle this option should be used although the bad pix thresh should then be higher than in the median noise option e ghost thresh As the name implies used to flag ghosts e peak thresh Also related to ghosts 21 7 Frequently Asked Questions and Problems In this section I list the questions that can arise from a use of SLICE and whose answer did not really fit in the flow of this introduction or are worth restating Q I have applied the 1tt action and subtracted the correction derived by SLICE yet after flat field correction with DivSky the long term transient is apparently still there what is happening A This is typically a property of DivSky which uses the previously existing map which in general was made prior to the long term transient correction and therefore still shows it to estimate the true sky If the long term transient is strong enough then it can be detected by SLICE as a flat field defect it is in the image but not in the cube itself because it has been removed and it will create a new image still affected by what seems to be a long term transient although now it is in the flat field a
74. it may catalogue any subset of SADs from an AOT Diagnostic Specific Data DSD Contains physical parameters of the camera tempera tures voltages wheel positions etc It is directly used only by instrument experts for in depth investigation of ISOCAM behaviour Such data that is of interest to the normal user is also held in the SCD Documentation of the DSD structure is beyond the scope of the CIA User s Manual 9 4 3 Calibration Data Structure CDS One generic structure is used to contain all of the different types of calibration data This is the Calibration Data Structure It contains a dynamic substructure or field named DATA and a standard set of fields DATA holds the actual calibration data dark images flat images point spread images etc The standard fields hold information related to the nature of the CDS structure itself e g the CDS name and size The CDS is almost a direct conversion from FITS format data to IDL data structure the important difference being that the actual image data are scaled so the BZERO and BSCALE keywords are discarded So the CDS differs from the SCDs SADs in that the CAM parameters are NOT presented in the CIA user friendly format but as raw values taken from the CAL G FITS files 9 4 4 Regular IDL structures Unlike the structures described in Sections 9 4 2 and 9 4 3 the structures described here are regular IDL structures Prepared Data Structure PDS Currently three flavours of the
75. masked for the entire duration of this CONFIGURATION This leads to the appearance of undefined pixels in the raster MOSAIC here they seem like white holes in the image Displayed by tviso CIA raster2fits raster pds name raster fits iraf 7 You can calibrate the other CONFIGURATIONS in a similar manner Start again at Step 4 and use a command like CIA another raster pds get sscdraster cleaned sscd 1 Chapter 4 Staring observation CAMOI1 4 1 Description of the observation The data used here is from a CAM staring calibration observation of HIC 73005 A staring observation is the simplest application of CAMOI there is only one pointing in the observation and hence only one STATE per CONFIGURATION Though you can have several CONFIG URATIONS per observation the data presented in this chapter is from an observation with a single CONFIGURATION The relevant parameters are LW7 filter 1 5 PFOV gain 2 and integration time 2 1 s 4 2 Data analysis It is assumed in this section that you have read Chapter 3 Generally concepts described in that section will not be re described here 1 Start a CIA session cia 2 Convert your CISP data product into SCDs with spdtoscd CIA spdtoscd cisp03001209 fits sscd dir cia_vers test nowrite 3 Get an overview of the SCDs CIA sscd info sscd deg 4 SCDs in the SSCD C88C030012090001 02022715032901 Seq channel mode fltrwhl pfov tin
76. most recently viewed Take a look at Figure 14 14 Note the following points e The image displayed is the raster MOSAIC of the input PDS in our example call above this is raster pds A cross hair is positioned on the pixel of maximum intensity The cross hair can be moved by clicking on the image or by clicking on the buttons left right up down center Clicking on max will return it to the pixel of maximum intensity Clicking on Raster FOV will place a grid representing the FOV of each EXPOSURE over the raster MOSAIC as display in Figure 14 14 e A plot of the history of the pixel under cross hairs should also appear this is not shown in Figure 14 14 Since you are viewing a raster observation each raster MOSAIC pixel can be made up of several EXPOSURE pixels one per each SCD or STATE of the observation For this reason a history of each EXPOSURE pixel is shown in the plot window Clicking on multi or single will switch to individual plots or to an overplot of EXPOSURE pixel histories respectively e Clicking the button mask will mark each masked pixel in the plot Clicking on SCD limits will place markers on the plots at the SCD boundaries e A postscript file of the plot xv_raster ps can be create by clicking on print e The button color will invoke IDL s XLOADCT 14 4 CUBE ANALYSIS 143 Click On The Raster Map Display Mask SCD limits Raster FOV Plot type Multi Left Right up
77. named and how they relate to CIA Data Structures see Part III for a more detailed account of the CIA Data Structures 9 1 Data product filename convention Observers who have not obtained their data via IDA will have to deal with the filename con vention that has been applied to the data product files on the ISO CD ROM This is due to a community requirement that the length of the filenames comply with the DOS convention of a maximum of eight characters with a three character extension The file datalist txt see Section C 2 2 lists the official names of the files on the CD ROM along with the actual abbre viated name used Note that in the CIA User s Manual so called observation data product files found in the CD ROM directory products pmmmmmmm nnnxxxyy are always referred to by their filename root even though the official name is the root with nnnxxxyy appended to it e g CIER refers to CIERnnnxxxyy 9 2 Data products as FITS files All ISO data products whether retrieved from IDA or delivered on an ISO CD ROM are FITS files The features of these FITS file are described in the ISOCAM Handbook The ISO Data Product Document lists the header and binary table keywords for all ISO files From the table of contents of the ISO Data Product Document you can see that the files are grouped in a particular way Section 9 3 will help you understand the grouping and find the FITS file details you want in the ISO Data Product Document 9 3 Rela
78. number i e the position of the SCD within the SSCD then an underscore then the creation date in the order year month day hour minutes seconds and centiseconds User defined names that fall short of the full name length will be automatically padded out with underscores Keep the following in mind when selecting a name e Do not forget to hit the return key after entering characters e If you press the Official button you will get the official name for your data e If you press Cancel button you will go back to the x handle slice window with your previous choice When you are satisfied press the Go button x_slicer will begin processing and a message will appear when it has finished You can then press the Quit button of the x handle slice window to exit x slicer or continue to process the data remaining in your ERD Chapter 13 Data calibration This chapter introduces you to the process of data calibration with CIA see Figure 11 1 Through out this chapter calibration will be described as being performed on PDSs which are of course derived from SCDs It is worth mentioning at this point that calibration can also be performed directly on SSCDs and SCDs and indeed more and more users are doing so for reasons described in Section 13 1 1 However it is probably best for the novice user to initially work with PDSs if it is possible to do so Note that the material in this chapter follows closely from Chap
79. of the time the fit and the exact solution apparently disagree This is because the so called fitted correction is not a fit to the exact correction but rather a correction derived assuming the long term transient is the combination of two exponentials whose parameters are fitted Therefore when these two corrections agree in shape you have a good sign that you have reached the right correction an offset is not important as it will be taken care of by your background subtraction On the contrary when they disagree it is a good indication that you should tune the parameters better Once you have reached a satisfactory determination it can be subtracted from your data using the commands suggested by SLICE Note that in some very complex case it may be interesting to subtract the fitted correction first and then iterate flat field determination and long term transient correction with the exact determination for the remaining residuals If for whatever reason you have determined the long term transient in any other way and stored it in a variable call my_1tt then the command to use to subtract it is I admit it is a bit complex and this is why the code itself prints it so that you can cut and paste it on the command line CIA c im param indfiltre 1 sub c im param indfiltre my ltt 21 5 4 Second flat field determination In a number of cases this will be the last one producing your final image In principle at
80. on the left and LW2 on the right Com pare with Fig 21 1 to measure the improvement 291 Distribution of jitter 294 Comparison of jitter computation methods 295 Definition of roll angle for the LW 328 Schematic of a Y axis raster 329 Schematic of a M 4 N 3 raster oriented with reference to the North axis 331 The roll angle a for each detector and for each value of the IDL ORDER system varia pler us darme e oe eese 333 Conventions for the standard astrometric keywords in a FITS header 334 xviii LIST OF FIGURES List of Tables 3 1 12 1 15 1 19 1 19 2 21 1 21 2 21 3 21 4 C 1 The CONFIGURATION parameters of the raster observation of the Antenna galaxys Xem Ru m RON RU PURUS Mee CE bee AG UR A PU 5 23 Conversion table for the variable names displayed by x handle slice 92 The calibration data and associated CDS mnemonic used for naming purposes 173 Slicing variables used in _ 5 223 Slicing variables used by CIA s automatic 5 226 The SLICE variables and their content 275 Observing setup for the NGC 2366 data 278 Our choice of parameters for the Perturbed Single Flat Field method 282 Our
81. p for LWS SWS and PHOT respectively for each instrument s data that you want copied If you are unsure choose A for all The drive is then re mounted and the transfer to disk is executed Now dismounting before final mount to read FITS data 4MOUNT I WRITELOCK volume is write locked AMOUNT I CDROM IS0 10058016 10058016 1 of 1 mounted on _ISOW40 DKA600 Ready for data transfers Enter from 1 to 4 characters one per instrument you want to copy for instance LS for LWS and SWS Alternatively you may copy all data example LPC or CLPS enter just A for all data a Product set NR P0087982 Proposer ID LMETCALF Product aocs58302914 gt Creating aocs58302914 fits on OLP1 IA STEPHAN TMP ACOPY S COPIED ISOW40 DKA600 PRODUCTS P0087982 58302914 AOCS FIT 1 copied to OLP1 IA STEPHAN TMP AOCS58302914 FITS 1 1350 blocks gt Product 583 gt Creating eoha583 fits on OLP1 IA STEPHAN TMP gt Product psta58303015 gt Creating psta58303015 fits on OLP1 IA STEPHAN TMP ACOPY S COPIED IS0W40 DKA600 PRODUCTS P0087984 58303015 PSTA FIT 1 copied to OLP1 IA STEPHAN TMP PSTA58303015 FITS 1 34 blocks DATALIST TXT finds 86 FITS files 10 2 EXAMINING THE AAR DATA PRODUCTS 69 10 1 2 Copying on UNIX fitsname is a general purpose routine for managing FITS data products in the UNIX environ ment To copy rename all the data products from an ISO CD ROM to
82. parameters for each CONFIGURATION 3 2 Data analysis 1 Start your CIA session cia If you work on a VMS system you may have to type IDL to begin your CIA session 2 The data products which are delivered to you on the CD ROM are in the format of extended FITS files The first thing you want to do is load all this data into your CIA session We will use spdtoscd to do this Firstly be sure that the directory you are working in is writable as CIA needs to be able to write and delete files on the current CONFIGURATION raster size filter wheel PFOV integration time gain arcseconds seconds 1 2 4 LW2 3 0 5 04 1 2 2x4 LW2 3 0 2 10 2 3 2x4 LW3 3 0 5 04 2 4 2x4 LW3 3 0 2 10 2 Table 3 1 The CONFIGURATION parameters of the raster observation of the Antenna galaxy 23 24 CHAPTER 3 RASTER OBSERVATION CAMOI directory for file saving sharing data with external executables etc Identify the location of the CISP data product The actual CISP data product name is the first argument to spdtoscd and the keyword dir is set to the directory containing the CISP file If you are working from the CD ROM you will need to copy your data products to disk space see Section 10 1 In our example the CISP file is called cisp02600506 fits and it is located in the subdirectory test of the CIA installation directory CIA spdtoscd cisp02600506 fits sscd dir cia_vers test nowrite When spdtoscd is finished we have c
83. pds You can view this MOSAIC with tviso a convenient modified version of IDL s TVSCL CIA tviso raster pds raster The window in Figure 3 3 will appear It contains the raster MOSAIC that is stored in raster pds raster You may wish to save the results of the data analysis You can do this with IDL s SAVE CIA save file raster_pds xdr raster pds Alternatively you can export the data to a FITS file This is more useful if you intend to perform further analyses with other analysis packages The following will export raster pds to an IRAF FITS file see also Section 18 2 3 2 DATA ANALYSIS 29 Frame Number 1 LORD LUT ww Horizontal cut Quit Qr Vectical cut Window Size E _ Mask _ Glitch Z00M RANGE Y SCALE gt p Temporal cut Aani Vif VL i I when Me Vu den Next frame Previous frame Pixel value Eubec1 8 106 4 0545 al Figure 3 2 A raster IMAGE after dark correction deglitching and stabilization The effects of the dark current have been removed and the glitch which was very apparent in Figure 3 1 is no longer visible A source which was previously obscured by the glitch has now become visible in the upper half of the IMAGE Displayed by x3d 30 CHAPTER 3 RASTER OBSERVATION CAMOI Figure 3 3 The calibrated raster MOSAIC Some pixels have been
84. raster window in ximage the tail of glitch 142 CHAPTER 14 IMAGE ANALYSIS AND DISPLAY 14 4 10 An example Simply starting ximage as CIA ximage raster gives the following output gt gt click on data mode multiple choices button choose overplot mode gt gt click on the sky view button gt gt click on overplot mode multiple choices button choose data mode gt gt click on the raster button gt gt move the mouse click the left button to select a pixel This tool can be used for example to test if one bright point in an image of a raster PDS is real or an artifact due to memory effects ghost or to glitches Figure 14 12 shows a point of the mosaic 46 19 which is a ghost one of the sky pointing has seen a bright source before and the downward transient correction is not perfect Figure 14 13 shows a point of the mosaic which is also an artifact one of the sky pointing contains the tail of long time glitch fader 14 4 11 xv raster xv raster the pre curser of ximage is another cube analysis tool What makes this tool different from the rest is that it is shows you the pixel histories in the context of the raster MOSAIC This gives you a better idea how each pixel of the PDS cube affects the outcome of the raster MOSAIC It can be invoked simply as CIA xv raster raster pds output Note from above that you need to supply the whole PDS to xv_raster The variable output will return an array of pixel history data
85. rules you must follow if you want to get a sensible result Some processes can be repeated without having to start from scratch with raw data 250 CHAPTER 20 ADVANCED DATA CALIBRATION 1 Save a copy of your PDS Some calibration methods are irreversible they change the data in CUBE and MASK and if you don t like the result you will need to begin again with fresh data 2 Dark correction deglitching and stabilization are all operations on CUBE These processes must applied before any other and in the correct order a Dark correction Example corr dark pds method model Caveat Irreversible Dark correction can only be performed once If you don t like the result of a dark correction then you have to start again with the original data b Deglitching Example deglitch pds method mm Caveat Irreversible but you could deglitch twice However there is a strong danger of over deglitching your data Best to begin again with the original data c Stabilization Example stabilize pds method s90 Caveat Irreversible but you could stabilize twice with non fitting methods such as s90 However it is not recommended 3 Reduction of CUBE to IMAGE Example reduce pds Caveat Reversible You can reduce as many times as you like Although processes following reduction will of course have to be repeated 4 Flat fielding Example corr_flat pds method library Caveat If the flat fielding is pe
86. select ADD TO to place this star in the working table for cross correlation Note that only pixel positions are currently identified for this star in the cross correlation table 14 6 IMAGE COMPARISON AND OVERLAYING 151 6 Now click with the left mouse button on this same star in the reference field The fields located below the Reference Field are now filled with appropriate values In this case the star s FITS astrometry header is used to calculate the equator ial position To associate these values to the pixel positions identified in the target field click on either the COMPUTE CENTROID RA DEC or COMPUTE RA DEC only buttons The former calculates a simple brightness weighted mean position before comput ing the equatorial co ordinates The latter does not compute any centroiding calculations This option is useful if you believe centroiding can negatively affect the calculations e g in crowded fields 7 Repeat for as many stars as desired The cross correlation table can be printed to a file or saved as an IDL save set ii To determine equatorial co ordinate offsets between two images Follow steps 1 7 to build the cross correlation table as listed above Then click on the COMPUTE DISTORTION COR RECTION button The jitter correction fields located to the right of the table are filled with the values for RA and DEC offsets In these fields CO is the RA offset and DO is the declination offset iii To deter
87. several FITS reading routines which are of interest MRDFITS and READFITS CIA comes with a modified version of MRDFITS the keyword option savemem has been added to save memory when reading large files 17 2 1 Reading an ISO data product The CIA routine extended fits read reads an extended FITS file into a conventional IDL structure not a PDS or SCD For example to read a CMOS data product remember that ISO data products are delivered in extended FITS file we do the following CIA ia extended fits read cmos14300601 fits header array CIA help hdr HDR STRING Array 209 CIA gt help array ARRAY STRUCT gt Anonymous Array 6 The output arguments header and array contain the header of the FITS file and the data from the FITS extensions We can look at lines in the header by indexing header CIA print header 0 SIMPLE T file does conform to FITS standard To look at the binary table CIA gt help array str Structure lt 1242100 gt 38 tags length 25896 refs 1 CHANNEL STRING INDEX LONG 1 TYPE STRING gt FLUX LONG 32 etc 204 CHAPTER 17 IMPORTING ISO DATA PRODUCTS TO CIA Each ISOCAM image is stored in a vector ISOCAM Handbook Section 2 3 4 Before we can reform and display them they need to be rescaled ISOCAM Handbook Chapter 7 CIA bscale array bscale CIA bzero array bzero CIA data float array array CIA for 0 5 d
88. slice according to the following variables e Entrance Wheel the position of the Entrance Wheel of CAM hole polarizers e Selection Wheel the wheel that reflects the light on the SW or LW channel WARNING the selection wheel on SW for example does not imply that the operating channel is SW Think about a LW dark measurement for example e LW lens Wheel the position of the LW lens wheel of CAM e LW filter Wheel the position of the LW filter Wheel of CAM e SW lens Wheel the position of the SW lens wheel of CAM e SW filter Wheel the position of the SW filter wheel of CAM e Observation Channel it tells you if you are observing with the LW or SW detector WARNING its value is independent of the selection wheel value e Beam Switching beam switch mode flags the cycle number the reference and the source e AOT the one of the three AOT of CAM that was used e Observation Type AOT CUS Calibration Up link System or other type of driven observations e Detector Offset No longer useful since there is only one offset per gain of the detector e On Board Process Normal mode accumulated mode or sampled mode of CAM detector reading e Integration Time value of the integration time e Detector Gain value of the detector gain 1 x2 or x4 e Observation mode OBS DARK WAIT IDLE GAP e On target Flag On if the telescope is on target and the qla_flag is OK see Section 19 6 7 e Raster Mode for raste
89. sliced your SPD SCDs you can convert them to a BS PDS in a similar manner to the other PDS flavours This time you use the routine get_sscdbs Note that there is one irregularity which sometimes arises in beam switch observations some observers have programmed their observations in reverse Section 19 3 tells you how get_sscdbs can be used to deal with this problem Assuming that bs_sscd is an IDL variable containing the name of your sliced SCDs SSCD then you can create a BS PDS named bs_pds with CIA bs_pds get sscdbs bs_sscd 13 1 5 CVF PDS The CVF PDS is designed to handle data from a CVF observation AOT 4 In principle the CVF PDS is created like the raster PDS of Section 13 1 2 though with the CIA routine get sscdcvf The following example follows that of Section 12 2 3 If the sliced SPD SCDs are not currently in memory they can be recovered from disk with sscd read CIA cvf sscd sscd read CSSC203056040001_96091918462427 dir scd dir The next step is to create the CVF PDS from the SPD SCDs the IDL string variable cvf_sscd contains the name of their SSCD CIA gt cvf_pds get_sscdcvf cvf_sscd del Note that the keyword del is set CVF observations yield large quantities of data and it is not wise to duplicate it in both an SSCD and a PDS del deletes the SSCD automatically upon creation of the PDS As for a raster PDS you can look at the structure of a CVF PDS with IDL s HELP CIA gt help
90. struct2sad 14 3 1 xdisp xdisp is a CIA routine to display and analyse single images They can be in several formats IDL 2 D array FITS format images the CCIM from an SAD or images in MIDAS format It can interactively display profiles of rows or columns display histograms create contour plots uses xcontour see Section 14 7 1 perform simple statistical analyses and compute the Fourier transform of an image A recently added feature is the option to interactively save a region of the displayed image as a FITS file This FITS file can then be used as an input to other analysis packages or other CIA routines such as isocont 14 3 1 1 The xdisp window Figure 14 4 shows the xdisp window All the operations described above can be performed by clicking on the following window buttons e quit Quit the application load Load an image Image formats can be fits MIDAS SAD or 2 D array e lut Modify the LUT i e call xloadct e get profile Examine lines or columns e get cursor Examine pixel values If image format is FITS PDS or SAD and if the header contains astrometric information then pixel position in the sky is given R A and Dec e plot histo Plot the histogram e plot contours Invoke xcontour e plot visu Invoke interactive widget to display a surface plot e plot load copy image to an ordinary IDL window e info Print the min max mean sigma of the image
91. the actions to perform red param structure holding the parameters of the routines im param a structure describing the current observation raster the cube of individual raster pointings the PDS image field and NOT the PDS raster field flat fielded correction the exact long term transient correction corrfit the fitted long term transient correction error error associated to the long term transient correction 276 CHAPTER 21 USING SLICE WITHIN CIA 21 4 Processing in SLICE 21 4 1 The SLICE syntax Here again SLICE differs from CIA In SLICE you do not manipulate directly the routines that perform the actions rather you describe with a set of structures the actions you want to perform and then tell SLICE to do them This philosophy is based on the necessity to perform the data reduction steps in the correct order and since there are many iterative processes involved this is very important and on the desire to be able to pipeline the data processing easily To describe the action i e data reduction steps you will perform you have access to two structures the red_param structure contains the parameters of the data reduction routines that will be involved in the processing and the act structure will contain the actions to be perform basically the act structure is a structure of boolean keywords one for each data processing step Once again it is not our intention here to supersede SLICE s manual We highly re
92. the contents of our raster PDS CHAPTER 3 RASTER OBSERVATION CAMOI CIA help raster pds str Structure b09cO 54 tags length 1399800 refs 1 RASTERCOL INT 4 RASTERLINE INT 2 M STEPCOL FLOAT 84 0000 N STEPLINE FLOAT 84 0000 RA RASTER DOUBLE 180 47414 DEC RASTER DOUBLE 18 878420 ANGLE RASTER DOUBLE 109 43000 RASTER ROTATION DOUBLE 199 43000 RASTER ORIENTATION STRING Y AXIS ASTR STRUCT gt ASTR STRUC Array 1 NX RASTER INT 116 NY RASTER INT 60 RASTER FLOAT Array 116 60 ete Now it is time to do some actual calibration However before we proceed let s take a look at some of the effects we want to eliminate from our data At the moment the most substantial portion of our data is held in a raster pds cube This is a cube of raw CAM IMAGEs taken from all the SCDs of the first clean SSCD We will use x3d to take a look at this cube of IMAGEs see Figure 3 1 CIA x3d raster pds Cube 1 9 106 134 000 x3d is a tool for browsing through the frames of a cube in our case the frames are CAM IMAGEs One IMAGE is displayed at a time The slider bar to the right of the displayed IMAGE can be used to flick through the cube A plot above the IMAGE displays the value of a selected pixel selection can be done by right clicking on the IMAGE throughout the cube In effect this is the history of that pixel during the entire CONFIGURATION Upon calling x3d displays the center frame o
93. the directory products CIA fitsname dir cdrom dest products copy Alternatively you can create links each link will have the official name from your hard disk to the data products on the CD ROM In this case invoke fitsname without any keywords CIA fitsname dir cdrom dest products Note that if you are accessing a CD ROM jukebox then you may have to mount your CD ROM before using fitsname This is simply done by doing an 1s on the desired CD ROM eg CIA 15 jukebox P038 03802817 AOCS FIT CIER FIT CSTA FIT IIPH FIT CDER FIT CISP FIT CUFF FIT IRPH FIT Now you may execute fitsname as described previously 10 1 3 Manual copying To manually copy your data products to disk space follow the steps below 1 Copy the files from products pmmmmmmm nnnxxxyy to a convenient directory for example data Now copy the calibration products from products pmmmmmmm nnn xxxyy others to data others Note that these directories are only examples and no convention exists for the actual directory names 2 Look in the file datalist txt for the official names of the files Rename the files accordingly Mostly this involves appending nnnxxxyy to the root e g cier fit to cier14300601 fit Finally rename all the files so as to have the extension fits rather then fit 10 2 Examining the AAR Data Products This section shows you some basic things you can do with your AAR data in a CI
94. this stage the main component of the long term transient has been removed and it is safe to use the DivSky method However remember that it is better to first reconstruct an estimate of the 286 CHAPTER 21 USING SLICE WITHIN CIA 0 10F 0 5 0 200 400 600 800 0 100 200 300 400 Figure 21 5 The long term transient corrections derived by SLICE The continuous curves are the exact corrections and the dashed ones the fitted corrections assuming the long term transient effect is a combination of two exponentials On the left the LW3 case and on the right the LW2 case Some oscillation appear on the LW2 exact curve but these are not obviously related to the raster scan period sky with another method Therefore we apply here two make_map actions the first one with the Perturbed Single Flat Field method using the parameter values in Table 21 3 and the second with the DivSky method using the parameter values listed in Table 21 4 The results after the long term transient correction and the new flat field are displayed in Figure 21 6 A comparison with Fig 21 1 shows the improvement in the image quality Further improvement can be achieved by looking at deviant pixels glitch impacts and so on This is explained in the next section 21 6 Bad pixels ghosts and sources Even though we have removed much of the problems that affect the data some may still remain For instance ghosts following observations
95. to be at the top of an ISOCAM image This is of course an arbitrary decision but to reverse it all the signs must be changed Do not forget to specify the equinox keyword IDL works by default in B1950 0 while ISO only uses J2000 0 Optionally you can pass in the CD matrix if you already know it It is defined as cos crota2 sin crota2 sin crota2 2 The easiest way to get the CROTA2 CDELT CD parameters is to ask the roll to crota2 routine to give you them Its calling sequence is crota2 roll to crota2 roll channel pfov cdelt cd pfov is the pixel field of view For example CIA crota2 roll to crota2 scd get roll scd lw 3 3 cdelt cd 336 APPENDIX E ISOCAM ASTROMETRY ANGLES AND COORDINATES E 3 2 From CIA structures to FITS images The main problem is to know which angle to supply to the fits header routine It obviously depends on the type of data E 3 2 1 Individual SCDs and the like This works also for an SAD that would be the average of an SCD for a reduced beam switch or CVF The only angle that is valid in that case is the ROLL or a The obvious way to transfer the data is to avoid unnecessary rotations during the transfer thus the Y axis will be Axis1 and the Z axis will be Axis2 see Figures E 1 and E 5 The angle that needs to be given to fits header CROTAQ2 is then the position angle of the North axis with respect to the Z axis Thus 9 360
96. works best with the contiguous data that is provided in an SSCD We use the same dataset as in Chapter 3 1 Create the SSCD and SCDs CIA spdtoscd cisp02600506 fits sscd dir cia_vers test nowrite CIA sscd info sscd deg 43 SCDs in the SSCD CSSC026005060101_98052614571745 Seq channel mode fltrwhl pfov tint gain offset size ra dec O LW IDLE LW2 6 0 25 20 1 512 1 1 LW IDLE LW2 6 0 2 10 2 512 1 2 LW OBS LW2 3 0 5 04 1 512 46 180 498 18 849 3 LW OBS LW2 3 0 5 04 1 512 24 180 490 18 871 4 LW OBS LW2 3 0 5 04 1 512 24 180 482 18 893 5 LW OBS LW2 3 0 5 04 1 512 24 180 473 18 915 242 CHAPTER 20 ADVANCED DATA CALIBRATION 6 LW OBS LW2 3 0 5 04 1 512 24 180 450 18 908 LW OBS LW2 3 0 5 04 1 512 23 180 458 18 886 etc Perform dark correction and deglitching on the SSCD CIA gt corr_dark sscd method vilspa CIA gt deglitch sscd Stabilize the entire SSCD Remember that there are 4 CONFIGURATIONS in this ob servation By performing transient correction on the SSCD we give the Fouks Schubert fitting method see Section 20 2 3 as much contiguous data as possible hence increasing the quality of the correction As recommended for the Fouks Schubert method we need to create an initial stabilized image This is derived by a simple median reduction of the data in the first SCD that contains good data A quick look at the output of s
97. 0 Rise Transit Set only NO Skip daylight NO Solar elong cut off 0 180 CSV spreadsheet output NO Table quantities y Accept default output cr y n 7 Select table quantities lt gt 7 Output reference frame J2000 B1950 Time zone correction UT 00 00 Output UT time format JD CAL BOTH Output time digits MIN SEC FRACSEC Output R A format HMS DEG Output high precision RA DEC YES NO Output APPARENT Airless Refracted Set units for RANGE output KM AU Suppress RANGE RATE output YES NO Minimum elevation 90 lt elv lt 90 Maximum air mass 1 lt a lt 38 Print rise transit set only N T G R Skip printout during daylight Y N Solar elongation cut off 0 180 Spreadsheet CSV format Y N 50 n 1 j2000 jd fracsec deg 2450849 03610000 19 7242037 24 7936940 2450849 03679444 19 7241060 24 7928154 2450849 03748889 19 7240081 24 7919368 E0E The lines between SOE and contain the ISO centred ephemerids of comet Tempel Tuttle Put them into a file here named eph txt and read them into an IDL structure CIA restore verb cia vers data cds template eph xdr 4 RESTORE Portable XDR SAVE RESTORE file RESTORE Save file written by SOTT ISOW41 Fri Jun 22 17 18 39 2001 RESTORE IDL version 5 0 vms alpha 4 RESTORE Restored variable TEMPLATE EPH CIA eph read_ascii eph txt template TEMPLATE E
98. 0 12 Obtaining the best calibration record from a CDS A CAL G library file contains many calibration records Remember that the CAL G library files are stored as CDSs in CIA see Section 15 3 These records are comprised of images e g OFLTs DFLTs DARKs etc and associated CAM parameters e g TINT FLTRWHL etc Obviously you would wish to choose the best record for your data CIA provides the routines find best and find best psf to do just this In simple terms the algorithm employed by these routines is they look at the CAM parameters in each calibration record and try to find the best match with the parameters of the observation data However some extra complexity 20 12 OBTAINING THE BEST CALIBRATION RECORD FROM A CDS SELECTION INSIDE OUTSIDE UNDO DISPLAY BACKGROUND MEDIAN PROFILES TEST SAVE O40 O41 042 O43 Of 44 EI S30 26900270 35885 020 021 022 023 O24 O19 018 017 O16 9 Click on frame number to select unselect it O45 O46 O47 O48 34 033 O22 031 025 02 027 0O28 014 013 012 01 Hi H2 04 05 06 O7 O9 WINDOW BREE PO PO SOS 40 88 mean 11 0058 sigma 1 47186 min 2 84723 max 20 0435 N mean sigma min max 4 00000 57183 581491 887974 02570 noc I MERN method I frame s 0 1 sigma gt 0 05 I MEDIAN metho I frame s 0 1 M sigma gt 0 05
99. 0 2 10 2 512 1 4 LW OBS 10 1 5 2 10 2 512 14 295 462 50 518 5 LW OBS 10 1 5 2 10 2 512 47 295 462 50 518 6 LW OBS LW10 1 5 2 10 2 512 48 295 437 50 547 7 LW OBS LW10 1 5 2 10 2 512 48 295 462 50 517 8 LW OBS LWi0 1 5 2 10 2 512 47 295 417 50 501 9 LW OBS LWi0 1 5 2 10 2 512 48 295 463 50 518 10 LW OBS LW10 1 5 2 10 2 512 47 295 488 50 488 11 LW OBS 10 1 5 2 10 2 512 48 295 463 50 518 12 LW OBS 10 1 5 2 10 2 512 28 295 508 50 534 13 LW IDLE LW10 1 5 2 10 2 512 24 295 463 50 518 14 LW IDLE LW2 6 0 25 20 1 512 1 CIA cleaned sscd sscd_clean sscd Out of 15 SCDs 19 5 ADVANCED SLICING OF CVF DATA CAMO04 215 4 are rejected due to mode 5 are rejected due to csh flag 6 are rejected due to qla flag In total 8 are accepted CIA sscd info cleaned sscd deg 8 SCDs in the SSCD C88C058046100001 98092617532800 seq channel mode fltrwhl pfov tint gain offset size ra dec O LW 085 LWi0 1 5 2 10 2 512 47 295 462 50 518 1 LW OBS LWi0 1 5 2 10 2 512 48 295 437 50 547 2 LW OBS LWi0 1 5 2 10 2 512 48 295 462 50 517 3 LW OBS LWi0 1 5 2 10 2 512 47 295 417 50 501 4 LW OBS LWi0 1 5 2 10 2 512 48 295 463 50 518 5 LW OBS LWi0 1 5 2 10 2 512 47 295 488 50 488 6 LW OBS LWi0 1 5 2 10 2 512 48 295 463 50 518 7 LW OBS LWi0 1 5 2 10 2 512 28 295 508 50 534 CIA gt bs_pds get_sscdbs cleaned_sscd Calibration can proceed as normal from this point 19 5 Advanced slicing of CVF data
100. 0 30 0 40 0 20 40 60 80 100 CUBE Figure 22 2 Comparison of jitter computation methods Jitter offsets computed with the gauss method are represented by the solid line and jitter offsets computed with the psf method are represented by the dashed line The data used in this example are from an observation using the 1 5 PFOV and the LW7 filter scd This keyword is used to select all the IMAGEs corresponding to a particular CAM STATE ISO pointing from the PDS CUBE By default all STATEs are selected display Setting display 1 will display an averaged fit per CAM STATE Setting display 2 will display the each fit of each IMAGE verb Setting verb 1 will output the averaged fit parameters per CAM STATE Setting verb 2 will output the fit parameters per IMAGE method Select fitting method The default is method gauss bsize See description of algorithm above The default depends on the selected method col24 Set to remove column 24 actually column 23 in IDL convention from IMAGEs before fitting nterms Only applies to method gauss Select the number of terms used in the gaussian fit 22 1 2 Applying jitter offsets The application of the jitter offsets du dv is the most difficult part of jitter correction Currently two methods exist and neither are really effective These methods are implemented with reduce and project_cube pro reduce We can attempt to correct the jitter by shifting each I
101. 026005060002 02022715011006 CSSC026005060003_02022715011203 C88C026005060004 02022715011401 Let s concern ourselves with the first clean SSCD Using sscd info compare the charac teristics of its component SCDs with those of CONFIGURATION 1 listed in Table 3 1 They should be the same As you might expect the first SSCD returned by sscd clean contains all the data from the first CONFIGURATION CIA sscd info cleaned sscd 0 deg 8 SCDs in the SSCD C88C026005060001 02022715010900 seq channel mode fltrwhl pfov tint gain offset size ra dec O LW OBS LW2 3 0 5 04 1 512 46 180 498 18 849 1 LW OBS LW2 3 0 5 04 1 512 24 180 490 18 871 2 LW OBS LW2 3 0 5 04 1 512 24 180 482 18 893 3 LW OBS LW2 3 0 5 04 1 512 24 180 473 18 915 4 LW OBS LW2 3 0 5 04 1 512 24 180 450 18 908 5 LW OBS LW2 3 0 5 04 1 512 23 180 458 18 886 6 LW OBS LW2 3 0 5 04 1 512 24 180 467 18 864 LW OBS LW2 3 0 5 04 1 512 23 180 475 18 842 4 So we have split our data into four clean SSCDs one for each CONFIGURATION Con tinuing with the first clean SSCD we now need to gather all the data in its component SCDs into one regular IDL structure get_sscdraster will do this for us CIA gt raster_pds get_sscdraster cleaned_sscd 0 Note that raster_pds is a regular IDL structure in CIA we call this class of structure a prepared data structure or PDS and in particular a PDS holding raster observation data is called a raster PDS We can use IDL s HELP to view
102. 0601_96091713370691 In this case there are two future SADs with a MOSAIC in each and a third SAD containing the contents of the AAR data product CSSP which you can ignore Listing the origin SADs catalogued in our second SSAD we see that there are many CIA print ssad elem CSSA000014300601_96091713360439 CSAD000014300601_96091713360476 CSAD000014300601_96091713360860 CSAD000014300601_96091713362332 CSAD000014300601_96091713362951 CSAD000014300601_96091713363160 CSAD000014300601_96091713363810 CSAD000014300601_96091713364023 etc CSAD000014300601_96091713360767 CSAD000014300601_96091713360985 CSAD000014300601_96091713362446 CSAD000014300601_96091713363048 CSAD000014300601_96091713363716 CSAD000014300601_96091713363912 CSAD000014300601_96091713364107 194 CHAPTER 16 DATA STRUCTURE MANIPULATION Each of these origin SADs contains one EXPOSURE from the CCIM file and one from the CMAP file and of course we can expect to have many such EXPOSUREs in a typical AOT This explains why there are so many more origin SADsthan future SADs So we have seen that there are two MOSAICs but many EXPOSUREs Our example is taken from a raster observation where two CONFIGURATIONs of CAM were employed one was performed with the LW3 filter and the other with the LW6 filter Each of these CONFIGURATIONS yields data from which a MOSAIC may be created A SAD contains more than just image data it also contains CAM parameters during the
103. 1 Importing FITS to CIA data structures This section describes the CIA routines for importing ISO data products to CIA data structures Section 17 2 describes routines for importing ISO data products or any extended FITS files for that matter to regular IDL structures 17 1 1 Assigning working directories To avoid tedious typing it is useful to assign the path names of the directories you will use in the calls to the CIA routines to IDL string variables The directories used in the examples in this chapter are assigned as follows In VMS product dir DKA200 MDELANEY 14300601 Scd dir DKA200 MDELANEY 14300601 scds sad dir DKA200 MDELANEY 14300601 sads tdtosn 14300601 and in UNIX product dir home mdelaney 14300601 scd_dir home mdelaney 14300601 scds sad_dir home mdelaney 14300601 sads tdtosn 714300601 Also the variable tdtosn is assigned the combined TDT and OSN number 201 202 CHAPTER 17 IMPORTING ISO DATA PRODUCTS TO CIA 17 1 2 SADs from AAR aa2sad To convert AAR data products to SADs in memory or on disk we use the procedure aa2sad Note that this routine is a command line version of the high level widget program sad display see Section 10 2 Using the assigned variables of Section 17 1 1 an example call is CIA aa2sad tdtosn ssad origin ssad future scd dir scd dir CIA sad dir sad dir arc dat product dir ack ack
104. 12 2 for examples of usage and also Sections 17 1 3 and 17 1 4 They both use three methods of slicing the default method Andy and gap_mode described by the Table 19 6 5 6 The default is new_method The main difference between these automatic slicers and x slicer is that they only create one SSCD per observation You can use x slicer to create any number of SSCDs per observation though in general it is only desirable to have one per CONFIGURATION 19 6 6 The x handle slice window This window is really the heart of the program from the user s point of view Most of the important actions are chosen here 19 6 6 1 Using the Advanced Slicing Menu This menu makes the manipulation of data very easy There is only one point to remember it ONLY allows to PERFORM SELECTIONS You have therefore to press the Unselect All SCDs button before using it Then choose the desired parameter in the menu and click on it Use then the Redisplay to see the result of your choice With this last button you can select the SCDs according to an AND logical rule for example LW2 filter AND 5 0 s of integration AND PFOV of 8 arcsec AND number of frames greater than 15 19 6 6 2 Using choice related buttons A choice in x_slicer is the list of the selected SCDs just before you hit the Redisplay button Except for the First Choice that is the result of the slicer_a_moi routine that was first displayed to you 226 CHAPTER 19
105. 14 IMAGE ANALYSIS AND DISPLAY CIA xphot raster pds raster psf dir home cia isocia ia PSF Take a look at Figure 14 1 It contains the image we supplied when invoking xphot i e a raster MOSAIC and a number of buttons The button Select Objects is highlighted so the first action you must take is to select the sources that you wish to do photometry on Click on Select Objects to begin the selection process pop up window will appear with some information read and dismiss Now you can choose your sources by clicking on the image In the example image displayed in Figure 14 1 there is clearly only one source Click on done when you are finished making the selection Now click on Photometric Method to choose the photometric method to apply to the source In our example here we will select the first option simple aperture photometry The next step is to determine the radius over which to perform the photometry Click on Select Radius and choose the simplest method 1 for all A new window will appear displaying a region of the image centered on the source Click on this window to indicate the desired radius Click done when finished The final step is to determine the background Clicking on Select Background will give you a choice of methods the simplest being 1 for all Again we will choose this option for our example here Now select a region of the displayed image for determination of the background click on t
106. 14 16 Following from the example in Section 14 3 1 2 we can overlay the xdisp generated FITS file which contains a zoom of the LW6 CAM raster image of Haro 3 with contours from the optical image of Haro 3 The keyword putinfo adds information to the window the intensity scale contour level values etc CIA isocont ima33x33 fits haro3 fits putinfo e This example is similar to the previous one except the keywords radec and scan are set Respectively these display the image oriented in standard astronomical fashion i e orienting the DEC axis upwards and labelling it in degs min secs and orienting the RA axis leftwards and labelling it hrs mins secs and indicate the individual positions of each CAM state in the raster CAM raster data structure must be present in the memory of the CIA session and the FITS image must be accessible on disk 14 6 IMAGE COMPARISON AND OVERLAYING 147 ISOCAM en a z lt 3Q0 6 30 SECONDS CENTER 10 45 21 50 DEC 55 57 35 1 EQUINOX J2000 ISOCAM Interactive Analysis GESA CEA Saclay Figure 14 16 isocont is used to overlay an optical image with contours from a CAM image The CAM image has been aligned and rotated to match the astrometry of the optical 148 CHAPTER 14 IMAGE ANALYSIS AND DISPLAY CIA isocont lw6 raster haro3 fits radec scan e The final example overlays image from two CAM configurations bo
107. 15 3 2 CDS and CAL G files The different types of calibration data i e CAL G files and the CDS name for each is given in Table 15 3 2 An example of how the subfields of DATA would look for a CDS containing DARK images is given below The index 7 refers to a particular record within the DATA field where the range of iis 0 lt i lt NUMBER subfield DATA i SPARE DATA i TINT DATA i EWHL DATA i SWHL DATA i GAIN DATA i OFFSET DATA i TEMPERAT DATA i TRMS DATA i TMIN DATA i TMAX DATA i VOLTAGE DATA i VRMS DATA i VMIN DATA i VMAX DATA i BUNIT DATA i BLANK i 2 IMAGE j description of contents empty array 36 1 bytes integration time of image CAMTU entrance wheel step number selection wheel step number electronics gain electronics offset mean temperature K RMS temperature K minimum temperature K maximum temperature K mean bias voltage mV RMS voltage mV minimum voltage mV maximum voltage mV pixel unit of intensity value corresponding to an undefined pixel dark images 7 0 data j 1 error We can create a CDS from a CAL G using calg2cds see Chapter 17 for a description of CIA conversion routines For convenience assign directory paths where the CAL G data products may be found see Section C 2 2 to IDL string variables If you are working with VMS CIA then you could make the following assignment CIA gt calg_dark_
108. 19 6 11 4 x slicer crashes x slicer crashes just in the beginning You should probably reread the section about x slicer customization because you prob ably made a mistake in entering your preferred directory x slicer crashes while creating SCDs You probably asked it to build a SCD with no frames or with a hole in it by misusing the merge buttons x slicer crashes while saving SCDs Are you sure you have enough place on your disk Or even the right to write on it Chapter 20 Advanced data calibration In this chapter of the CIA User s Manual different calibration methods and algorithms are briefly reviewed and their implementation in CIA is described Detailed information on calibration algorithms can be found in the technical reports listed in Appendix K Probably the most helpful report for a novice is the ISOCAM Handbook This report contains a broad treatment of calibration algorithms 16 and the technical reports will be referenced throughout the chapter where appropriate Since more up to date algorithms and routines are continually being developed it is also a good idea to refer to the on line help or cia help see Section 2 3 2 20 1 Before reading this chapter Throughout this chapter references are made to the Prepared Data Structure PDS Sec tion 15 5 and the CIA calibration routines Chapter 13 Familiarity with both topics is as sumed You will also frequently encounter references such as IMAGE C
109. 2 LW2 LW2 LW2 OD 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 NONUNNNNNNNNNNNNNNNN gt RP RP RP RP 78 78 78 78 74 TT 78 78 78 78 78 78 78 70 77 PRP RP RP RP RP RP RP RP RP RP RP RR 53 We need to extract and concatenate the correct SCDs in order to make a raster PDS containing data acquired with the same entrance wheel parameter First we clean the SSCD and eliminate all SCDs that have obviously too few IMAGES to be valid In this example cleaned_sscd splits the SSCD into two segments an unnecessary side effect due to limitations in the design of the SSCD and the CAM05 observing mode We can however easily recombine the SSCD segments CIA cleaned sscd sscd_clean sscd minimum 20 CIA gt sscd_del sscd CIA sscd_concatene cleaned sscd 0 cleaned_sscd 1 2 The next step is to create a new SSCDs containing SCDs from a single entrance wheel parameter setting This we can do with scds select CIA sscd_polari scds select cleaned sscd 0 polari entwhl POLARIZOR 1 CIA sscd info sscd polari scds pol seq 12 SCDs in the 5950 entwhl mode fltrwhl POLARIZOR POLARIZOR POLARIZOR POLARIZOR POLARIZOR PO
110. 20 These guidelines are based on the joint experiences of all support astronomers of Centre ISO francais du CEA Saclay and they are very likely to evolve as new methods are available A general help on x cia is presented in Section 23 2 13 3 2 Quick Look analysis with x cia It is strongly advised to read Chapter 8 of the ISOCAM Handbook in order to inform oneself about the calibration of ISOCAM images The final calibrated images are indeed the result of many attempts Nevertheless it is very useful to get a rough idea of what ISOCAM has detected or to check that saturation has not occurred In that sense the following commands can be considered as the standard quick look analysis 1 Start a CIA session 2 Type the following on the CIA command line CIA gt x_cia Taken from Claret A 1996 ISOCAM Data Analysis with _ v2 2 Sections 1 3 13 83 DATA CALIBRATION WITH X_CIA 103 3 Choose AOT Type Default is raster scan 4 Data Load SSCD 5 Data AOT Info 6 Process Default 7 Use all available functions of View and Tools menu to explore calibrated data 8 Tools Hardcopy to create hardcopies 9 Data Save IDL File If result is worth saving 10 CIA Quit 13 3 3 Guidelines for using x cia 13 3 3 1 Calling sequence Full calling sequence of x cia is X cia data my data histo my histo indark my dark inflat my flat logfile my_logfile noinit noinit where data is a PDS al
111. 4 4 Miscellaneous auxiliary calibration data This section describes some miscellaneous auxiliary data files that are distributed with CIA All of this data may be found in the directory cia_vers data cds CVF position data is stored in the files conv_filter_lw_table xdr conv_filter_sw_table xdr ORBIT data are stored in the files 15 4 AUXILIARY CALIBRATION DATA 179 orbit dat orbit fits orbit hdr orbit unix xdr orbit vms xdr The file orbit fits is the ISO ORBIT data product It contains the orbital parameters of ISO The data in this file is also contained in orbit dat and its FITS headers are contained in orbit hdr The data is also contained in the IDL savesets orbit_unix xdr and orbit_vms xdr These files are for the UNIX and VMS version of CIA respectively Dark model see Section 20 2 1 auxiliary data is stored in the files cam_reset_level xdr activation_utk save mod19980320 xdr coeff_shortterm_drift_tint xdr Note that the dark model also uses data in the ORBIT files listed above xphot see Section 14 1 4 needs the aperture correction data stored in the file xphot_aper save xradial see Section 14 1 2 needs the radial coefficients stored in the file xcoef inc standardpsf It contains the variable standardpsf a Fourier transformed PSF model for ISOCAM s field of view with a good spatial sampling rate This PSF model can for example be used to compute quickly the correction factor of an aperture photo
112. 4 s that have passed since the 1 November 1995 at 0100 Its value is updated every image in the TDF F2 format e ITK stands for Instrument Time Key its value is the number of CAMTU or CAM Time Unit roughly 0 14 s that have passed since the 1 November 1995 at Oh00 Its value is updated every image in the TDF F2 format e FMT stands for TDF FORMAT Its value is the number of TDF formats one every 2 seconds sent since the 16 November 1995 18h36 It is of course updated every TDF format See the ISO Data Product Document Appendix B for additional information on these keys 19 6 4 3 Slicing TDF Taken all that has been said in the last two subsections into account slicing TDF files is just the same as slicing ERD files Remember that you cannot use any coordinate information like IIPH or ORBIT It is also impossible to check the contents of a TDF versus the commanded position 19 6 5 Selecting slicing variables The problem of selecting slicing variables has been widely addressed in Section 12 3 but since it is one of the most important things that the user has to do when using x_slicer we wish to recall some points here First of all Table 19 6 5 gives the complete list of the slicing variables that can be used with x slicer In order to avoid confusion the official names of variables in the telemetry files in SCDs and DSD are also given Let us now discuss some of these variables x slicer name telemetry name CAM wheels
113. 5444142 Now take a look at the history of scd CIA print scd get history scd date 05 Apr 1997 15 44 37 node IS0W40 user LMETCALF procedure x slicer V II 1 algorithm user CIER FIT 1 ISOW40 DKC100 IA LEO WORK MKN297 IIPH FIT 1 ISOW40 DKC100 IA LEO WORK MKN297 NO COMPACT STATUS HISTORY NO ORBIT FILE starting record 1850 ending record 1963 cia otf END date 05 Apr 1997 15 44 38 node ISOW40 user LMETCALF procedure erd2spd V 3 1 algorithm default CSCDMKN297L6P100_97040515443475 END The output we get in the above example is in a user friendly format Within CIA structures the history is stored in a coded fashion in the field PROCESS In the example above the PRO CESS was automatically converted into a readable history In the following example we take a look at the actual contents of PROCESS CIA process scd get process scd Have a look at process CIA print process 1 260639077 ISOW40 LMETCALF x slicer V II 1 user 0 31 1 260639079 ISOW40 LMETCALF erd2spd V 3 1 default 0 53 As you can see from the above it is not in a very user friendly format To make it into a readable history we can use process2history CIA history process2history process CIA print history etc In fact it is this more readable history which can be found in the field HISTORY in the PDS For example CIA print raster pds history date 28 8ep 1998 14 50 33 node bermuda user mdelaney procedure darklibrary
114. 56040039 96091919440421 Refer to Section 15 5 3 for guide to the CVF PDS architecture Again before proceeding it is recommended to save the newly created CVF PDS CIA save filename cvf_pds dat cvf pds xdr 100 CHAPTER 13 DATA CALIBRATION 13 2 Calibrating the PDS So you have your PDS created and want to calibrate it There is a core set of CIA calibration routines that perform corrections which are required for all CAM data regardless of AOT e g dark correction deglitching stabilization flat fielding These routines are described in the Section 13 2 1 Further sections describe routines that perform AOT dedicated processing e g raster MOSAIC creation So to do a complete calibration of your data first read Section 13 2 1 followed by one of Sections 13 2 2 13 2 4 or 13 2 5 depending on the flavour of your PDS 13 2 1 Core calibration The first three core calibration routines that we will meet are corr dark deglitch and stabi lize These routines perform the following core corrections i dark correction ii deglitching iii stabilization All have a choice of different methods though for simplicity we will just use the default here Though we use a raster PDS in our example below these routines accept any flavour of PDS as input 1 Following from Section 13 1 2 we will restore our previously saved raster PDS and perform the first calibration steps on the IMAGEs in Iw6_raster cube CIA restore lw6_ra
115. 6010001 96082811465989 As expected the only SSCD in memory is erd_sscd Now take a look at what SCDs are in memory CIA print scd list CSCD143006010001_96082811470062 CSCD143006010002 96082811473689 etc CIA gt print sscd_elem erd_sscd CSCD143006010001_96082811470062 CSCD143006010002_96082811473689 etc Again as expected the only SCDs in memory are those that correspond to erd sscd For convenience assign the names of the SCDs in memory to an IDL variable CIA scds sscd elem sscd 4 Remember that erd scds is an array of names of ERD SCDs i e SCDs which contain EOI and RESET frames CIA help scd get eoi erd_scds 0 Expression INT Array 32 32 This means that we must convert them to SPD SCDs using erd2spd There are two ways of doing this the first uses the routine frames to image and the second erd2psd a Using frames to image is probably the simplest way to convert ERD SCDs to SPD SCDs CIA spd sscd frames to image erd sscd ack ack Since a new SPD SSCD spd sscd is created we can delete the old ERD SSCD and its SCDs CIA sscd del sscd b Find the number of ERD SCDs that we have created 80 CHAPTER 12 DATA SLICING CIA nscd sscd get nscd erd sscd CIA print nscd 41 Using a simple IDL loop we can convert all the ERD SCDs to SPD SCDs in one line CIA spd_scds strarr nscd CIA for i 0 nscd 1 d
116. 8355 CIA gt Cube 31 0 4 13 7431 CIA gt stat raster_pds image 4 31 0 28 Image dimensions 28 29 Number of frames 9 Total number of pixels 7308 Minimum Maximum Mean Median RMS 11 7170 15 5912 14 0292 14 0416 0 248977 CIA gt x where raster_pds image 1t 14 0292 6 0 248977 CIA raster pds npix x 0 CIA raster scan raster pds 20 8 CALIBRATING AN SSCD 241 Figure 20 1 Comparison of standard vs improved processing of observations using the small Fabry mirror Left MOSAIC of standard processing Right MOSAIC of improved processing The uneven appearance of the MOSAIC is gone 20 3 Calibrating an SSCD In most cases users are only interested in calibrating a PDS and the SSCD is quickly discarded However there are certain situations where the calibration of an SSCD is very more useful Mostly these cases require that contiguous data over as a long a time scale as possible is avail able for input to the calibration routines Remember that a PDS does not necessarily hold contiguous data see Section 13 1 1 for details of the limitations of a PDS In particular data from polarization observations are best calibrated using SSCD see Chapter 8 for an example of this Here we have an example of how and why you would calibrate an SSCD This example illustrates the usage of the VilSpa dark correction and the Fouks Schubert transient correction method see Section 20 2 3 This transient correction method
117. 9 13 CIA print bs pds ref image 3 7 11 15 You can use x3d to look at EXPOSUREs CIA x3d bs pds image Use the slider to select the desired EXPOSURE e g for the second source EXPOSURE set the slider to 5 9 Finally we can create the beam switch MOSAIC This is simply done CIA reduce bs bs pds reduce bs adds up all the source EXPOSUREs and subtracts the reference EXPOSUREs taking the MASK into account of course To display this MOSAIC CIA tviso bs pds raster You should see the same image as in Figure 6 2 Note that for historical reasons the beam switch MOSAIC is placed in the field RASTER 10 You may wish to save the results of the data analysis You can do this with IDL s SAVE 44 CHAPTER 6 BEAM SWITCH OBSERVATION ap QOeAmlss dl 165059 Frame Number LOAD LUT lt Horizontal cut Quit G Vertical cut Window Size so Mask J Glitch 200M RANGE T SCALE gt Next frame Previous frame Pixel value Eube 8 14 125 185 4 zi Figure 6 1 x3d display of a calibrated BS PDS CUBE The button mask has been activated Undefined image pixels are blank In the plot window crosses mark where pixels are undefined in the their time history 6 2 DATA ANALYSIS 45 169 312 0 194057 2 Figure 6 2 tviso display of beam switch MOSAIC Undefined pixels in column 24 are blank 46 CHAPTER 6 BEA
118. A data structure cataloging SCDs from an AOT or CON FIGURATION staring observation A particular CONFIGURATION of CAM 1 A staring obser vation is simply one pointing of CAM STATE The finest time division of CAM activities and a subdivision of an OP MODE All CAM parameters are fixed during a STATE but parameters may change between STATEs This depends on the observation type e beam switch AOT 3 raster and micro scan AOT 1 CAM pointing changes with S TATE e CVF AOT 4 CVF position changes with STATE e polarization AOT 7 5 the entrance wheel changes with STATE TDF Telemetry Distribution File or FORMAT The data stream down linked from ISO 311 TDT sequence number Target Dedicated Time The contiguous time spent on the obser vation of a source in an ISO revolution including integration time and overhead times An AOT is a subset of a TDT The TDT sequence number identifies a TDT within a revolution lt TDT OSN gt Combination of TDT number and OSN number Uniquely identifies an AOT tracking observation A particular CONFIGURATION of CAM 1 A tracking ob servation is used for observing Solar System objects CAM pointing changes to track the target object 312 APPENDIX A GLOSSARY Appendix B CIA command short list This quick reference guide lists core CIA commands and groups them by processing level User s may find it useful to make a copy of this chapter and keep it near to hand
119. A example of usage and a very brief description is given for each command Only the more important arguments in the calling sequence are used for an exhaustive list see the header documentation in the on line help Section 2 3 2 Examples of usage for almost all the commands here appear in the Quick Start Guide or elsewhere in the CIA User s Manual B 1 Data preparation slicing This section lists routines that are used for generating an SPD SSCD per CAM CONFIGURA TION Further routines are listed for creating the appropriately flavoured PDS from the SPD SSCD For a thorough treatment of slicing go to Chapter 12 spdtoscd creates an SPD SSCD per observation from an SPD FITS product This is probably the simplest and fastest way to create an SPD SSCD CIA spdtoscd cisp02600506 fits spd sscd dir nowrite x slicer is a widget program that can create an SPD SSCD per CAM CONFIGURATION from either an ERD or SPD FITS product Very flexible and interactive approach to slicing CIA x slicer erdtoscd creates an ERD SSCD per observation from an ERD FITS product CIA erdtoscd cier02600506 fits erd sscd dir nowrite frames to image converts an ERD SSCD to an SPD SSCD CIA spd_sscd frames to image erd sscd sscd info lists all the SCDs of an SSCD CIA sscd info spd sscd 313 314 APPENDIX B CIA COMMAND SHORT LIST sscd clean breaks an SPD SSCD into one SPD SSCD per CAM CONFIGURATION CIA
120. A session The most useful CIA routine for displaying the AAR products is sad display Simply type CIA sad display windows 1 to invoke it A pop up window will appear requesting a CCIM file After some processing time during which SADs see Section 9 4 2 are created in memory from the CMAP CCIM and CMOS files a widget will appear with an image window and a control panel see Figure 10 1 In fact will work with any FITS file that contains the key FILENAME in its primary header 5fitsname uses a unix find program to compile lists of files you may not be allowed to execute find on your jukebox Note that UNIX notation is used in the above instructions In VMS products pmmmmmmm nnnxxxyy would be something like DKA600 000000 PRODUCTS PMMMMMMM NNNXXXY Y 5Found on the ISO CD ROM in the directory products pmmmmmmm nnnxxxyy See Section 9 3 3 70 CHAPTER 10 FIRST LOOK AT THE DATA CCIM Data Image Zeem Glitches Sources _ Grid Data Image Zem Siateshes D Sources B Grid CSBD000055100433 97100617154663 MKN309 Ch Mode Wave PFOV EWhl LW OBS de Hole LWSM 2 1 SWhl Tint 5 Previous SAD Next Future Colour 550000055100433 97100617155526 BMCBREEN ngc999 Ch Mode Wave PFOV EWhl LW OBS 7 8 6 0 Hole Figure 10 1 sad display windows The upper window displays an EXPOSURE from a
121. ACE ROUTINES 189 16 1 1 4 sad init This function initialises an SAD in memory An example following from the Section 16 1 1 3 is CIA sadi sad init 143006010101 ssid ssadi ack ok CIA help sadi ok SAD1 STRING CS5AD143006010101_96080513124826 OK INT 1 The first argument is the combined lt TDT OSN CN STATE gt number that is used to name the returned SSAD ssad1 that is created by ssad_init see Section 15 2 1 The keyword ssid specifies the name of the SSAD we have initialised in Section 16 1 1 3 and to which 1 now belongs A keyword argument not used here is source which can be used to pass an already existing SAD to sad_init so to copy its parameters to sad1 16 1 1 5 cds_init This function initialises a CDS in memory An example of its usage follows CIA gt cdsi cds_init ccgswdark ack ack CIA help cdsi ack EXPRESSION STRING CCD8123456789012 95012318401579 ACK INT 1 The first argument is the type of the returned CDS 1 The possible types of CDSs are given in Table 15 3 2 16 1 2 structure extract This function simply extracts data from a CIA structure and places it in a regular IDL structure A simple example for an SCD is CIA regular struct scd extract CSCD143006010110 97092611592803 Now we can use IDL s HELP to look at the guts of regular struct CIA help regular struct str etc 16 1 3 structure put Procedure to place a val
122. ADU gain sec assuming you have not already manually changed the units Likewise for a CVF PDS and general PDS you will need to add IMAGE UNIT This can be done with IDL s CREATE STRUCT CIA new pds create struct temporary old pds raster unit CIA ADU gain sec Use IDL s TEMPORARY to prevent making a duplicate copy of the structure while ap pending the new tag Upgrading CIA 1 0 structures There is a small bug in the ASTR substructure of the CIA 1 0 CVF PDS the type of the field ASTR CDELT is float instead of double This problem will only manifest itself when you attempt to load a CIA 1 0 CVF PDS when you have a definition of the ASTR structure already in memory eg ASTR in a CIA 3 0 raster PDS CIA restore old cvf xdr RESTORE Unable to restore structure tag type disagrees with existing definition ASTR STRUC Temporary variables are still checked out cleaning up Though small this problem is awkward to solve Open a second CIA session in parallel with your first In this new session load the CVF PDS as attempted above 347 348 APPENDIX I UPGRADING OLD CIA STRUCTURES CIA restore old cvf xdr verb 4 RESTORE Portable XDR SAVE RESTORE file RESTORE Save file written by LANDRIUGSAPIO1 Thu Jun 12 17 11 05 1997 RESTORE Restored variable OLD CVF CIA help old cvf astr str Structure ASTR STRUC 8 tags length 96 CD DOUBLE Array 2 2 CDELT FLOAT Array 2
123. AP data product of the same CONFIGURATION CONFIGURATION A subdivision of an AOT A CONFIGURATION is comprised of a set of chronologically ordered OP MODEs associated with the same mandatory OP MODE OBS where CAM is configured to observe a celestial source See also STATE CPSL CAM Point Source List n AAR level data product containing a catalogue of AA detected point sources CSSP CAM Source SPectrum An AAR level data product containing the measured spectrum for each point source detection CUFF CAM User friendly File An AAR level data product containing a log of messages from AA processing CUS Calibration Uplink System It provides a more flexible interface to command the instru ment and allows more freedom in the setting of instrument and observation parameters CVF Circular Variable Filter CVF observation Also AOT 4 This is a spectral observation CAM pointing remains con stant but the CVF wavelength changes CVF data structure See PDS DARK In the context of calibration data it refers to a dark current image It is also an OP MODE when CAM is performing a DARK measurement data preparation Also know as slicing See entry for slicing data product Refers to the FITS files on your CD ROM data product version number Identifies the version of a data product data structure A programming term referring to a record of data organised in to a single addressable entity The structure may contain data of mixed types b
124. APTER 2 ABOUT CIA This will probably be a link You can use the 1s 1 command to see what it is pointing to if you wish though it is not particularly important Inspect this file to see what cia vers is defined as Alternatively this information will probably be printed when CIA is started Now you can copy the file cia vers unix user init pro In order for CIA to find it you should put it in your HOME bin directory To change the default version you can either copy and modify the cia file or alias cia to include the version though remember you can also specify it on the command line By default no log files are written You can turn on the logging by reinstating the commented out line set logfile in user init pro Then they are written to the directory specified by the environment variable logfile_dir which in turn is set to the environment vari able cia logfile dir These variables are set in respectively cia and setup cshrc camia if it exists If it does exist and sets cia logfile dir you will need your own version of cia to change the logfile directory otherwise you can simply define cia logfile dir yourself 2 4 CIA caveats Some points to note about IDL and CIA e Do to a bug concerning nesting structure in IDL 5 5 CIA will not run under IDL 5 5 e You have to include the obsolete directory in the IDL_PATH in order to run pickfile and CIA routines using this program CIA should only be run in a directory where the user ha
125. Advanced Use of CIA 209 Introduction This part of the CIA User s Manual is intended as a guide to advanced use of CIA It is hoped that it will help you to get the most out of your CAM data e Chapter 19 is an advanced guide to data slicing It follows from Chapter 12 e Chapter 20 takes a closer look at the core calibration routines and at AOT dedicated processing It also gives guidelines on how to determine the best calibration for your data It follows from Chapter 13 e Chapter 21 serves as an introduction to the SLICE package and its long term transient correction or LTT and variable flat field correction or VFF algorithms e Chapter 22 describes second order corrections for CAM data e Chapter 23 is an advanced guide to x cia it follows from Section 13 3 211 212 Chapter 19 Advanced slicing This chapter describes and gives examples of advanced slicing 19 1 Advanced slicing options This section describes advanced slicing keyword options that are available to the automatic slicers spdtoscd and erdtoscd corrected Since OLP version 6 1 improved of values of RA DEC and ROLL namely CRA CDEC and CROLL are provided in the data products refer to the ISOCAM Handbook for caveats on their use These improved values are obtained by applying a correction for errors in focal lengths sensor alignments and sun position Set the keyword corrected to access these corrected values new otf Setting this keyword
126. CDSs from CAL G file calg2cds 0 0 17 2 Importing FITS to regular IDL data structures 17 2 1 Reading an ISO data product 17 2 2 Extracting key from ISO data product 18 Export of CIA data structures 18 1 Export to the spectral analysis package 18 2 Export to external packages 18 3 Export for archiving Advanced Use of 19 Advanced slicing 19 1 Advanced slicing options 19 2 Saturation warnings during slicing 184 185 185 187 187 188 189 189 190 190 190 190 191 191 192 192 193 194 195 195 196 197 197 198 199 201 201 201 202 202 202 202 203 203 204 205 205 205 206 209 X CONTENTS 19 3 A beam switch observation 214 19 4 Advanced slicing of beam switch data CAM03 214 19 4 1 Concatenating intermediate SCDs in a beam switch observation 214 19 5 Advanced slicing of CVF data CAM04 215 19 5 1 Up and down CVF observation 215 19 5 2 Mixed LW and SW CVF observation 216 19 6 Advanced slicing with x slicer 218 19 6 1 Files directories and x slicer customization 218 19 62 Shcimg Sore ii wg Be eek Oy Oe Rl be ae ed eet 219 19 6 3 Slicing ERD f les less 220 19 624 S
127. CRPIX FLOAT Array 2 CRVAL DOUBLE Array 2 CTYPE STRING Array 2 LONGPOLE FLOAT 180 000 PROJP1 FLOAT 1 00000 PROJP2 FLOAT 2 00000 This time there are no problems no other ASTR definition exist in the new CIA session Now export old_cvf as a FITS file and exit your second CIA session CIA gt struct2fits old_cvf name cvf fits CIA gt exit 101 DELANEY Returning to your original CIA session you can now load the FITS file into IDL memory as a CVF PDS fits2struct will initialize new CVF PDS fill it with data from the old CVF PDS and in so doing correct the type of ASTR CDELT CIA fits2struct cvf fits hdr new_cvf CIA gt help new_cvf astr str Structure ASTR STRUC 8 tags length 104 CD DOUBLE Array 2 2 CDELT DOUBLE Array 2 CRPIX FLOAT Array 2 CRVAL DOUBLE Array 2 CTYPE STRING Array 2 LONGPOLE FLOAT 180 000 PROJP1 FLOAT 1 00000 PROJP2 FLOAT 2 00000 e Due to an error in the original FITS astrometry definition which was propagated into the ASTROLIB astrometry routines and the CIA routines roll_to_crota2 and fits_header CIA 1 0 CVF ASTR CD was defined incorrectly and the CD matrix in FITS files created ICIA is distributed with debugged versions of the ASTROLIB astrometry routines extast putast ad2xy xy2ad The location of the CIA version of these routines is placed before the ASTROLIB location on IDL s IPATH if CIA is installed correctly to ensure that they are compile
128. CRVAL1 Dec CRVAL2 Angle CROTA2 ResolX CDELT1 ResolY CDELT2 equinox my equinox Some explanations are required for those unfamiliar with FITS To fully grasp them please consult also Figure E 5 NAXISn are the number of pixels on the 2 axes 1 is the horizontal rightward axis and Axis2 is the vertical upward axis In order to perform some astrometry one need to know the coordinates o of a reference pixel in the image the plate scale and an orientation angle CRPIXn are the pixel coordinates of that reference pixel usually the central one while CRVA Ln are its a 9 coordinates in degrees CROTA2 it the position angle of the celestial North axis with respect to the Axis 2 counted positively eastward i e it is different from the definitions of angles used in CIA where the North axis was the reference Figure E 5 clearly shows that difference it is the angle called 0 An incorrect value in CROTA2 is the first error possibility when you find that fits header does not work CDELTn are the plate scales on both axes In the FITS header they will appear in degrees per pixel which are the units you must give to fits header Getting the units wrong is a second source of error when the routine does not work The CDELT parameters must be given with the correct signs These are LW SW CDELT1 gt 0 CDELT1 lt 0 CDELT2 lt 0 CDELT2 lt 0 The reason that CDELT 2 is negative is that pixel 0 0 is defined
129. CVE 49 10 1 sad display windOWs is foedo 9 fo e 3 4 70 11 1 Overview of CIA 74 12 1 x slicer windows nk ER um 8T 12 2 x handle slice window 91 xGcla WIN AEP penis ah d cR Gu Abo ei eir BOY geom bod E hes 104 14 1 The xphot window 115 14 2 cv display window l1 ek RESOSY er Pe Pa UE EEG aS ge 118 Phe xevi window o2 fe Moher Cu Id ec ee ee es EG Pug 120 14 4 The xdisp windows 9 auus bie odo eu Rem Ree ie P Dan 123 14 5 xselect_frame window 125 14 6 The whole xcube 127 14 7 The Plot Window 4 4E R9 Ge See ee ee EOM RU RO S 129 14 8 The Frame Window 130 T4 9 x3d windows cede Siar owt og ene os qe XE E Eis 133 T4 window woe esi ie Ghose ebd Edge 135 14 11The main window of ximage 139 14 12The raster window in ximage a downward transient 141 XV xvi LIST OF FIGURES 14 13 The raster window in ximage the tail of glitch llle 141 I4 4xv raster window Sioa eis ere Sigg Se Ee ee BPs 143 14 1 5show frame windOW ix p dd iege doe Bee gee Ane ERR Ee ECTS 145 14 16isocont is used to overla
130. ERD data in the CDER file 5CAM parallel SPD data are delivered in CPSP 9 3 RELATING DATA PRODUCT TYPES TO FILENAMES 63 9 3 4 Auxiliary data products Additional information such as pointing and spacecraft position and velocity are also pro vided on the CD ISO CD ROM users can find IRPH and IIPH in the directory products pmmmmmmmy nnnzrxyy and ORBIT in products pmmmmmmm others IIPH Instrument Instantaneous Pointing History Contains instantaneous pointing information for CAM during your AOT Note that this is an essential file for calibrating your data Users have experienced difficulty with very old IIPH files i e generated by OLP prior to version 4 0 see Section E 2 for details The IIPH equivalent for CAM parallel mode is CIPH IRPH Instrument Reference Pointing History Contains reference pointing information for CAM during an AOT It contains similar though less comprehensive information as the IIPH CIA uses the IRPH when analysing CAM parallel mode data The IRPH equivalent for CAM parallel mode is CRPH ORBIT Contains information on orbital parameters for all ISO revolutions up to and at the very least including the revolution during which your AOT is performed Strictly speaking this is a CAL G file but to follow the organisation of the ISO Data Product Document it is placed in this section 9 3 5 Calibration Data Products Calibration data are available in two formats CIA s Calibration Data Structures CDS a
131. ESTIONS AND PROBLEMS 291 10 0000 Figure 21 1 The raster maps using a standard CIA procedure see text for details Left panel shows the LW3 data while the right panel shows the LW2 data Both data sets are affected by periodic patterns due to bad flat field determination as well as long term transients 35 0000 10 0000 25 0000 8 00000 Figure 21 6 The results of the long term transient correction and variable flat field determi nation Variable flat field was performed using the DivSky method with parameter setup as indicated in Table 21 4 LW3 is on the left and LW2 on the right Compare with Fig 21 1 to measure the improvement 292 CHAPTER 21 USING SLICE WITHIN CIA Chapter 22 Second order corrections This chapter describes second order calibration routines Depending on your data such cali bration maybe entirely unnecessary All second order correction routines should be used with care 22 1 Jitter correction Spacecraft jitter can cause a slight displacement of a source from IMAGE to IMAGE This displacement is limited to about 0 5 Kessler et al 1996 Figure 22 1 shows the distribution of jitter offsets for one particular observation Jitter is most noticeable when a point source is observed You can check for jitter in your data by using x3d to flick through the IMAGEs in CUBE while watching for sub pixel shifting of a source Jitter correction is performed in two steps Firstly the jitter o
132. Estimating the total source flux 114 14 1 4 Photometry measurements with xphot 114 14 1 5 Other methods for photometry 116 T42 CVF imag analysis BO ae e des 117 142 1 cvf display ecas Ere Ros Aem Ge e De B 117 1422 3 Rp REO RR E Sow Ru bU RAE ERR 118 14 3 2 D image analysis ee 122 14 3 T zxdisSp i uis nre Sete RR EE ACC ERE Osea e Yo 122 14 8 2 sad display and struct2sad 124 I44 Cubeanalysis i i go Re ug a ee EG 125 14 4 1 Extracting images from cubes with xselect frame 125 14 4 2 Extracting images from cubes with xsubcube 125 14 4 DOE c Le FOU ee Moa he Gh ae 126 14 44 Frame Window o det ADR ee e ROW pO HG oe 129 14 4 5 Cube analysis with x3d 132 viii 14 5 14 6 14 7 14 8 14 4 6 x3d calibration aid IAAT xv tenp fe oc sd ei oe ome Pre mene pP eu ed 144 8 ximage eR UA 14 4 9 Raster visualization 14 410 Anexample 220000 14 4 11xv_raster 14 4 12 Cube animation with xmovie Simple image I4 5 L EVIS esr et BE ee ae ees 14 5 2 Cube display with show_frame Image comp
133. H et al 1998 Cosmic Ray effects on the ISOCAM LW detector Experimental Aston omy vol 10 page 279 290 Landsman W B 1995 The IDL Astronomy User s Library in Astronomical Data Analysis Software and Systems IV ed R A Shaw H E Payne J J E Hayes ASP Conference Series Vol 77 p 437 Lari C et al 2001 A new method for ISOCAM data reduction I Application to the European Large Area ISO Survey Southern Field method and results MNRAS vol 325 page 1173 1189 Miville Deschenes M A et al Optimizing ISOCAM data processing using spatial redundany A amp AS vol 146 page 519 530 NOST 1983 Definition of the Flexible Image Transport System NASA Science Office of Standards and Technology NOST standard 100 1 0 NASA Goddard Space Flight Center 355 356 BIBLIOGRAPHY Okumura K 1998 ISOCAM PSF Report Technical report ESA 1998 http www iso vilspa esa es users expl_lib CAM Okumura et al 1998 Ghosts in ISOCAM images Technical report ESA 1998 http www iso vilspa esa es users expl_lib CAM Ott S et al 1997 Design and Implementation of CIA the ISOCAM Interactive Analysis System ASP Conference Series Vol 125 Ott S et al 1998 Data Analysis with ISOCAM Interactive Analysis System preparing for the Future ASP Conference Series Vol 145 Pantin E and Starck J L 1996 Deconvolution of Astronomical Images using the Multireso lution Maximum Entropy Method Astron Astrop
134. IMAGEs in CUBE are divided by the FLAT Finally places the FLAT in the PDS field FLAT For all the different methods of flat fielding the impact on the PDS is the same The field FLAT is filled with the FLAT used for flat fielding Either the reduced EXPOSUREs are flat fielded IMAGE is modified as is the default or if the keyword cube is set then the IMAGEs are flat fielded CUBE is modified 1 2 3 4 method library or method calg method CAL G or library FLAT correction In a similar manner as for the CAL G DARK get sscdstruct uses find best see Section 20 12 to find the most suitable DFLT and OFLT images from the CIA CDSs and places the product of these images i e the FLAT in the PDS field CALG FLAT When the the keyword method is set to method library corr flat uses the FLAT in CALG FLAT If the corr_flat keyword cube is set then the flat fielding is performed on the IM AGEs otherwise it is performed on the EXPOSUREs called routine flat library method oflat method CAL G or library OFLAT correction This is exactly the same method as for method library except that only an optical flat field correction is performed called routine flat library method oflat method CAL G or library DFLAT correction This is exactly the same method as for method library except that only a detector flat field correction is performed called routine flat library method auto method Autom
135. ISOCAM Interactive Analysis User s Manual Version 5 0 Document Reference Number A1I 96 5226 Dc Edited by Matt Delaney Stockholm University amp Stephan Ott ISO Data Centre with contributions from ISOCAM Interactive Analysis Team ESA CEA IAS IPAC amp RAL March 1 2002 i Acknowledgements Contributions to the CIA User s Manual were made by all members of the ISOCAM Interactive Analysis Team and in particular members of the ISOCAM Instrument Dedicated Team and the ISO Data Centre Villafranca del Castillo and the Service d Astrophysique Saclay Some sections of this document are extracted from internal documentation acknowledgments can be found in footnotes to these sections Alain Abergel IAS Babar Ali IPAC UR Bruno Altieri ESA Jean Louis Augu res CEA Herve Aussel CEA Jean Philippe Bernard IAS Andrea Biviano OAT ESA Joris Blommaert ESA Olivier Boulade CEA Francois Boulanger IAS Mark Calabretta ATNF WCS library Catherine Cesarsky CEA ESO ISOCAM PI Diego Cesarsky IAS MPIA Pierre Chanial CEA Vassilis Charmandaris CEA Ranga Ram Chary UCLA Arnaud Claret CEA Alain Coulais IAS Matt Delaney ESA UCD SO Christophe Delattre CEA Thomas Deschamps CEA Francois Xavier D sert IAS Pierre Didelon CEA David Elbaz CEA Ya l Fuchs CEA Pascal Gallais ESA CEA Ken Ganga IPAC Ren Gastaud CEA Steve Guest ESA RAL George Helou IPAC
136. K and 1 selects the complex MASK for configuring MASK see Section 2 3 4 In both cases the first bit of the MASK value determines if a pixel is good or bad 0 is good and 1 is bad The simple representation only uses this first bit so the MASK will be filled with ones or zeros The complex representation uses the remaining 7 bits to record more information about the masked pixel For example if MASK 16 16 0 2 then pixel 16 16 is dead in the first IMAGE of DATA If MASK 16 16 0 4 then the same pixel has been deglitched In the complex representation a pixel is still only considered to be bad if the first bit of the corresponding MASK pixel is 1 so this means odd MASK values are bad and even are good regardless of what other status the pixel may have A list of possible types or keywords of flagged pixels is given in the table below Section 16 4 describes how the MASK may be manipulated Type integer IDL cube type keyword rank description bad 1 bad pixel dead 2 dead pixel glitch 4 deglitched pixels memory 8 pixel affected memory effects transients blind 16 blind pixel i e not illuminated qla 32 pixels flagged as bad by qla or csh spare 64 spare ext_source 128 extended source 5Do not modify this field directly It contains a reference copy of your CAM image data 15 2 OBSERVATION DATA STRUCTURES 167 19 20 HK Substructure containing house keeping parameters For every IMAGE FRAME pair there is a set of hous
137. LARIZOR POLARIZOR POLARIZOR 1 PB OBS OBS OBS OBS OBS OBS OBS OBS LW2 LW2 LW2 LW2 LW2 LW2 LW2 LW2 CSSCOOOOOOPOLAR1_00020117462500 3 Fh 4 tint gain size m raster n raster 10 10 10 10 10 10 10 10 NONON OPNS 9 M ON 1 PB 78 78 78 74 78 78 70 78 1 RP RP RP RP eB 54 CHAPTER 8 POLARIZATION OBSERVATION CAM05 DEDICATED CAM99 8 9 10 11 POLARIZOR 1 OBS LW2 3 0 2 10 T 78 2 2 POLARIZOR 1 OBS LW2 3 0 2 10 1 64 1 2 POLARIZOR 1 OBS LW2 3 0 2 10 1 78 1 2 POLARIZOR 1 OBS LW2 3 0 2 10 1 78 1 2 3 Clearly from the above listing sscd polarl contains only those SCDs with the entrance wheel set to POLARIZOR 1 However for each of the 4 raster positions we have 3 SCDs To merge these 4 triplets we can use scd concatene CIA CIA CIA CIA CIA CIA CIA CIA CIA CIA CIA CIA CIA CIA CIA CIA tmp scd concatene scds 0 scds 1 scd del scds 0 amp scd del scds 1 scdO scd concatene tmp scds 2 Scd del tmp amp scd del scds 2 tmp scd concatene scds 3 scds 4 scd del scds 3 amp scd del scds 4 scd1 scd concatene tmp scds 5 Scd del tmp amp scd del scds 5 tmp scd concatene scds 6 7 scd del scds 6 amp scd del scds 7 scd2 scd concatene tmp scds 8
138. LICING 19 6 2 3 Problems with STD files Since they were mostly produced on VAX VMS you may have problems to read them under UNIX If this occurs x slicer will inform you of this kind of problem What you have to do is e Log onto a V X system e Begin an IDL session e Restore the STD file e Save it again using the XDR keyword e FTP it to your UNIX system do not forget the BIN command Or more simply slice it directly on your VAX system Otherwise except for pointing information STD is the internal format that is used by x slicer for the data representation Therefore no problem should occur 19 6 3 Slicing ERD files Edited Raw Data files ERD were preferred to Standard Processed Data files SPD for histor ical reasons mostly because the pipeline has made the least alterations to ERD with respect to the telemetry To slice an ERD properly you also need e an IIPH file e a CSTA file e and sometimes an IFPG file 19 6 3 1 files IIPH files are Instantaneous Instrument Pointing History files they contain the information about the RA DEC and ROLL angle of the satellite during the observation as well as information about satellite jitter They are calculated for the prime instrument IIPH files have evolved as the pipeline software OLP has been refined In the distant past it was possible to accidentally receive an IIPH file from another instrument while CAM was prime see Section E 2 Now tha
139. LTRWHL in the CIA data structure see Section 15 2 1 168 CHAPTER 15 CIA DATA STRUCTURE HIGH LEVEL ARCHITECTURE 15 2 3 Set of SCDs SSCD As you now know the SCD corresponds to a single STATE of the camera However a meaningful observation is a collection of STATEs such as all the STATEs of a raster observation or all the wavelength measurements of a CVF scan The SSCD is a catalogue of SCDs either ERD SCDs or SPD SCDs It may be a catalogue all the SCDs in an AOT or a subset of an AOT though it is usually wise to create SSCDs that catalogue SCDs from a CONFIGURATION only In itself it holds no data other than CONFIGURATION parameters and the names of the SCDs attached to the CONFIGURATION The SSCD contains the standard fields of Section 15 2 1 and the following additional fields 10 11 NSCD The number of SCDs belonging to this SSCD Type integer RASTER COLUMNS Number of steps in a raster observation in the M direction Type integer RASTER LINES Number of steps in a raster observation in the N direction Type integer M STEP SIZE Step size of raster in M direction Type float Unit arcsecond N STEP SIZE Step size of raster in N direction Type float Unit arcsecond RASTER ORIENTATION The reference for a raster orientation Type string Possible values are value description UNDEFINED not a raster observation or incomplete SCD NORTH raster performed w r t
140. M SWITCH OBSERVATION CIA save file bs_pds xdr bs pds Alternatively you can export the data to a FITS file Again we take advantage of the BS PDS compatibility with the raster PDS and use raster2fits CIA gt raster2fits bs_pds name bs fits Additionally you can correct the beam switch MOSAIC in MOSAIC for distortion see Section 20 15 5 Chapter 7 CVF observation 4 7 1 Description of the observation The data used here is from a CAM calibration raster observation of the object HIC94890 This observation is comprised of only one CONFIGURATION this makes it simpler than the raster observation described in Section 3 The CONFIGURATION parameters are LW channel CVF1 filter wheel 6 PFOV and a gain of 2 As we will later see this CONFIGURATION has 20 STATEs where meaningful observation data was accumulated by CAM In a raster observation the pointing changes from STATE to STATE but as we will see later in a CVF observation the CVF wheel position changes 7 2 Data analysis It is assumed in this section that you have read Chapter 3 Generally concepts described in that section will not be re described here 1 Start a CIA session cia 2 Convert your CISP data product into SCDs with spdtoscd CIA spdtoscd cisp05805004 fits sscd dir cia_vers test nowrite 3 Remove unwanted SCDs with sscd clean CIA cleaned sscd sscd clean sscd Out of 22 SCDs 2 are rejecte
141. MAGE or CUBE if the keyword cube is set Of course at any stage in CIA processing you can fill column 24 yourself with corr col24 For example to fill column 24 in CUBE and IMAGE corr col24 pds To fill only CUBE or only IMAGE then set the keyword cube or image respectively 20 15 Advanced projection The projection is method is probably the best choice for creation of the raster MOSAIC see Section 20 4 for other methods There are a number of useful options available for projection and indeed the CIA projection routines have other purposes aside from the create of raster MOSAICs 20 15 1 Distortion correction There are a couple of options when it comes to projection and distortion correction These can be specified as raster_scan keywords Note that projection is actually performed by the lower level routine projette and the C executable projection Keywords given to raster_scan are passed to projette so it may be a good idea to take a look at the online help for both these routines e Perform the projection but create the raster MOSAIC without distortion correction CIA gt raster_scan raster_pds nodisto e Perform the projection using drizzling CIA gt raster_scan raster_pds shrink 0 7 e Perform distortion correction but give an alternative distortion file CIA raster scan raster pds dist file lw3asri0 dis 20 15 2 Weighted mean option Normally the weighting factor per EXPOSURE pixel is computed as a
142. MAGE by du dv before av eraging to an EXPOSURE The limitation of this method is the shifting algorithm So far all attempts to perform sub pixel shifting causes some smoothing and this defeats the purpose of removing the jitter After jitter computation with compute jitter jitter correction with reduce can be attempted with 296 CHAPTER 22 SECOND ORDER CORRECTIONS CIA reduce pds jitter CIA tviso pds image 0 project cube This routine attempts to correct jitter by applying du dv as an astrometric correction In the example below the entire PDS CUBE is projected onto a MOSAIC image In attempting to make a crude correction of the jitter a magnification factor of 3 is applied to IMAGEs before projection CIA project cube pds mosaic jitter mag 3 CIA tviso mosaic 22 2 Field of view distortion FOV distortion affects the 12 6 and to a lesser extent the 3 lens For raster PDSs FOV distortion correction is integrated with the project raster MOSAIC creation method see Sec tion 20 4 However any CAM image can be corrected for FOV distortion with corr field In order for this routine to apply the appropriate distortion correction some CAM parameters must be supplied along with the image pfov channel and filter wheel CIA corr field image fltrwhl 1w10 channel lw pfov 3 0 corr field re projects the image in order to correct the distortion As a result of this the output image maybe rebinn
143. O N_RASTER i NFO M RASTER i description SCD name RA DEC ROLL CRPIX1 CRPIX2 Gain offset integration time entrance wheel selection wheel filter wheel pixel field of view on board accumulations raster position in N direction raster position in M direction iis the index to the data for an individual SCD 15 5 8 CCIM subfi CCIM IMAGE fitarr 32 32 nsed fltarr 32 32 nscd fltarr 32 32 nscd CCIM RMS CCIM NPIX eld type reference Section 15 2 1 1 Section 15 2 2 6 Section 15 2 2 Section 15 2 2 8 Section 15 2 2 4 Section 15 2 2 5 Section 15 2 1 11 Section 15 2 1 12 Section 15 2 2 9 Section 15 2 2 10 Section 15 2 1 10 Section 15 2 1 13 Section 15 2 2 3 Section 15 2 2 1 Section 15 2 2 2 description of contents EXPOSURE in detector co ords RMS error on above No of IMAGE pixels per EXPOSURE pixel 185 The RMS or standard deviation is independent of the number of samples taken e g reducing normal Gaussian distributed IMAGEs will lead to a values close to 1 in CCIM RMS Care has to be taken to distinguish between standard deviation c and standard error the standard error is defined by 1 and therefore goes asymptotically to 0 for a large n 15 5 9 ASTR The ASTR substructure can be found in the raster PDS CVF PDS and the BS PDS This substructure contains MOSAIC e g RASTER in raster PDS astrometry information CIA gt help bs_pds astr str Structu
144. Or CROTA2 360 ROLL E 3 2 2 Reconstructed raster images In that case the meaningful angle is RASTER_ROTATION The individual images can no longer be distinguished therefore their ROLL is irrelevant to the raster it is only coincidentally meaningful for Y axis rasters Here also the simplest way to store the image is to have the M axis be Axis and the N axis be Axis2 Thus from Figures and E 5 one has 0 90 8 or CROTA2 90 RASTER_ROTATION Appendix F What is new in CIA 5 0 F 1 New and improved algorithms e New distortion correction We have now significantly improved distortion corrections You can apply them not only for rasters as before but also for CVFs and staring obser vations Distortion coefficients exist now for all 6 PFOV LW configurations and for some 3 PFOV LW fixed filters With the routines project bs project cvf project struct You can now correct your beam switch cvf or staring observations for distortion effects magnify them or if you corrected the astrometry with xcorr astro for shifts induced by the CVF get even a properly aligned image stack where you can derive easily a spectrum e Improved projection via the routine raster_scan featuring Reduced flux loss More projection methods New WCS library Improved astrometric accuracy Creation of the raster via weighted mean options wcalg wauto and wmap Easier access
145. PDS exist a CVF PDS a raster PDS also known as a raster data structure and a general PDS This data structure is formed from an SSCD and its SPD SCDs and is used to hold all the sliced i e prepared data that you need to perform calibration Usually the data in a PDS correspond to a single CONFIGURATION See Section 15 5 for more on details of PDSs Further details on the DSD may be found in the ISO Data Product Document or the CAM Parameter Characteristic Document 1991 ref ISOCAM ST110 Chapter 10 First look at the data This chapter will introduce to a CIA session and show you how to display AA computed images Reading the earlier chapters of the CTA Basic Guide may be helpful for fully understanding the text here though not necessary if you are impatient for a glimpse of your images and not worried too much about the details at this time 10 1 Copying data products from ISO CD ROM to hard disk If you are working with ISO CD ROMs rather than data obtained from the IDA then it is recommended that the data products are copied to hard disk prior to analysis though if you just intend at this time to view the AAR data with sad display then skip to Section 10 2 Some of the CIA routines strictly require a naming convention for the data products and the filenames of the products on the CD ROM use only an abbreviated version of the full name official name see Section 9 1 Routines described below will copy your data products to hard
146. PH The variable eph should have the following format 20 8 TIPS ON CIA DATA CALIBRATION 249 CIA help struct eph Structure lt 91 1 0 gt tags length 72 refs 1 JD DOUBLE Array 3 RA DOUBLE Array 3 DEC DOUBLE Array 3 Now you are ready to create the final mosaic which is contained in the tag RASTER To benefit from from superresolution due to the motion of the comet a magnification factor of 3 is chosen CIA project sso sso pds eph eph magnify 3 CIA tviso sso pds raster Figure 20 3 Comparison of standard vs improved SSO processing Left SSO without correction for proper motion Right SSO including correction for proper motion Note that this result benefits from super resolution due to the motion of the comet Due the rebinning of the pixels the flux per pixel is reduced by the factor 9 20 8 Tips on CIA data calibration If you have read the last sections then you will have realized that a great choice of calibration methods exist in CIA Though some broad recommendations can be made on which methods to use the only way you can be sure that you are achieving good calibration is through experi mentation and experience A good way to know what is happening is to go through each calibration step check your result and then proceed with the next step However you do have to be careful how you proceed CIA requires that certain calibration processes follow a particular order There are broad
147. RPOSE CATEGORY CALLING SEQUENCE INPUT OPTIONAL INPUT KEYED INPUT KEYED OUTPUT OUTPUT OPTIONAL OUTPUT EXAMPLE ALGORITHM DEPENDENCIES COMMON BLOCK SIDE EFFECT RESTRICTION CALLED PROCEDURE SEE ALSO MODIFICATION HISTORY Note that the space and the upper case are compulsory but you can add extra characters i e KEYED INPUTs is correct but not KEYED Input e The field of the keyword CATEGORY is defined in the file category txt This file can be found in the document directory of the CIA distribution The CATEGORY field should also specify whether the routine is User or Internal for example CATEGORY III 2 User e The compulsory keywords are NAME PURPOSE CALLING SEQUENCE INPUT OUTPUT EXAMPLE MODIFICATION HISTORY D 4 AUTOMATIC INCLUSION OF NEW PROCESSING ALGORITHMS IN CIA 325 The order of keywords is not important except that MODIFICATION HISTORY must be the last It is recommended that NAME is the first Format for INPUTS OUTPUTS KEYED INPUTS KEYED OUTPUTS They can be described as you wish but it is better to follow the following rules in order to have a nice table Put your general comments if needed on the first lines without inside Then at the beginning of a line name of the variable type of the variable description of the variable The separators are
148. S FITS file and the second is the output ISO FITS file CIA palomar to iso dss 10 45 22 55 57 35 fits haro3_iso fits You can now use your FITS file as input to isocont for displaying see Section 14 6 2 or xdisp for analysis see section 14 3 1 14 6 2 isocont We include a only brief description of isocont with examples here For a fuller guide to its use your attention is drawn to Claret A Charmandaris V Gastaud R 1997 A Learning Guide for ISOCONT v1 0 isocont accepts two arguments either one can be a FITS image or a CAM image in a raster data structure Because isocont needs astrometry for both images the CAM image must be contained in a raster data structure and the entire structure must be passed as an argument Whichever image is the first argument its astrometry is used as a reference and the image of the second argument is rotated or aligned so positions and orientation of both images match The first argument is plotted in grey scale or colour and the second as contour levels Figure 14 16 shows an optical image in grey scale overlayed with contours from a CAM image Examples of isocont usage are e Following the examples of Section 12 2 and Section 13 2 our first example here displays an optical image of Haro 3 overlayed with a CAM raster image The keyword mag magnifies the image for readability CIA isocont haro3 fits lw6 raster mag 4 This command produced the image in Figure
149. SE node OF e Add your machine name in the case statement with the proper definition of the variables default data directory default orbit file and default orbit directory 19 6 1 3 Bypassing the default data directory In any case if you wish to bypass the value of the default data directory you can use the keyword here to start slicing in the directory where you are working with CIA Just enter CIA x slicer here 19 6 1 4 Output Directory Even if you are working on a VAX VMS system the x slicer will never write to the ARC_DAT directory If you do not give it a directory where it can write the data he will by default try to write them to the current directory i e where CIA is currently running 19 6 2 Slicing STore Data 19 6 2 1 S Tore Data files STD STore Data files contain data from the ground calibration They therefore contain no astrometry information They are IDL SAVE files They contain the following variables e CAM CUBE the data cube of CAM frames e IMT_BLOCK an array containing the block of CAM telemetry except the frames e IMT_RECORD a structure to find the variables in the IMT_BLOCK e plus copies of values of important variables STD files contain only CAM telemetry 19 6 2 2 Slicing STD files Since STD files contain only CAM telemetry none of the IIPH IPFG CSTA and ORBIT file are usable Otherwise these files are sliced the same way as other types of files 220 CHAPTER 19 ADVANCED S
150. SOCAM s processor It can take three values Normal Accumulated or Sampled If you work in Accumulated mode frames coadded CIA will take into account this kind of observation Remember two things in your SCD the field TINT displays the effective integration time i e the number of accumulated frames times the integration time of the camera for each frame On the other hand the field CAL TINT gives you the integration time for one individual frame in CAMTU e Observation Mode This variable tells you if you are observing OBS idle IDLE waiting for a good configuration WAIT or in GAP or DARK mode If you are looking at CUS data see the value of the Observation Type variable it is very likely that you will get an IDLE value instead of an OBS for your observations This can be corrected in the SCDs CIA scdi sscd elem sscd CIA for i 0 n elements scdi 1 do scd put MODE OBS scdi i 19 6 5 4 Observation related variables These variables are again simple to understand Only two remarks have to be made e the F RAST variable in telemetry can only be on 1 or off 0 But in an SCD its value can be RASTER MICRO_SCAN STARING TRACKING or UNDEFINED see Section 15 2 1 The definition of a micro scan is very loose M step size and N step size lt 16 M step size or N step size lt 8 This means that
151. Scd del tmp amp scd del scds 8 tmp scd concatene scds 9 scds 10 scd del scds 9 amp scd del scds 10 scd3 scd concatene tmp scds 11 Scd del tmp amp scd del scds 11 And now we have a more usefully arranged SSCD CIA seq 0 1 2 3 Sscd info sscd_polar1 pol 4 SCDs in the SSCD C88C000000POLAR1 00020118111300 entwhl mode fltrwhl pfov tint gain size m raster n raster POLARIZOR 1 OBS LW2 3 0 2 10 1 234 1 26 POLARIZOR 1 OBS LW2 3 0 2 10 1 230 2 a POLARIZOR 1 OBS LW2 3 0 2 10 1 226 2 2 POLARIZOR 1 OBS LW2 3 0 2 10 1 220 1 2 For other entrance wheel settings e g POLARIZOR 2 return to Step 2 and create another SSCD e g sscd polar2 8 2 4 Freeze the data in a PDS The SSCD sscd_polar1 appears to contain data from a regular raster hence we use a raster PDS CIA _ 1 1 get sscdraster sscd_polar1 Reduce the IMAGEs to EXPOSUREs in the usual way CIA reduce pds polar 8 3 CALCULATE STOKE PARAMETERS 55 8 2 5 Flat field correction For the flat field correction we must use special flat fields These are distributed with CIA as IDL savesets and must be restored manually CIA help pds polari pfov pds_polar1 fltrwhl Expression FLOAT 3 00000 Expression STRING LW2 CIA restore file cia_vers data cds fl_lw2_p1_6 xdr Now use corr flat to do the flat correction supplying the restored flat field image CIA corr flat pds polari inflat flat
152. Section 2 4 for restrictions of use 16 1 6 structure read Function to read a named structure from disk A pointer to the structure is returned Our example below follows from that of Section 16 1 5 CIA scd scd read CSCD143006010105_96080110071423 dir scd dir ack ack 16 1 7 structure list Function to return the a list of structures of type structure in memory Following from Sec tion 16 1 6 our example lists the SCDs in memory CIA print scd list num no scds ack ack CSCD143006010105 96080110071423 etc CIA gt help no_scds ack NO_SCDS INT 42 ACK INT 1 The keyword num returns the number of scds found 16 1 CIA DATA STRUCTURE INTERFACE ROUTINES 191 16 1 8 structure del Procedure to delete a specified structure of type structure from memory A simple example which follows from Section 16 1 7 is CIA scd del CSCD143006010105 96080110071423 Using a little of IDL s power we can delete all the SSCDs in memory with the following example CIA sscds sscd list CIA help sscds SSCDS STRING Array 4 CIA for i 0 3 do sscd del sscds i CIA print sscd listO UNDEFINED Note that when deleting SSCDs or SSADs the whole set of SCDs or SADs is erased from memory 16 1 9 structure info Procedure to display a list of fields of a structure of type structure and their values The routine varies for each structure but in general the example below is valid for all flavou
153. Tools xv raster Launch xv raster See Section 14 4 11 Tools isocont ISOCAM axis Launch isocont See Section 14 6 2 Tools isocont RA DEC axis As above but the MOSAIC is displayed with EAST left and NORTH up Tools x_isocont Launch x isocont See Section 14 6 3 Tools hardcopy gif Create a print file of the isocont display see above in GIF format Tools hardcopy ps Create a print file of the isocont display see above in postscript format Some additional corrections can be performed But second order corrections should not be used unless standard processing has been performed first to assess the quality of data Remove Dark Residuals Dark pattern residuals can be removed using filtering in Fourier space Note that this secondary order method updates ISODATA IMAGE whereas first order dark correction updates ISODATA CUBE Smooth Undefined Values It may happen that ISODATA RASTER contains unde fined values if there is no stabilized value for example Undefined values can be replaced with smoothed values as determined from neighboring pixels FOV Distortion The FOV distortion effect can be corrected the trapezoidal shape of pixels is then taken into account in order to re sample the image Distortion can be assumed to be linear around the central pixel linear method or can be calculated from ground calibration measurements calibration method Spectral Deconvolution The measured spectrum can be de
154. Topic CIA USER ASTRON CIA HELP CIA USER T NAME SAD DISPLAY PURPOSE Widget program to display SSADS AA products CATEGORY 11 2 User CALLING SEQUENCE SAD_DISPLAY arg dir dir select select windows windows x amp ize xsi20 velze ysize trace trace all all hurry hurry raxdec radec mosh most ok ok INPUTS None OPTIONAL INPUT PARAMETERS arg string One of NothingL 2 5AD SSAD TDTOSN wildcards can be used if the specification is mot unique a selection menu is ised See examples prints the calling sequence KEYED INPUTS dir string Directory to tired files default is current directory SAD SSAR TOTOSN TOTOSN B default Relevant only when nao argument given windows integer Number of graphics windows in display 1 data 2 data error 3 data error exposure default 4 data error exposure mask X aize of graphics windows Y size of graphics windows Trace mouse events and report positions arsi values by default Normally this feature iB only activated in zoom windows RA DEC are given as Siexagesimal or DOecimal Just read the images and forget the rest Only meaningful if reading from the AA this is now the default see all Read all the AA products Maximum mimber of SADs bo create Hover select string saize integer ysize integer trace logical radec string hurry logical all logical most integer Fig
155. UBE MASK etc These are general references to fields and substructures in the PDS 20 2 Core calibration You have already been introduced to the core calibration routines in Section 13 2 1 Here we will look a little deeper into the different correction methods that are available for each of the core calibration routines The chosen correction method can be specified with the keyword method Each method is implemented by a low level routine which may accept keywords for tuning algorithm parameters T hese keywords may be specified when calling the core calibration routine and IDL will pass the keywords to the low level routine ie using IDL s keyword inheritance functionality To find out what tuning can be done for a particular method look for the relevant low level routine in the online help Section 2 3 2 20 2 1 Dark correction In principle dark correction is a simple procedure a matter of subtracting the DARK image from the IMAGE The hard part is choosing a good DARK All dark correction methods are handled by corr_dark Technical details of some of these methods can be found in the references given below In general corr dark performs the dark correction in the following manner 1 Obtains the DARK as instructed by the user setting of the keyword method or indark 231 232 4 5 CHAPTER 20 ADVANCED DATA CALIBRATION Divides the IMAGEs of CUBE with ADU SEC COEFF so as to normalize CUBE to ADU sec ga
156. V 1 1 algorithm Find best ccglwdark 97031713382678 END 16 6 MANIPULATING CIA DATA STRUCTURE HISTORY 199 date 28 8ep 1998 14 48 27 node bermuda user mdelaney procedure spdtoscd V 2 0 algorithm default 7 cisp02600506 fits undefined undefined undefined undefined undefined undefined 2 2 2 47 END date 28 8ep 1998 14 50 34 node bermuda user mdelaney procedure flat library V 1 1 algorithm Find best ccglwoflt 98050815090326 END date 28 8ep 1998 14 50 35 node bermuda user mdelaney procedure flat library V 1 1 algorithm Find best ccglwdflt 98031519384439 END date 28 8ep 1998 14 50 28 node bermuda user mdelaney procedure get sscdraster V 4 0 algorithm default CSSC026005060001_98092814502500 model 1 END 16 6 2 Replacing the history You can use the routines history2process and structure put to convert a history into a format suitable for insertion into the field PROCESS in CIA data structure Following from the example in the previous section we can recreate the contents of the variable process and then insert into our SCD CIA process history2process history CIA scd put process process scd 200 CHAPTER 16 DATA STRUCTURE MANIPULATION Chapter 17 Importing ISO data products to CIA CIA provides dedicated FITS handling routines to create the data structures from the data products distributed on the ISO CD ROMs or retrieved from the IDA This chapter provides an overview of their use 17
157. VF PDS the most interesting information is spectral rather than spatial Since each EXPOSURE was observed at a different wavelength spectra can be obtained directly from EXPOSURE pixels Since most observers would like their calibrated spectra to be in janskys CIA provides conv_flux to convert the EXPOSURES pixels from ADU to mJy CIA conv flux cvf_pds image To get a look at the calibrated CVF data use cvf display CIA cvf display cvf pds By using the mouse to point and click on selected EXPOSURE pixels you can display spectra See Section 14 2 1 for more information on cvf display 13 3 Data calibration with x cia This section serves as an introduction to the widget interface to CIA x cia 13 3 1 Introduction x cia was designed to help visitors of Centre ISO francais du CEA Saclay to become familiar with ISOCAM data analysis quite rapidly Once ISOCAM data has been sliced prepared a standard quick look analysis can be performed Section 13 3 2 ISOCAM data are mainly affected by the dark current cosmic ray impacts transient effects and flat field errors Correction of one of these effects is usually the result of several runs Section 13 3 3 Note however that not all CIA processing steps are available to x cia Some advanced functionality is available but should only be used by advanced users of x cia Section 23 Guidelines to help the user to choose the most appropriate processing methods are given in Section
158. VMS 2 0 3 5 4 fa aba ae x RR a e bd a 8 67 10 12 Gopying on UNIX a eee Eo Ep 69 10123 Manual copying uestem be oe GR Rae eke a e 69 10 2 Examining the AAR Data Products 69 11 Introduction to CIA data analysis 73 11 1 CIA Processing Overview 73 12 Data slicing 77 12 1 Data slicing methods 2 eR a eee Ge Oe aE EP Rex ey 77 12 2 Automatic data slicing 77 12 2 1 General slicing tips ooa ee 77 12 22 Slicing a raster observation AOT 1 78 12 2 3 Slicing a CVF observation 4 84 12 3 Data slicing with x_slicer 86 12 3 1 29tartidg xcslicer 2s Got epp Peete nme dog YR qua 86 12 3 2 Selecting ariile cist ass Bhd one ine eae RS euo d RO 88 12 3 3 Selecting slicing variables 89 CONTENTS vii 12 34 Rain the slicet s xxt EUG EIWUEGEGRSUERGX AE 90 12 3 5 The x handle slice window 90 12 3 6 Selecting SCDs and SSCDs 93 12 3 7 Choosing names 2 2 2 2 rss 94 13 Data calibration 95 13 1 Creating PDS from an SSCD 95 PDS caveats dI mom IRE BRE RE a bee ee elk a ae Rs 95 13 1 2 raster PDS A 2 2 52 x Less Bee a hae ag ae 96 19 L 3 general PD Sr 22 she Fa Age eke a
159. W CVF1 LW CVF1 LW CVF1 LW CVF1 LW CVF1 LW CVF2 LW CVF2 LW CVF2 LW CVF2 LW CVF2 SW DARK SW DARK SW DARK SW DARK SW DARK SW CVF SW CVF PPP Pe 0 C1 Oc1010O0O0000101 NNNNNY DANAOANONNNN 10 10 10 10 10 10 10 10 10 10 10 10 28 10 02 02 02 02 ONNNNNNNNN 512 512 512 15 16 15 16 37 15 16 16 1 49 1 1 41 10 174615 174615 174615 174615 8 174615 999999 174615 174615 174615 174615 8 174615 999999 174615 6 174615 999999 999999 174615 174615 22 22 23 21 22 99 22 23 22 22 22 99 22 22 99 99 22 22 sscd_clean will discard the intermediate states and split the raw two for each LW segment and one for the SW segment CIA gt cleaned_sscd sscd_clean sscd Out of 140 SCDs 10 are rejected due to mode 14 are rejected due to csh flag 13 are rejected due to qla flag In total 125 are accepted 27 Sep 1998 15 47 05 00 SSCD_CLEAN v 2 3 lt Splitting SSCD into 3 segments I gt 285012 285012 285011 285011 285012 999999 285012 285012 285012 285011 285012 999999 285012 285011 999999 999999 285011 285012 00 01 99 98 01 99 03 02 00 99 00 99 02 94 99 99 99 00 217 SSCD into three SSCDs The variable cleaned sscd is a string array co
160. a it simply informs the slicer to cia routine where they may be found 19 6 10 Getting your SSCDs To work on your data after having exited x slicer there are two ways to proceed e use the routine sscd read to read them back from the disk This is time consuming if you have big SSCDs e enter the command CIA print sscd list Pick the sscd that you are interested in 19 6 11 Frequently Asked Question 19 6 11 1 Why use x slicer Because it is beautiful uh 19 6 11 2 slicer compilation e x_slicer doesn t compile You should probably reread the section about x slicer customization because it does compile 230 CHAPTER 19 ADVANCED SLICING slicer_to_cia doesn t compile That is because COMMON block is not defined You probably exited IDL after having sliced a big file and you tried to read your sli file Reread carefully Section 19 6 8 or type the following commands CIA COMMON slicer data Oh my God what a stupid name for a variable CIA run slicer to cia 19 6 11 3 x slicer did not do what I told it to do x slicer did not write SCDs where it was told to Are you sure that you hit the return key after entering the path in the directory field Remember that the default directory is the one from which you launched x slicer x slicer did not create all the SCDs I asked for You probably asked for SCDs of 1 frame as they do not exist they were not created
161. a structure 1 NAME The data structure name is unique to each data structure The name is used to store the structure on disk and as a pointer to the structure in memory If you use x slicer to slice your data then you have some choice in the selection of your own naming scheme see Section 12 3 7 Otherwise if you are creating the data structures using IDL conversion routines of Chapter 17 then each data structure is named by either one of two conventions as follows If the TDT and OSN is defined then the following convention is followed CDDDttttttooccss_yymmddhhmmssdd where Further details on the DSD may be found in the ISO Data Product Document or the CAM Parameter Characteristic Document 1991 ref ISOCAM ST110 15 2 OBSERVATION DATA STRUCTURES 163 variable DDD tttttt 00 cc SS yy mm dd hh mm SS dd definition structure type abbreviated name i e SCD SAD SSC D SSA D TDT number observation sequence number configuration number state number year of creation month of creation day of creation hour of creation minute of creation second of creation 0 01 second of creation Otherwise if neither TDT or OSN is defined then the convention is CDDDYYMMDDHHMMSS yymmddhhmmssdd where variable DDD YY MM DD HH MM SS mm dd hh mm SS dd definition structure type abbreviated name i e SCD SAD SSC D SSA D UTC at beginning of data year UTC at beginning of data mon
162. ach CONFIGURATION To accommodate this re pointing of the satellite solar system observations are implemented as a one dimensional raster with the first raster point identification M RASTER being incremented for each re pointing 5 2 Data analysis It is assumed in this section that you have read Chapter 3 Generally concepts described in that section will not be re described here 1 Start your CIA session cia 2 Convert your CISP data into SCDs with spdtoscd CIA spdtoscd cisp81200202 fits sscd dir cia_vers test nowrite During the slicing CIA will print the following warning pixel 14 17 is affected by saturation at SCD 3 with the average value 4093 00 value for End of Integration 3 readouts This indicates that pixel 14 17 was saturated and its photometry has to be carefully assessed 3 Get an overview of the SCDs 35 CHAPTER 5 SOLAR SYSTEM OBJECT OBSERVATION CAMOI CIA sscd info sscd parameter ENTWHL mode fltrwhl pfov CIA tint gain size m raster n raster 44 SCDs in the SSCD CSSC812002020001_02022616563501 seq ENTWHL mode fltrwhl pfov tint gain size m raster n raster 0 HOLE IDLE LW2 6 0 25 20 1 1 1 1 1 HOLE IDLE LW2 6 0 2 10 1 2 1 1 2 HOLE IDLE LW2 6 0 2 10 2 1 2 1 3 HOLE IDLE LW2 6 0 2 10 2 1 2 1 4 HOLE OBS LW2 6 0 2 10 2 3 2 1 5 HOLE OBS LW2 6 0 2 10 2 1 2 1 6 HOLE OBS LW2 1 5 2 10 2 2 2 1 7 HOLE OBS LW7 1 5 2 10 2 10 2 1 8 HOLE OBS
163. adct doesn t preserve x and y This results in a wrong cursor position if widgets are called together with xloadct original wessph2xy doesn t return output as scalar if input was a scalar original wessph2xy doesn t return output as scalar if input was a scalar xy2ad contains patches for correct call to wcsxy2sph computation of xsi eta and correction for ISOCAM distortion ad2xy contains patches for correct call to cons_ra extast contains patches for correct computation of CD matrix putast contains patches for correct computation of CD matrix fxbparse uses LONG for dimensions new fxaddpar replaces type by type because there is a procedure type in the astro nomical library mrd struct uses openr lun filename delete free lun lun instead of rm or delete so it works under all operating systems mrdfits was optimised for reading big cubes new get equinox contains a patch for proper check on existence of EPOCH keyword 345 346 APPENDIX H PATCHED ASTROLIB AND IDL ROUTINES IN CIA Appendix I Upgrading old CIA structures Structures created in previous versions of CIA will need some upgrading to be 10096 compatible with CIA 3 0 1 1 Upgrading CIA 2 0 structures e Since the MAR98 version of CIA there are unit tags in the PDSs CUBE_UNIT IM 1 2 AGE_UNIT and RASTER_UNIT If you want to use conv flux on a BS PDS or raster PDS then you will need to add RASTER_UNIT to a pre MAR98 PDS and fill it with
164. affected data would not be distributed by ESA 19 6 3 2 files The Instrument Focal Plane Geometry file contains the rotations that allow transformation of the pointing information given in an IIPH from one instrument to another this may be useful if you wish to attempt to work on non scientifically validated data see Section E 2 Since the focal plane of ISO will not change during the mission it has been written to a CDS Calibration Data Set The IFPG file is therefore no longer useful use the CDS instead 19 6 3 3 files files are Compact STAtus files They contain a summary of the CAM CONFIGURATION This information can be used to check that the observation was properly done Since the x slicer displays what is in the data you can do that job too Therefore you may omit it when slicing especially if you know what should be in the data 19 6 3 4 ORBIT files This file is used to calculate the position and speed of ISO with respect to the Earth The position may be useful if you are observing very nearby solar system objects because of the parallax Though the speed is useful for the ISO spectrometers the spectral resolution of ISOCAM s CVF mode is of the order of 4 50 and so is not affected by ISO s speed You may therefore omit the ORBIT file 19 6 4 Slicing TDF TDF files are Telemetry Distribution Files They contain CAM telemetry for the prime and parallel mode TDF files are used by instrument
165. ally of the order or slightly larger than the radius of the PSF that is because the size of the filtering box will be 2x flat smooth window 1 Note that smaller values of this parameter usually translate in small values of the residual noise but that is some sort of artifact since for small windows most of the signal is removed e flat thresh this is a percentage that is used to discard pixels from the flat field compu tation The histogram of the pixel readouts is built and we discard the bottom and top flat thresh percent This is useful when you have strong sources and structures in the field to avoid having them propagate in the flat field e nplanes once the filtered and histogram selected cube of readout is built the flat field cube is built with a running average of nplanes In that case nplanes should be larger than Nep but not as large as if no filtering had occurred Now that the definition of the parameters is clear let s proceed Table 21 3 lists the pa rameters setup for both filters Note that both filters share the same tdt number since they were obtained in a single AOT For this data set the sources are not very important so the flat thresh value is not critical As always in SLICE choosing the parameters values requires some tuning and you ll need more than one try to arrive at a good choice of parameters The commands line to issue to SLICE are thus in the LW3 case CIA red_param set_red_param tdt 65801627
166. and additional substructures are listed here e CAL G see Section 15 5 6 e CCIM see Section 15 5 8 e INFO see Section 15 5 7 Standard PDS fields of Section 15 5 1 raster PDS specific fields These fields are listed in the table below along with a brief description and where appropriate a reference to the SSCD or SCD field where the data originates or otherwise the type of the data in the field 15 5 PREPARED DATA STRUCTURE PDS 183 subfield description reference or type RA RASTER RA of MOSAIC centre Section 15 2 3 13 DEC RASTER DEC of MOSAIC centre Section 15 2 3 14 ANGLE RASTER spacecraft ROLL Section 15 2 3 15 RASTER ORIENTATION orientation reference Section 15 2 3 6 _ rotation w r t celestial axes Section 15 2 3 7 M STEPCOL M step size of raster Section 15 2 3 4 N STEPLINE N step size of raster Section 15 2 3 5 RASTERCOL no of pointings in Section 15 2 3 2 M direction RASTERLINE no of pointings in Section 15 2 3 3 N direction NX_RASTER no of columns in MOSAIC integer NY_RASTER no of lines in MOSAIC integer RMSRASTER RMS error on MOSAIC fltarr nx raster ny raster NPIXRASTER no of IMAGE pixels fltarr nx raster ny raster per MOSAIC pixel RASTER MOSAIC image fltarr nx raster ny raster RASTER_UNIT MOSAIC units string 15 5 5 BS PDS The BS PDS is a beam switch observation dedicated data structure Its substructures and fields are listed below You will n
167. ard ACSA was added as slicing criteria e We upgraded to the latest version of the ASTROLIB e CIA runs under all IDL versions 5 0 5 2 5 3 and 5 4 F 3 Bug fixes There are several bug fixes and many more small improvements e isocont doesn t crash any more if the image dimension is a prime number e sliders were introduced into xphot to help to get the cuts right e simul_source can now also generate distorted PSFs by setting the use_disto keyword Also the raw option sources are added in ADUs not ADUGs was added F 3 BUG FIXES 341 The new option pol in spdtoscd ignores now the beam switch flag whose spurious changes during polarisation observation led to unnecessary slicing The central wavelength used for color correction was outdated This is fixed now imagette2fits works now properly for observations with different pixel field of view and for mixed SW LW observations the chattiness of low revel routines was reduced la mul works now also for cube x image 342 APPENDIX F WHAT IS NEW IN CIA 5 0 Appendix G Warning messages in CIA This chapter contains a selection of common CIA messages G 1 Error messages All error messages are expected to be self explanatory G 2 Warning messages Such messages contain warnings only To judge the final quality of the data processing users should take them into account The following message indicate that a saturation occurred The photometry for this exposure an
168. arison and overlaying 14 6 1 Obtaining images from the Digitized Sky Survey 1436 2 150 0nb edie 2 Gone de ee ge Bod eee 14 0 9 x 1SOCOBLU 1 2 dne oria ale oe ae a qui pn 14 6 04 xcorr strO ie Creating hardcopy plots 14 7 1 Using xcontour 14 7 2 Screen dumps with ps color Redirecting graphics to the postscript device 14 8 1 Avoiding postscript problems III Data Management 15 CIA data structure high level architecture 15 1 15 2 15 3 15 4 15 5 InttoduetlOma cse s enun im cec e e Observation data structures 15 2 1 Standard fields of observation data structures 15 2 2 Science CAM data SCD 15 2 3 Set of SCDs SSCD 15 2 4 Science Analysed Data SAD 15 2 5 Set of SADs SSAD Calibration Data Structure CDS 15 3 1 Standard fields of the 05 15 3 2 CDS and CAL G files 153 8 eds display Auxiliary calibration data 15 4 1 Theoretical PSFs 15 4 2 Observed PSFs 2 15 4 3 House keeping and CAM wheels data 15 4 4 Miscellaneous auxiliary calibration data Prepared Data Structure 5 15 5 1 Standard fields of the
169. arkwardt craigm lheamail gsfc nasa gov fuse of cmps_form pro by Craig B Markwardt craigm lheamail gsfc nasa gov 14 4 CUBE ANALYSIS 129 gt gt Znin o 2 455 Min 4 07121 Max 14 9491 Puto Mean Figure 14 7 The Plot Window 14 4 3 4 Plot Window This window has buttons and a display region it is described in 14 7 The display region plot the Intensity of a selected pixel versus the time see next window The last two buttons of the previous window enable to choose the scale and range of the Z axis Here the buttons enable to choose the Z range of the plot lt to scroll the plot along the Z axis text widget you can type the Z range the plot will be displayed according to these values text widget you can type the intensity thresholds the plot will be displayed according to these values e toggle button if Auto is selected then an automatic intensity scaling of the plot is done between the minimum and maximum value of the plot and NOT the whole cube If Fixed is select the minimum and maximum values don t change when an another pixel is selected toggle button to plot the sum or the mean when a rectangle is chosen in the Frame Window see advanced features 14 4 4 Frame Window This window has buttons and a display region it is described in 14 8 This window displays one selected image of the cube In this window the upper rectangle contains the title o
170. aster positions or on the full readout cube The latter option is obtained by adding docube on the slice_pipe command line Although this is in principle better the differences at this stage are try using it at the start of your SLICE session and you will see that SLICE uses the Sliding Mean Flat Field method instead 21 5 A WORKED EXAMPLE 281 small They are most noticeable for the DivSky method or at the bad pixels removal stage see sec 21 6 Let us begin with the Perturbed Single Flat Field method One should note that this is really to allow determination of the long term transient Since a subtraction is done this is not strictly speaking a flat field correction and should not be used as such This method requires 4 parameters to be set with set red param tdt flat smooth window flat thresh and nplanes e tdt this is ISO specific number which is located in the tdtosn field of your raster data structure How to extract this information is explained in section 21 4 1 In our example its value is 65801627 meaning the observation number 27 of that proposal executed on revolution 658 as the 16 observation of the revolution e flat smooth window this is a number of pixels used to define a smoothing window for the filter that is going to be applied to the individual 32x32 images to remove the large scale structure before the flat field construction Since it is working on 32x32 images it should be a rather small number typic
171. aster PDS the image is taken from RASTER and the astrometry from ASTR e a file containing data in one of the following formats FITS IDL saveset xdr MIDAS format bdf You can start ximage without any arguments and then load an image from a file using the file menu 14 4 8 2 Banner ximage provides five menus in its banner File Scale Tools Zoom Help The contents of each of these menus is described below File Load you can load an image with the various input formats described above just by clicking Save you can save the image as a PostScript GIF TIFF or FITS file with or without the color bar Plot image you can plot the image with or without the color bar this image is in a new window This window is kept after exiting ximage The title of this plot is given by the sub window labelled title in the ximage widget see below Quit exit ximage Scale linear logarithmic equalization of histogram Taken from Chanial P amp Gastaud R 2000 Ximage manual 14 4 CUBE ANALYSIS 137 Tools maximum the maximum value and its location is echoed in the terminal and dis played in the intensity and coordinate sub window profile not implemented histogram the histogram of the image is plotted in a new window contour the contours of the image are drawn in a new window 3d surface the surface in perspective 3D of this image is drawn in a new window Zoo
172. astrometry of the reference field To counter this problem set the keyword reverse when calling get sscdbs to re reverse the fields in the created BS PDS CIA bs pds get sscdbs bs sscd reverse 19 4 Advanced slicing of beam switch data This section introduces the CIA user to advanced concepts in slicing beam switch data 19 4 1 Concatenating intermediate SCDs in a beam switch observation The slicing keyword bs may be of interest to observers with beam switch data Beam switch observations have an intermediate step which occurs while slewing and data accrued in this step will appear in an SCD Normally this SCD is considered to contain invalid data though in reality this data may be quite usable Setting bs will make spdtoscd merge this SCD with the previous SCD and users can later manually unmask this invalid data for inclusion in the finally MOSAIC image In addition using the keyword bs has the added advantage of making it possible to use sscd clean on the SCDs Compare the following with Chapter 6 CIA spdtoscd cisp05804610 fits sscd dir cia vers test nowrite CIA sscd info sscd 15 SCDs in the SSCD CSSC058046100101_98092617482201 Seq channel mode fltrwhl pfov tint gain offset size ra dec 0 LW IDLE LW2 6 0 25 20 1 512 1 1 LW IDLE LW2 6 0 2 10 2 512 1 2 LW OBS LW2 6 0 2 10 2 512 1 3 LW OBS LW2 6
173. at each position using a model method inv method This method uses a fitting technique called the IAS or INVERSION Model This method is one of the more successful fitting methods routine called corr_transient_inv PDS side effects Unstable pixels are flagged in MASK and CUBE is modified reference ISOCAM Handbook Chapter Data processing methods Section Simplified model of the CAM LW response 20 2 CORE CALIBRATION 237 5 6 method vision method This method uses a model called the VISION Model This is a rather severe method and can kill off much of your data It does produce very nice clean images though it is good at eliminating ghosts from raster MOSAICs routine called corr transient vision PDS side effects Unstable pixels are flagged in MASK and CUBE is modified reference ISOCAM Handbook Chapter Data processing methods Section Vision model method med method Global transient removal using a median filter Best for extracting faint point like sources from a raster observation with many pointings See the online documen tation Section 2 3 2 for detailed restrictions and caveats routine called corr_transient_med PDS side effects Undefined pixels are flagged in MASK and CUBE is modified reference online help 20 2 4 Reducing IMAGEs to EXPOSUREs All the IMAGEs accrued in each CAM STATE or ISO pointing can be averaged to an EXPO SURE Although different averaging metho
174. at the top of the window allow you to toggle between the CCIM and CMAP EXPOSUREs zoom the image in the window and toggle between the data image of the EXPOSURE its RMS error and an image representing the integration time per pixel in the EXPOSURE also called the weight of the EXPOSURE e Clicking on the zoom buttons activates mouse tracking to display pixel values with coor dinates Pop up windows with error messages may appear when you try to attempt an undefined action you may regard these as more warning messages than errors but just hit O K and ignore them 72 CHAPTER 10 FIRST LOOK AT THE DATA Chapter 11 Introduction to CIA data analysis This chapter provides an overview of the CIA data analyses processes vis vis data structures and data products 11 1 CIA Processing Overview Figure 11 1 provides an overview of the general analysis procedure you will follow to reduce your data products to presentable images The three processing steps are data preparation data calibration and image analysis amp display subsequent chapters describe each of these steps Note that the level of processing depends on the data products you begin with e g you don t need to perform data calibration on the images in the AAR data products Also remember that CIA data structures are the kernel of CIA processing Each processing step is performed on a particular data structure and the structures evolve as the data are further reduced Hope
175. at there are two kinds of stabilization methods i Masking methods that simply flag unstable pixels in MASK ii Fitting methods that modify CUBE in an attempt to compensate the data for the transient response of the CAM detector The fitting methods may also flag pixels Technical details of some of these methods can be found in the references given below and in the technical reports listed in Appendix K Before proceeding with the details of the different methods stabilize has an important keyword option that the user should be aware of This is the keyword timeline It determines how the timeline is generated for the transient correction It can be set to one of the following string values arrival uses the BOOTTIME available for OLP 7 0 products or UTK tint creates the timeline from the integration time No telemetry drops are taken into account scd creates the timeline from the integration time However telemetry drops within an SCD are taken into account 236 1 2 3 4 CHAPTER 20 ADVANCED DATA CALIBRATION method fs method This method uses a fitting technique based on the Fouks Schubert model It is best used on an SSCD see Section 20 3 for an example There are three different implementations of this algorithm an IDL version a Fortran version and a C version To use the IDL or Fortran version set the stabilize keyword idl or fortran Some points to note about these different implementations
176. at you can either use a single flat field image or provide a flat field cube with as many planes as the data cube Assuming you have this flat in variable my flat to use this method you need to type CIA red_param set_red_param tdt 65801627 CIA act set_act make_map CIA gt slice_pipe flat_in my_flat For some reason SLICE will erase your flat variable during its processing so before starting slice_pipe make a copy of your flat field cube Now the three specifically SLICE methods are e Sliding Mean flat field where the flat field is derived from the mean of the data cube taken on a sliding window of readouts This is satisfactory if your sources are not very strong but if this is not the case or if you have large scale structures you run the risk of seeing some of these structures go into the flat field which is not advisable You could protect yourself from that using a large sliding window typically around 10 time Nexp but this is not always possible i e when you have a small number of raster positions Perturbed Single Flat Field although this does not sound much better this is actually an improved flat field method Here using a median filter on the individual raster point ings the object structures are removed before an auto like flat is computed Then still using a sliding window over the readouts perturbations to this single flat field are com puted The interesting point here is that since the obje
177. ata Display History Display the history of current data Note that the history is size limited to a 512 string array Data AOT Info Display useful informations such as the optical path the detector gain the integration time View Cube Temporal Mean Display the temporal mean of the data cube View Cube Temporal Analysis Display one by one all frames of the data cube The temporal flux variation recorded by a given pixel mouse selected is also displayed View Cube All Frames Display simultaneously all frames of the data cube View Mask Temporal Mean Display the temporal mean of the data mask View Mask Temporal Analysis Display one by one all frames of the data mask The number of times that a pixel is masked can be assessed by plotting the temporal cut of the mouse selected pixel View Mask All Frames Display simultaneously all frames of the data mask This can give an idea about the temporal evolution of the effective area View Image Averaged Frames Display simultaneously the averaged frames corre sponding to each configuration of the camera ISODATA IMAGE The numbers of frames displayed along the horizontal and vertical axis correspond to the M and N parameters of the raster map View Image Mean Of All Frames Display the mean of the above averaged frames View Cal G Dark Current Display the dark current frame extracted from the calibration database View Cal G Flatfie
178. ata not with the closest wheel index 20 12 2 find best psf This routine is used in a similar way to find best but of course it is just for PSF calibration data See the on line help or cia help for more details 20 13 Unit conversion and colour correction This section describes how pixel units are handled in CIA and how users may convert their images into milli janskys mJy 20 13 1 Propagation of pixel units within a PDS The IMAGE pixels values in a freshly made uncalibrated PDS will be in units of ADUs CIA print pds cube unit ADU After dark correction the same pixels will be in units of ADU gain second 20 13 UNIT CONVERSION AND COLOUR CORRECTION 257 CIA corr dark pds CIA print pds cube unit ADU G s After CUBE reduction these units will propagate into the EXPOSUREs CIA reduce CIA print pds image unit ADU G s Likewise for MOSAIC creation CIA raster scan pds CIA print pds raster unit ADU G s 20 13 2 Conversion to milli janskys At any stage in the processing you can attempt to convert pixel values to milli janskys mJy with conv flux For example after raster MOSAIC creation you can convert raster pixels into mJy with CIA conv flux pds raster CIA print pds raster unit mJy pix Likewise cube IMAGE pixels may be converted by setting the keyword cube and image EXPOSURE pixels by setting image conv flux uses the low level conversion routine adu to
179. ata is organised within CIA Examples are included to aid description of the data structures These examples include CIA manipulation routines which may be unfamiliar to you in such cases refer to Chapter 16 for a description 15 1 Introduction The data is managed in a CIA session by purpose designed data structures These structures may seem large and unwieldy to the novice user but experience has shown that they are the most convenient way of managing CAM data AOTs can be quite complex and many CAM FRAMEs may be accrued during a typical observation In addition the many CAM parameters that exist need to be stored along with the FRAMEs IMAGEs EXPOSUREs and MOSAICs The structures attempt to neatly store all these data CIA makes the structures as transparent as possible to the user by using dedicated manipulation routines see Section 16 or by allowing you to transform CIA data structures into ordinary IDL structures PDS It is hoped that this chapter will help you work with the structures but not get too deeply involved with their architecture The structures are broadly grouped into those containing observation data see Section 15 2 and those that contain calibration data see Section 15 3 15 2 Observation data structures Observation data structures are designed to hold actual image data from a CAM observation and detailed information about CAM and ISO parameters during the observation Where possible the data is presented in f
180. atic creation of the FLAT In the case of a raster observation a very good FLAT can be determined from the actual observation data To use this auto FLAT set the keyword method to method auto The algorithm used to create the auto FLAT is as follows a A median image is derived from the EXPOSUREs in IMAGE or if the keyword cube is set a median image is derived from the IMAGEs in CUBE b The resulting median image is then normalized by dividing by its mean c This FLAT is placed in the PDS FLAT and flat fielding proceeds in the usual way 20 2 CORE CALIBRATION 239 called routine flat auto 5 method manu method A FLAT may be manually from observation data In this case corr_flat calls the routine flat builder to interactively aid you in obtaining the perfect manual FLAT see Section 20 10 After exiting flat builder your custom FLAT is placed in the PDS field FLAT and flat fielding proceeds as normal called routines flat builder 6 inflat my flat method Flat fielding with your own FLAT If you happen to have your own FLAT e g say my_flat then you can pass this to corr flat by setting inflat my flat Again my_flat will be placed in FLAT and the flat fielding procedure will continue as usual called routine N A 20 2 6 Flat fielding and wheel jitter Due to wheel jitter CAM images can become considerable shifted in the horizontal or instrument y axis direction This shift is caused by a misalign
181. ber of CONFIG URATIONs do Two of the SSCDs correspond to the LW3 and LW6 CONFIGU RATION and the remaining SSCD contains calibration data CIA gt print cleaned_sscds C388C143006010001 98060117273666 C88C143006010002 98060117274484 C388C143006010003 98060117275395 We will ignore the calibration data few observer s will have such data and concen trate on the LW3 and LW6 CONFIGURATION We will use sscd info to find which SSCD corresponds to which CONFIGURATION CIA sscd info cleaned sscds 0 deg 16 SCDs in the SSCD CSSC143006010001_98060117273666 seq channel mode fltrwhl pfov tint gain offset size ra dec O LW OBS LW3 6 0 5 04 2 512 31 161 244 55 967 1 LW OBS LW3 6 0 5 04 2 512 20 161 282 55 983 2 LW OBS LW3 6 0 5 04 2 512 20 161 319 56 000 3 LW OBS LW3 6 0 5 04 2 512 20 161 356 56 016 etc CIA gt sscd_info cleaned_sscds 1 deg 16 SCDs in the SSCD CSSC143006010002_98060117274484 seq channel mode fltrwhl pfov tint gain offset size ra dec O LW OBS LW6 6 0 5 04 2 512 51 161 244 55 967 1 LW OBS LW6 6 0 5 04 2 512 20 161 282 55 983 Note that sscd_clean does not work with beam switch data unless the spdtoscd keyword bs is set see Section 19 4 1 12 2 AUTOMATIC DATA SLICING 83 2 LW OBS LW6 6 0 5 04 2 512 20 161 319 56 000 3 LW OBS LW6 6 0 5 04 2 512 20 161 356 56 016 etc Now we will store the name of each SSCD in appropriately named variables CIA lw3_sscd cleaned sscds 0 CIA lw6_ssc
182. by clicking in the plot to choose a time then clicking on the button zoom and then select block You can launch multiple copies of xcube Please properly quit xcube This program uses pointers which swallow memory and this memory is freed when you nicely exit 14 4 3 2 Button Description The figure 14 6 shows organization of the xcube widget The widget can be divided in 3 zones Banner the Plot Window and the Frame Window e Banner buttons The old routine xcube was declared obsolete and replace by this new routine 14 4 CUBE ANALYSIS 127 ISOCAM CIA Cube Analysis File Window Color Zoon Axis Znin o Zmax 455 Min 5 01200 Max 11 5776 Auto Hean Hin 7 01484 9 82807 futo Cube 16 16 3601 9 44956 Title Ms1 FILL Lui 3 360 860 Next Frane Previous Frane Tenporal cut _ Horizontal cut _ ertical cut Bad pixels N Symbols Hasking Inage Cube Scale Hist Figure 14 6 The whole xcube 128 CHAPTER 14 IMAGE ANALYSIS AND DISPLAY e Plot Window visualize the time history of a pixel or a cut in the image The intensity is plot versus a Z axis Z can be time or X or Y of the camera e Frame Window visualize one image of the cube 14 4 3 3 Banner The banner is the upper sub window from left to right The banner contains three scrolling menus File Window Color Tool
183. cd method model deglitch Deglitch IMAGEs in CUBE using the method mm CIA deglitch pds_scd method mm B 3 DATA VISUALIZATION 315 stabilize Stabilize IMAGEs in CUBE using the method s90 CIA stabilize pds scd method s90 corr jitter Compute the jitter in the IMAGEs in CUBE The arrays DU and DV are filled with the jitter offsets CIA corr jitter pds scd reduce Reduce the IMAGEs in CUBE to EXPOSUREs in IMAGE CIA reduce pds scd corr flat Perform flat field correction on the EXPOSUREs in IMAGE CIA corr flat pds scd raster scan Useful only with a raster PDS It creates the raster MOSAIC and places it in RASTER In the example here a projection method is used CIA raster scan raster pds method project reduce bs Useful only with a BS PDS It creates the beam switch MOSAIC and places it in RASTER CIA reduce bs bs pds flux Converts data from ADU gain sec to janskys CIA conv flux pds B 3 Data visualization This section summarizes some of the visualization routines that appear in Chapter 14 tviso The simplest display routine Accepts any 2D numeric array e g a raster MOSAIC CIA tviso raster pds raster x3d Examine a 3D numeric array i e a cube of images To display CUBE with the option of indicating masked pixels CIA x3d pds isocont Displays an image and overlays with contours from another image Screen output can be directed to a postscri
184. cds 1w6 CIA sscd write lw6 sscd dir 1w6 scd dir To slice the LW3 CONFIGURATION we would reload the SPD SCDs CIA spd sscd sscd read spd sscd dir scd dir And then use scd find to delete all SCDs not corresponding to the LW3 CONFIG URATION and save the remaining LW3 SSCD and SCDs in their own subdirectory When you are completely finished slicing you can delete the original unsliced SSCD from disk CIA sscd remove spd sscd dir scd dir 12 2 3 Slicing CVF observation AOT 4 An example of automatic slicing of a CVF observation is given in this section This example is not so thorough as that of Section 12 2 2 since in principle slicing is similar for data products of all observation types Note that this is a relatively simple example for a more complex CVF observation comprising of multiple segments see Section 19 5 Here the slicing process begins at SPD level with the CISP file 1 Start a CIA session 2 Assign the IDL variables path and scds path with the directory holding the data products and the destination directory for the SCDs example is given for both UNIX and VMS CIA path DKA200 MDELANEY CVF OBS CIA scd dir DKA200 MDELANEY CVF OBS SCDS OR CIA path home mdelaney cvf_obs CIA scd dir home mdelaney cvf_obs scds 3 Both these directory paths and the IDL variable cvf sscd are passed to spdtoscd CIA spdtoscd cisp20305604 fits c
185. ce and the background level in the image Upon calling it displays a plot window of integrated flux against distance from source The user selects the interval for the background fit and enters it upon request An example call to flux sum could be CIA flux sum cvf pds image 2 flux flux rms background background rms CIA print flux background In the example the input argument is the 3rd image in the IMAGE of a PDS The subsequent arguments are outputs and return the estimated flux the RMS on the flux the estimated background level and the RMS on the background 14 1 4 Photometry measurements with xphot xphot is a photometry tool specifically designed to work with CAM data A detailed description of xphot can be found in Sauvage amp Aussel 1997 Here we will give a simple walk through example 1 Invoke xphot with CIA xphot raster pds raster As you might guess from the syntax above we are going to perform photometry on the MOSAIC of a raster PDS You can supply any kind of CAM image to xphot though you may prefer it to contain a source If you wish to do PSF fitting and the theoretical PSFs see Section 15 4 1 for a description have not been installed in the correct location then you can specify an alternate path with the keyword psf dir 14 1 GENERAL ANALYSIS ROUTINES 115 Xloadct Select Objects Pins fune Positions Save Current Tables QUIT Figure 14 1 The xphot window 116 CHAPTER
186. celestial axes Spacecraft Y_axis raster performed w r t spacecraft axes RASTER_ROTATION Angle of rotation of CAM with respect to celestial axes Type double Unit decimal degrees NUMBER OF REFERENCES The number of reference fields in a beam switch ob servation undefined for other AOTs Possible values range from 1 to 4 Type integer NUMBER OF CYCLES The number of cycles in a beam switch observation unde fined for other AOTs RA_REFERENCES RA coordinates of the reference fields in a beam switch observation undefined for other AOTs Type double IDL array Unit decimal degrees DEC REFERENCES DEC coordinates of the reference fields in a beam switch obser vation undefined for other AOTs Type double IDL array Unit decimal degrees S T hese fields are only defined for a raster or micro scan observation See Appendix E for more on CAM angles vis vis the raster observation 15 2 OBSERVATION DATA STRUCTURES 169 12 WAVELENGTH START WAVELENGTH END WAVE INCREMENT Ap plies to CVF AOT only WAVELENGTH START and WAVELENGTH END are the wavelengths at the beginning and end of a CVF scan WAVE INCREMENT is the number of wheel steps between each position in the scan Type float WAVELENGTH START WAVELENGTH END and byte WAVE INCREMENT Unit microns 13 RA RA J2000 of the centre of the final MOSAIC that is constructed from the EXPO SUREs in the SCD Type double Unit decimal degrees
187. commend that before you use SLICE you read its user s manual as well as M A Miville Desch nes paper to familiarize yourself with the concepts involved In this section we will only show an example of how to perform a perturbed flat field correc tion on your raster data to illustrate the main feature of SLICE s syntax sec 21 5 provides a more detailed example The principle of this flat field derivation is that to first order the flat field is assumed to be constant over the whole raster and that temporal variations are treated as small perturbations to this flat field It is also assumed that for a given readout departures from this single flat field are dominated by high frequencies To get them a smoothed version of the datacube is made which is subtracted to the original data One then derives the perturbation of the flat field from a sliding mean on the modified cube The parameters for this flat fielding method are flat smooth window The window size used to smooth each readout nplanes the number of readouts used in the sliding mean flat thresh the percentage of data discarded from the flat field computation both in the top and bottom part of the distribution flat smooth window is generally a small number as it is applied to 32x32 images and the window size is actually 2x flat smooth window 1 nplanes need not be larger than the number of exposure per raster point as most of the signal has been removed finally flat thresh lar
188. comparison and overlaying Astronomers may find it useful to do counterpart comparison and overlaying of images In general the counterpart will be optical and the Digitized Sky Survey see Section 14 6 1 provides suitable optical images for comparison with CAM images If you have your own source of images in FITS format then these will do just as well CIA provides isocont see Section 14 6 2 to help you visualise overlays of FITS images and or CAM images 14 6 1 Obtaining images from the Digitized Sky Survey Your site may possess the DSS CD ROM set If so ask your CIA administrator for informa tion on accessing the CD ROMs Alternatively DSS images can be retrieved from the on line ESO ST ECF archive e To obtain a DSS image from the on line ESO archive point your WWW browser at http arch http hq eso org cgi bin dss 14 6 IMAGE COMPARISON AND OVERLAYING 145 Figure 14 15 show frame window 146 CHAPTER 14 IMAGE ANALYSIS AND DISPLAY You need to supply the astronomical name of the object which you wish to overlay or its position in RA J2000 and DEC J2000 and the size of the image in arcminutes Be sure to request an image of size comparable to the CAM image you wish to compare it to Further instructions can be found on the Web page Note that the images from the DSS are in a particular FITS format GSSS astrometry and need to be converted to ISO FITS with the CIA routine palomar to iso The first argument is the DS
189. convolved of the spectral resolution of the CVF filter Note that the spectral sensitivity is corrected by selecting the Jy pixel units instead of ADU s pixel Appendix A Glossary AA Auto Analysis As the name implies this is an automatic analysis of the telemetry data to produce AAR AAR level Auto Analysis Results AAR level refers to the AA processing level that produces AAR In the CIA User s Manual generally used to refer to the data products CCIM CMAP CMOS ADU Analogue to Digital Units AOT Astronomical Observation Template In the ISOCAM Observer s Manual an AOT is defined as Representation of an observing mode in the proposal generation software PGA In the CIA User s Manual an AOT refers to a period of observing during which only one AOT type is employed i e one of either AOT 1 AOT 3 AOT 4 5 In this sense AOT and observing mode are synonymous see glossary entries for observing mode Note the following characteristics of the AOT e An AOT can be identified by its lt TDT OSN gt number e One set of data products exists for each AOT e An AOT corresponds to a set of CONFIGURATIONS which in turn contains several OP MODEs comprised of one or more STATEs AOT 0 This AOT is reserved for CUS use AOT 1 An AOT type where a raster micro scan staring tracking observation may be per formed AOT 2 Not currently used AOT 3 An AOT type where a beam switch observation is performed AOT 4 An AOT ty
190. count what you entered 228 CHAPTER 19 ADVANCED SLICING For people used to the previous version x slicer now fills in the 12 characters with some s if your name is shorter 19 6 7 On Target Flag Two kinds of On Target Flag OTF exist The first one is the F2 OTF FLAG that is in the telemetry This is one of the slicing variables This flag is set on when the pointing of the satellite is less than 10 arcsec away from the intended position This is of course much less than the ISO real pointing precision but not far off its initial specifications This flag was of some use at the beginning of CIA because some routines did not separate on target frames from other ones This is no longer the case Moreover some routines work on a cube without holes transient correction for example By using this variable you will create such holes For these reasons this slicing variable should no longer be used The second one is the Enhanced OTF see Section 19 6 7 2 An important point to understand is that the OTF that appears in SCDs is not this F2 OTF FLAG but a combination of it plus some other variables This point will be addressed now 19 6 7 1 OTF in SCDs QLA FLAG In your SCDs the cube hk qla_flag variable plays a big role It tells you if your frame is in good shape this role is played by the F2 QLA telemetry variable coded on one bit if the pointing is correct for normal mode 2_ FLAG and for sample or accumulated mo
191. ct structures have been removed the size of the window can be smaller than in the previous case Variable flat with Sky Divided or DivSky this is the ideal method where an estimation of the sky is divided out of the images before the flat field is computed This estimate of the ideal sky is done by smoothing the current value of the map Therefore it is immediately clear than a first flat field correction must have occurred through the make map action This is also the reason why two successive make map actions using DivSky with the same settings can produce different maps although the differences are small This is however the method which can be affected by the long term transient or affect its determination Therefore we recommend to use it after the long term transient has been removed and in combination with another flat field method to produce the first version of map You will rapidly see that in general as far as flat fielding quality goes the DivSky method produces apparently the best results compare for instance Fig 21 2 with Fig 21 3 However it is also the method whose interaction with the long term transient correction is the most complex and which can lead to strong artifacts Therefore in this section we describe them both but in sec 21 5 3 we will use the Perturbed Single Flat Field method to determine the long term transient correction Also note that SLICE now gives you the choice of working either on the cube of r
192. cted if selected if selected if selected if selected if selected if selected if selected if selected f selected selected selected selected selected m gt gt gt gt Table 12 1 Conversion table for the variable names displayed by x_handle_slice 12 8 DATA SLICING WITH X SLICER 93 e Print the slicer display will print the big table in a text file The name of this file will be the official name of the observation with the prn extension For example 1380302 prn e Save Slicer File save the work done under a form re usable by the slicer see Handling big datasets in the Advanced Slicing section e Quit Quit x slicer e Back to last Choice will redisplay the state of the slicing operation before you hit the button Redisplay e Back to First Choice will redisplay the first screen that appeared just after the slicing proposal was computed e Back to Main Window will go back to the x_slicer window WARNING you will lose all the slicing work that has been done unless it has been saved The Advanced Slicing pull down menu contains the list of all lens filters gains and integra tion times used in your data plus a Number of exposures menu 12 3 6 Selecting SCDs and SSCDs Let us now begin the real slicing work Looking at the big board where the proposition of slicing has been displayed you will soon notice that there are many SCDs that you will not need especially in the beginning of observatio
193. ction 14 4 3 and Section 14 4 5 respectively e Overlay your CAM images with optical images or indeed any counterpart images isocont See Section 14 6 2 e Interactively choose contours for contour plots with xcontour See Section 14 7 1 e Create postscript plots 76 CHAPTER 11 INTRODUCTION TO CIA DATA ANALYSIS Chapter 12 Data slicing This chapter introduces you to CIA data preparation or data slicing see Figure 11 1 It follows logically from Chapter 11 and it is assumed that you have read that chapter 12 1 Data slicing methods As already stated in the Chapter 11 you have a choice of starting with ERD or SPD data products and as you might expect the data preparation is different for each data product You may use CIA s conversion routines to automatically create CIA data structures from either data products or you may use the widget based program x slicer to slice the ERD data Both these methods are described below in Section 12 2 and Section 12 3 respectively 12 2 Automatic data slicing Guidelines and examples of slicing data from different CAM observations can be found in the following sections In these sections you will be introduced to several new routines the most important of which are erdtoscd and spdtoscd these routines are used to convert ERD data products to ERD SCDs and SPD data products to SPD SCDs respectively A general guide to these routines may be found in Chapter 17 All other CIA routines use
194. cvf_pds cvf pds xdr 112 CHAPTER 13 DATA CALIBRATION 13 4 3 calib bs The BS PDS is calibrated with calib bs Again it is called in a very similar manner to calib raster Like calib_raster calib_bs does create a MOSAIC image though in this case this is simply the difference between the on source field and the reference field e As with the other calibration routines you specify the general calibration treatments e g CIA calib bs bs pds deglitch tcor dark flat calg CIA stab s90 e Then to create the beam switch MOSAIC you set the keyword bs CIA calib bs bs pds bs Of course as in the calib_raster examples all the keywords can be set at one call of calib bs For compatibility reasons the beam switch MOSAIC is stored in the field RASTER To view the result of your calibration you can use x3d and xsnr or simply CIA tviso bs pds raster 13 4 4 calib struct The general PDS is calibrated with calib struct This is in fact the same routine that calib raster calls to do calibration on a raster PDS The main difference between the routines is that calib raster has the additional capability to create a raster MOSAIC Read Section 13 4 1 keeping the following points in mind e As when using calib_raster you specify the type of calibration you want though of course there is no keyword raster with calib struct e g CIA calib struct staring pds deglitch tcor dark f
195. d cleaned sscds 1 Now we can discard our original SPD SSCD and save our sliced SSCDs CIA sscd write lw3 sscd dir scd dir CIA sscd write lw6 sscd dir scd dir CIA sscd remove spd sscd dir scd dir We begin by finding those SCDs we don t want with scd find and then deleting them We will choose to keep the LW6 CONFIGURATION CIA unwanted scds scd find fltrwhl 1w6 NOTFIND Searching for FLTRWHL NOT LW6 Found 22 occurrences CIA gt scd_del unwanted_scds We must delete also SCDs containing calibration data or data accrued when CAM is not in OP MODE OBS CIA gt unwanted_scds scd_find mode obs NOTFIND Searching for MODE NOT OBS Found 3 occurrences CIA gt scd_del unwanted_scds To avoid confusion it is best at this stage to rename the IDL variable containing the name of our modified SSCD CIA lw6_sscd spd_sscd The SSCD in memory is automatically modified when some of its SCDs are deleted CIA print sscd get nscd lw6_sscd 16 The final step is to save the modified SSCD that we have created It is best to keep this SSCD and the sliced SPD SCDs separate from the unsliced ones Create a new directory to hold the sliced data and for convenience create the following IDL string variables to hold their names CIA 1w6 scd dir DKA200 MDELANEY 14300601 scds 1w6 84 CHAPTER 12 DATA SLICING OR CIA lw6 scd dir home mdelaney 14300601 s
196. d depending on the severity of the saturation also subsequent exposures has to be carefully assessed pixel 14 17 is affected by saturation at SCD 4 with the average value 4095 00 value for End of Integration 3 readouts G 3 Information messages Such messages are for information only Users don t have to take any actions The following messages informs that only a best match for the calibration parameters could be obtained CCGLWOFLT_99060112290500 not exact matching for TINT 36 lt gt 15 record index 44 CCGLWDFLT_98031519384439 not exact matching for SWHL 220 88 PFOV 192 lt gt 448 36 lt gt 15 record index 14 The following message informs that for the dark correction the library dark instead of the modelled dark was used No SW model switching to library The following message informs that no distortion calibration for this optical configuration exists and no distortion correction was performed 343 344 APPENDIX G WARNING MESSAGES IN CIA can t correct distortion file The following message informs that a value indicated the current AOT isn t set This is often the case for calibration or parallel observations Unexpected AOCT value AOCT 0 I hope it is a calibration observation Appendix H Patched ASTROLIB and IDL routines in CIA xloadct wcssph2xy wcssph2xy xy2ad ad2xy extast putast fxbparse fxaddpar mrd struct mrdfits get_equinox original xlo
197. d by library flat field error Bottom Right Sky flat fielding and projection weighted by sky flat field error 262 CHAPTER 20 ADVANCED DATA CALIBRATION e Project the images contained in the FITS files and place the results in the file result fits CIA spawn cia_exect projection i2 input o result fits e Take a look at the result CIA result readfits result fits hdr CIA gt tviso result 0 20 15 4 Back projection The projection routines also allow the possibility to back project a raster MOSAIC to a set of EXPOSUREs A real source will due to the many effects present in the ISOCAM data have a somewhat different signal in each EXPOSURE The purpose of the data calibration is to minimize or eliminate this difference Knowledge of how the signal varies from EXPOSURE to EXPOSURE is not contained in the raster MOSAIC so the back projected EXPOSUREs will each contain an averaged source signal Figures 20 7 and 20 8 illustrate this point Note that this averaged source signal is really an idealized signal i e it assumes that the same source has the same signal in each EXPOSURE This assumption can be useful for testing the quality of the data analysis The less the signal in the original EXPOSUREs deviates from the idealized signal in the back projected EXPOSUREs the better the quality of the data calibration The back projection is performed with the routine back_project You should of course only perform back p
198. d due to mode 0 are rejected due to csh flag 2 are rejected due to qla flag In total 20 are accepted CIA print cleaned sscd CS8C058050040001 98052617314573 AT 48 CHAPTER 7 CVF OBSERVATION In contrast to the raster observation in Section 3 we here we have only one CONFIGU RATION and consequently only one SSCD As in Section 3 we need to create a PDS from the SSCD For CVF data the equivalent routine to do this is get_sscdevf Note that the resulting CVF PDS differs somewhat from a raster PDS CIA gt cvf_pds get_sscdcvf cleaned_sscd Now we can calibrate the data in cvf_pds Mostly everything that was discussed in the previous data analysis examples apply here with the exception that there is no MOSAIC for the CVF PDS As in Section 4 2 this data does not need stabilization correction Again as in the previous data analysis examples you don t have to execute all the commands below at once you may like to examine the data with x3d between each command CIA gt corr_dark cvf_pds CIA gt deglitch cvf_pds CIA gt stabilize cvf_pds CIA gt reduce cvf_pds CIA gt corr_flat cvf_pds An additional step that we will perform is the conversion of the EXPOSUREs from ADU to milli jansky mJy CIA gt conv_flux cvf_pds To view the results of your calibration use cvf_display CIA gt cvf_display cvf_pds The cvf_display Figure 7 1 window displays an EXPOSURE co added IMAGE and a pl
199. d first L2 UPGRADING CIA 1 0 STRUCTURES 349 from CIA data structures is incorrect by a 180 degree rotation Some programs are immune to this error saoimage and skyview for example because they use the FITS CDELT and CROTA2 keys to determine astrometry 1 To update your CVF PDS first you need to know the roll of the spacecraft during the observation in question To do this look in the IIPH delivered with your CVF observation data products and find the value of the key instroll You can do this by simple listing the contents of the IIPH file with VMS s type or if you work in UNIX you can use more or the CIA routine readhdr CIA print readhdr iiph03800232 fits key instroll iiph03800232 fits 229 970 2 Now run the routine struct update to fix the ASTR CD CIA new cvf struct update old cvf ro11 229 970 3 You may also want to correct FITS files you have created from CIA 1 0 CVF PDS To do this you will need to re export your files see Section 18 350 APPENDIX I UPGRADING OLD CIA STRUCTURES Appendix J Reporting problems and suggestions J 1 Problems with CIA software If you encounter problems with CIA which were verified by the local contact point then please submit a Software Problem Report SPR The template can be found in the following section or in the doc subdirectory of the CIA installation as spr txt Fill in the appropriate template and e mail it to helpdesk iso vilspa esa es P
200. d in automatic slicing are described in Chapter 16 However it is hoped that from the examples in the following sections the function of these routines should be clear 12 2 1 General slicing tips When using the automatic slicers the following steps can be applied to data of all types of observations 1 Make ERD SCDs from the ERD data products with erdtoscd and then convert them to SPD SCDs with erd2spd OR Make SPD SCDs from the SPD data product with spdtoscd 2 Save a copy of the unsliced SPD SSCD and its SPD SCDs They can be discarded when you are finished slicing 78 CHAPTER 12 DATA SLICING 3 At this stage you can choose to use the fast and less flexible sscd clean a Use sscd clean to split the unsliced SSCD into its component sliced SSCDs one per CONFIGURATION b Save each of the sliced SSCDs OR the slow and flexible scd find and scd del to slice the SPD SCDs a Identify the SPD SCDs which contain STATEs that you don t want to keep i e when CAM is not in OP MODE OBS Use info or scd find Delete the unwanted SCDs with scd del b If more than one CONFIGURATION exists in the AOT identify STATEs of a config uration you want to keep Delete the SCDs corresponding to the remaining STATEs Again use sscd info or scd find to help you do this c Save your sliced SPD SCDs by saving their SSCD with sscd write though in a different directory to where the original SCDs are stored
201. dark and background can be done in a CIA session with the following IDL commands CIA restore my_file xdr verb CIA on_source isodata cube isodata from 0 isodata to 0 CIA gt off_source isodata cube isodata from 1 isodata to 1 CIA gt source reduce_cube on_source CIA background reduce cube off source CIA image source background CIA tviso image 13 83 DATA CALIBRATION WITH X_CIA 109 13 3 3 6 Example 4 testing several methods for transient correction Let s assume that the user wants to try different transient corrections on the previous data see above Example 3 Hence the sequence of commands would be the following 1 Start a CIA session 2 Start x cia simply type x_cia on the CIA command line 3 Choose AOT Type default is raster scan 4 Data Load IDL File my file xdr 5 Process None 6 Transient SAP Model Fitting 7 Process Selected 8 Use all available functions of View and Tools menu to explore calibrated data 9 Transient 90 Of Final Flux measured 10 Process Selected 11 Use all available functions of View and Tools menu to explore calibrated data 12 Tools Hardcopy to create hardcopies 13 14 CIA Quit 13 3 3 7 Example 5 loading IDL data structure from memory If an IDL data structure is already in memory it is possible to load it directly into the x cia session by typing CIA gt x_cia data my structure The mod
202. data DARK DATA STRUCT CCGLWDARK_STRUC Array 5 CIA help dark data str Structure CCGLWDARK STRUC 17 tags length 8572 SPARE BYTE Array 36 TINT INT 2 EWHL INT 308 SWHL INT 112 GAIN BYTE 1 OFFSET BYTE 1 TEMPERAT FLOAT Array 10 TRMS FLOAT Array 10 TMIN FLOAT Array 10 TMAX FLOAT Array 10 VOLTAGE FLOAT Array 10 VRMS FLOAT Array 10 VMIN FLOAT Array 10 VMAX FLOAT Array 10 BLANK LONG 32768 BUNIT BYTE Array 12 IMAGE FLOAT Array 32 32 2 5Now that you have CDSs in memory as an alternative to manually playing around with the data structure refer to Section 15 3 3 for a description of the CIA routine cds_display 15 3 CALIBRATION DATA STRUCTURE CDS 175 303 0 308 8 508 6 308 4 308 2 308 0 TEMPERAT Array TRMS Array THIN Array TMAX Array VOLTAGE Array VRMS Array VMIN Array VMAX Array IMAGE H Image CDS Exit _ Figure 15 1 cds display window As you can see from the above no user friendly strings exist in the CDS as stated earlier calibration is handled internally by CIA so usually the innards of the CDS are not touched by a CIA user For completeness an example of how to convert the coded data into a readable format follows If we want to know the pixel unit of intensity it has to be converted to string CIA print string dark data 0 bunit ADU G s Furthermore data such as GAIN are in telemetry coded format and also have to be converted CIA p
203. de 2_ 80 Its value is defined as scd cube hk gla flag F2 QLA 2 x F2 0TF_FLAG 4 F220TF SUM From this definition a good value of the scd cube hk qla flag is 7 19 6 7 2 Enhanced OTF A button in the x handle slice window asks you if you wish to use the Enhanced OTF The default configuration is Yes This OTF relies on the OTF It makes a fit on the RA and DEC coordinates of the IIPH and finds out the best positions near the source it then flags them on To discriminate this method from the other the value 128 is added to the scd cube hk qla_flag Good values are then 143 You should not use this feature if you work with real micro scans see M RASTER comment or if your data has suffered telemetry drops This is because the fit of RA and DEC will not work their variations are too small between two raster positions and the routine may crash 19 6 8 Handling big datasets Sometimes you will face a huge amount of data in one file By huge we mean something like more than 5000 frames Since x slicer uses a large amount of memory for its widgets this reduces the available memory and introduces the risk of a crash when handling large data sets This problem can be avoided by use of the big file button option in the main x slicer window see Figure 12 1 Clicking on this button will cause x slicer to work in a mode where it minimizes the amount of data that it loads into memory e g when it needs to kn
204. del CSCD143006010105_96080110071423 CIA struct get sscdstruct sscd no qla flag 2 bad CSH flag The CSH flag will remain bad either due to a failure to uplink commands for ISOCAM and ISOCAM consequently did not reach its commanded position or more likely to an incorrectly generated CSTA file You can try to recover the data by using the no_csh_flag option of get sscdstruct CIA spdtoscd cispxxxxxxxx fits sscd CIA cleaned sscd sscd clean sscd CIA struct get sscdstruct cleaned sscd no csh flag strong saturation If spdtoscd warns you about saturation events see Section 19 2 the data has to be inspected carefully to assess the impact on photometry If it is only a short and mild saturation affecting only one pixel the data might still be usable In case of strong saturations affecting several pixels the observation is lost 272 CHAPTER 20 ADVANCED DATA CALIBRATION Chapter 21 Using SLICE within CIA This chapter explains how to use the long term transient correction or LTT and variable flat field correction or VFF algorithms implemented in SLICE which can significantly improve raster data reduction 21 1 Preface For the examples in Section 21 3 and 21 4 1 we assume that you are reducing a raster observation and that the data are currently stored in a PDS called data Section 21 5 presents a worked example in more depth Warning Figures 21 1 21 2 21 3 and 21 6 are best viewed in colo
205. dified reference ISOCAM dark current calibration report method library or method calg method CAL G DARK correction A DARK may be selected from the CAL G DARK library As you probably know by now the most up to date CAL G files are dis tributed in CDSs with the CIA system When you create a PDS from a SSCD get sscdstruct automatically uses find best see Section 20 12 to place the most suitable DARK from the latest CIA DARK CDS into the PDS field CALG DARK see Section 15 3 2 With this method CALG DARK is applied to the CUBE called routine darklibrary PDS side effects The field DARK is filled with the DARK used for correction The IMAGEs in CUBE are dark corrected i e CUBE is modified reference ISOCAM Handbook Chapter Data processing methods Section CALG dark 20 2 CORE CALIBRATION 233 4 indark my_dark method Correcting with your own DARK If you have your own DARK obtained for example from CAM in OP MODE DARK then it can be supplied to corr dark via the keyword indark This DARK would then used to apply correction in the normal way called routine N A PDS side effects The field DARK is filled with the DARK used for correction The IMAGEs in CUBE are dark corrected i e CUBE is modified reference N A 20 2 2 Deglitching The core calibration routine deglitch serves as an interface to all the available low level deglitch ing routines As for the other core calibration routines you
206. dir DKA200 MDELANEY P0007780 14300601 OTHERS COGLWDAR If you are using UNIX the notation is CIA calg dark dir home mdelaney p0007780 14300601 others CCGLWDAR 15 3 CALIBRATION DATA STRUCTURE CDS 173 CAL G file CDS mnemonic description basic calibration libraries CSCGCROSS CROSS SW noise cross talk decorrelation matrices CHCGCONV house keeping interpolation values CCG DEAD DEAD dead pixel map CCG DARK DARK dark current exposures CCG DFLT DFLT detector flat field library CCG OFLT OFLT optical flat field library CCG SPEC SPEC filter amp CVF spectral data CSWCVF SWCVF SW CVF description CLWCVF1 LWCVFI LW description CLWCVF2 LWCVF2 LW CVF2 description CCG SLP CVF spectral line profile CCG PSF PSF point spread function library IFPG IFPG focal plane geometry GAIN gain conversion table CWHEELS CAM wheels information table ORBIT ISO orbital parameters higher level calibration libraries AMDIST polynomial distortion coefficients XFLT distortion flat CCGLWDMOD LWDMOD parameters for the time dependent dark corrections CCG TRANS TRANS model transients CCG LINEAR LINEAR linearity correction library CCG FRAME FRAME detector astrometric calibration CCG GLITCH GLITCH glitch Model CCG STRAY STRAY non dark local light model One CAL G file or CDS exists for each CAM detector replace with LW or SW for full name See Section 15 4 for an alter
207. ds exist in general the routine reduce reduces the IMAGEs as follow 1 2 3 4 The sets of IMAGEs corresponding to individual STATEs are extracted by reduce from Each set of IMAGEs along with MASK is passed to reduce cube reduce cube averages the IMAGEs to an EXPOSURE taking MASK into account pixels which have been flagged because of instability glitches etc ignored The EX POSURE its RMS image and the weight image are returned Each pixel of the weight image equals the number of IMAGE pixels which have been averaged to create a corre sponding pixel in the EXPOSURE The EXPOSURE its RMS image and the weight image are placed into the PDS fields IMAGE RMS and NPIX respectively The default averaging method is to take a mean of the IMAGEs Other available methods are described below 1 median method Takes the median of the IMAGEs to form the EXPOSURE routine called la median PDS side effects The PDS fields IMAGE RMS and NPIX are filled 238 CHAPTER 20 ADVANCED DATA CALIBRATION 20 2 5 Flat fielding Currently several methods of flat fielding all of which are handled by corr flat are available In general corr flat performs flat fielding as follows 1 2 3 Looks at the keywords method and inflat to determine how the FLAT is to be selected Divides each EXPOSURE of IMAGE with the selected FLAT or if the corr_flat keyword cube is set then the
208. e of sso pds image CIA sso pds image 0 reduce cube sso pds cube mask sso pds mask Flat correction finishes off this reduction step CIA corr flat sso pds So to view the results of your calibration use xdisp or tviso specifying the first frame of sso_pds image as input CIA tviso sso pds image 0 You should see the same image as in Figure 5 1 The effects that we saw earlier in Step 7 with x3d smearing and pointing inaccuracies will have combined to make the comet look somewhat blurred in this final MOSAIC image Now we can convert the MOSAIC image from ISOCAM units into milli janskys mJy CIA conv flux sso pds As in the end of Section 3 2 you can save the data using IDL s SAVE CIA save file sso pds xdr sso pds Export to FITS may be performed with CIA s imagette2fits This routine will place the data in the field sso pds image and hence the MOSAIC image into the primary array of the FITS file CIA imagette2fits sso pds name tempel tuttle fits rank 0 We use the keyword rank to specify the MOSAIC image i e the first frame of sso pds image This image is written to the file tempel tuttlel fits An improved method to analyse solar system observations can be found in section 20 7 5 2 DATA ANALYSIS 39 Figure 5 1 tviso display of the MOSAIC image from a solar system object observation 40 CHAPTER 5 SOLAR SYSTEM OBJECT OBSERVATION CAMOI Chapter 6 Beam sw
209. e sso_pds cube In this observation the data does not need stabilization correction however feel free to experiment with this CIA gt corr_dark sso_pds CIA gt deglitch sso_pds Now we have a nicely calibrated PDS You might want to check this with x3d This time you can click on the button mask to see which pixel have been masked by the calibration routines 38 ro 10 CHAPTER 5 SOLAR SYSTEM OBJECT OBSERVATION CAMOI The final step in the data reduction is to create the MOSAIC image This is really just the average of all the IMAGEs in sso pds cube and we can easily derive it with CIA s reduce cube Before doing so we must do a fake reduce as we normally do when reducing a raster PDS BS PDS or CVF PDS This is really just to keep the PDS consistent with the data reduction step by making sure important fields such as sso_pds image_unit are updated CIA reduce sso pds Now you can take a look at the EXPOSURE obtained for each SCD or pointing of the spacecraft These images should look very much alike all will be slightly smeared This smearing is due to steps ISO makes as it re points i e the tracking is of course not smooth You will also notice that the position of the object may be slightly different from EXPOSURE to EXPOSURE This second effect is due to the inaccuracies of ISO s pointing CIA x3d sso pds Now we will do the real reduce manually The MOSAIC image is stored in the first fram
210. e Flat Field method with parameters set as in Table 21 3 In this observa tion using 1tt_thresh 6 is quite satisfactory In that case the commands to issue to SLICE are for the LW3 case CIA param set red param tdt 65801627 flat smooth window 4 nplanes 60 CIA flat thresh 10 1tt thresh 6 CIA act set act ltt CIA slice_pipe Figure 21 5 plots the long term transient corrections that are derived with this setup On Be aware that putting the two actions in your call to set_act will not work at the 1tt action is executed before the make map one 21 5 A WORKED EXAMPLE 285 0 05 0 00 0 05 0 10 0 200 400 600 800 Figure 21 4 An example of artifacts obtained with a incorrect long term transient determination the signal oscillates and the number of complete oscillations is roughly equal to half the number of raster legs The dashed line represent the fitted correction see text for details In fact these artifacts were generated while using the DivSky flat field method in the long term transient determination for the LW3 image The LW2 image shows similar problems the graph there are two curves a continuous one and a dashed one The continuous on is the exact correction as derived from the data It will by definition see M A Miville Desch nes paper end with a zero value i e no long term transient correction for the last raster position The dashed one is a fitted correction Most
211. e called useful below In the banner of the cube analysis window there are three scrolling menus e File New window creates a new window which is a copy of the cube analysis window and stays when you quit this widget for further comparison Save saves the cube analysis window as a PostScript or encapsulated PostScript file Quit closes this cube analysis window and unselect the raster button e Plot Single in the window there is only one graph The whole time history of each useful pixel camera is plotted The part of the time history which contributes to the sky pixel is bold scd limits are drawn around this part and the scd number is written at the beginning of this part All the plots are on the same graph overplotting This can be useful when there is a gain variation Multi The whole time history of each useful pixel is drawn each on a different plot The part of the time history which contributes to the sky pixel is bold scd limits are drawn around this part Compare for each useful pixel camera only the part which contributes to the sky pixel is drawn The plots are put one after the other on the same graph the abscissa is not the real time e Options Bad Pixel interpolation toggle button pixels flagged as bad are interpolated by the neighbour valid pixels or kept as it Flat Correction toggle button during standard data reduct
212. e data are delivered in the CISP file ISO CD ROM users can find this file in the directory products pmmmmmmm nnnzxayy 9 3 3 Automatic Analysis Results AAR The AA processing produces several data products CCIM CMAP and CMOS files primarily contain processed images The remaining files contain by products of the AA for example CGLL contains a list of glitches detected by AA processing ISO CD ROM users can find this file in the directory products pmmmmmmm nnnxrryy e Primary AAR level data products containing ISOCAM images CCIM Contains AA computed EXPOSUREs in detector coordinates CMAP Contains AA computed EXPOSUREs in astronomical coordinates CMOS CAM Mosaic Contains MOSAICS constructed by AA from EXPOSUREs con tained in the CMAP data product of the same CONFIGURATION e By products of AA processing CUFF CAM User friendly File Contains a log of messages from AA processing CGLL CAM Glitch List Contains a list of AA detected glitches CJAM CAM Jitter Memory and Stabilisation information CPSL Contains a catalogue of AA detected point sources CSSP Contains the measured spectrum for each point source detection Used in multi filter observations only Refer to the ISOCAM Handbook and the ISO Data Product Document for a comprehensive account 3Currently x_slicer see Section 12 3 uses the CSTA data product though it will work without it CAM parallel ERD data are delivered in the CPER file and CAM Diagnostic
213. e fft Computes the Fourier transform and display either the power spectrum the phase the real part or the imaginary part e pan the image The zoom factor can be 2 4 8 1 2 1 4 1 8 e ROI select Interactively select a region of the displayed image e ROI save Save the selected region as as a FITS file 14 8 2 D IMAGE ANALYSIS 123 40 50 10 45 27 6 55 56 48 3 1 4 19268 Figure 14 4 The xdisp window 124 CHAPTER 14 IMAGE ANALYSIS AND DISPLAY 14 3 1 2 xdisp examples 1 Following the examples of Section 12 2 we can examine the raster image in the raster PDS Figure 14 4 was created with this example CIA xdisp lw6 raster raster raster2hdr lw6 raster In the above example the routine raster2hdr is used to create a FITS header from the PDS Iw6 raster and this header is passed directly to xdisp This is particularly useful for getting coordinates of raster MOSAIC pixels see description of menu option get cursor in Section 14 3 1 1 We can also use the select and save a selected region of the image to a FITS file see description of menu option ROI Since we have supplied a FITS header to xdisp then the FITS file will contain astrometry for the selected region For example click on ROI select and use the mouse to drag the selection zone over the region you are interested in saving watch the screen for information After selection is completed click on ROI save to save the select
214. e keeping parameters These are listed in the table below The index i refers to the HK parameters for the i th IM AGE FRAME pair As an example if you wish to obtain the UTK of the first IM AGE FRAME pair in the SCD you could try the following CIA print scd get hk 0 utk scd of interest subfield description K iDMA COUNTER Rank of IMAGE FRAME K i UTK Uniform Time Key of frame K i QLAFLAG Quick Look Analysis flag K i NGLITCH number of glitches detected in IMAGE K i denotes stability of IMAGE FRAME 0 unstable 1 stable K i D jitter offsets in spacecraft x axis K i D jitter offsets in spacecraft y axis CAL Structure containing original encoded parameters taken directly from the data product FITS file see ISO Data Product Document The following example displays the contents of CAL CIA cal scd get cal CSCD143006010214 96090420275809 CIA help cal str Structure CAL STRUC 2 14 tags length 28 TYPE INT 33 COMMANDER INT 1 TELEMETRY INT 100 DEID INT 1 MODE INT 1 OBC INT 0 TINT INT 36 EWHL INT 308 SWHL INT 88 PFOV INT 360 FCVF INT 275 GAIN INT 1 OFFSET INT 1 SPARE INT 0 With the exception of CAL OBC see Section 15 2 2 3 these parameters are of little use to the typical user Their values as they appear in CAL are encoded CIA decodes these values into a user friendly and readable format In some cases the parameter name itself is changed to make it more intelligible e g FCVF is F
215. e main menu are e CIA ISOCAM e CIA Log File e CIA Help l Taken from Claret A 1996 ISOCAM Data Analysis with _ v2 2 Sections 4 amp 6 297 298 CHAPTER 23 CIA Info CIA Quit Data Load SSCD Data Load IDL File Data Save SAD Data Save SAD Fits Data Save IDL File Data Change Sign Data Reload Original Data Data Display History Data AOT Info View Cube Temporal Mean View Cube Temporal Analysis View Cube All Frames View Mask Temporal Mean View Mask Temporal Analysis View Mask All Frames View Image Averaged Frames View Image Mean Of All Frames View Cal G Dark Current View Cal G Flat field View Cal G PSF View Cal Used Dark Current View Cal Used Flat field View Result Reconstructed Raster Map View Result Monochromatic CVF Frames View Result All CVF Frames View Ra Dec Roll Info View Change LUT X_CIA REFERENCE GUIDE 23 2 HELP ON X CIA Dark None Dark User Input Dark Cal G Dark Model Dark VilSpa Deglitch None Deglitch Manual Deglitch Particle Impact Deglitch Temporal Deglitch Spatial Deglitch Temporal amp Spatial Deglitch Multiresolution Median Transient None Transient 9096 Of Final Flux modelled Transient 9096 Of Final Flux measured Transient IPAC Model Fitting Transient IAS Model Fitting Transient SAP Model Fitting Transient
216. e raster steps are less than half the array which ensures high redundancy In principle SLICE does not need a very high level of redundancy but you will be better off with high numbers typically when the median overlap factor is higher than 3 4 Figure 21 1 shows the raster maps obtained with a standard CIA procedure dark correction using the dark model deglitching with the multiresolution method transient correction with the Fouks Schubert model and flat field correction using the auto method i e the flat field is derived from the data As can be seen clearly on the figure we are suffering from periodic patterns which are due to the use of a single flat for the whole raster and also from long term transients both in LW3 and LW2 On can also see that due to the strong flux decrease between the LW3 and LW32 raster we have a falling then increasing long term transient in the LW2 raster At that stage we are ready to use SLICE An important point to understand is that the variable flat field and the long term transient are not independent effects one must have no flat field residuals in order to accurately determine the long term transient and one would need ideally a long term transient corrected data set to pin point variations due to the flat field Since 21 5 A WORKED EXAMPLE 279 determining both simultaneously is not yet possible we are going to use an iterative process In most cases very few iterations are necessary This i
217. e time plot linear linear linear log etc You can choose minimum and maximum of intensity and the unit of time seconds or frame index You can choose the scale of the intensity of displayed cube linear logarithmic by histogram equalisation and the minimum and maximum of intensity You can put the plot in a different window and save it as postscript file You can put the image in a different window and save it as postscript file e Upgrade to the latest SLICE version SLICE V1 2 New features of this SLICE version are actions on bad pixels ghost and sources have been integrated into one task the docube option permits now to perform operations which were previously per formed on raster positions can be now done on individual frames This is quite interesting for bad pixels the error map is now generated by default e The faint source detection tool PRETI is now distributed with CIA Calling syntax is CIA reduce faint source cisp_file fits raster e Improvement of display routines for tviso the min_real and max_real keywords work now properly and accept also 0 as input Furthermore a new keyword source list was added which puts a star on the positions given by source list ximage uses now the distortion correction on the sky view pixel history of a raster e Improved functionality of sscd clean 340 APPENDIX F WHAT IS NEW IN CIA 5 0 The new keyword min scd rejec
218. e user to select which frame to display as well the range of intensity to use The panel at the bottom left of the display region is used to select the beam The left panel in the display region shows the current frame in the reduced data cube The right panel highlights the pixels currently in the aperture In both panels clicking mouse buttons has the following effects mouse LEFT Selects the pixel and makes it part of the aperture The right panel highlights the pixel If the pixel is already selected clicking on it has no effect mouse RIGHT Removes the pixel from the aperture mouse MIDDLE While held down shows the spectrum of the current pixel in the plotting window and the astrometry information in the bottom panel see below When released the previous spectrum if an aperture has been selected is displayed again The last part of the GUI displays the following information about the current pixel selected 1 its row and column on the array 2 its intensity in the currently displayed frame 3 the equatorial co ordinates corresponding to the pixel s spatial location in J2000 equinox Additional information and buttons in the last row are 1 the RESET APERTURE button which resets the aperture and 2 the number of pixels in the current aperture 14 2 2 2 Defining and editing apertures First you need to the select the beam either sky or source The apertures are defined by pressing the LEFT mouse button on the pixel whic
219. e ye EOE eh GAS i ee ee EC 97 131A BS PDS iso arte ah Bootle Boies Ws noe RR SUE 98 13 125 CVF PDS sume Gt eee ERREUR ae eG 98 13 2 Calibrating the PDS 100 13 21 Corevcalibration 100 13 2 2 Raster MOSAIC creation 101 13 23 Staring analysis e x Ges Rok ge4xap ugs Soh eto ee X Ed BS 101 13 2 4 Beam switch MOSAIC creation 101 13 2 5 CVE analysis xor os ee Re ee ees 102 13 3 Data calibration with _ 102 13 31 Introduction inu eats Sie ee YA ee ee 102 13 3 2 Quick Look analysis with x_cia lens 102 13 3 3 Guidelines for using x cia 0 103 19 9 4 X cla caveats vut es pog dope Rig aon DI UR ote d ae P dn 110 13 4 Calibrating a PDS the old way 110 13 4 1 calib raster m Latet a Rus a 110 134 2 calib evE Re EROS Mee wise X for enki 111 19 413 calibEbs xc see ee RIS qe Ae bos RED RAE 112 13 4 4 calibzstr ct uae ER RS 112 14 Image analysis and display 113 14 1 General analysis routines 113 14 1 Estimating S N in a cube or image 113 14 1 2 Energy radial profile aoaaa 22e 114 14 1 3
220. ean the SSCD Cleaning the SSCD is a difficult task that must be done manually A single polarization obser vation may be a combination of multiple rasters with changing entrance wheel at each raster point id or each raster position The steps for this observation are as follows 1 First list all the SCDs in the observation CIA gt sscd_info sscd pol 50 SCDs in the SSCD CSSC356005010001_00020117264901 seq entwhl mode fltrwhl pfov tint gain size m_raster n_raster 0 HOLE OBS LW2 6 0 25 20 1 1 1 1 1 HOLE IDLE LW2 6 0 2 10 1 11 1 1 2 HOLE OBS LW2 3 0 2 10 1 102 1 1 3 HOLE OBS LW2 3 0 2 10 1 92 2 1 4 HOLE OBS LW2 3 0 2 10 1 92 2 2 5 HOLE OBS LW2 3 0 2 10 1 97 1 2 6 HOLE OBS LW2 3 0 2 10 1 13 1 1 7 POLARIZOR 1 OBS LW2 3 0 2 10 1 78 1 1 8 POLARIZOR 2 OBS LW2 3 0 2 10 1 77 1 1 9 POLARIZOR OBS LW2 3 0 2 10 1 1 1 1 10 POLARIZOR OBS LW2 3 0 2 10 1 78 1 1 11 POLARIZOR 1 OBS LW2 3 0 2 10 1 78 1 1 12 POLARIZOR 2 OBS LW2 3 0 2 10 1 77 1 1 8 2 DATA ANALYSIS 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 POLARIZOR POLARIZOR POLARIZOR POLARIZOR POLARIZOR POLARIZOR POLARIZOR POLARIZOR POLARIZOR POLARIZOR POLARIZOR POLARIZOR POLARIZOR POLARIZOR POLARIZOR POLARIZOR POLARIZOR etc KB OBS OBS OBS OBS OBS OBS OBS OBS OBS OBS OBS OBS OBS OBS OBS OBS OBS LW2 LW2 LW2 LW2 LW2 LW2 LW2 LW2 LW2 LW2 LW2 LW2 LW2 LW
221. ecsi gns Rea oxoxROR gsx Kk ded 296 23 x cia reference guide 297 23 T Advanced use ob dL E GR ORO DEG ys XS eee ee Eee RB 297 23 1 1 Executing IDL commands from within x_cia 297 23 L2 Buffer variables 297 23 2 Helpczon x us deut erue et a ue wel ea ed 297 23 2 1 hist otcommandss 23 o Ea RIAM QU RIS ote ue Det 297 23 2 2 Short description of commands 300 A Glossary 305 B H CONTENTS CIA command short list 313 1 Data preparation slicing 313 B 2 Data calibration le ex eh ae Neh eee Ro Rud dus 314 B 3 Data visualization 42x o3 od od 315 B 4 FITS input output routines 316 epa erm eg RR ee d Re Ro piada Geen d 317 The ISO CD ROM 319 C Mounting the CD ROM oso xm Ru ee RE RR 319 CUL Sided sake de Uu E UAE Re REUS RE Eo SS REDE B RON 319 GE UNIX ee Arr Dd dte cg ERE S 319 C 2 Overview of the CD ROM Contents 319 C 2 1 Where to find the ISO documents 320 C 2 2 Where to find the Data 320 C 2 3 Where to find nice images 1 321 Guidelines for writing CIA routines 323 Del Introductions
222. ed To prevent rebinning set the keyword norebin Chapter 23 x cia reference guide This chapter completes the description begun in Section 13 3 of x cia It includes a description of the advanced features of x cia and a useful reference guide to x cia s commands 23 1 Advanced use of cia Two advanced features of x cia are described in the following sections 23 1 1 Executing IDL commands from within x cia It is possible to execute directly an IDL command in the CIA Command Line window of the x cia screen You will be prompted in the message window if the command has been correctly executed Note that the result of the IDL command is visible either in the IDL window from which x_cia was launched or in a graphics window This was mainly designed to display or plot some interesting fields of the PDS structure without quitting x cia session It can also be useful to modify manually some fields of the PDS structure such as the coordinates or the mask See Section 15 5 for more details on the architecture of the PDS 23 1 2 Buffer variables Three variables can be used as buffers DUMMY TEMP and OLD At the beginning of the session they are initialized to zero but can contain any kind of data including part of the PDS structure Temporary results can be stored in them in order for example to plot the difference between the latest result and an older one 23 2 Help on x cia 23 2 1 List of commands Available functions of th
223. ed in RESPONSE i SENSITIV When a CVF PDS is initially created with get sscdcvf the sensitiv ity and straylight correction factors are taken from a CDS and placed in RESPONSE SENSITIV This is similar in principle to the way the CAL G FLATs and DARKs are handled conv flux applies the correction and converts the EXPOSURE pixels to mJy CIA conv flux cvf pds image The EXPOSURE pixels are now converted mJy The CVF PDS field IMAGE UNIT is updated to reflect this Note that this operation is not reversible To re perform sensitivity correction use reduce to recreate the EXPOSUREs and hence refill IMAGE 20 6 2 Photometry on faint point sources For a CVF observation the source positions moves depending on the wavelength slightly within each CVF segment For changes from one segment to another there are two LW segments or a switch of the detector channel the source position can jump several pixels This behavior makes photometry on CVFs more difficult To overcome this compute spectrum computes for each wavelength or CVF step for a point source the best fitting of a PSF and uses this information to compute a shift corrected flux spectrum After a the CVF observation has been fully reduced and the approximate position of a point source has been determined e g by cvf display cvf spectrum can be called CIA compute spectrum cvf pds x pos y pos psf dir SAPIO1 DKA200 CIA DATA PSF sort wave flux est flux est fl
224. ed region You will get a screen message like File ima33x33 fits written The FITS file can be used as input to isocont see Section 14 6 2 2 Using xdisp for viewing FITS data in CIA CIA haro3 image readfits haro3 iso fits haro3 header CIA xdisp haro3 image haro3 header 3 xdisp is designed to directly read SADs from disk or memory and display the contents of the CCIM DATA field CIA xdisp CSAD143006010205 96090620365354 14 3 2 sad_display and struct2sad In section Section 10 2 we used sad display to create SADs from CCIM and CMOS data products and then display their corresponding EXPOSUREs and MOSAICs We were only concerned with viewing AAR data products at that time However since we can create an SSAD from a PDS using struct2sad we can use sad display to view the results of our own calibration Here are some examples 1 Any PDS can be converted to an SSAD and viewed with sad display CIA sad display struct2sad any pds all The optional keyword all tells struct2sad to make a complete as possible set of SADs If any pds is a raster PDS or BS PDS then clicking on the button future see Figure 10 1 will display the raster or beam switch MOSAIC 2 If you only want to see the raster or beam switch MOSAIC then set the keyword raster CIA sad display struct2sad raster pds raster 14 4 CUBE ANALYSIS 125 CANCEL APPLY Figure 14 5 xselect frame window 14 4 Cube
225. egularly spaced pointings on the sky allowing a MOSAIC to be constructed of a region of sky greater then CAM s FOV Reference fields See beam switch observation RESET This refers to CAM CCD read out i e FRAME taken before integration begins see MOSAIC revolution number An orbit of ISO The Revolution Number identifies a particular orbit SAD Science Analysed Data A CIA data structure used to hold a CIA computed EXPO SURE see origin SAD or MOSAIC see future SAD or similar data from the AAR data products saturation Occurs when a CAM detector is exposed to a bright object and pixels reach their full well capacity Sensitivity of affected pixels can seriously be altered for succeeding observations See entry for CLEAN SCD Science CAM Data Comes in two flavours ERD SCD and SPD SCD The former is a data structure used to hold EOI and RESET FRAMES from the CIER and the latter to hold either CIA computed IMAGEs or IMAGEs from the CISP slicing Slicing data refers to the CIA process of producing SPD SCDs one per STATE and SSCDs one per CONFIGURATION from the bulk AOT data contained in the CISP or CIER data products SPD level Standard Processed Data SPD level refers to the pipeline level of processing that produces SPD For CAM SPD is synonymous with CISP SSAD Set of Science Analysed Data A data structure cataloging SADs from an AOT or any subset of an AOT e g CONFIGURATION SSCD Set of Science CAM Data
226. el 20 15 ADVANCED PROJECTION 265 20 15 5 Distortion correction for staring beam switch and CVF observations Using the routines project struct project bs and project_cvf also the exposures of staring observation can be corrected for distortion In the following example we discuss distortion correction for a staring observation Beam switch and CVF observations are treated in a similar way Assuming the EXPOSURE in has been generated the usual way it will have 32x32 pixels see left picture in Figure 20 9 CIA help staring pds image lt Expression gt FLOAT Array 32 32 Distortion correction and magnification by the factor 2 will be performed by the following command CIA project struct staring pds magnify 2 The resulting EXPOSURE has now 67x65 pixels CIA help staring pds image lt Expression gt FLOAT Array 67 65 Additionally the original fields IMAGE RMS and NPIX were renamed to OLD IMAGE OLD RMS and OLD_NPIX while RMS and NPIX contain now information corresponding to IMAGE As usual export to a FITS file is done with CIA s imagette2fits This routine will place the data in the PDS field IMAGE into the primary array of the FITS file Figure 20 9 Comparison of standard vs distortion corrected staring observations Left Standard 32x32 EXPOSURE Right Distortion corrected and magnified EXPOSURE 67x65 pixels Due the rebinning of the pixels the flux per pixel is reduced by the factor 4
227. eld was performed using the DivSky method with parameter setup as indicated in Table 21 4 LW3 is on the left and LW2 on the right Compare with Fig 21 1 to measure the improvement 21 6 1 Removing bad pixels The method used to identify remaining bad pixels while protecting the sources a problem in previous versions of SLICE is quite simple 1 2 in the docube option the cube is flat fielded and a local cube of raster positions is created a local copy of the sky map is created and is smoothed successively with a window of 5 7 and size_filter pixels This smoothed map is used to identify and remove ghosts in the flat fielded datacube If the error map has not been provided see below then a mean error level is computed at that stage Each plane of the cube is projected on the sky and for the defined pixels the difference map is computed between that image and the smoothed map created above We now have a cube of residuals four sky maps are created summing the residuals that fall in the following intervals above a given positive threshold bin 1 between 0 and this positive threshold bin 2 between 0 and a given negative threshold symmetrical from the positive one bin 3 and below this negative threshold bin 4 The number of readouts that fall in these four bins is also computed An empty sky map is created and for each sky pixels and if more that two thirds of the readouts fall in bins 3 and 4 a
228. enerated with the commands CIA tviso raster pds npixraster CIA tviso raster pds rmsraster To obtain the standard error of the MOSAIC image a measure of the quality of the calcula tion of the mean in this case the quality of calculated MOSAIC pixel values we simply divide the MOSAIC image by the square root of the MOSAIC weight image CIA raster std err raster pds rmsraster sqrt raster pds npixraster Note however that this is for purposes of illustration only unlike reduce we do not take account of unusable pixels in the MASK 268 CHAPTER 20 ADVANCED DATA CALIBRATION Figure 20 10 The RMS image that correspond to Figure 3 3 wl 5 Et ip T Cum E lt j amp amp ak i a 4 a E 4 M gt me a p s E p pu E E e A b b d D a a e PLATE Figure 20 11 The weight image that correspond to Figure 3 3 20 18 HOW TO SAVE SPOILED OBSERVATIONS 269 20 18 How to save spoiled observations The data of some observations are spoil and therefore resists standard CIA data reduction The most common reasons are 1 telemetry drops 2 bad raster point IDs 3 target not acquired 4 bad QLA flag 5 bad CSH flag 6 strong saturation CIA provides the functionality to recover some of these observations However only experi enced CIA users should try these steps
229. ent pixel or a vertical cut of the column of the current pixel The current pixel is changed simply by clicking on the image and the type of plot displayed in the plot window is changed by choosing one of the buttons temporal cut horizontal cut or vertical cut If the x3d input is a PDS or the CUBE and MASK are explicitly supplied then the masked pixels in each image of the CUBE may be viewed Clicking on the button mask will activate this feature In the plot window all the masked pixels will be marked with an and in the image display window masked pixels will not be displayed i e will show up blank in the image A newly added feature of x3d is the button glitch see Section 14 4 6 14 4 6 x3d asa calibration aid x3d is an excellent tool for examining cubes of CAM images and looking at the temporal and spatial behaviour of a pixel in CAM observation It can even perform manually deglitching This makes it a very useful calibration aid To preform the functions described below x3d should be given a PDS or a CUBE and MASK as input e Dark correction In a PDS that has not been dark corrected a regular pattern of dark and bright lines are apparent in the IMAGEs in the CUBE When viewing the CUBE with x3d click on the button vertical cut to display a plot of pixel intensity against image line 14 4 CUBE ANALYSIS 133 pen Frame Number 1 LORD LUT ww Horizontal cut Quit 6 Vertical cut Mindow Size J Glitch
230. er clearly displays all the states within an AOT and their associated parameters presenting you with an overview of how the observation had been performed 12 3 1 Starting x slicer In order to produce your first SCDs all you need is to 1 Start a CIA session 2 Invoke x slicer CIA x slicer The Figure 12 1 shows the first window of the x slicer The top half of the window displays information about the files that you are slicing the bottom part allows you to choose the variables according to which you will slice your file and to ask the slicer to perform some actions Let s go on for our first slicing It will be divided into a few steps e Load the file to be sliced Taken from Aussel H 1996 ISOCAM Data Preparation with X slicer v2 1 Section 2 you want to force x slicer to choose the current directory as the default location it searches for data then type CIA x slicer here Alternatively VMS users may redefine arc dat define arc_dat 12 8 DATA SLICING WITH X SLICER 87 File to Slice c ier47400671 fits Directory Starting at Ending at END 11 Units NONE File on off Attitude File 1 1 fits Instrument CAM Directory edrom cdrom0 Automatic Find on off IFPG File NEEDED Directory Use CDS if needed On Off Automatic Find On Off Com
231. er documents exist in parallel directories but these are only relevant to other ISO instruments C 2 2 Where to find the Data Products The file datalist txt is an important reference document making a hardcopy for future ref erence is advisable In addition to some archive information such as data product owner and lot ID number it contains a list of all the data products using their official names along with their version number and their location on the CD ROM see an sample listing of datalist txt below Chapter 9 explains what the different types of data products actually contain and what the official filename convention is Here we will merely point you to your data Owner identification ID VMARINI Name Dr Valerie Marini Address ESA Vilspa P 0 Box 50727 City 28080 V d Castillo Madrid Country Spain Lot identification Serial number 601210411878 Lot ID 123 Seq NR 12 CD label I0012312 Product sets Product set NR P0007780 Proposer ID VMARINI aocs14300601 0577 products p0007780 14300601 aocs fit ccglwdark 0481 products p0007780 others ccglwdar k fit ccglwdead 0443 products p0007780 others ccglwdea d fit etc The directory products will contain one subdirectory per observation or TDT of the form products pmmmmmmm where pnmmmmmm refers to the product set number Each of these subdirectories contain further subdirectories of the form products pmmmmmmm nnn prod uct
232. es an entire CIA data structure to a FITS file The data is stored in extensions CIA struct2fits raster pds name raster_archive fits fits2struct recovers the output of struct2fits It initializes the appropriate PDS and then fills it as best it can It is very useful for upgrading the architecture of an obsolete PDS CIA fits2struct raster_archive fits hdr raster pds recovered B 5 ONLINE HELP 317 B 5 Online help The various methods of getting online help are listed here cia html invokes a HTML based CIA help ciainfo is an alias for the old style widget olh where the headers of all CIA s IDL routines may be found Note that alias must be compiled before attempting to invoke ciainfo cia help invokes a searchable dedicated CIA help invokes IDL s hyper help 318 APPENDIX B CIA COMMAND SHORT LIST Appendix C The ISO CD ROM This appendix provides a guide to mounting and using the ISO CD ROM It is not applicable to observers who have obtain their data via IDA It shows you where your data the documents listed in Section 1 2 and some much less technical information nice JPEG images of ISO and its instruments can be found on the CD C 1 Mounting the CD ROM Mounting the CD should be a simple job The commands below should work on your system Your system administrator can provide the devicename and advice if you have any problems 11 VMS To mount the CD ROM on a VMS machine try the fol
233. es when your data are very old coordinates are given but no information about which instrument they describe appear in the IIPH You then have two choices work on your data hoping the coordinates are CAM ones or wait until your data are reprocessed with a more recent version of the pipeline If you choose the first case and find a big discrepancy i e more than 30 arcsec between ISO coordinates and what you asked for you can be sure that your coordinates are not CAM ones In all cases remember that even if ISO is telling you that it didn t observed where you asked for it has truly observed it The next fields give the name and directory of the Compact STAtus file CSTA again see Chapter 9 With this file the slicer checks if all commands sent to the camera during your observation were well followed l he same remarks about Automatic Find apply here too The last fields of the top part of the window display the name of the orbit file This file contains the position and speed of the satellite around the Earth The same remarks about Automatic find apply here too 12 8 DATA SLICING WITH X SLICER 89 12 3 3 Selecting slicing variables The next step to slice a file is to select the variables according to which the slicer will slice the file For each record in your file the slicer will check the chosen variables If their value has changed it will flag the record and propose to build a new SCD starting from this record You can
234. etical PSFs for all possible optical CONFIGURATIONS of CAM along with the observed PSFs described in Section 15 4 2 can be found in the subdirectory psf at the official CIA ftp site In order for CIA routines to find these PSFs they should be placed in the directory cia_vers data cds The PSFs same may be shared between all CIA versions at your site by using soft links The theoretical PSFs are derived from the theoretical camera model the model includes the transmission profile of a particular filter by assuming a flat spectral shape of the source Note that no stray light features have been considered Though originally generated for use by xphot many observers have found them invaluable for advanced analyses such as jitter correction Section 22 1 and source simulation The theoretical PSFs are stored as IDL save sets and with filenames of the following format Fixed filters e g psf_lw5_6p0as save pst filter pfov save CVF filters e g psf_9p5mic_3p0as save psf wavelength in microns pfov save Extracting a PSF can be easily done with the aid of make psf name Just give this routine a PDS and it return the correct filename for the appropriate PSF library CIA restore make psf name raster pds verb RESTORE Portable XDR SAVE RESTORE file RESTORE Save file written by aussel challenger Sat Nov 29 23 24 43 1997 RESTORE IDL version 5 0 sunos sparc RESTORE Restored variable PSF If you don t have a PDS t
235. even though you have corrected for short term transient there is still a transient artifact at the start of the raster In that case it is better to completely mask the stabilization frames at the start of the raster If this is not done then the flat field and long term transient determination may fail mostly because the sliding means cannot work on the very first and last frames by construction This is what happens here for the LW2 case and thus we have masked the first 14 frames of the LW2 raster see Table 21 2 21 5 2 Choice of flat field methods parameters For this first flat field determination we are working with data that are potentially still heavily affected by a long term transient we must thus use a method which is not too affected by that SLICE has 6 different methods to build the flat field the first three of which are common with CIA The first one computes a single flat for all the data and is equivalent to the auto method of CIA If you need to apply it you can do it with CIA red param set red param tdt 65801627 The second one corresponds to the calg method of CIA However it is worth stating how it is used since this is not described in SLICE s original manual Simply set CIA red param set red param tdt 65801627 library flat 280 CHAPTER 21 USING SLICE WITHIN CIA The third one corresponds to the inflat method of CIA where you provide a flat field to SLICE The only enhancement here is th
236. experts only and can be accessed at the ISO Data Centers at VilSpa Saclay or Orsay Telemetry was delivered only during PV phase by ESA If CAM is parallel the TDF also contains the telemetry of the prime instrument 222 CHAPTER 19 ADVANCED SLICING 19 6 4 1 TDF files types Here it becomes a bit tricky the CAM microprocessor is of the 680x0 family Its internal coding of integers follow the IEEE convention the most significant byte is the first Telemetry is received at VilSpa ESA Station and processed on VAX computers These like all based on Intel microprocessor follow the opposite convention the least significant byte is the first lot of byte swapping is therefore done Two kinds of TDF files are created they are distinguished by their extension e LTDF files are sequential files LTDF e TDFG files are indexed files TDFG All CAM slicers work on LTDF files If you feed them with a TDFG file it will be converted into the LTDF format first without you being told about This operation can last some time 19 6 4 2 Loading TDF TDF files are very big files normally covering a whole orbit It may be a challenge for your computer just to load the entire file Therefore it is better to select a part of a telemetry file rather than the whole Three keys may be used for this in order to pick the interesting part of the TDF they are UTK ITK and FMT e UTK stands for Uniform Time Key its value is the number of 1 2
237. f a PDS CIA image_subcube xsubcube cvf_pds image 14 4 3 xcube xcube is a widget program to display cube of images The intensity is a function of space and time I z y t The idea is to display an image for a fixed time to I to and to plot the time evolution for a given pixel zo yo I zo yo You can switch back and forth between the two windows You select one pixel in the Frame Window and immediately the plot of this pixel is drawn in the Plot Window and if a time is selected in the Plot Window the given image is displayed in the Frame Window The parameters of the plot display can be modified in two ways By clicking which useful for a quick look or by typing its value which is good for precise comparison When one parameter appears in different places e g a slider and in a text widget its value will be updated every where 14 4 3 1 Starting The input of xcube can be one cube in various formats e 3 D IDL array e IDL raster structure the cube is raster cube e IDL cvf structure the cube is cvf cube You can start xcube without arguments with one argument or two arguments Zero 32768 infinite or NaN values of the input are considered as undefined values and don t affect the intensity scaling of the display When you click inside the plot you choose the displayed image below When you click inside the image you choose one pixel which time history will be plotted over You can zoom the plot first
238. f the cube i e raster pds cube 106 The are several effects evident in this IMAGE Dark current causes the alternately dark and bright horizontal lines The bright pixels at the bottom left of the IMAGE is a cosmic ray glitch this is clear from the sharp spike in the history of the currently selected pixel raster pds cube 1 9 106 Additional effects which are not obvious in Figure 3 1 are i pixel to pixel non unformity which need to be corrected by flat fielding ii instability in the detector which can be corrected by a variety of fitting and masking routines We will now attempt to remove all these effects There is a set of CIA calibration routines that we will use to do all the necessary data calibration and correction Chapter 20 contains more detail on these routines Almost all of these routines accept any flavour of PDS the one exception is the raster MOSAIC creation routine This is of course specific to a raster PDS All these routines have a choice of different methods though for simplicity we will just use the default here Now we will perform the following corrections i dark correction ii deglitching iii stabilization 3 2 DATA ANALYSIS 27 Frame Number 1 LORD LUT v Horizontal cut Quit Qr Vectical cut Window Size 1 Mask J Glitch 200M RANGE Y SCALE gt p Next frame Previous frame Pixel value
239. f the image with the selected filter and the selected pixel field of view then buttons to select the intensity of the displayed image 130 CHAPTER 14 IMAGE ANALYSIS AND DISPLAY Title Msi FILL USE Hin 6 59823 26 5618 Auto lasa amp 16 16 1841 12 1503 Figure 14 8 The Frame Window text widget you can type the intensity thresholds the image will be displayed according to these values Next Frane Previous Frane Temporal cut _ Horizontal cut _ ertical cut _ Bad pixels Suynbols Hasking Cube H Linear 2 A toggle button if Auto is selected then an automatic intensity scaling of e im age is done between the minimum and maximum value of the displayed image and NOT the whole image If Fixed is selected the minimum and maximum values don t change when moving the image Just over the displayed image are displayed the coordinates and intensity of the selected pixel On the upper left there is a rectangle frame which contains 5 buttons to select the displayed image e il text widget it displays the first index and the last index of the displayed image If they are identical the image of this index will be displayed if not the mean of the images in this range will be displayed These indexes can be modified by the users In the Plot Window the arrow s which give s the selected image s is are updated e toggle button if
240. faint point sources 245 20 7 Analysis of solar system objects 246 20 8 Tips on CIA data calibration 249 20 831 PIS History uu ei A gh anus ux eure ak 251 20 9 Dealing with dead pixels 2 2 251 20 10Making custom FLATs with flat builder 252 20 10 1 Building simple flat 252 20 10 2 Advanced features 253 20 11Background subtraction 254 20 120 btaining the best calibration record from CDS 254 20 12 find best eR oe ele ea deed IRURE 256 20 12 2find best psf 256 20 13Unit conversion and colour correction 256 20 13 1 Propagation of pixel units within a PDS 256 CONTENTS xi 20 13 2 Conversion to milli janskys 257 20 13 3 Color correction 5 sosca acea Eoo Rd RU ee 257 20 14A note on the infamous column 24 258 20 15Advanced projection 2 ee n i an a ee 259 20 15 1 Distortion 259 20 15 2 Weighted mean option ooo 259 20 15 3 Coadding images of different 260 20 15 4 Back projection ca a p moraa a be wee ROT URGE ee ee 262 20 15 5 Distortion cor
241. ffsets from IMAGE to IMAGE are computed and secondly these offsets are applied to the data 22 1 1 Computing the jitter offsets In general the jitter computation is performed as follows 1 All the IMAGEs corresponding to a single ISO pointing are extracted from the PDS CUBE These IMAGEs are averaged and the maximum pixel avg Jmax avg is taken as first estimate of the position of the center of the source 2 A sub cube is made from the IMAGEs centered on CU B E imax avg Jmax avg and of radius jitter where jitter is the expected maximum jitter amplitude and defaults to 1 pixel For each frame of this sub cube the maximum pixel imax jmax y is determined 3 A new sub cube SU BCU B E is again made from the IMAGES this time each sub cube frame SU BCU B Efi j k is centered on CU B E isax x k and of radius bsize 4 A fit is applied to each sub cube frame SU BCU BE j k and the center of the fitted function is taken as the observed source location isource Jsource k 5 The jitter offsets in the x and y axis du and dv are taken as actives bi thedian es median jsource k du and dv are placed in the PDS fields DU and DV 293 294 CHAPTER 22 SECOND ORDER CORRECTIONS 80 60L H 20r 0 LE 0 3 0 2 0 1 0 0 0 1 0 2 jitter offset arcseconds Figure 22 1 Distribution of jitter
242. fields of the CDS The standard fields of the CDS are listed here 1 NAME Generally of the form CCGCCTTTT yymmddhhmmssdd where Variable Definition CC Channel i e LW or SW TTTT CDS name of calibration data see Table 15 3 2 yy year of creation mm month of creation dd day of creation bh hour of creation mm minute of creation ss second of creation dd 0 01 second of creation Note that if the CDS is created from a CAL G file i e using calg2cds see Section 17 1 5 then creation refers to the date of creation of the CAL G file and NOT the date the CDS was created Should the CDS have been created using cds init see section 16 1 1 5 then creation refers to the date cds init was executed 2 SPARE This is an empty array Type byte 1 D IDL array There are some exceptions to this rule e g the GAIN CDS is named CCIGAIN_yymmddhhmmssdd 172 CHAPTER 15 CIA DATA STRUCTURE HIGH LEVEL ARCHITECTURE 3 CREATION Contains details about CDS owner and CDS creation time Type string array DATA The calibration data is held in this substructure DATA is in fact an array of IDL structures Each structure holding a calibration record and the number of structures in the array is equal to NUMBER See Section 15 3 2 NUMBER Refers to the number of records in the substructure DATA A record cor responds to a single independent calibration measurement such as a DARK image and associated CAM parameters
243. fined 2 19 2 119 END date 26 May 1998 17 21 54 node bikini user mdelaney procedure flat library V 1 1 algorithm Find best CCGLWOFLT_98041510080669 END date 26 May 1998 17 21 55 node bikini user mdelaney procedure flat library V 1 1 algorithm Find best CCGLWDFLT 98031519384439 END date 26 May 1998 17 21 40 node bikini user mdelaney procedure get sscdstruct V 2 1 algorithm default 55 030012090001_98052617205667 undefined undefined END date 13 Jul 1998 10 53 06 node bikini user mdelaney procedure corr dark V 3 9 algorithm cube cube gain tint dark model END etc If you use CIA routines to convert your PDS to a FITS file the text in HISTORY is saved in the FITS header see Chapter 18 20 9 Dealing with dead pixels Usually the only dead pixels you need worry about are the four of the SW detector and col umn 24 of the LW detector However if you have very heavily glitched or unstable data then after calibration a pixel may be masked all the way through a set of IMAGEs from a STATE When reduced to an EXPOSURE such pixels will be effectively dead in that EXPOSURE Their corresponding value in NPIX will be zero in other words they will have a zero weight raster scan ignores all dead pixels if EXPOSUREs do not overlap where a dead pixel occurs then a blank spot will appear in the MOSAIC Note the following when you are dealing with data from the LW detector after calibration but before building the raster MOSAIC
244. fore attempting to create a PDS This step is very much dependent on how the observation was programmed 3 Now freeze the SSCD into a raster PDS with get sscdraster and reduce the IMAGEs to EXPOSUREs with reduce 51 52 CHAPTER 8 POLARIZATION OBSERVATION CAM05 DEDICATED CAM99 4 Flat field the EXPOSUREs Firstly identify the entrance wheel associated with each EXPOSURE Restore the appropriate flat field image from the set of polarization flat fields distributed with CIA and supply as input to corr flat 5 Convert the pixel values to milli janskys mJy with conv flux and perform photometry on the EXPOSUREs and or MOSAIC images One may use xphot for this purpose 6 Use get polar weight to determine the polar weight factors and comp stokes to calcu late the Stokes parameters 8 2 2 Slice and perform core calibration Slicing is straightforward CIA spdtoscd cisp35600501 fits sscd dir cia_vers test nowrite Core calibration may be done with the usual routines except we work directly on the SSCD rather than on a PDS Later you will see that the PDS will contain SCDs extracted from different points within the SSCD or observation i e the PDS will not contain contiguously acquired data Because core calibration works best on contiguous data we must postpone the freezing of the data in a PDS until after the core calibration CIA gt corr_dark sscd CIA gt deglitch sscd CIA gt stabilize sscd 8 2 3 Cl
245. fully all this will become clear as you read on 1 Data preparation also known in CIA as data slicing refers to the process of extracting data from the ERD or SPD data products translating telemetry coded parameters into a user friendly format and placing all these data in an SPD SCD There are two paths to producing SPD SCDs from data products a Automatic data slicing e ERD data products are converted into ERD SCDs with erdtoscd and then the ERD SCDs are converted into SPD SCDs with erd2spd e SPD data products can be directly converted into SPD SCDs with spdtoscd These methods are referred to as automatic data slicing methods see Section 12 2 b Data slicing with x slicer ERD data products can be directly converted into SPD SCDs with the widget based interface x_slicer see Section 12 3 x slicer requires more effort to use than the automatic slicers but has greater flexibility and a nice user interface However regardless of which way you choose to create SPD SCDs the end result should be the same See the referenced sections for the method of your choice 73 74 CHAPTER 11 INTRODUCTION TO CIA DATA ANALYSIS Data Product Type CIA Processes CIA Data Structure ERD SCD SPD SCD Data Calibration PDS PDS amp SAD Image Analysis amp Display Figure 11 1 Overview of the processing steps in CIA beginning with either ERD SPD or AAR data product type The data structure employed at each level of
246. function of the good number of readouts stored in CCIM NPIX see Section 15 5 8 However in case of systematic errors e g due to the flat field uncertainty at the edge of the detector or unstabilized pixels better results might be achieved by using the standard error instead This error is computed as ER LE CCIM RMS vn CCIM NPIX The following script gives some examples how to use the different methods raster short 2 x 2 raster has been dark corrected deglitched transient corrected and flat fielded with the library flat field for all but the last example which was flat fielded with a sky flat field standard error 20 1 260 CHAPTER 20 ADVANCED DATA CALIBRATION Standard treatment EXPOSURE pixels are weighted by CCIM NPIX to create the top left MOSAIC of Figure 20 6 CIA raster scan raster pds Projection using standard variation of each pixel held in CCIM RMS to create the top right MOSAIC of Figure 20 6 CIA raster scan raster pds weight Projection using the flat field error contained in flat field library to create the lower left MOSAIC of Figure 20 6 CIA raster scan raster pds weight wcalg Projection using the flat field error computed from the sky flat auto flag to create the lower right MOSAIC of Figure 20 6 CIA raster scan raster pds weight wauto The resulting MOSAICs are shown below The top left mosaic demonstrates clearly the negative effects of an equal wei
247. g comments the CIA User s Manual Suggestions for improving this manual may be submitted to the address in Section J 2 Chapter 2 About CIA 2 1 History and Purpose of CIA Once upon a time there was a prototype system that went by the name of ICE or ISOCAM Calibration Environment Soon thereafter CIA CAM Interactive Analysis came into existence inheriting some modules from ICE CIA was to be an evolution from e the minimum system completed 15th April 95 e over the operational system used during the operations of ISO e to the astronomical data processing system used during the post operational phase of ISO It may be instructive to list the functional requirements of the operational system which was to run on VMS e calibrate ISOCAM e monitor the health of ISOCAM e perform any sort of investigation requested for problem diagnostics e assess the quality of ISOCAM data products e debug validate and refine the pipeline e provide a test bed for algorithmic developments It should be clear that given the driving forces behind the design the evolution into a different kind of beast altogether an astronomical data processing system would not be without its difficulties Some users unaware of the operational raison d tre are puzzled by certain features of the system CIA has now acquired a wealth of astronomical data processing routines and user friendly widget based programs Though the end of ISO s operationa
248. ge CIA Quit Quit x_cia Data Load SSCD Load from disk a SSCD data structure Default input file is the current directory An history of data is then initialized Note that an AOT type must be chosen first default chosen AOT is CAMOI Raster Scan The date and time of loading file into memory are saved in the data history Data Load IDL File Same as above but data are loaded using an IDL restore command This file must have been saved by Data Save IDL File The restored variables are history string array containing the data history ihist next line index to be written in history and isodata PDS structure containing the data All new processes will be added to the history of the current data Date and time are of loading file into memory are saved in the data history 23 2 HELP ON _ 301 Data Save SAD Save the current SAD Data Save SAD Fits Same as above except that the output format is extended fits Data Save IDL File The saved variables are history string array containing the data history ihist next line index to be written in history and isodata IDL structure containing the data The date and time of saving are written in the data history Data Reload Original Data Before trying another processing it may be necessary to reload the original data set in order to avoid to deglitch it twice for example Note that history of data is not reset to the original history D
249. ger than 5096 makes no sense Before starting the operation one last information must be known the TDT of your obser vation Although this is in principle not necessary when using SLICE in CIA modifying this would mean modifying SLICE which we do not want to do This information is quite easy to get as it resides in your raster structure in the field data tdtosn Note however that it is an integer in the raster structure and that SLICE requires a string Note also that for revolutions smaller than 100 you should add the leading 0 to that string or SLICE will not find your data In our example observation 45 taken on revolution 83 we would for instance have CIA print data tdtosn 8301045 CIA tdt 08301045 21 5 A WORKED EXAMPLE 277 Therefore to perform the SLICE processing we now have to first set the reduction param eters structure with CIA red_param set_red_param tdt tdt flat_smooth_window 6 nplanes 30 CIA gt flat_thresh 10 Two things are worth mentioning here First it is mandatory that the result of the set_red_param function go into the variable red_param as this is how SLICE will access the parameters If you use another name the SLICE structure will not be updated correctly Second you can see that the name flat field method appears nowhere on the command lines This is because the method is entirely determined by the set of parameters used Thus make sure you have read the manual and understand the s
250. ghting of good and badly flat fielded pixels Additionally there is the option wmap which permits the user to supply his own weight error map 20 15 3 Coadding images of different astrometry A very useful function of the projection routines are their ability to combine or coadd images that have different astrometry This can be used to combine raster MOSAICs from different observations of the same object To take advantage of this functionality we need to go quite low level and use the C executable projection Here is an example on how to combine raster MOSAICs from different observations of the same object Note that the same could be done for MOSAICs from different AOTs For example a raster MOSAIC and a beam switch MOSAIC Also more than two MOSAICs can be used e Save both raster MOSAICs as FITS files giving them each a name with an individual sequence number We use conv flux to make sure the data are calibrated to the same units CIA conv flux raster pds CIA conv flux raster pds2 CIA raster2fits raster pdsi name input1 CIA raster2fits raster pds2 name input2 CIA 1s input fits inputi fits input2 fits 20 15 ADVANCED PROJECTION 261 Figure 20 6 Comparison of standard projection vs weighted projection Top Left Library flat fielding and standard projection Top Right Libra flat fielding and projection weighted by variance of each pixel Bottom Left Libra flat fielding and projection weighte
251. h IFPG file is to be taken into account if one is needed see below Click on the Load ERD button A pickfile menu will appear Select your file and click the OK button at the end The fields in the top half of the x slicer window are now filled with information Let s have a look at it The first six fields contain information about the file we are going to slice i e its name directory type and so on you can note that for an ERD file the starting and ending fields are automatically set to Beginning and End we will always slice an entire ERD This is not the case for a TDF file for example Going down the next fields give the name and directory of the attitude file This file is the Instantaneous Instrument Pointing History IIPH see Chapter 9 It allows the slicer to compute the astronomical coordinates of each SCD that you will produce If the Automatic Find was not On or if the slicer couldn t find it automatically you would have been asked to give the IIPH file corresponding to the ERD with a Pickfile menu At this point hitting the Cancel button aborts the search of the IIPH file and the field displays No Attitude Control This has no effect on the rest of the slicing but the fields of Right Ascension Declination and Roll will be left undefined in your SCDs and SSCD you will be unable to re project properly your data on the sky after having fully treated them A worst case aris
252. h is to be included in the aperture Either the frame display or the aperture display windows are clickable To remove a pixel from an aperture position the cursor over the pixel and press the RIGHT mouse button 14 2 2 3 Quick Look This is useful when you are curious about how the spectrum form a particular pixel looks but don t necessarily want to include the pixel in the aperture Position the cursor over the pixel of interest and press and hold the MIDDLE mouse button The spectrum will remain visible while the button is held down The astrometry information at the bottom panel is also updated 14 2 2 4 Some miscellaneous tips 1 All messages if any from xevf are displayed in your IDL session window If you become confused check to see if there were any messages from xcvf 2 When using TOTAL sum of all spectra in the aperture to compute the current spectrum the user should take care to use the same size aperture for both the SKY and the SOURCE beam for the sky subtraction to be meaningful Currently xcvf warn users if a discrepancy exists and the user tries to export the current spectrum However the output file will be written after the warning message is displayed 122 CHAPTER 14 IMAGE ANALYSIS AND DISPLAY 14 3 2 D image analysis Currently two routines exist for 2 D image analysis These are xdisp and sad_display You have met sad_display before see Section 10 2 so here we will give only a brief example of its use with
253. he QLA flag and the latter means you take 7 as a good value The criteria for deciding if CAM is on target are stricter for the latter than for the former See also QLA flag PDS Prepared Data Structure Three flavours exist a CVF PDS a raster PDS also known as a raster data structure a beam switch PDS and a general PDS These data structures are created from an SSCD with one of the routines get sscdcvf get sscdraster get sscdbs and get sscdstruct respectively They are used to hold all the sliced i e prepared data that you need to perform normal calibration Usually the data in a PDS corresponds to a single CONFIGURATION pipeline processing refers to the OLP processing that creates the ERD SPD and AA data products delivered on an ISO CD ROM or retrieved from the IDA PMA Post Mission Archive See IDA polarization observation See AOT 5 QLA flag Quick Look Analysis flag Originally this was a telemetry flag but in CIA it is redefined as a combination of the original QLA flag the original OTF flag and the OTF sum flag What this all actually means is that the QLA flag has two values which may be taken to indicate a good FRAME 7 or 143 Both of these indicate the CAM is on target but the latter is a stricter criterion than the former See also OTF 310 APPENDIX A GLOSSARY raster data structure See PDS raster observation A particular CONFIGURATION of CAM AOT 1 A raster is per formed when CAM makes a series of r
254. he calibration routines Finally we make the single EXPOSURE from the cube CIA reduce staring pds and perform flat field correction on the EXPOSURE CIA corr flat staring pds So to view the results of your calibration use xdisp or tviso to take a look at the single EXPOSURE CIA tviso staring pds image You should see the same image as in Figure 4 1 4 2 DATA ANALYSIS 33 27 2860 0 827425 Figure 4 1 tviso display of an EXPOSURE from a staring observation 34 CHAPTER 4 STARING OBSERVATION 01 9 As in the end of Section 3 2 you can save the data using IDL s SAVE CIA save file staring_pds xdr staring pds And export to a FITS file with CIA s imagette2fits This routine will place the data in the PDS field IMAGE into the primary array of the FITS file CIA imagette2fits staring pds name staring fits Additionally you can correct the staring EXPOSUREs in IMAGE for distortion see Sec tion 20 15 5 Chapter 5 Solar System Object observation CAMO1 5 1 Description of the observation The data used here is from a CAM solar system object observation of the comet Tempel Tuttle This observation is comprised of four CONFIGURATIONs These four CONFIGURATIONS together make up the entire observation or AOT Each CONFIGURATION has roughly the same duration of 180 seconds In order to keep the object in ISOCAM s FOV multiple 5 6 re pointings of ISO had to be made during e
255. he data are filtered and removed from the readouts before computing the flat field However it is the sky image in map that is smoothed and then divided out Note that contrary to flat smooth window the size of the smoothing box is really size_filter It is rather hard to determine a priori what should be the value of this parameter One way to proceed is to try different values and judge from the results from a size slightly larger than the PSF to one larger than the raster step Remember to inspect not only the resulting map but also the flat field it is generally in the flat field that you can judge the success of your parameter choice If features reminiscent of your source appear in the flat field cube then size filter is wrong see section 21 4 1 to see how to recover the flat field cube Table 21 4 summarizes our choice of parameters for the DivSky method while figure 21 3 shows the results for the two filter For this particular method and in the LW3 case the command lines are CIA red_param set_red_param tdt 65801627 flat_thresh 10 nplanes 60 CIA gt size_filter 15 divsky CIA gt act set_act make_map CIA gt slice_pipe An obvious improvement is seen here the gradient of emission which is in fact the long term transient is much smoother now and pointing imprints have disappeared In your data reduction session we suggest that you play around with both flat field methods before you select the one to u
256. he lower left point of the region first followed by the upper right point You can now click on the button compute to perform the photometric calculation according to the parameters that you have chosen above The results will be printed to screen in the following order the source object number its pixel coordinates the aperture radius the background value and the source flux When you quit xphot click on quit these data will be saved to disk To recover CIA restore xphot_tables_final save verb RESTORE Portable XDR SAVE RESTORE file RESTORE Save file written by mdelaney bikini Thu Oct 9 17 07 40 1997 RESTORE IDL version 5 0 sunos sparc RESTORE Restored variable ID RESTORE Restored variable X RESTORE Restored variable Y RESTORE Restored variable RAD RESTORE Restored variable BKG RESTORE Restored variable FLUX lt gt CIA print flux 106 678 14 1 5 Other methods for photometry measurements This sections describes some other basic tools to perform photometry measurements photom_psf High level routine to perform PSF photometry As an example we restore the test dataset a raster observation of M51 and perform PSF photometry on the companion Calling syntax 14 2 CVF IMAGE ANALYSIS 117 CIA restore cia vers test raster mb lw3 octO1 xdr CIA photom psf raster psf flux x out y out 11 25 image 8 background background bsize 5 CIA print psf flux x
257. he raster MOSAIC will also be aligned to ISOCAMs axes This method is of most advantage when raster steps sizes are not multiples of pixel sizes the initial rebinning of the EXPOSUREs make the raster step size a whole multiple of the pixel size called routine raster_cam 20 5 Creation of the beam switch MOSAIC After all the calibration steps described in Section 20 2 have been performed on a BS PDS then the beam switch MOSAIC maybe created This is done with reduce_bs The process is simple 20 6 CVF ANALYSIS 245 1 Determine which of the EXPOSUREs of the BS PDS field IMAGE are source pointings and which are reference pointings The source and reference EXPOSURES are indexed by REF IMAGE and SRC IMAGE respectively 2 The source and reference EXPOSURES are coadded with reduce cube taking into account NPIX The coadded reference image is subtracted from the coadded source image The result is the beam switch MOSAIC 3 RASTER is filled with the beam switch MOSAIC NPIXRASTER is filled with the total number of EXPOSURE pixels that are used to compute each pixel in RASTER RMSRASTER is filled with the RMS image of RASTER 20 6 CVF analysis Dedicated CVF analysis is described in this section 20 6 1 Sensitivity and straylight correction Sensitivity correction or more simply conversion from ADU to milli janskys mJy can be per formed by dividing the EXPOSUREs in IMAGE by the sensitivity correction factors stor
258. he raster PDS fields RASTER RMSRASTER and NPIXRASTER to FITS primary array CIA raster2fits raster pds name raster fits The keyword option iraf will make the FITS file compatible with the IRAF preferred CD matrix format of CDx_y instead of CD00x00y and only RASTER will be written to the Tt appears that MIDAS has a problem reading CIA generated FITS files see Section 2 4 205 206 CHAPTER 18 EXPORT OF CIA DATA STRUCTURES primary array Since the BS PDS is compatible with the raster PDS this routine also works for beam switch CAMO03 data To save a beam switch MOSAIC to a FITS file CIA raster2fits bs pds name bs fits These FITS files can be viewed with packages like SAOIMAGE CIA saoimage raster fits imagette2fits writes each frame or EXPOSURE in the PDS field IMAGE along with the corresponding frame of RMS and NPIX to the primary array of an individual FITS file This makes it useful for exporting CVF observation and staring observation CAMOI data contained in an CVF PDS or general PDS In the example below 20 FITS files are created with a prefix taken from the keyword name CIA help cvf_pds image Expression FLOAT Array 32 32 20 CIA imagette2fits cvf pds name cvf fits CIA 1s cvf fits cvfi fits cvfi3 fits cvf1i7 fits cvf20 fits cvf6 fits cvfiO fits cvf14 fits cvf18 fits cvf3 fits cvf7 fits cevfii fits cvfib fits cvf19 fits cvf4 fits cvf8 fits
259. he result is displayed in the reference window 7 You can flip back any time to the original EXPOSUREs by clicking DISPLAY OBS Also you can invoke XLOADCT by clicking on xloadct When you are finished click on DONE Your FLAT will be in the array built flat 20 10 2 Advanced features Another great feature of flat_builder is that you can choose regions of an EXPOSURE to be included excluded from you FLAT This is useful if you are short of signal free EXPOSUREs 1 Choose an EXPOSURE by clicking the appropriate button under Click on frame number to select unselect it The selected EXPOSURE is displayed in the main window 2 With the right mouse button click on two points on the currently selected EXPOSURE which enclose the region you desire to be excluded included 3 A box will appear outlining the region Now click on INSIDE or OUTSIDE depending on whether you want to include the selected region or exclude it 4 Click on SELECTION DISPLAY to display all the selected EXPOSUREs including the regions of those partially selected in the reference window 254 CHAPTER 20 ADVANCED DATA CALIBRATION 5 You can now make the FLAT by clicking on MEAN the option to use a median is not available when partial EXPOSURES are used in the selection 6 You test the flat by clicking on TEST FLAT 7 When you are finished click on DONE The most recently computed FLAT will be returned by flat builder 20 11 Background subtracti
260. hemg LDE uper oL REI ens 221 19 6 5 Selecting slicing variables 222 19 6 6 The x handle slice window s n 225 19 67 On Larget Elag tat ue ge Re c iet roe ee Eh rete 228 19 6 8 Handling big datasets 228 19 6 9 The save slicer file button and slicer_to_cia 229 19 6 10 Getting your SSCDs 229 19 6 11 Frequently Asked Question 229 20 Advanced data calibration 231 20 1 Before reading this 231 20 2 Gore Calibration RO EUR oe 231 20 2 1 Dark correction 0 0 231 20 2 27 Beghlitching ohh ay te OP ee 233 20 2 9 SUaDIIZALIOIE mcd iren ek at ea ee e oe UE NUTS A 235 20 2 4 Reducing IMAGEs to EXPOSUREs 237 20 20 Flatsheldings zar uet ek E at 238 20 2 6 Flat fielding and wheel jitter 239 20 2 7 Small mirror and unilluminated pixels 240 20 3 Calibrating an SSCD Ra EROR ER 241 20 4 Raster MOSAIC creation 243 20 5 Creation of the beam switch 244 20 6 CVF analysis dev kw ee Rao E ER d 245 20 6 1 Sensitivity and straylight correction 245 20 6 2 Photometry on
261. hen you can supply the necessary parameters directly CIA restore make_psf_name aot raster wavelength 6 75 pfov 3 0 CIA fltrwhl 1w2 ack ack verb 15 4 AUXILIARY CALIBRATION DATA 177 Have a look at the restore variable PSF CIA gt help psf str Structure lt 81418 gt 7 tags length 1809884 refs 1 PFOV FLOAT 3 00000 WAVELENGTH STRING LW2 RESOLUTION FLOAT 0 142857 STEP FLOAT 0 142857 IMA FLOAT Array 32 32 441 XCEN FLOAT Array 441 YCEN FLOAT Array 441 The field IMA holds the actual theoretical PSF image and XCEN and YCEN the coor dinates of the center of the PSF The step size and resolution of the PSFs are given by STEP and RESOLUTION respectively 15 4 2 Observed PSFs The observed PSFs were generated by replacing theoretical computed PSFs with the best fitted observed PSFs At the date of publication of the CIA User s Manual observed PSFs had only been generated for the 1 5 PFOV The observed PSFs are stored as IDL save sets and have the following shape psf filter pixel field of view simul save e g psflw10_lp5as_simul save They have the same structure as the theoretical PSFs described in Section 15 4 1 The observed PSFs may be retrieved along with the theoretical PSFs 15 4 3 House keeping and CAM wheels data CAM wheels position data are distributed in CIA as simple text files These files are used by rou tines such as convert_wheel_back to determine the wheel posi
262. hich the OTF was off For the raster PDS the additional field RASTER is built from IMAGE and contains the constructed raster map Dark correction updates the CUBE of the PDS and glitch and transient corrections update both CUBE and MASK of the PDS Then CUBE can be averaged into IMAGE excluding all bad pixel values flagged in MASK This averaging operation is optional but is performed 106 CHAPTER 13 DATA CALIBRATION automatically whenever a flat field correction is required Chapter 20 gives a more detailed description the various calibration methods and their impact on the PDS It is possible to correct data for dark current and glitches only save the result in an IDL file and then try different methods for transient and flat field corrections reloading the dark and glitch corrected data before each new try So the user must take care of what corrections have been performed on a data set including the order in which they have been performed before trying any other correction It is possible to perform the same correction more then once on a data set though novice users are strongly discouraged from doing so It is recommended that you carefully read Chapter 20 if you really want to try this The DATA STATUS window displays some useful information about the number of times that each calibration was performed on the data set If the data set is loaded from an IDL restore command then all indices start at 900 instead of zero in
263. hys Suppl Ser 118 575 585 Starck J L et al 1999 Faint source detection in ISOCAM images A amp AS vol 138 page 365 379 Starck J L Murtagh F Bijaoui A 1998 Image Processing and Data Analysis The Multi scale Approach Cambridge University Press Tiphene D et al Modelling transient effects in the IR array of the short wavelength channel of ISOCAM the camera onboard the ISO satellite Experimental Astronomy accepted 357 ww ea ref sa INN Mamm crems mj ET Tur Sd8is 095 UM SONVHLN3 LIA sdeis gp uni MAS sus SIN IRIS Ears omm 91 1700 6 M1 FANE EJ EE 5 9 SFH OM 50015 Qpz NOILLO3T13S ZIN 5 S 9 lt t S amp e b E si 5 E sJeill MS SIM ena onem o rm ws a E 50815 sua
264. ields of the structure in a user friendly format i e the original telemetry coded parameters have been converted to more readable strings The structures containing observation data are outlined below and a reference to a more detailed section is given with each description Science CAM Data SCD ERD SPD Level The SCD structure has two flavours one is used to hold data of ERD product type the ERD SCD and the other of SPD product type the SPD SCD The ERD SCD contains all the EOI and RESET FRAMES from a 161 162 CHAPTER 15 CIA DATA STRUCTURE HIGH LEVEL ARCHITECTURE single STATE and in addition parameters describing that STATE e g coordinates lens filter etc The SPD SCD differs primarily in that it holds IMAGEs which have been computed from the EOI and RESET FRAMEs These IMAGES are either directly taken from the CISP data product or computed by CIA from the FRAMEs in ERD SCD See also Section 15 2 2 Set of Science CAM Data SSCD The SSCD is primarily designed to catalogue a set of SCDs either ERD SCDs or SPD SCDs but not both together belonging to the same CONFIGURATION and variables which describe that CONFIGURATION However it may be used to catalogue any number of STATEs from all the STATEs in a AOT down to a single STATE See also Section 15 2 3 Science Analysed Data SAD AAR Level Contains two EXPOSUREs one calibrated in detector coordinates from the CCIM data product and the other calibrated in celes
265. ified value of my structure will then be returned at the end of session If a string array is already in memory the user can also use it as its own data history by typing CIA gt x_cia data my structure histo my history The string array my_history will then be returned at the end of x cia session If the array my_history is not empty at the beginning of the session new information will be appended to it 13 3 3 8 Returning to home institute You may choose to return to your home institute with the results of your calibration either in XDR format using Data Save IDL File or in FITS format using Data Save SAD Fits 110 CHAPTER 13 DATA CALIBRATION 13 3 4 x cia caveats x cia can only load a PDS from an IDL save file if the PDS has the variable name isodata The name of the actual save file is not limited in anyway and may be independent of the saved PDS 13 4 Calibrating a PDS the old way This section describes the calibration of a PDS using the old calib raster calib bs calib cvf and calib struct routines While these routines continue to work they are no longer maintained within CIA Use of the core calibration routines described in Section 13 2 1 gives the user more flexibility and control over the calibration process You are strongly advised to use the core calibration routines 13 4 1 calib_raster We present here an example calibration of a raster PDS with calib raster This example follows from Section 13
266. ignification of the parameters before you start any processing Now that the parameters are set you need to specify the action to perform for flat fielding this is called make_map so in a similar fashion you will update the action structure CIA act set_act make_map Here again it is absolutely mandatory to store the result of the set_act go into the variable called act So far we have only updated the structures that describe the actions to perform to actually start the processing type CIA gt slice_pipe At the end of this processing the resulting map can be found in the field map and is displayed by SLICE s visualizing tool If you wish to recover the flat cube used in the flat fielding start the slice pipeline with CIA gt slice_pipe flat_out flat_out 21 4 2 Error computations Even though this is an important aspect of the data reduction up to April 2000 SLICE in its default setup did not compute the error map associated to the sky map This operation is performed at the make_map stage You can easily check what is your setting by doing CIA help error map If the result is UNDEFINED then you need to read the following lines otherwise the error map is computed by default at the flat field correction stage If the error map is not computed by default simply add the keyword error_map to the call to set red param and proceed as before The error map should now be available in the variable error map Therefo
267. implements the improved OTF slicing criteria bs This keyword is only useful for beam switch observation data see Section 19 4 1 method nomode A new slicing mode nomode has been introduced to erdtoscd and spdtoscd which slices only according to hardware changes of ISOCAM such as filter integration time etc and ignores changes of flags set by the commanding software such as the observing mode old_data This flag should be set when slicing CAM parallel data CIPH that has been produced by OLP versions prior to OLP 8 0 Data produced by OLP 8 0 do not need this flag 19 2 Saturation warnings during slicing spdtoscd warns users of saturated data with messages like pixel 14 17 is affected by saturation at SCD 3 with the average value 4093 00 value for End of Integration 3 readouts This indicates that pixel 14 17 was saturated and its photometry has to be carefully as sessed Using the complex mask option IMASK 1 and setting the option sat mask these pixels can be masked out too 213 214 CHAPTER 19 ADVANCED SLICING 19 3 A beam switch observation caveat Some observers have programmed their beam switch observations in reverse this means that the reference field is programmed as the source field and vice versa This problem will become apparent when the final beam switch MOSAIC image is viewed as contained in the BS PDS field RASTER The MOSAIC will appear inverted and will incorrectly have the
268. in Subtracts the IMAGEs of CUBE by the DARK Places the DARK in the field DARK Finally updates the field CUBE_UNIT to ADU sec gain The different dark correction methods are detailed below One important point about dark correction that the user should be aware of corr dark treats all the data in CUBE as having the same integration time However since it is possible to create a general PDS from data of mixed CONFIGURATIONS then the situation may arise where the integration time is not constant throughout the CUBE In these cases corr dark will fail To avoid this problem the user should perform dark correction on the SSCD before creating the PDS 1 2 3 method vilspa method DARK model correction Computes a dark which depends on the detector temperature time since activation and the integration time called routine dark_vilspa PDS side effects The field DARK is filled with the DARK used for correction The IMAGEs in CUBE are dark corrected i e CUBE is modified reference Ott S and Roman P 2000 In preparation method model method DARK model correction Compute the best dark for given revolution and time since activation of ISOCAM This is a new method and seems to yield excellent results It is also the default method called routine darkmodel PDS side effects The field DARK is filled with the DARK used for correction The IMAGEs in CUBE are dark corrected i e CUBE is mo
269. ing to the same configuration It is normalized on the 10 x 10 central pixels Flatfield Manual Flatfield is built interactively by selecting only a subset of all frames of the data cube and normalized on the 10 x 10 central pixels This method is well adapted whenever only background is observed at the edge of large raster map If a source is detected in each frame it is also possible to exclude interactively some rectangular zones before computing the flat field Process None Set to None all calibration fields Dark Deglitch Transient and Flatfield Process Default Execute the default data processing Warnings are displayed if some corrections have already been made to the current data set Note that these warnings concern only the current x cia session So before processing data the user should have a look at the data history in order to check that some data corrections have not been yet processed see above Data Display History 304 CHAPTER 23 X_CIA REFERENCE GUIDE Process Selected Execute the user selected data processing Warnings are displayed if some corrections have already been made to the current data set Note that these warnings concern only the current x cia session So before processing data the user should have a look at the data history in order to check that some data corrections have not yet been processed see above Data Display History Tools xv_temp Launch xv_temp See Section 14 4 7
270. ion only the fields raster image and raster raster of the raster PDS structure are flat corrected and not the whole cube raster cube field This option corrects the cube for flat fielding and makes easier the comparison of the flux histories SCDs Limits toggle button this is valid only for the plot options Multi and Single in the compare mode the SCD limits are always drawn 14 4 CUBE ANALYSIS 141 E E NE s MAE j 5 V Figure 14 12 The raster window in ximage downward transient M51 FILL LW3 3 pixel 57 85 of the roster map 29 29 9 1 29 10 1 1 13 29 1 14 T T T T T T T E 4 11 29 10 1 1 13 29 1 14 121 1 101 E E N 5 M n i 11 FA TRE J D IE A Lan d d j In a A M M T CLR an yo rofl yyy o I 1 bot nl 4 A E ANTEE nir P qu E M 4 H LU Uy U 29 29 9 1 T p gt _ 6 1 I o 20 40 60 80 100 Figure 14 13 The
271. is section The default configuration should be adequate for the novice user You can customize the way CIA is set up with your own CIA startup file user_init pro This file should follow the rules for IDL batch files You should copy the default startup file delivered with CIA modify it yourself and then instruct CIA to use it Currently the Unix version of this file has the following content startup file for running CIA with IDL can be modified at willing but must call the CIA start file 0 cia vers unix cia start DO NOT CHANGE LINES ABOVE if you really need logfiles modify this line SET LOGFILE noerror nosession nomaster newsite mysite define whether you would like to have informational messages printed 1 quiet 1 define how many many lines you would like to have in the command buffer ledit input 100 define the display order lorder 0 CIA s mask easy or complex mask 0 change CIA s warning level ciaerr xwarn 0 add the contributions to the path for VMS if version os_family eq vms then Ipath path EXPAND_PATH CIA_DIR CONTRIB if version os family eq vms then print if version os family eq vms then print Contrib routines have been added decide whether you like to have the user s or the full help for cia help if version os family eq unix then setenv help dir cia vers ia help he
272. itch observation 6 1 Description of the observation The data used here is from a CAM beam switch calibration observation of HIC 96901 A beam switch observation comprises of a cycle of ISO pointings an on source pointing followed by an off source or reference pointing There can be one or several cycles in a single observation Since each pointing is a STATE we will have at least two STATEs per CONFIGURATION The CONFIGURATION and in this case the STATE parameters are LW10 filter 1 5 PFOV integration time of 2 10 s and gain 2 In the observation analysed here there are 4 cycles 6 2 Data analysis It is assumed in this section that you have read Chapter 3 Generally concepts described in that section will not be re described here 1 Start a CIA session cia 2 Convert your CISP data product into SCDs with spdtoscd CIA gt spdtoscd cisp05804610 fits sscd dir cia_vers test nowrite 24 SCDs in the SSCD CSSC058046100101_02022715110701 seq channel mode fltrwhl pfov tint gain offset size ra dec 0 LW OBS LW2 6 0 25 20 1 512 1 295 463 50 518 1 LW IDLE LW2 6 0 2 10 2 512 1 295 463 50 518 2 LW OBS LW2 6 0 2 10 2 512 1 295 463 50 518 3 LW OBS LW2 6 0 2 10 2 512 1 295 462 50 518 4 LW OBS LW2 1 5 2 10 2 512 1 295 463 50 518 5 LW OBS 10 1 5 2 10 2 512 14 295 462 50 518 6 LW OBS 10 1 5 2 10 2 512 26 295 463 50 518 7 LW OBS 10 1 5 2 10 2 512 21 295 462 50 518 8 LW OBS LW10 1 5 2 10 2 512 27 295 437 50 547
273. ition angle of the M axis of the raster i e the angle between the celestial North and the M axis counted positively eastward In a Y axis raster RASTER ROTATION cannot be specified by the observer at PGA level and a very simple relation links it with the roll angle B a 4 90 Or RASTER ROTATION ROLL 90 These rasters can be reconstructed quite straightforwardly as the camera s axes are parallel to the raster axes E 1 3 Rasters referenced to the celestial North axis These rasters are called in short North axis rasters An example of such a raster is shown in Figure E 3 These rasters can generate quite some confusion First to program them the observer had to specify the raster s position angle but could only supply one comprised between 0 and 180 in PGA this parameter was called orientation angle Therefore there is a 180 uncertainty between programing and reality see Section E 2 to remove that uncertainty In this section we are only concerned with reality what has actually been performed Second before reconstructing the raster images have to be rotated by a certain angle This angle is not written in the data but has to be derived from the two others that we know already o and Figure E 3 show these angles their definition follows e a is the roll angle of the mosaic also called ANGLE_RASTER in CIA structures e 3 is the true RASTER ROTATION comprised between 0 and 360 e is
274. j 0 Na EL m Variance g AUT 2j T 20 3 j 0 Standard Deviation v Variance 20 4 Standard Error 20 5 VN 20 17 ERROR HANDLING IN CIA 267 The CIA routine reduce see also Section 20 2 4 not only averages the IMAGEs to EXPO SUREs it also creates corresponding RMS images and weight images Each pixel or element of the weight image contains the total number of IMAGE pixels that have been averaged to the EXPOSURE pixel Each pixel in the RMS image contains the standard deviation of this sample of IMAGE pixels To summarize using some CIA pseudo code raster pds image i j k average raster pds cube i j raster pds from k raster pds to k l raster pds npixli j total raster pds cube i j raster pds from k raster pds to k raster pds rms i j stdev raster pds cube i j raster pds from k raster pds to k l Similarly the routine raster scan see also Section 20 4 not only creates the raster MOSAIC image but also a corresponding RMS image and weight image In this case each RMS pixel contains the standard deviation of all IMAGE pixels that sample the same sky pixel as the MOSAIC pixel Again in CIA pseudo code raster pds npixraster i j k total raster pds cube i j raster pds rmsraster i j k stdev raster pds cube i j The RMS and weight images that correspond to Figure 3 3 are given in Figures 20 11 and 20 10 These figures were g
275. k model Deglitch None No deglitch correction is performed Deglitch Manual Glitches are removed interactively Deglitch Particle Impact The glitch detection is based on the analysis of cosmic particle impacts Deglitch Temporal All values higher than 3 times the temporal standard deviation of a pixel are considered as a glitch Note that this method is not robust when data are not stabilized Deglitch Spatial All values higher than 3 times the spatial standard deviation of a pixel are considered as a glitch Deglitch Temporal amp Spatial Both temporal and spatial informations are used This method doesn t allow to detect either strong glitches which remain for many exposures or successive glitches hitting the same pixel within a short time interval Deglitch Multiresolution Median default Multiresolution information is taken into account to detect all significant small scale structures not due to the noise As no structure can be detected either in the first frame or in the last one these frames are completely masked and lost This method gives good results even when data are not well stabilized Transient None No transient correction is performed 23 2 HELP ON X CIA 303 Transient 90 Of Final Flux modeled A model based on ground calibrations is used to determine the number of stabilized frames given the first and last flux values of pixels Transient 90 Of Final Flux measured defau
276. king anywhere on the image Clicking the button max will centre the region on the image pixel of maximum intensity The buttons marked 121 3x3 etc can be used to quickly set the size of the region Alternatively more customized sizes can be set with the sliders A plot of the history of each pixel in the region will be displayed by clicking on the button display There are two possible plot types overlay of all histories click on single or individual pixel history plots click on multi 136 CHAPTER 14 IMAGE ANALYSIS AND DISPLAY e A postscript file of the plot xv temp ps be create by clicking on print e The button color will invoke IDL s XLOADCT 14 4 8 ximage This section introduces you to CIA s ximage This is a widget program for astronomical image display and was designed to mimic SAO image What makes this tool different from the rest is that it provides you with a variety of plotting input and output options and that it can show you the pixel histories in the context of the raster MOSAIC This gives you a better idea how each pixel of the PDS cube affects the outcome of the raster MOSAIC and permits especially to verify that apparent sources are not remainders of glitches or transients 14 4 8 1 Getting started The input of ximage can be one image with or without astrometry in various formats e 2 D IDL array e 2 D IDL array with a FITS header 2 D IDL array with and an IDL astrometry structure a r
277. l RMS 40012 9055527209724122 QR 9764279154675056 2450872 597734112 OM 235 270989149082 W 172 5002736828059 IN 162 4865753794343 A 10 33833822975773 4 9783396846902 ADIST 19 70024854404794 PER 33 24178 N 029650223 ANGMOM 023464539 DAN 18 20692 DDN 98041 L 62 4267664 B 2 2510269 TP 1998 Feb 28 0977341 Physical amp non grav parameters KM SEC A1 amp A2 in AU d 2 GM n a RAD n a 1 1 580981E 9 A2 9 186416D 11 Mi 10 M2 16 ki 25 k2 10 PHCOF n a COMET comments 1 soln ref JPL J985 69 2 k1 25 0 k2 10 0 ref for magnitude laws is ICQ 1998 Handbook ORO AG a kkk kk kk Select A lpproaches Elphemeris F tp 11 R edisplay S PK lt cr gt Observe Elements Vectors o e v 0 Coordinate center lt id gt coord geo iso Starting UT ex 1995 Nov 17 01 54 JD 2450849 0361 Ending UT 1998 Jul 16 01 43 JD 2450849 0380 Output interval ex 10m 1h 1d 1m Current output table defaults ICRF J2000 0 Reference frame Time zone correction UT 00 00 Time format JD Time digits output FRACSEC 248 CHAPTER 20 ADVANCED DATA CALIBRATION R A format DEG RA DEC extra precision YES Apparent coord type AIRLESS Range units AU Suppress range rate NO Minimum elevation 90 Maximum airmass 38 000
278. l lifetime has passed CIA will continue to develop as an astronomical data analysis tool well into the future 3 4 CHAPTER 2 ABOUT CIA 2 2 System requirements The CIA system requires the following resources e IDL must be installed on your system Versions earlier than IDL 5 0 are not supported by CIA 4 0 e CIA needs disk space of 375 MB Additional disk space of 450 MB is required for the optional theoretical PSFs available from the CIA server e Users should have access to 250 MB of swap space bigger data sets might require up to 1GB of swap space e CIA is supported to run on a Sun Sparc Solaris 2 5 or later and VMS Alpha 6 2 or later and is known to run on DEC ALPHA OSF1 Digital Unix HP UX Hewlett Packard Unix and x86 Debian Linux 2 0 Users are invited to build the CIA executables and modify IDL coded modules so as to enable CIA to run on alternative platforms 2 3 Getting started Before proceeding make sure that CIA is properly installed on your system and that your system meets the necessary requirement s for running CIA see Section 2 2 2 3 1 How to Start CIA This section is by its nature dependent on system configuration which can vary In case of problems consult your system manager though Section 2 3 4 describes CIA customization and may provide pertinent information on On VMS idl If this fails try typing 1915 On Unix cia version Here there is an optional parameter version
279. lat calg CIA stab s90 e Similarly you can also use x3d and xsnr to look at your resulting data However there is no MOSAIC in a general PDS to view Chapter 14 Image analysis and display This chapter serves as an introduction to the variety of image and cube analysis and display routines in CIA They are categorised as e General analysis routines that perform some type of statistical or photometric analysis xsnr xradial flux sum xphot e CVF image analysis routines cvf display e mage analysis routines that only operate on 2D images xdisp e Cube analysis routines that operate on a cube yielding temporal and spatial information a pixel xselect frame show frame xsubcube xcube x3d xv temp xv raster xmovie Image display routines that just display images or images from cubes tviso and xv sscd Image overlaying routines isocont x isocont Finally routines to aid you to produce postscript output xcontour ps color white black ps_open ps print 14 1 General analysis routines These routines are xsnr for S N estimation xradial for determining the energy radial profile and flux sum for estimating the flux from a point source 14 1 1 Estimating S N in a cube or image Estimating S N in a CAM image is a useful indicator of the strength of source detection or the quality of a calibration xsnr is a widget based program that is used to interactively choose noise and signal regi
280. ld Display the flat field extracted from the calibration database View Cal G PSF Display the point spread function extracted from the calibration database View Cal Used Dark Current Display the dark current frame used for the data calibration process View Cal Used Flatfield Display the flat field used for the data calibration process 302 CHAPTER 23 X CIA REFERENCE GUIDE View Result Reconstructed Raster Map Display the reconstructed map of the raster View Result Monochromatic CVF Frames Display one by one the averaged monochromatic frames of the CVF scan View Result All CVF Frames Display one by one all frames of the CVF scan in order to visualize the transient effects from one CVF wheel position to another View Ra Dec Roll Info Display the Ra Dec Roll parameters of each averaged frames corresponding to each configuration of the camera Ra Dec Angle Rotation and Orientation of the reconstructed raster map are also displayed in the message window View Change LUT Adjust the color table Dark None No dark correction is performed Dark User Input A user dark frame can be given as input in order to subtract the dark current The user must then use the following command to start the session CIA x_cia indark my dark my_dark is a 32x32 array Dark Cal G The dark frame is extracted from the calibration database Dark Model default The dark frame is created from the dar
281. lease give adequate information to allow for the conditions which caused your problem to be reproduced Keeping a journal IDL s JOURNAL command should help Also please keep your data and be prepared to make it available to us it may be required to reproduce and correct the reported bug Please keep in mind that CIA V5 is the legacy version of CIA and in principle no manpower for further maintainence is available CIA is delivered as source code so if you have the necessary IDL knowledge and can fix the bug or implement the required additional functionality on your own then please furnish us with the modifications By doing so you will not only help us but the CIA community as a whole J 1 1 Template for a Software Problem Report ISOCAM INTERACTIVE ANALYSIS SOFTWARE PROBLEM REPORT NUMBER SPR TITLE ORIGINATOR ISSUE DATE VERSION ENVIRONMENT 351 352 PRIORITY PROBLEM DESCRIPTION ANALYST ANALYSIS DATE PROBLEM ANALYSIS RECOMMENDED SOLUTION ITEMS TO BE CHANGED TESTS TO BE RUN ESTIMATED EFFORT IMPLEMENTOR ASSIGNMENT DATE ITEMS CHANGED TESTS DONE ACTUAL EFFORT BOARD DECISION DATE STATUS CLOSING DATE FIX DELIVERY COMMENTS APPENDIX J REPORTING PROBLEMS AND SUGGESTIONS J 2 Comments on this document Comments on the CIA User s Manual and or suggestions for improvements are always welcome They may be submitted to helpdesk iso vils
282. libration steps 51 8 2 2 Slice and perform core calibration 52 8 2 3 Clean the SSCD 52 8 2 4 Freeze the data a PDS 54 8 2 5 Flat field correction 22 22 55 8 2 0 Photometry 2 e bcs ie eite et Dose irn MERE le PRO ti e SR I ELSE 55 8 3 Calculate Stoke parameters 55 CIA Basic Guide 57 9 The data products and CIA data structures 61 9 1 Data product filename 61 9 2 Data products as FITS 0 61 9 3 Relating data product types to filenames 61 93 1 Raw d ta prod cts ze ROGO Rer Ge teg 62 9 3 2 Standard Processed Data SPD 62 9 3 8 Automatic Analysis Results 62 9 3 4 Auxiliary data products 63 9 3 5 Calibration Data Products 63 9 4 Relating Data Product Types to CIA Data Structures 64 9 4 1 What is Data Structure 64 9 4 2 Structures containing Observation Data 65 0 4 9 Calibration Data Structure CDS 66 9 4 4 Regular IDL structures 66 10 First look at the data 67 10 1 Copying data products from ISO CD ROM to hard disk 67 10 L 1 C pying on
283. litches This method needs a steady background level and so is not very suitable for very unstable data routine called deglitch sig PDS side effects Glitches removed from IMAGEs in CUBE i e CUBE is modified Glitched pixels also flagged in MASK reference ISOCAM Handbook Chapter Data processing methods Section Temporal deglitch 234 CHAPTER 20 ADVANCED DATA CALIBRATION 3 method tcor method Deglitch tcor This method attempts to attempts to avoid the problems that arise from instability by initially transforming the cube into a zero mean cube and then deglitching The algorithm is a Compute the zero mean D of the input CUBE cube in D dg cube in i jg cube in x i 1 cube_in d 1 b Compute S t the N sigma clipping of D c For each IMAGE of in i Compute where IM AGE median I M AGE 5 ii Compute S_p the N sigma clipping of I ii A glitch has occurred if abs D x f gt St and abs T gt Sp iv Replace glitched pixels cube out i j k by cube_out i j k 1 m Some criticisms of the tcor have been made under certain conditions it will eradicate your data As with all deglitching techniques it should be used with care routine called deglitch tcor PDS side effects Glitches removed from IMAGEs in CUBE i e CUBE is modified Glitched pixels also flagged in MASK reference online help 4 method mm method Deglitch MM
284. ll readouts below the level in the smoothed map the sky value is taken from the readouts in bin 3 if more that two thirds of the readouts fall in bins 1 and 2 all readouts above the level in the smoothed map the sky value is taken 288 CHAPTER 21 USING SLICE WITHIN CIA from the readouts in bins 1 and 2 if more than two thirds of the readouts fall in bin 2 and 3 rather normal distribution around the mean level the sky value is taken from the readouts in bins 2 and 3 If none of these combinations represents more than two thirds of the readouts the sky value is taken from the readouts in bins 2 and 3 as well 7 As what we have now is a map of residuals we add the smoothed map that had been subtracted to create them and this is the new sky map From this description I hope it is rather clear that the parameters to the methods will be those related to the ghost filtering the smoothing window for the map and the threshold for the bad pixels detection Here they are listed in more details e size filter This is the size of the smoothing window applied to the sky map both to determine location of ghosts and to identify bad pixels Although it is in principle equivalent to the keyword in the DivSky flat field method if the map contains undefined values and it generally does it is already smoothed twice with windows of 5 and 7 pixels further smoothing is thus not required and at least small windows should be used e bad pix thresh This
285. lowing command mount noass over id media cdrom undefined fat fixed none 512 devicename C 1 2 UNIX Mounting a CD ROM on a UNIX machine really depends on the UNIX OS you are running It is best to ask your system administrator for advice C 2 Overview of the CD ROM Contents From the top of the directory tree one can find two subdirectories aboutiso and products Those directories below the former lead you to the ISO documentation and general ISO informa tion and those below the latter to the data products Also there are two text files in the root directory README and datalist tzt which contain details of the relationship of the product list and the directories below products 1 As is usual with VMS you can only access the CD ROM from the window in which you mounted it Refer to the ISO Satellite Handbook for a detailed listing of the contents of an ISO CD ROM and a description of the directory tree structure 319 320 APPENDIX C THE ISO CD ROM C 2 1 Where to find the ISO documents The directory aboutiso docs contains two subdirectories relevant to ISOCAM observers These are aboutiso docs cam and aboutiso docs iso The former contains the documents specific to ISOCAM and the latter documents for all ISO observers Each of these directories contain subdirectories named idum and obsman where the IDUMs and the Observer s Manuals see Section 1 2 can be found in both postscript format and original ATEX Oth
286. lp prg else setlog cia help dir CIA_DIR HELP_PRG if version os family eq unix then setenv cia_help_dir cia_vers ia_help help_prg else 2 8 GETTING STARTED 11 setlog cia help dir CIA_DIR HELP_PRG if version os_family eq unix then cia html help cia vers ia help help html prg else cia html help cia dir HELP HTML PRG decide if you want to run in 24 bit display mode Note that this doesn t work for IDL 5 0 DEVICE GET VISUAL DEPTH d if d eq 24 then device true color 24 if d eq 24 then device decompose 0 decide if you want to run the alias file run alias Generally the things you may want to configure are e The default CIA version if more than one version is installed This only applies to VMS CIA For same in UNIX see Section 2 3 1 e CIA s logfile behaviour By default no logfiles are produced e QUIET system variable which defines how informational messages are handled e EDIT_INPUT system variable which enables keyboard line editing e ORDER system variable Since CIA 2 0 ORDER defaults to 0 the IDL default though in CIA 1 0 it defaults to 1 If you want later versions of CIA to display images in the same manner as CIA 1 0 then change ORDER to 1 e MASK configuration see Section 15 2 2 18 The MASK can be set to be simple or complex by setting the system variable MASK to 0 or 1 respectively MASK defaults
287. lp you either then you are encouraged to submit a report of your problem or even better if you have managed to solve the problem yourself then send us the fixes Please keep in mind that CIA V5 is the legacy version of CIA and in principle no manpower for further maintenance is available Instructions for the submission of both Software Problem Reports and Software Change Requests can be found in Section J 1 18 CHAPTER 2 ABOUT CIA Part I Quick Start Guide Introduction The Quick Start Guide contains real life examples of data analysis with CIA for different AOTs ie observation types It is written for the novice user and attempts to avoid unnecessary details of the CIA system However some CIA and ISOCAM concepts will be introduced STATE CONFIGURATION AOT SCD SSCD IMAGE EXPOSURE MOSAIC See the glossary Appendix A for definitions of these terms 21 22 Chapter 3 Raster observation CAMO1 3 1 Description of the observation The data used here is from a CAM calibration raster observation of the Antenna galaxy This observation is comprised of four CONFIGURATIONs These four CONFIGURATIONS together make up the entire raster observation or AOT Each CONFIGURATION has a fixed raster size of 2 x 4 pointings and a PFOV of 3 0 However other parameters do change indeed this is what gives rise to the different CONFIGURATIONS such as the filter wheel integration time PFOV and gain Table 3 1 summarizes the
288. lt Pixel values exceeding the last value 10 of the difference between the first and last values are considered as not stabilized Transient IPAC Model Fitting The model developed at IPAC is used to fit the transient behavior of the detector Transient IAS Model Fitting The model developed at IAS is used to fit the transient behavior of the detector Transient SAP Model Fitting The model developed at SAP is used to fit the transient behavior of the detector Transient Remove Ghosts An adaptive filtering is applied in order to determine if an upward or a downward transient is detected If an upward transient is detected then the last detected flux is considered as the stabilized one If a downward transient is detected then the stabilized flux is assumed to be equal to that of the background This method allows to remove all ghosts in the reconstructed raster map but it can lead to a loss of information if a source is superimposed on a downward transient for example Flatfield None No flat field correction is performed Flatfield User Input A user flat field can be given as input in order to correct the flat field effect The user must then use the following command to start the session CIA x cia inflat my flat my_flat is a 32x32 array Flatfield Cal G default Flatfield is extracted from the calibration database Flatfield Auto Flatfield is obtained by computing the median of all frames corre spond
289. lue of one field and a numeric value of another CIA print scd find mode fltrwhl OBS Lw6 Searching for MODE 0BS Searching for FLTRWHL LW6 CSCD143006010202_96082815180230 CSCD143006010203 96082815180271 CSCD143006010204_96082815180303 CSCD143006010206_96082815180373 CSCD143006010208_96082815180442 CSCD143006010210_96082815180510 CSCD143006010212_96082815180578 CSCD143006010214_96082815180644 CSCD143006010216_96082815180710 CIA gt print scd_find gain 1 Searching for GAIN 1 CSCD143006010001_96082815175532 The keyword list allows you to supply a list of structures restricted to the same type In CSCD143006010205_96082815180339 CSCD143006010207_96082815180407 CSCD143006010209_96082815180478 CSCD143006010211_96082815180542 CSCD143006010213_96082815180610 CSCD143006010215_96082815180676 CSCD143006010217_96082815180742 the following example scd find supplies a list to a second call to scd find CIA print scd find gain 1 CIA list Searching for MODE 0BS Searching for FLTRWHL LW6 Searching for GAIN 1 STRUCT FIND Failed to find GAIN UNDEFINED 16 1 11 structure elem This is a function to return the list of SCDs in an SSCD or a list of SADs in an SSAD it doesn t exist for the SCD SAD or CDS Calling structure elem is quite simple CIA print sscd elem CSSC143006010001_96082814202966 num noscds scd find mode fltrwhl OBS
290. ly validated Just get the latest OLP 10 data and re run your CIA processing E 2 1 Incorrect astrometry in beam switch data If after calibration the beam switch MOSAIC BS PDS field RASTER appears inverted and incorrectly has the astrometry of the reference field then read Section 19 3 E 2 2 Astrometry inaccuracies If the coordinates appear to be systematically wrong by a small amount then the problem is probably due to the repositioning accuracy of the lens and filter wheels This can have a bigger effect than ISO s pointing accuracy The error induced by the wheel jitter is typically less than 2 pixels but can reach up to 3 pixels For example for a 6 PFOV measurement the coordinate error can be up 3 6 4 22 E 2 3 Roll image orientation and ORDER It is important to understand the effect of the IDL ORDER system variable on the displaying of images The difference is simply that an image will be displayed with pixel 0 0 in the upper left corner when ORDER 1 and in the lower left corner when ORDER 0 However this simple difference can cause confusion when it comes to changing the orientation of a CAM image Figure E 4 should help you understand the effects of ORDER and how to correctly change the orientation of a CAM image E 3 Using FITS in CIA new problems E 3 1 FITS convention and IDL s astrolib IDL comes with an astronomical package called the Astrolib which is in fact quite handy most projectio
291. m Reset Zoom 2 Zoom 4 Zoom 1 2 Zoom 1 4 Help has not yet been implemented in the current version of ximage 14 4 8 3 min max and thumbnails The upper panel of the ximage window just below the menus contains the following features e Intensity scaling Min Max text widget you can type in this text window the intensity thresholds the image will be displayed according to these values Fixed Auto toggle button if Auto is selected then an automatic intensity scaling of the image is done between the minimum and maximum value of the displayed image and NOT the whole image If Fixed is select the minimum and maximum values don t change when moving the image Reset button use the minimum and maximum values of the whole image to scale the image e thumbnaill contains the full image reduced to thumbnail size useful for big images or zoomed images e arrows to move the cursor useful for small images e thumbnail2 displays a zoomed image controlled by 1 dragging the mouse on the image around the mouse position 2 dragging the mouse outside the image around the cursor 14 4 8 4 Mouse Mode This panel contains a multiple choice button on the left which changes the menus and buttons on the right e Data mode Raster toggle button creates a new window valid only when the input is a raster PDS see below 138 CHAPTER 14 IMAGE ANALYSIS AND DISPLAY Statistic
292. mage get CIA x ss 1 CIA y ss 2 14 8 REDIRECTING GRAPHICS TO THE POSTSCRIPT DEVICE 155 If x is greater then y then CIA device xsize n x y ysize n yoffset 27 n bits 8 color where n is less then 27 cm If x is less then y then CIA device xsize n ysize n y x yoffset 27 n y x bits 8 color where n is less then 18 cm The DEVICE keywords xsize ysize and offset determine the x axis size y axis size and offset from 0 0 of the postscript device area The keyword color is set for obvious reasons 5 Output the graphic CIA tv image get 6 Close the postscript device CIA device close CIA set plot x 156 CHAPTER 14 IMAGE ANALYSIS AND DISPLAY Part III Data Management 157 Introduction The purpose of this part of the CIA User s Manual Data Management is to provide a guide on how data is managed by CIA e Chapter 15 presents an overview of the data architecture of the CIA data structures e Chapter 16 describes the CIA data structure user interface and handling routines e Chapter 17 describes CIA routines for converting FITS data products into CIA data structures and ordinary IDL data structures e Chapter 18 shows you how to export import CIA data structures to from FITS 159 160 Chapter 15 CIA data structure high level architecture This chapter describes the high level architecture of CIA data structures It is hoped that it will help you to understand how your d
293. ment between the lens and Fabry mirror and the optical axis The extent of this shift can be up to 2 pixels This shift also has the effect of making flat field correction with the library or CAL G optical flat field images invalid These flats expect a well aligned lens wheel As an alternative data from suitable raster observations may be flat fielded with the auto method However this is not an option for data from raster observations with few raster points and certainly not an option for data from other AOTs A further alternative is suggested here 1 Create a rough auto flat from the IMAGEs in a PDS This will be used to measure the shift caused by the lens wheel CIA flatauto flat auto raster pds 2 Use xdisp to measure the shift in the auto flat image CIA xdisp flatauto The shift will be apparent in two ways e You may see dark columns at the edge of the image The number of such columns helps to indicate the shift in the image e Features in the rough auto flat should correspond to features in the library flat Any misalignment will indicate the amount of shift You can check the library flat with CIA xdisp raster pds calg flat You might also want to try correlating flatauto with the library flat raster pds calg flat using the IDL Astronomy User s Library routine correl optimize However this routine does not usually work well with images of this nature Whichever method you choose you need t
294. metry by scaling it spatially to the PSF at a given effective wavelength A standardpsf is the ISOCAM PSF at 15 um through the 1 5 PFOV lens Each camera pixel is divided into 7x7 sub pixels This is equivalent to a PSF at 105 jum through the same lens The image size is 2048 x 2048 covering an area far beyond the physical detector area in order to include almost 10096 of the flux of a point source The spatial scaling factor f is given by f 105 PFOV 1 5 180 CHAPTER 15 CIA DATA STRUCTURE HIGH LEVEL ARCHITECTURE 15 5 Prepared Data Structure PDS The PDS was designed for convenience to neatly hold all the data that you need for calibration see Chapter 13 and Chapter 20 in one structure However as described in Section 13 1 1 there is a cost for such convenience Because of individual differences between AOTs four distinct PDSs currently exist a raster PDS a CVF PDS a BS PDS and a general PDS As might be expected the first three are specific to a raster CVF and beam switch observation respectively The latter is a multi purpose PDS The routines get_sscdraster get sscdcvf get_sscdbs and get_sscdstruct are used to create the appropriate PDS from an SSCD containing sliced SPD SCDs In the following sections Sections 15 5 1 15 5 2 15 5 3 and 15 5 4 each of the PDSs are described Tables listing the fields and substructures of the PDS are given and where appropriate a reference to where you may find the original data i
295. mine and apply ISOCAM jitter correction and distortion coefficients Follow steps 1 7 above Then follow the steps for determining the RA DEC offset example ii above Now click on the APPLY JITTER CORRECTION to change the astrometry structure in the ISO data structure This change is not permanent until you quit xco rr_astro Hitting the RESET button will restore the original astrometry data structure The old values are kept in an additional tag OLD ASTR The buttons fields etc of xcorr_astro The main window is shown in Figure 14 18 The various buttons fields and the hidden mouse commands or short cuts are described in this section There are four main sections of the widget 1 Buttons on the top row QUIT exit xcorr_astro COLOR manipulate color tables for display HELP display a short help file REFRESH refresh all plots 2 The display panels The images in the two display windows can be controlled via the but tons and editable windows provided to the right and left of the display windows Zooming capabilities are provided for the object display The FWHM option for the reference window is used by the centroiding routine 3 The selection panels These are all the fields between the display panel and the table These fields provide information on the currently selected stars When all values are filled in the user can add the current star to the table by clicking on the appropriate button 4 The cross correlati
296. mjansky This is a useful routine for finding the sensitivity of a particular CAM filter For example to convert a signal of 1 ADU observed with the LW2 filter to mJy CIA print adu to mjansky 1 lw2 0 429129 20 13 3 Color correction The conversion to mJy as performed by conv_flux Section 20 13 2 is given for the reference wavelength of each filter and assumes a source with spectral shape that follows a F A A law In reality your source may have a spectrum that deviates from this law To correct this effect CIA provides the routine corr colour You can use this routine to calculate a correction factor that may be applied manually to the results of your CAM photometry There are three ways to calculate this correction factor 258 CHAPTER 20 ADVANCED DATA CALIBRATION e If you believe your source has a blackbody spectrum then you just need to supply the blackbody temperature CIA corr colour l1w10 bb 5000 filter ok LW10 Filter LW10 ref wavelength 12 0000 correction factor K at ref wave 1 27288 by which you should divide the ISOCAM flux to obtain the actual flux density of your source In this case one would correct all flux density measurements of this source by dividing by the correction factor 1 27288 If your source has a power law spectrum then simply supply the power law exponent CIA corr colour 1 10 power 2 Finally if you have constructed a SED possibly from CAM measurements in
297. n origin SAD The lower window displays the MOSAIC from a future SAD A grid indicating image orientation and coordinates is overlayed on the MOSAIC 10 2 EXAMINING THE AAR DATA PRODUCTS 71 All the images that you will now view with sad display are held in SADs within IDL s memory These images are either EXPOSUREs or MOSAICs the former is taken from the CMAP file and the latter from the CMOS file Within CIA origin SADs hold EXPOSUREs and future SADs hold MOSAICs If you think that a MOSAIC is built from a set of EXPOSUREs then this terminology makes more sense Now try browsing through your data Initially the sad display image window will display an EXPOSURE from the CMAP e You can flick through the EXPOSURES in your AOT by clicking on the button Next SAD or by using the slider e You can also display MOSAIC from the CMOS file by clicking on the future button After the future button is selected clicking on the button Next SAD allows you to flick through all the MOSAICs in the CMOS e Clicking on the button colour will load the colour table e By clicking on the button Grid a grid is displayed indicating the astronomical coordinates of the current image e Glitches and point sources detected by AA may be displayed by clicking on the buttons Glitches and Sources These will be marked on the displayed image Note that this func tionality will only be activated when the keyword all is set when invoking sad display e Buttons
298. n routines are already coded However these assume that we are manipulating images that have a FITS header As you know by now with structures there is no need of a FITS header Therefore to use these astrometric routines we have to create a FITS header appropriate to the images This is done with the idl routine fits header The syntax of the call is the following with standard FITS keywords given with each parameter E 3 USING FITS IN CIA NEW PROBLEMS N N LW order 1 rot reverse image roll E N LW order 0 rot image 180 roll 333 Z N a E SW order 1 rot image roll E 0 Z SW order 0 rot reverse image 1 80 roll Figure E 4 The roll angle a for each detector and for each value of the IDL ORDER system variable The IDL command to rotate an image to the standard astronomical convention is shown for each case Note that the ROT function rotates clockwise 334 APPENDIX E ISOCAM ASTROMETRY ANGLES AND COORDINATES Figure E 5 Conventions for the standard astrometric keywords in a FITS header Axisl is the horizontal rightward axis Axis2 is the vertical upward axis CROTA2 0 the rotation angle described in the header has a definition which is different from that of a or 8 It is the position angle of the North axis counted positively eastward from Axis2 E 3 USING FITS IN CIA NEW PROBLEMS 335 my header fits header Nx NAXIS1 Ny NAXIS2 RefX CRPIX1 Ref Y CRPIX2 RA
299. n the SSCD and SCDs Sections 15 2 1 to 15 2 5 Note that the PDS differs from other observation data structures i e SSCDs SCDs and SSADs SADs in that it is not accompanied by dedicated CIA manipulation routines You may treat the PDS like any other IDL structure 15 5 1 Standard fields of the PDS The fields that are present in all flavours of PDS are listed in the table below along with a brief description and where appropriate a reference to the SSCD or SCD field where the data originates or otherwise the type of the data in the field 15 5 PREPARED DATA STRUCTURE PDS subfield OBSERVER TARGET TDTOSN SSSCD NAME AOT SAD_NAME NSCD CHANNEL FLTRWHL WAVELENGTH NBR FRAME _ PFOV TINT GAIN ADU SEC COEFF FROM i TO i MASK OTF RMS CUBE CUBE UNIT IMAGE IMAGE_UNIT DV UTK TABFLATCOEFF BOOTTIME TEMPERATURE DARK FLAT HISTORY description ID of observer target name TDTOSN number input SSCD name observation type output SAD name number of SCDs CAM channel CAM filter wheel wavelength of MOSAIC no of IMAGES in CUBE no of IMAGES per SCD pixel field of view integration time gain normalisation factor ADU to ADU gain sec IMAGEs in INFO SCD_NAME i are CUBE FROM i TO i mask of CUBE On Target Flag RMS error on EXPOSUREs no of IMAGE pixels per EXPOSURE pixel cube of IMAGEs from SCDs IMAGE units cube of
300. namic substructure or field named DATA see Section 15 3 2 and a standard set of fields see Section 15 3 1 DATA holds the actual calibration data dark images flat images point spread images etc The standard fields holds information related to the nature of the CDS structure itself e g the CDS name and size The CDS is almost a direct conversion from FITS format data to IDL data structure the important difference being that the actual image data is scaled so the BZERO and BSCALE keywords are discarded So the CDS differs from the CIA data structures in that the CAM parameters are NOT presented in the CIA user friendly format but as raw values taken from the CAL G FITS files The CDS is not a structure that you need to be very familiar with It is mainly handled by CIA routines For example get sscdraster will automatically obtain OFLT DFLT and DARK CDSs from the CAL G archive for calibration of a PDS However if you wish to view the data in a CAL G file you can use CIA conversion routines to make a CDS and convert the CAM parameters to a readable format we give an example later in this section Alternatively you may use IDL s Astronomy Library routines and follow the guidelines in the ISOCAM Handbook In any case a detailed explanation of the CAL G FITS files can be found in the ISO Data Product Document You may also need to refer to the ISO Satellite Handbook and ISOCAM Handbook for explanation of CAM parameters 15 3 1 Standard
301. native handling of these data Table 15 1 The calibration data and associated CDS mnemonic used for naming purposes see Section 15 3 1 for CDS naming convention A missing entry denotes that a CDS does not exist for that CAL G file in the case of the CWHEELS and ORBIT data these are handled in an alternative manner by CIA see Section 15 4 Refer to Section 9 3 5 for further information on CAL G files and where they may be sourced Also your attention is drawn to the ISOCAM Handbook and ISO Data Product Document for more detailed information on the contents of CAL G files 174 CHAPTER 15 CIA DATA STRUCTURE HIGH LEVEL ARCHITECTURE Now convert the CAL G file named K fits in our example to a CDS CIA dark calg2cds K dir calg dark dir n rows 5 creating CDS CCGLWDARK 96052312123000 Listing the CDSs in memory we see that two CDS exist CIA print cds list CCIGAIN 9504417414600 CCGLWDARK 96052312123000 CCIGAIN 9504417414600 is simply a conversion table for the raw values in 1 CIA gain cds get data 9504417414600 CIA help gain str Structure CCIGAIN STRUC 1 tags length 12 GAIN FLOAT Array 3 CIA print gain 1 1 00000 2 00000 4 00000 CCGLWDARK_96052312123000 or dark is the CDS we are interested in We can extract the DATA substructure with cds get and display its contents CIA dark data cds get data dark CIA help dark
302. nd CAL G FITS files 9 3 5 1 CIA s Calibration Data Structures All the necessary calibration data are delivered with CIA in the form of CDSs see Section 9 4 3 and Section 15 3 with notifications of updates available by email Access to the CDSs is usually handled by CIA routines and so is mainly transparent to the user For CIA users this is the best way to deal with calibration data It is certainly the most convenient and has the added advantage that CDSs will contain the most up to date calibration data 9 3 5 2 CAL G FITS files Alternatively calibration data in the form of CAL G FITS files can be sourced from an ISO CD ROM or the IDA at http www iso vilspa esa es Of course the ISO CD ROM files are current when the CD is pressed while the archives will always be up to date Note however that CAL G files from both these sources are official ESA releases of calibration data There will usually be some time lag between the release of ESA official CAL G files and the ISOCAM consortium official CDSs that are part of CIA Again another reason why CIA users generally only use CIA s CDSs On the ISO CD ROM the CAL G files are found in the directory products pmmmmmmm others and in subdirectories below this This should be clear from datalist txt Generally the names of the CAL G files begin with CCG where refers to the either the SW or LW detector 64 CHAPTER 9 THE DATA PRODUCTS AND CIA DATA STRUCTURES These files contain da
303. nd not in the cube The solution is to apply another flat field method right after the long term transient correction to create an estimate of the sky free of the long term transient component and then use DivSky Q I m trying to run the 1tt action on my raster and I encounter an IDL crash in the routine correct ltt Apparently it is trying to access a non existing element of table si_raster Why is that 21 7 FREQUENTLY ASKED QUESTIONS AND PROBLEMS 289 A Since April 2000 approximately the correct 1tt function uses the raster variable to find and mask ghosts before it evaluates the long term transient component Obviously this requires that the variable exists si raster is a variable that holds its dimensions correct ltt does not create it only make map does Make sure it has run before the 1tt action see also sec 21 5 3 290 CHAPTER 21 USING SLICE WITHIN CIA 35 0000 10 0000 25 0000 6 00000 Figure 21 2 The resulting maps for the Perturbed Single Flat Field determination Note that the map orientation has changed as SLICE always produces maps with North up and East left Imprints of the individual raster pointings are still visible 35 0000 10 0000 25 0000 6 00000 Figure 21 3 The resulting maps for the first flat field determination with the DivSky Note that the emission gradient produced by the long term transient is much smoother now and pointing imprints are mostly gone 21 7 FREQUENTLY ASKED QU
304. ng a filename for the output file choosing the window to dump to the file setting the orientation of the postscript output to landscape passing a title for the plot and placing the ESA CEA copyright notice in the plot window margin Note that ps_color needs sufficient room in the margins for placing text If the margins are not wide enough for this you may have to adjust your plot see the IDL User s Manual Each of these keywords appear in the following example in the order in which they have just been described CIA ps_color filename test ps win 0 landscape title hhh CIA gt copyright Alternatively ps_color can produce GIF output CIA gt ps_color filename test gif win 0 gif You may find the routines white and black useful with ps_color These routines change the colour table to black on white and back to white on black respectively making your output plot more readable and saving printer toner 14 8 Redirecting graphics to the postscript device ps open and ps print are used to redirect graphics to the postscript device The former opens an output postscript file and the latter closes the file Several keywords can be supplied to ps_open In the example below the output file to be created contains a plot in portrait orientation in colour and in encapsulated format Without setting the keywords it defaults to landscape grey scale and normal postscript respectively The name of the output file is idl
305. ng above These messages appear when there is slight mismatch between the CONFIGURATION of the best available calibration data and the data in the PDS Now that you have created a PDS it might be nice to take a look at its architecture Section 15 5 4 will guide you through its structure Since it is simply an ordinary IDL structure as opposed to a CIA data structure we can use IDL s HELP to look at its innards CIA help 1w6 raster str Structure 134390 54 tags length 2294560 refs 1 RASTERCOL INT 4 RASTERLINE INT 4 M STEPCOL FLOAT 96 0000 N_STEPLINE FLOAT 96 0000 RA RASTER DOUBLE 161 34459 DEC RASTER DOUBLE 55 959980 ANGLE RASTER DOUBLE 321 73999 RASTER ROTATION DOUBLE 411 73999 RASTER ORIENTATION STRING SPACECRAFT Y AXIS ASTR STRUCT gt ASTR STRUC Array 1 NX RASTER INT 80 NY RASTER INT 80 RASTER FLOAT Array 80 80 RASTER UNIT STRING 8 RMSRASTER FLOAT Array 80 80 NPIXRASTER FLOAT Array 80 80 AOT STRING RASTER TARGET STRING HAROO3 13 1 CREATING A PDS FROM AN SSCD 97 OBSERVER STRING LMETCALF TDTOSN LONG 14300601 CHANNEL STRING LW PFOV FLOAT 6 00000 TINT FLOAT 5 04018 GAIN FLOAT 2 00000 FLTRWHL STRING LW6 WAVELENGTH FLOAT 7 75000 NSCD INT 16 NBR FRAME INT 348 FROM INT Array 16 TO INT Array 16 TAB FRAME INT Array 16 ADU SEC COEFF FLOAT Array 16 TABFLATCOEF FLOAT Array 16 CUBE FLOAT Array 32 32 348 CUBE UNIT STRING ADU MASK BYTE Array 32 32 348
306. ning all the data from the up and down LW CVF scan can be created from this single big SSCD CIA cvf pds get sscdcvf cleaned sscds 0 After calibration a spectrum from an up and down CVF scan should look something like that in Figure 19 1 Note that the same procedure can be applied to an up and down SW CVF scan The difference is that there will only be 2 SSCDs to concatenate since there is only one SW CVF segment 19 5 2 Mixed LW and SW observation The example in this section is from a CVF observation that is comprised of a mixture of LW and SW data This is what the output SSCD of spdtoscd looks like CIA gt sscd_info sscd 140 SCDs in the SSCD C55C849007010101_98092714013700 seq channel mode fltrwhl pfov tint gain offset size ra dec LW IDLE LW CVF2 1 5 25 20 1 512 1 999999 99 999999 99 1 LW IDLE LW CVF2 1 5 2 10 1 512 1 999999 99 999999 99 2 LW IDLE LW CVF2 1 5 2 10 2 512 4 174615 21 285011 96 3 LW OBS LW CVF2 1 5 2 10 2 512 2 174615 22 285012 12 4 LW OBS LW CVF2 1 5 2 10 2 512 1 999999 99 999999 99 5 LW OBS LW CVF1 1 5 2 10 2 512 117 174615 22 285012 00 6 LW OBS LW CVF1 1 5 2 10 2 512 15 174615 22 285012 00 19 5 ADVANCED SLICING OF CVF DATA CAMO04 44 45 46 47 48 87 88 89 90 91 92 93 94 95 96 LW LW LW LW LW LW LW LW LW LW LW SW SW SW SW SW SW SW OBS OBS OBS OBS IDLE OBS OBS OBS OBS OBS IDLE IDLE CLEA IDLE OBS OBS OBS OBS LW CVF1 L
307. noscd Two SSADs are created one being the origin and the other the future SSAD see Sec tion 15 2 4 The keyword ack returns a logical value 1 for a successful call and 0 for a failure Other useful optional keywords to aa2sad are nowrite and noscd Setting nowrite will inhibit writing of SADs to disk Setting noscd as in the example inhibits the default creation of SCDs along with the SADs 17 1 3 SCDs from SPD spdtoscd To convert SPD data products to SPD SCDs in memory or on disk we use the procedure spdtoscd Using the assigned variables of Section 17 1 1 an example call is CIA spdtoscd cisp14300601 fits sscd dir product dir CIA scd dat scd dir ack ack The output parameter sscd will contain the name of the created SSCD Only one SSCD exists for a single CISP file The SSCD and its SCDs will be written to scd dir permission to write on this directory is required The keyword ack returns a logical value 1 for a successful call and 0 for a failure For more advanced users the UTK start time and end time of required SCD records can be passed as keywords to spdtoscd see the on line help or cia help In addition the keyword bs may be of interest to observers with beam switch data Beam switch observations have an intermediate step which occurs while slewing and data accrued in this step may appear in an SCD Normally this SCD is considered to contain invalid data Setting bs will make spdtoscd merge this SCD with
308. ns and between two filters or PFOV when more than a configuration was programmed in a single AOT You now have to select which SCDs you want to create and in which SSCD you want to put them Let s examine together a concrete example Figure 12 2 displays the x handle slice for an observation of a cluster This observation consists of two 4x4 rasters made with the 3 arcsec PFOV lens one with the LW2 filter the other with the LW3 filter Of course we are going to make two different SSCDs with these ERD one for the LW2 filter the other for the LW3 filter Let s begin with the LW2 raster e First click on the Unselect All SCDs button All the toggle buttons on the Make your Choice column are now unselected and the Merge buttons are insensitive e Select the lwfil_2 item in the Advanced Slicing Filters menu All LW2 SCDs are now selected e Click on the Redisplay button You are back in the same situation as before making any choices but the x handle slice window displays only data obtained with LW2 filter A way to select the frames of our LW2 observation is to use the number of exposures all the lens and filters motions take only a few frames but in this case at least 30 frames have been taken per raster position Here it goes again e First click on the Unselect All SCDs button 94 CHAPTER 12 DATA SLICING e Select the Nb Exposure item in the Advanced Slicing menu A window appea
309. nsient is much smoother now and pointing imprints are mostly gone Another important point is that the 1tt action tries to remove ghosts before it estimates the long term transient component this is made using the variable raster which contains the flat fielded cube of raster pointings see table 21 1 However the action itself does not create this cube Therefore make sure it exists it generally will since it is created and updated by the action before you start the 1tt action In fact the long term transient determination adds only one new parameter 1tt_thresh This is similar to the flat thresh parameter in the sense that it allows to discard pixel readouts from the long term transient determination Typically this is used to discard pixels observing strong sources where the short term transient correction may not have been ideal On each raster position pixels 1tt threshxoc above the mean level of the current sky position are discarded note that the exact meaning is opposite to that of flat_thresh a higher value implies that less pixels are discarded In fact SLICE displays the raster positions with the discarded pixels masked good way to check that your setting is correct is that the location of strong sources are masked In our example for the LW3 band and to a lesser extent for the LW2 band as well the DivSky method always produces the oscillating artifacts and thus we are going to use the Perturbed Singl
310. ntaining the names of three SSCDs with data from observations done with the first CVF LW segment the second CVF LW and the CVF SW segment The LW data from both segments can be combined into one SSCD for conversion to a CVF PDS CIA sscd concatene cleaned sscd 0 cleaned sscd 1 CIA lw cvf pds get cvfstruct cleaned sscd 0 218 CHAPTER 19 ADVANCED SLICING 19 6 Advanced slicing with x slicer This section completes the guide begun in Section 12 3 to x slicer This section addresses all possibilities that are allowed by x_slicer They have been classified by the order of appearance of the windows in which they occur 19 6 1 Files directories and x slicer customization 19 6 1 1 Where are the data By default x slicer will search some directories for the data It uses three variables to do so When you use the Automatic Find to get all the files these variables are used They are all defined at the beginning of the x slicer pro routine see Section 19 6 1 2 to customize these settings Here is how it goes for ERD e Start a pickfile IDL program routine in the default data directory e From this point the user searches for ERD and selects it Let us call the selected directory user directory x slicer looks for the IIPH file in the user directory If needed and if a CDS is not to be used x slicer looks for the IFPG file in the user directory If the file is not in the user directory x
311. o o TN ES ES orm D ua Figure 20 7 Original pixel histories of the same source Note the source has different signal levels in different SCDs After reduction this will propagate into different signal levels in the corresponding EXPOSUREs 264 CHAPTER 20 ADVANCED DATA CALIBRATION M M o o N 2 1 e ES MOS Pata ke ae e or 4 or 4 c Bn 4 4 Jg NT 72 NP 1x or jx 40 Jo 1 L 4 L 4 CI L JN L L E L im H 4 H 4 B 40 L 4 ch o F 1 1 L 40 L 4 L JO L Jo g E JS sc 49 oL Ja oL 4 L 4 H 4 AT 7 AT Ok 4 orm 4 4 C a 1 1 1 1 1 1 1 1 1 1 pot i pa 11 ANTI ae ix o o 5 M zx ES E E E ul Ts N e o ER SER NES ae rou orm 4 GL al OL d git 19 gl 19 L dx L x Sr Je f 19 B TN In TN L 15 L 4 L 4N L al ca L o F 10 F TO L j L l1 10 wf 40 wt 40 OR 40 Ok 0 eL 4 4 L o gt Jo af 1 af 1 orm 4 a 4 oa 4 Blasi pu Li LI 1 poi tos Figure 20 8 Back projected pixel histories of the same source Note how the signal is of the same averaged or idealized lev
312. o data n data n bscale n bzero n The scaled data must now be reformed to the appropriate image size To find the actual image size in rows and columns we look at the fields NAXIS1 and NAXIS2 respectively CIA naxisi array naxis1 CIA gt naxis2 array naxis2 CIA gt print naxisi naxis2 80 80 80 79 79 79 80 80 80 80 80 80 To reform the images CIA nb naxisi naxis2 1 CIA gt for n 0 5 3 do tviso CIA reform data 0 nb n n naxisi n naxis2 n When image sizes differ only the first nb entries in each row of data are valid Each sky image stored in the CMOS is accompanied by a second equally sized image indicating the noise level per pixel and a third image indicating the statistical weight assigned to each pixel So in the example above we only display every third image i e the sky images 17 2 2 Extracting a key from an ISO data product With the procedure by extended fits key we can read selected keys directly from a FITS file this is more efficient then unnecessarily reading the entire file T he following extracts the data from a CISP file and displays an image CIA by extended fits key cisp14300601 fits data cispdata CIA help data DATA INT Array 1024 813 CIA tviso reform data 0 32 32 Refer to the ISO Data Product Document to find where the data is stored in the binary table i e cispdata Chapter 18 Export of CIA data str
313. o measure the x y positions of the limits of the image containing good quality data 240 CHAPTER 20 ADVANCED DATA CALIBRATION 3 It is important to mark the boundary of the image beyond which the pixel values are invalid due to the shift Given the x y positions determined previously flag edges will generate a mask with 1 where pixels are valid and 0 elsewhere CIA raster pds npix CIA flag edges x1l yl x2 y2 raster pds nscd raster pds npix When the raster MOSAIC is created the invalid pixels will not be included 4 Now correct the library flat with register flat This routine will shift the optical flat and multiply it by the detector flat CIA my flat register flat x1 raster pds 5 Supply the corrected flat to corr flat CIA corr flat raster pds inflat my flat That completes the flat field correction 20 2 7 Small mirror and unilluminated pixels To avoid straylight for the 3 PFOV normally the small Fabry mirror was used However this results into an incomplete illumination of the detector For raster observations standard treatment leads to an uneven MOSAIC as illuminated and unilluminated pixels are averaged left picture in Figure 20 1 In order to produce a publishable MOSAIC right picture in Figure 20 1 the unilluminated pixels should be masked out before the MOSAIC is generated as shown in the example script below CIA gt x3d raster_pds image CIA gt Cube 4 28 4 14
314. o spd scds i erd2spd erd scds il CIA ssid erd sscd del Note that the del keyword is set to delete the ERD SCDs from memory as the SPD SCDs are created Finally for convenience create a new variable containing the name of the SPD SSCD Since the name of SSCD is unchanged this is achieved by copying the variable erd_sscd CIA spd_sscd erd_sscd Now we can remove the old ERD SSCD and its ERD SCDs from disk We can use sscd_remove to do just that CIA gt sscd_remove erd_sscd dir scd_dir Finally we can save the SPD SSCD with its new SPD SCDs CIA gt sscd_write spd_sscd dir scd_dir As an alternative to the above steps using spdtoscd we can directly create SPD SCDs in memory and on disk with the SPD data product CIA gt spdtoscd cisp14300601 fits spd_sscd dir product_dir CIA gt scd_dat scd_dir ack ack Our next step is to find the data we require from the SSCD The whole AOT is contained within the SSCD but we may be only interested in a single CONFIGURATION In our example here we have a raster observation with two CONFIGURATIONS one for filter wheel LW6 and one for LW3 and in addition several S ATEs corresponding to internal calibrations and detector stabilisation We can get an overview of the contents of the SSCD with sscd_info it lists important parameters of all the SCDs in an SSCD These parameters are simply values extracted from fields within the SCD structure The content
315. observation As an example we can see which MOSAIC corresponds to which filter by extracting information from the corresponding future SAD The future SAD s field WAVELENGTH holds the wavelength of the filter wheel used in the CAM CONFIGURATION from which AA has created the MOSAIC Likewise the same field in the origin SAD will hold the wavelength of the filter wheel used in CAM STATE The value of a field in a CIA data structure can be extracted with the CIA routine sad get CIA print sad get wavelength CSAD000014300601_96091713370625 7 75000 CIA print sad get wavelength CSAD000014300601_96091713370548 14 5000 We can also extract the image from the SAD and place it in a regular IDL array The field CMAP DATA in the future SAD contains the MOSAIC from the CMOS data product The same field in the origin SAD contains an EXPOSURE from the CMAP data product CIA lw3_image sad get cmap data CSAD000014300601_96091713370548 CIA lw6_image sad get cmap data CSAD000014300601_96091713370625 CIA tviso lw6_image lw3_image In the above example the MOSAICs are extracted from the SADs and their ratio is displayed The CIA routine tviso is a useful variation on IDL s TVSCL the difference being that it displays the image in a convenient size Though you have created SSADs in memory they have not been saved to disk You may wish to save them rather than having to go
316. of bright sources are not corrected by either the 1tt or make map action Pixels that have been hit by strong glitches may still show up in the map as holes or fake sources This section thus describes the tool that SLICE offers to correct these artifact For those that may already be familiar with SLICE the current version April 2000 has introduced a dramatic change by merging the ghost bad pixels and source actions into one the bad pixels action A note of warning though if your observation contains bright small scale structure it may be affected by the bad pixels action It is important that you compare in detail the result of the action to the previous state of the map in map before to make sure you understand what has been done An important point to mention before we begin the action described here affect neither the data cube variable c nor the cube of raster pointings contained in raster It will create new versions of the sky map in map the error map error_map and the redundancy map redun Also note that as for the make map action you have the possibility to work on the full cube or simply on the cube of raster pointings here I recommend working on the full cube by placing the docube keyword on the call to slice pipe 21 6 BAD PIXELS GHOSTS AND SOURCES 287 35 0000 10 0000 6 00000 Figure 21 6 The results of the long term transient correction and variable flat field determi nation Variable flat fi
317. offsets as computed with the gauss method Data are from a CAM calibration observation of HIC 73005 also featured in Chapter 4 In this particular case the jitter standard deviation is observed to be about 0 065 The routine corr jitter is the user interface to the jitter computation routines It can operate on any flavor of PDS It is recommended that the PDS is at least dark corrected and deglitched before attempting to compute the jitter To perform the default jitter computation corr jitter can be run as CIA corr jitter pds You can view the computed jitter offsets with CIA plot pds du CIA plot pds dv However jitter computation is an interactive process and a choice of methods and options are available The available jitter computation methods are described below See Figure 22 2 for a comparison of jitter computation methods 1 method gauss method Apply a 2D elliptical gaussian fit to each IMAGE See the on line help for fit_2dgauss for a description of the gaussian function routine called fit_2dgauss PDS side effects DU and DV are filled with the jitter offsets du dv for each image 2 method psf method Select the best fitting theoretical PSF from a library routine called fit_psfstruc PDS side effects DU and DV are filled with the jitter offsets du dv for each image Some useful keyword options are 22 1 JITTER CORRECTION 295 0 20 0 10 0 00 0 10 du 0 20
318. old a MOSAIC The SAD contains all the standard fields of Section 15 2 1 and in addition extra fields asso ciated with the type of analysis which may have been performed on the data For example the PFOV is a standard field because it is relevant to an image regardless of the processing but the unit of pixel intensity TUNIT is dependent on how the data has been processed Below is a list of fields that are considered most useful to the CIA user 1 CCIM CMAP Both CCIM and CMAP each contain further subfields To access these replace DATA with either CCIM or CMAP 170 CHAPTER 15 CIA DATA STRUCTURE HIGH LEVEL ARCHITECTURE subfield description of contents type DATA TUNIT pixel unit of intensity string DATA RA RA J2000 of CRPIXI double DATA DEC DEC J2000 of CRPIX2 double DATA ROLL roll of the DATA double DATA CRPIX1 reference pixel float DATA CRPIX2 reference pixel float DATA DATA actual image data float IDL array DATA RMS_IMAGE error on image float IDL array DATA WEIGHT IMAGE exposure time secs per pixel integer IDL array DATA MASK as for SCD Section 15 2 2 byte IDL array The subfields RA DEC ROLL CRPIX1 and CRPIX2 contain parameters that follow the FITS convention RA DEC and ROLL are in decimal degrees CRPIX1 and CRPIX2 are the x and y coordinates of the reference pixel respectively and refer to the centre of the reference pixel e g for reference pixel 16 16 CRPIX1 16 5 and CRPIX2 16 5 N_RASTER As for SCD see
319. ommand without any arguments or by setting the help flag in the command line call produces a short help on xcorr astro syntax Required and optional data 1 ISO data structure with or without astrometry information the data target field 2 A reference image with pre determined astrometry information the reference field 3 Catalog of point sources optional Step by Step guides i To build a cross correlation source table between two image Please consult usage syntax and required data sections above before starting Once xcorr astro gets going follow these steps 1 Set the display settings which suit your needs The min max floor ceiling buttons etc can be used for this 2 Visually examine images to find stars common to both images 3 Select one of these common stars in the target field by clicking on it with the left mouse button The current selection fields located immediately below the target field are popu lated by approximate values for the star s position The IX and IY fields are integer pixel references in IDL convention The X and Y values are fitted centers see next step we use the CIA positioning convention for these 4 Improve the star s centroid by either using the FIT PSF or the CENTROID option FIT PSF will use an appropriate PSF library to find the best fit PSF match to your input star CENTROID performs a simple brightness weighted mean 5 Once satisfied with the centroiding
320. on The routine bkg builder can be used to manually select IMAGEs from the cube of a PDS or indeed any cube for use in determining a background subtraction frame Since this requires that you have to have IMAGES containing no source then it is most likely that bkg builder will only be of use to you if you have raster observation data To invoke bkg builder CIA background frame bkg builder raster pds cube Four windows will appear DATA displays all the frames IMAGES in the input cube VIEW displays all the frames in the input cube which you have selected BACKGROUND displays the current background frame BUILDER is the main window and the user s interface to bkg builder see Figure 20 5 To make a background frame with bkg builder follow this basic procedure 1 Select some frames by clicking on the panel of buttons under the legend Click on frame number to select unselect it As you select your choice of frames will appear in the window VIEW and the currently selected frame will be displayed in the bottom of the main window 2 Now click on the buttons mean or median and the a mean or median frame of the selected frames will appear in the window background 3 Some statistical information about your background frame will appear in a subwindow of the main window When you are happy with your background frame you can click on done to exit Now you chosen background frame will be in the variable background frame 2
321. on table This includes the table itself the buttons at the bottom row and the jitter correction information to the right of the table The tables can be edited The available edit operations are the buttons available below the table To edit first select the appropriate row by clicking on it and followed by the edit operation 152 CHAPTER 14 IMAGE ANALYSIS AND DISPLAY Figure 14 18 xcorr astro window 14 7 CREATING HARDCOPY PLOTS 153 14 7 Creating hardcopy plots 14 7 1 Using xcontour This is a widget based program for producing contour plots for screen display and postscript output You call it with a single image specifying the window for display keyword win if you wish CIA xcontour raster pds raster win 1 xcontour presents you with a widget for choosing contour levels and contour label sizes e Click on apply when you want to see your contours plotted e Click on annotate to display a widget which allows you to annotate text and graphics to your plot This is usual for marking regions of interest in an image options gives you the postscript output options available Options for producing an output file are available in the drop down menu under the push button file e Click on print to screen dump the graphics window to a postscript file 14 7 2 Screen dumps with ps color ps color produces a postscript or GIF format file from the contents of an IDL graphics window This routine has keywords for specifyi
322. ons in an image The image could be the MOSAIC RASTER in a raster PDS or an EXPOSURE IMAGE in all PDSs As an example we can take the fifth EXPOSURE from a CVF PDS CIA xsnr cvf pds image 4 113 114 CHAPTER 14 IMAGE ANALYSIS AND DISPLAY The xsnr window presents three subwindows to the user Two contain a display of the noise region selected and of the signal region selected Selection of these regions can be made with sliders The third subwindow displays the full image the first in a cube with both noise and signal regions indicated by a white box Clicking on the button Calcul with calculate the ratio of the signal and noise in each region 14 1 2 Energy radial profile An estimation of the energy radial profile of a source may be done with the widget based program xradial CIA radial profile xradial raster pds raster The brightest pixel in the image is automatically selected and the user may interactively set the radius of the disk around that pixel from which the profile is taken with a slider The integrate button toggles on off integration of the profile Calculated data are displayed in a subwindow of the widget and is also the output of the function In the preceding example radial profile contains these data when xradial is quitted To reproduce the profile try CIA plot radial profile 14 1 3 Estimating the total source flux The routine flux sum will help estimate the total flux from a point sour
323. order to indicate that the data status is unknown A data history has been designed in order to help the user to keep a track of all processes which have been applied to the data set This history can be displayed on the screen at any time using Data Display History The file data_history tat is then created in the current directory log file of the whole x cia session can also be displayed on the screen using CIA Log File This log file contains all commands warning and error messages of the current x cia session The default name of the log file is cia log tzt This file is automatically overwritten at the beginning of a x cia session unless the noinit keyword is used The user can also choose another name by typing CIA x cia logfile my_name txt 13 3 3 4 Example 2 testing several methods for flat field correction Let s assume that the user wants to correct data from dark current glitches and transient effects and then try several methods for the flat field correction Assuming that a user input frame an IDL 32x32 array called my flat in the following is used as flat field it is necessary to load it using the INFLAT keyword Hence the sequence of commands would be the following 1 Start a CIA session 2 Start x cia with the following CIA x_cia inflat my flat 3 Choose AOT Type default is raster scan 4 Data Load SSCD 5 Data AOT Info 6 Process None 7 Dark User Input 8 Deglitch
324. ostscript format and can be found in the directory doc just below the top of the CIA distribution This documentation includes Technical reports These are referred to in the CIA User s Manual where relevant and are also listed in Appendix K User guides These are guides to individual routines and written by the authors of the routines Again they are referred to throughout the CIA User s Manual and are also listed in the bibliography is a registered trademark of Research Systems Inc 1 2 CHAPTER 1 ABOUT THE CIA USER S MANUAL Note that the documentation is only as up to date as the installed version of CIA The ISO web site is the best source of current documentation http www iso vilspa esa es The ISO Explanatory Library contains full documentation on ISO and the ISO instru ments http www iso vilspa esa es users expl lib expl lib html The CAM Instrument Page contains specific information on CAM http www iso vilspa esa es users expl lib CAM top html 2 ISO documents Throughout the CIA User s Manual references are made to the doc uments below They are retrievable browsable from the ISO web site and IDA address above or found on an ISO CD ROM see Chapter C along with the data products The ISO handbooks especially the e ISOCAM Handbook e ISO Satellite Handbook The Observer s Manuals e ISOCAM Observer s Manual e ISO Observer s Manual The IDPD e ISO Data Product Document 1 3 Reportin
325. ot of the values of a selected pixel throughout the cube of EXPOSUREs As stated earlier the EXPOSURE is derived from the data in a STATE or SCD and since the CVF wheel position changed as the STATE changed then the EXPOSUREs are pictures taken over a range of wavelengths This means the plot is a CVF spectrum of the sky covered by a selected pixel You can select pixels by right clicking on the EXPOSURE The CVF spectrum is calibrated to flux in mJy per square pixel against wavelength in microns 6 You may wish to save the results of the data analysis You can do this with IDL s SAVE CIA gt save file cvf_pds xdr cvf_pds Alternatively you can export each of the CVF EXPOSUREs in IMAGE to individual FITS files see also Section 18 2 7 2 DATA ANALYSIS 49 iun 5 8 to NONIS maE TUA Figure 7 1 The left panel of the cvf display window displays an EXPOSURE from cvf_pds The right pane displays a plot of the CVF spectrum of a pixel selected from that EXPOSURE CIA imagette2fits cvf pds name cvf fits CIA 1s cvf fits cvf1 fits cvfi3 fits cvf17 fits cvf20 fits cvf6 fits cvf10 fits cvf14 fits cvf18 fits cvf3 fits cvf7 fits cvf11 fits cvf15 fits cvfi9 fits cvf4 fits cvf8 fits cvfi2 fits cvf16 fits cvf2 fits cvf5 fits cvf9 fits Additionally you can correct the CVF EXPOSUREs in IMAGE for distortion see Sec tion 20 15 5 50 CHAPTER 7 CVF OBSERVATION Chapter 8 Polarization observa
326. otice that the BS PDS is almost identical to the raster PDS This is purely for reasons of simplicity and compatibility of previously written processing routines e CAL G see Section 15 5 6 e CCIM see Section 15 5 8 e INFO see Section 15 5 7 e ASTR see Section 15 5 9 e Standard PDS fields of Section 15 5 1 e DS PDS specific fields These fields are listed in the table below along with a brief description and where appropriate a reference to the SSCD or SCD field where the data originates or otherwise the type of the data in the field Again you will see similarities with the architecture of the raster PDS some fields common to both have fixed values Note that the MOSAIC in this case is the source reference EXPOSUREs 184 CHAPTER 15 CIA DATA STRUCTURE HIGH LEVEL ARCHITECTURE subfield RA RASTER DEC RASTER ANGLE RASTER RASTER ORIENTATION RASTER ROTATION M STEPCOL N STEPLINE RASTERCOL RASTERLINE NX_RASTER NY_RASTER RMSRASTER NPIXRASTER RASTER RASTER UNIT BEAM SRCIMAGE REF IMAGE NCYCLES description RA of MOSAIC centre DEC of MOSAIC centre spacecraft ROLL not used by BS PDS rotation w r t celestial axes not used by BS PDS not used by BS PDS not used by BS PDS not used by BS PDS no of cols in MOSAIC no of lines in MOSAIC RMS error on MOSAIC no of IMAGE pixels per MOSAIC pixel MOSAIC image MOSAIC units beam switch flag index to so
327. ou can begin by selecting EXPOSUREs which you want to make up your FLAT Clearly it is best to pick those with no signal Selection is made by clicking on the buttons located in the main window under the title click on frame number to select unselect it 2 When you have finished the selection click on SELECTION DISPLAY to display all selected EXPOSUREs in the reference window 3 You can experiment with a median of all the selected EXPOSUREs click on MEDIAN or a mean click on MEAN Each time you click the resulting FLAT is displayed in the display region of the main flat builder window Also statistical information on your FLAT and messages will appear in the text sub windows in the main window 4 Now click on TEST FLAT bottom of main window This applies your FLAT to the EXPOSUREs and the results are displayed in the reference window 5 You can also examine row and column profiles of the computed FLAT by clicking on the button PROFILES Along with a new plot displaying profiles the following message will appear in the CIA command window Left mouse button to toggle between rows and columns Right mouse button to Exit One important note All other flat builder functions are suspended while profiles are being viewed Exit this function by clicking on the FLAT with the right mouse button 6 You can compare your FLAT with the CAL G FLAT Click on CAL G and then on T EST FLAT The CAL G FLAT is applied to the EXPOSUREs and again t
328. out y out background 302 396 11 1429 24 4762 6 83221 photom_aper High level routine to perform aperture photometry Calling syntax CIA gt photom_aper raster aper_flux 13 83 radius 7 CIA gt print aper_flux 556 184 As NGC 5195 isn t a point source the flux derived by PSF fitting is lower than the flux derived by aperture photometry 14 2 CVF image analysis 14 2 1 cvf_display cvf_display is a widget program which you can use to calculate a CVF spectrum Assuming that you have a calibrated CVF PDS then cvf_display can be simply invoked with CIA gt cvf_display cvf_pds flux cvf_spectrum A single window will appear with an image display subwindow and a plotting subwindow see Figure 14 2 A mean of the EXPOSUREs in the CVF PDS field IMAGE is initially displayed 1 Clicking on a pixel will display its spectrum arbitrary flux units against wavelength mi crons in the plotting subwindow 2 Clicking on a point on the plot with the right mouse button will display the EXPOSURE at that wavelength 3 A region of an EXPOSURE can be selected and the mean spectrum of its pixels plotted Click on a pixel in the EXPOSURE with the middle mouse button and then release the button over another pixel A mean spectrum of the subarray defined by these pixels will be displayed 4 If you have kept the SAD created by get sscdcvf when you first made the CVF PDS then the current spectrum in the plot window can be placed in
329. ow the values of some keys in FITS file instead of loading the entire file into memory it simply reads the required keys The behavior of x slicer in this mode is transparent to the user 19 6 ADVANCED SLICING WITH X_SLICER 229 In addition to using the big file button you may want also want use the save slicer file button located in the x handle slice window see Section 12 2 See Section 19 6 9 for details on its use 19 6 9 The save slicer file button and slicer to cia If you have a large data set you can simply slice them with x slicer working on data with the x handle slice window Only the last action is different Instead of hitting the Continue Button hit the Save Slicer file button You will then be asked to enter a SSCD name Once you have done that hit the OK button x_slicer creates a file named sscd_name sli This file contains the current x_slicer settings but no data To use this file just quit x_slicer and even quit CIA in order to free all memory Then enter CIA again and type CIA gt COMMON slicer_data toto CIA slicer to cia sscd name sli dir sli directory of sscd name sli CIA sscd dsd save All your SCDs will be created just as if you never exited x slicer Moreover SSCD and DSD variables will contain your created SSCD and DSDs IMPORTANT check first that the data you entered in x slicer are still in the same directo ries because the sli file does not contain any dat
330. pa esa es Appendix K Technical reports Abergel A et al 1996 IAS model for ISOCAM LW transient correction Altieri B et al 1998 CAM Calibration Explanatory document Altieri B et al 1998 ISOCAM Faint Source Report Aussel H 1998 ISOCAM LW Channel Field of View Distortion Aussel H 1996 ISOCAM Data Preparation with X_SLICER v2 1 Biviano A 1998 ISOCAM Calibration Error Budget Report Chanial P and Gastaud R 2000 Ximage manual Chanial P and Gastaud R 2001 Xcube manual Chanial P and Gastaud R 2001 CIA HTML HELP Claret A 1996 ISOCAM Data Analysis with X CIA v2 2 Claret A et al 1998 A Learning Guide for ISOCONT v2 0 Claret A and Dzitko H 1998 ISOCAM Glitch Library Galais P and Boulade O 1998 Report on Trend Analysis of CAM daily Calibration Gastaud R 1999 Technical Note on the error of raster in the software CIA Leech K and Pollock A M T 2000 ISO Satellite and Data Handbook ESA document SAI 99 082 DC Moneti A et al 1997 Reference Wavelengths for ISO CAM and PHOT filters Roman P and Ott S 1999 Report on the behaviour of ISOCAM LW Darks Sauvage M 1996 Angles and ISOCAM LW v2 0 Sauvage M et al 1999 X PHOT a crude photometric package for CIA v3 0 Sauvage M 2000 An Introduction to SLICE inside CIA V0 9 Saxton R 1999 ISO Data Product Document ESA document ISO SSD 9111A 353 354 APPENDIX K TECHNICAL REPORTS Siebenmorgen R et al 1996 Addendum to the
331. pact Status COMPACT STATUS HISTORY Automatic Find On Off Directory Orbit file ORBIT FILE Automatic Find On Off Directory Entrance Wheel Observation Channel Integration Time Raster Position Selection Wheel Beam Switching Detector Gain Raster Position F Lens Wheel A O T Observation Mode 1 Sequence Number F LH Filter Wheel Observation Type Target Flag Parallel Mode SW Lens Wheel Detector Offset Raster Mode Parallel Aperture SH Filter Wheel On Board Process LOAD TDF LOAD ERD LOAD Store Data SLICE Figure 12 1 x slicer window 88 CHAPTER 12 DATA SLICING e Select the variables to perform the slicing e Run the slicer e Select the SCDs that you really want to build 12 3 2 Selecting a file The first step to produce our SCDs is to select the file that contains the data This is done very easily by following these few steps e Check that the three Automatic Find buttons are On pressed This should be the case since it is their default configuration These three buttons tell the x slicer to look for the files it will need at some default places If you meet problems using these buttons take a look at Section 19 6 1 e Check that the Use CDS if needed button in the IFPG file area is On pressed This again should be the case since it is its default configuration This button allows you to select whic
332. pe where a CVF observation is performed AOT 5 An AOT type where a polarization observation is performed Most polarization ob servations were performed by the CUS as AOT 99 observations AOT 99 CUS observations mainly for calibration purposes 305 306 APPENDIX A GLOSSARY beam switch observation Also AOT 423 an observation where CAM alternates pointings on a target object with pointings on the selected fields of empty sky surrounding the object The empty sky fields are known as the reference fields There can be up to four reference fields in such an observation beam switch structure See PDS calibration data Calibration data refers to data such as flat fields DARK images etc For the user this is synonymous with CAL G data calibration data structure The calibration data structure CDS is the CIA data structure which is used to hold calibration data CAL G General term referring to CAM calibration data products Note that in the CIA User s Manual the terms CAL G and calibration data are synonymous In CIA CAL G files are converted to Calibration Data Structures CDSs and the user does not work directly with them calibration data Refers to flat field images dark images etc CAMTU CAM Time Unit 1 CAMTU 0 14000498 seconds CDS Calibration Data Structure A CIA data structure for holding CAL G data products See entry for CAL G CCIM CAM Calibrated IMage n AAR level data product containing AA comp
333. plot PALOMAR TO ICI sort Mag sort the data with Increasing viae DISPLAY MOUSE INPUTS Lr SHOW CALO ciliking on a pixel of the image with The left or The rigth button will uem Citing On a piva with the middle button wil the on pixel ww m mean cele spectrum of pixels that are between the piel where you push the mille EVICE TO s button and the pixel where You relegin It AD PEI n gr the inci METI VERI M Figure 2 3 typical cia help display 8 CHAPTER 2 ABOUT CIA In addition there is cia html using a browser to display html files with hyper links both to the code source and to the routines listed in the SEE ALSO section of the header There are two possibilities to call cia html If you don t know the name of the procedure CIA cia html will offer you a list of all routines az routines and several categories processing graphics input output miscellaneous and the astronomical library see Figure 2 4 To get deeper into the help choose a category you are interested in and click on the appropriate link A new page will appear presenting you the list of routines for this category see Figure 2 5 If you know the name of the routine then you can access the help directly via CIA cia html x3d Figure 2 6 gives you a typical display In addition to the usual buttons you have a previous routine and a routine field
334. pose to build a new SCD each time such an event happens Of course you don t want to produce all these SCDs because some of them will be made of frames taken while the camera is adjusting its configuration for the observation that you have requested One other point is that some observations have been made with different configurations in a single AOT for example you will find in the same file the same raster on the same sky position done in two different filters If they are put in the same file it does not mean that they should be assigned to the same SSCD All these data manipulations are done on the x handle slice window that appears in front of you 12 3 5 The x handle slice window Figure 12 2 shows the x handle slice window It is divided in two parts each of them is divided in two parts too Let s go through them The top of the window presents the results of the slicing process On the left a large table gives you the values of the variables you choose for each SCDs that the x slicer has found remember that the slicer creates a new SCD each time one of the variables changes The titles on the top have been shortened so Table 12 1 presents the conversion On the right of this table are two columns of buttons named SCD and Merge SCD where stands for the number of the SCD Note that by default all SCDs are selected and none are merged together There are scroll bars that allow you to align the row containing the informations abou
335. positions 19 6 ADVANCED SLICING WITH X_SLICER SCD names DSD names entrance wheel F1 W1 POS ENTWHL F1 BLOCK HKI1 POSI1 CAL EWHL selection wheel F1 W2 POS SELWHL F1 BLOCK HK1 POS2 CAL SWHL LW lens wheel F1 W3 POS PFOV F1 BLOCK HK1 POS3 CAL PFOV LW filter wheel F1 WA4 POS FLTRWHL F1 BLOCK HKI1 POS4 CAL FCVF SW lens wheel F1 W5 POS PFOV F1 BLOCK HK1 POS5 CAL PFOV SW filter wheel F1 W6 POS FLTRWHL F1 BLOCK HK1 POS6 CAL FCVF detector and its configuration observation channel F2 IM ORIG CHANNEL F2_BLOCK F2IMORIG CAL DEID integration time F2 INT TIME TINT F2 BLOCK F2INTTIM CAL TINT detector gain F2 ADC GAIN GAIN F2 BLOCK F2ADCGAI CAL GAIN detector offset F2 ADC OFFS OFFSET F2 BLOCK F2ADCOFF CAL OFFSET electronic setup on board process F2 IM PROC OBP MODE F2 BLOCK F21MPROC CAL OBC observation mode F2 AOT OPM MODE F2 BLOCK F2AOTOPM CAL MODE observation related variables observation type F2 AOT OBS COMMANDER F2 BLOCK F2AOTOBS CAL COMMANDER A O T F2 AOT AOT AOCT F2 BLOCK F2AOTAOT beam switching F2 0PER FLAG BEAM F2 BLOCK F2BMSWFL raster mode F RAST F RASTER F2 BLOCK F RAST M raster position M RAST M RASTER F2 BLOCK M RAST N raster position N RAST N RASTER F2 BLOCK N RAST parallel mode F2 PARALLEL TELEMETRY F2 BLOCK F2PARALL CAL TELEMETRY specific variables On Target Flag F2 0TFLAG CUBE HK QLA FLAG F2 BLOCK F20TFLAG sequence number F2 AOT CNF STN F2 BLOCK F2AOTCNF 223 SCD names column sometimes contain
336. ppear and present you with a list of groups from the supergroup you chose Again choose the group you want and a widget containing a list of routines will appear Each cia_help widget contains a set of buttons These provide the following functions e Done button of any of the widgets will close that widget and its dependents e Find button will invoke the find function which returns a list of routines from the current supergroup or group widget that contain the supplied strings For searches for multiple items e g for all routines which all supplied strings just enter these separated by blanks e Quit button quits cia help e Help button invokes a description of the current widget HX z AAPSAD Quit Close Current Topic CIA ADDY en ee ensuite irc t A gau J LIN CIA ASTRON CIA HELF CIA MISC CIAPROCESS CIA USER ao CIA GRAPHICS CIA CIA ARRAY EXTRA ASTR STRLIC INIT Figure 2 1 typical ciainfo display for UNIX SAD_FIND SAD_FIRST AD GET AD IMAGE AD INFO AD INIT AD LIST AD MOVE NEXT PREVIOUS 5A D QUERY READ AD WRITE SCALE UNDEF CD amp SELECT SCD_APFEND CD CHECK SCD_CONCATENE 5cn 320 DEL DLUPLICATE cn FiMD SED FIRST 320 GET CD INIT CD LIST cCD MOVE CD MEXT 320 PREVIOUS cCDp PUT 5cCD QUERY CHAPTER 2 ABOUT CIA Current
337. processing is given in the right most column 11 1 2 CIA PROCESSING OVERVIEW 75 Data calibration refers to the process of dark correction deglitching stabilisation and flat helding of your ISOCAM images Though this is a highly complex process CIA handles it transparently via a set of high level routines The basic process of calibration is a After slicing your data you make a PDS from the set of SPD SCDs you have created in the slicing process This is done with one of the CIA routines get_sscdraster get_sscdstruct or get sscdcvf depending on your observation type b You perform the calibration on the PDS with the core calibration routines see Sec tion 13 2 1 Note that it is possible and sometimes desirable to directly perform calibration on the SPD SCDs All the core calibration routines are capable of accept ing an SSCD as input Chapter 13 describes in more detail the process of calibration When your data are cali brated you are ready to move on to the final step of image display Image analysis amp display refers to the presentation of your calibrated ISOCAM images You can also display the images in the AA calibrated images in the AAR data product directly this you may have already done in Chapter 10 CIA provides routines to do the following e Display images from the AAR data products with sad_display See Section 10 2 e Examine the temporal behaviour of pixels with xcube and x3d See Se
338. ps 154 CHAPTER 14 IMAGE ANALYSIS AND DISPLAY CIA ps open portrait color encapsulated Any plotting routine that does not specify a window can be used to direct output to idl ps CIA contour raster pds raster The following call to ps print will rename idl ps to raster ps and print it to a printer named e15 deleting the old file in the process CIA ps print raster delete printer e15 14 8 1 Avoiding postscript problems In this section we will address some problems which are commonly experienced when printing to IDL the postscript device e Because postscript device pixels are scalable any output graphics will expand to fill the entire postscript device area This means that a square image on your screen could end up as a rectangle in your postscript file e Modifying the colour table by using XLOADCT for example can mess up the IDL s internal colour table This can give unexpected results when you output colour graphics to the postscript device The above problems can be avoided by following the procedure in the example below 1 Create a new window by calling WINDOW with keyword retain set to 2 CIA window 1 retain 2 2 Display your image CIA tviso image Modify the colour table with XLOADCT as you desire 3 Save the displayed image using TVRD CIA image get tvrd 1 4 Now open the postscript device CIA set plot PS Determine the graphic size CIA ss size i
339. pt file Useful for multi wavelength comparison Requires that images are accompanied by astrometry It will take FITS files and raster PDSs and BS PDSs as input Supposing that you have a DSS FITS image and a CAM raster of the same field of view then to overlay the optical and IR data 316 APPENDIX B CIA COMMAND SHORT LIST CIA isocont raster pds dss fits xcube Click on a raster MOSAIC pixel and get its time history CIA xv raster raster pds ximage Display cube of images to verify deglitching and transient correction CIA xcube pds show frame More cube display With this routine many frames are displayed simultaneously Display all the EXPOSUREs in image CIA show frame pds image cvf display Interactively plot spectra from a CVF observation CIA cvf display cvf_pds sad display Routine to display the final pipeline auto analysis products CIA sadN display B 4 FITS input output routines This section summarizes some of those routines that appear in Chapter 18 raster2fits writes the raster PDS fields RASTER RMSRASTER and NPIXRASTER to a FITS primary array CIA raster2fits raster pds name raster fits Works also for a BS PDS CIA raster2fits bs pds name bs fits imagette2fits writes each frame or EXPOSURE in the PDS field IMAGE along with the corresponding frame of RMS and NPIX to the primary array of an individual FITS file CIA imagette2fits pds name cvf fits struct2fits writ
340. r If you don t want to print the relevant pages of this manual you can print the last pages of the original document An Introduction to SLICE inside CIA 21 2 brief description What is SLICE It stands for Simple and Light ISoCAM Environment and was developed by M A Miville Desch nes mamd ias u psud fr from the original ground base calibration data reduction package ICE Because it is Light it is not as complete as CIA but is rather restricted to the raster data reduction It contains two specific tools that are not present in CIA and that can allow significant improvements on the observation quality when dealing with rasters long term transient correction or LTT and variable flat field correction Rather than duplicating these tools in CIA it was chosen to provide access to SLICE in CIA As integration of two software packages into one can prove delicate the version of SLICE you can access in CIA is frozen with respect to the version developed by M A Miville Desch nes Through him you can probably obtain a more recent version of SLICE but we cannot guarantee its compatibility with CIA For the examples in sec 21 3 and 21 4 we assume that you are reducing a raster observation and that the data are currently stored in a PDS called data Sec 21 5 presents a worked example in more depth Sec 21 6 describes the principles of further data quality enhancement tools that remove bad pixels and ghosts while protecting the sources Finally
341. r micro scan or staring observations e M Raster Position the position along the first axis of the raster e N Raster Position the position along the second axis of the raster e Sequence Number order number of the SCD in the observation sequence e Parallel Mode CAM is parallel or prime instrument 90 CHAPTER 12 DATA SLICING We suggest you use all the variables at least once during your first slicing it will greatly help you to understand how ISOCAM operates After some time you will discover that some variables are not useful to you and that you can leave them aside if you remember how you programmed your observations For example if you are working on a raster you won t need the Beam Switching Flag As a matter of fact there is only one detector offset per detector gain therefore the offset is useless Some other variables may look useless to you but may be indeed of some interest especially if you are working on data from the PV phase For this particular first slicing don t use the enhanced OTF in case of telemetry drops More will be told about this in Section 19 12 3 4 Run the slicer The next step is to click on the SLICE button The Slicer will run on the data you have selected according to the variables you have selected Let s see what this means The slicer will read your file the CIER FIT you have selected and flag each record where the value of one of the selected variables has changed It will pro
342. raster proj CIA gt lw6_raster_mm lw6_raster raster This process of calibration and re calibration may be repeated ad nauseum Finally to save your work use IDL s SAVE with the xdr option for portability CIA save filename lw6_raster lw6_raster lw6_raster_mm lw6_raster_spat xdr 13 4 2 calib_cvf Calibration of the CVF PDS is similar to that of the raster PDS in Section 13 4 1 the only difference being that you don t have a MOSAIC to create from the EXPOSUREs in a CVF observation Our calibration example here follows the slicing example of Section 13 1 5 1 It is good idea to begin by making a copy of the uncalibrated CVF PDS CIA original cvf pds cvf pds The CVF PDS is passed to calib cvf for calibration CIA calib cvf cvf pds stab s90 deglitch spat flat calg CIA dark calg Our call to calib cvf specified using the spatial deglitching technique the 890 method of stabilisation and the CAL G DARK and FLAT As for a raster PDS see Section 13 4 1 you can inspect the quality of the calibration of your CVF PDS with x3d and xsnr However there is no MOSAIC image to be displayed You can reassign cvf pds with the uncalibrated CVF PDS and run calib_cvf on it again CIA cvf pds original cvf pds CIA calib cvf cvf pds deglitch mm stab s90 flat And you can finally save your work with IDL s save CIA save filename
343. re the example above becomes CIA red param set red param tdt tdt flat smooth window 6 nplanes 30 CIA flat thresh 10 error map CIA act set act make map CIA slice_pipe 21 5 A worked example This section is intended to lead you through a classical use of SLICE on a raster observation We once again recommend that you read M A Miville Desch nes paper before using SLICE Remember to add the keyword at all your flat fielding calls otherwise set_red_param will reset it at 0 278 CHAPTER 21 USING SLICE WITHIN CIA Table 21 2 Observing setup for the NGC 2366 data Filter N M AN AM Tin PFOV S xt LW3 28 24 5 5 75 0 75 0 2 1 LW2 14 12 5 5 75 0 75 0 5 04 35 0008 Sp 0009 MZ a nooor Figure 21 1 The raster maps using a standard CIA procedure see text for details Left panel shows the LW3 data while the right panel shows the LW2 data Both data sets are affected by periodic patterns due to bad flat field determination as well as long term transients 21 5 1 The data set The dataset we are using here is a 5x5 Y axis raster on the irregular galaxy NGC 2366 It is performed in LW3 followed by LW2 Table 21 2 provides more details on the observing setup These are raster with non integer step sizes 12 5 pixels in all direction This is not a problem for SLICE which does exact projection back and forth from sky to detector T he important point is that th
344. re ASTR STRUC 8 tags length 104 CD DOUBLE Array 2 2 CDELT DOUBLE Array 2 CRPIX FLOAT Array 2 CRVAL DOUBLE Array 2 CTYPE STRING Array 2 LONGPOLE FLOAT 180 000 PROJP1 FLOAT 1 00000 PROJP2 FLOAT 2 00000 The data in the ASTR substructure follows the FITS convention the field CRVAL contains RA and DEC coords and the field CD contains the CD matrix 186 CHAPTER 15 CIA DATA STRUCTURE HIGH LEVEL ARCHITECTURE Chapter 16 Data structure manipulation In Chapter 15 we looked at the high level architecture of the CIA data structures and how to create them As a user you don t have to worry too much about the low level architecture of the data structures Manipulation such as extraction of data is done via CIA routines In this chapter the use of these routines are described and examples are given 16 1 CIA data structure interface routines Below is a list of user interface routines that are used to manipulate CIA data structure In usage replace the stem structure with the name of the data structure type The following sections provide examples and descriptions of their use A formal and more comprehensive description of each routine and additional routines of this type not documented here can be found in the on line help see Section 2 3 2 e structure init initialises a structure in memory and returns its name e structure extract extracts the data from a structure and places it in a regular IDL struc ture e s
345. ready in memory also referred to as ISODATA in x cia histo is a string array containing the history of the PDS 512 lines max indark is a 32x32 array containing the optional user input dark frame inflat is a 32x32 array containing the optional user input flat field logfile is the log file name of the x cia session default name is cia_log txt noinit prevents initialisation of the log file new information is appended to it Since all keywords are optional the calling sequence can be reduced to x cia only If DATA and HISTO are used then their modified values are returned at the end of the x cia session see below Example 5 x cia screen is displayed in Figure 13 1 It is important to read all error warning and info messages displayed in the text window at the bottom of the x cia screen 13 3 3 2 Example 1 testing a calibration method other than the quick look anal ysis The default calibration method consists of using the dark model the multi resolution median MM deglitching method the so called 90 Of final flux s90 method to detect non stabilised values and a CAL G or library flat field Note that it may be better to use the auto flat field if you have a raster with sufficient background and not very extended sources This global calibration is well adapted to get a rough idea of what ISOCAM has detected So let s assume now that the user would like to test another calibration method other than the default one such as
346. reated a multitude of data structures in CIA s mem ory There is a simple bi level hierarchy to these structures At the bottom level there is one data structure for all the data from a CAM STATE these data structures are called SCDs In the case of a raster observation a STATE is time spent in each pointing of CAM So in our example there are at least 32 SCDs 4 CONFIGURATIONS each of a 2x4 raster adding up to 32 In order to keep track of all these SCDs there is a top level data structure this is called an SSCD It holds relatively few data as its primary function is to catalogue its component SCDs The variable sscd returned by spdtoscd contains the unique SSCD name CIA gt print sscd 55 026005060101_02022714575801 This name is used to address the data CIA Note that the SSCD and SCD not regular IDL structures but are implemented in CIA using handles Remember from Section 3 1 that there is more than one CONFIGURATION in our exam ple data set This can be illustrated with the routine info This routine operates on the SSCD and lists the characteristics of its component SCDs In doing so it is also listing the characteristics of all the STATEs in the AOT CIA sscd info sscd deg 48 SCDs in the SSCD C88C026005060101 02022714575801 seq channel mode fltrwhl pfov tint gain offset size ra dec 0 LW IDLE LW2 6 0 25 20 1 512 1 180 498 18 849 1 LW IDLE LW2 6 0 2 10 2 512 1 180 498 18 849 2 LW OBS LW2
347. rection for staring beam switch and CVF observations 265 20 16Faint source data reduction with PRETI css 266 20 17Error handlingin 266 20 18How to save spoiled observations 269 21 Using SLICE within CIA 273 Preface 10e uerb fel aay dI ed VW Ee es spe Wes ivre 273 21 2 A brief description ue bebe A OG 4 PE Oe y xe Lae 273 21 3 Organization of data in 6124 274 21 4 Processing in SLICE 276 21 4 1 The SLICE syntax 276 21 4 2 Error Computations 277 21 5 A worked example tee pa A ke RUPES 277 21 5 t sets toe tale pao ee on Poh ROS MUR ER Tp 278 21 5 2 Choice of flat field methods parameters 279 21 5 3 Long term transient determination 283 21 5 4 Second flat field determination 285 21 6 Bad pixels ghosts and sources 286 21 6 1 Removing bad pixels 22A 287 21 7 Frequently Asked Questions and 288 22 Second order corrections 293 22 1 Jitter correction 2 0 0 293 22 1 1 Computing the jitter 22e 293 22 1 2 Applying jitter offsets 22e 295 29 2 view distortion s
348. rformed on EXPOSUREs as is the default then the flat field correction can be reversed by re reducing CUBE to IMAGE so as to create fresh EXPOSUREs If the flat fielding was performed on IMAGEs by setting the corr_flat keyword cube then the correction is irreversible 5 Raster MOSAIC creation Example raster_scan raster_pds method noproj Caveat Reversible You can recreate the raster MOSAIC ad nauseum 6 Beam switch MOSAIC creation Example reduce_bs bs_pds Caveat Reversible You can recreate the beam switch MOSAIC ad nauseum After any of the above steps the results can be evaluated before proceeding with the next step Routines to aid you are tviso to do general CAM image displaying see Section 14 5 1 x3d to look at the characteristics of the cubes CUBE and IMAGE see Section 14 4 5 xsnr to do S N analysis of an IMAGE EXPOSURE or MOSAIC see Section 14 1 1 20 9 DEALING WITH DEAD PIXELS 251 20 8 1 PDS history In general operations performed on a PDS are recorded in HISTORY It lists routines and calibration methods applied to the PDS CIA print staring pds history date 26 May 1998 17 21 49 node bikini user mdelaney procedure darklibrary V 1 0 algorithm Find best CCGLWDARK 97031713382678 END date 26 May 1998 17 19 22 node bikini user mdelaney procedure spdtoscd V 2 0 algorithm default 7 cisp03001209 fits undefined 7 cia vers test undefined undefined undefined unde
349. rint convert gain fix dark data 0 gain 2 However displaying the actual DARK image is easy CIA tviso dark data 0 image 0 For an example on how to extract PSF images from a CDS see Section 16 5 15 3 3 cds display If you simply want to browse through the contents of a CDS then cds display is the tool for you It may be called by simply typing cds display without arguments on the CIA command line cds display will then prompt you for the name of a CDS stored on disk Alternatively if you have a CDS in memory it may be supplied as an argument For example to view the DARK CDS created in the previous section 176 CHAPTER 15 CIA DATA STRUCTURE HIGH LEVEL ARCHITECTURE CIA cds display CCGLWDARK 96052312123000 All the parameters from the CAL G FITS file are graphically displayed see Figure 15 1 Clicking on the appropriate button will display plots of voltage and temperature statistics Clicking on IMAGE M Image will display a window containing the data image alongside its error image and allow you to scroll through all the images in the CDS You can also load an additional CDS from disk by clicking on CDS 15 4 Auxiliary calibration data Along with CDSs there are additional calibration data contained in the CIA distribution These are described in the following subsections 15 4 1 Theoretical PSFs CIA is distributed with a limited number of theoretical PSFs for use by xphot A full set of 210 theor
350. rk current exposure detector flat image optical flat image dead pixel map and PSF libraries A listing of the files is provided below and full details can be found in the ISOCAM Handbook Chapter 3 and Section 2 3 5 The files are listed here by their official name again datalist txt will help you associate these names with the files on the CD ROM see Section 9 1 e Basic calibration files File Description CSCGCROSS CAM SW noise cross talk decorrelation matrices CHCGCONV house keeping interpolation values CCG DEAD dead pixel map CCG DARK CAM dark current exposure CCG DFLT CAM detector flat field library CCG OFLT optical flat field library CCG SPEC CAM filter amp CVF spectral data CSWCVF CAM SW CVF description CLWCVF1 CAM LW CVF1 description CLWCVF2 CAM LW description CCG SLP CAM CVF spectral line profile CCG PSF CAM point spread function library IFPG focal plane geometry CWHEELS CAM wheels information ORBIT ISO orbital parameters SW or LW e Higher level calibration libraries File Description of contents CCGLWDMOD CAM LW parameters for the time dependent dark corrections CCG TRANS CAM model transients CCG LINEAR linearity correction library CCG FRAME detector astrometric calibration CCG GLITCH CAM glitch model CCG STRAY CAM non dark local light model SW or LW 9 4 Relating Data Product Types to CIA Data Structures In Section 9 3 we ha
351. rm a more sophisticated calibration Most users will only briefly encounter CIA data structures as a means to prepare or slice their data before placing the data in the more user friendly PDS and proceeding with calibration CIA data structures are accessed via IDL handles However CIA does provide routines that make their manipulation transparent so for the most part you never need to dig deeply into the nature of these structures A similar suite of such routines exists for each data structure For example you may use such routines as scd get to extract data from an SCD sad get to extract data from an SAD and so on Part III serves as a reference guide to these data structures and manipulation routines It also intended to be a companion to the CIA Basic Guide and to help you should you have difficulty with understanding references to CIA data structures that appear here We will broadly group the CIA structures into those containing observation data see Sec tion 9 4 2 and those that contain calibration data see Section 9 4 3 For a more detailed look at these structures see Chapter 15 9 4 2 Structures containing Observation Data These structures are used to contain actual images from an observation and data about an observation Science CAM Data SCD ERD SPD Level The SCD structure has two flavours one is used to hold data of ERD product type the ERD SCD and the other of SPD product type the SPD SCD The ERD SCD contains all
352. rojection on a fully calibrated PDS after creation of the raster MOSAIC CIA gt raster_scan raster_pds CIA gt help raster_pds image lt Expression gt FLOAT Array 32 32 8 CIA gt back_project raster_pds images CIA gt help images IMAGES FLOAT Array 32 32 8 You can use stat to check the quality of your calibration The lower the RMS the less the averaged or idealized signal deviates from the real calibrated and corrected signal CIA gt stat images raster_pds image Image dimensions 32 32 Number of frames 8 Total number of pixels 8192 Minimum Maximum Mean Median RMS 10 7506 14 4660 0 144570 0 00626230 0 902133 20 15 ADVANCED PROJECTION 263 E N N A Q T T or 4 or 4 12 rS 13 N ix or 4x orm 40 lo o jo 4 4 4 jo lea dno 4 4 40 4 zh o 4 1 40 4 JO Jo w w orm d 49 o o J Mi 4N gt A Ook 4 orm 4 E 12 324 TE MT b rid eL p ub up 7 1 1 ix EN M E 2 ul ul e uo ul e o T T orm 4 19 19 M 527 N x a le dx Je 8 19 TN TN 15 4 4 4N ap sas Jo 10 j 10 10 19 19 OF 4 0 o
353. rs Assuming an SCD called CSCD143006010001_96082815175532 in memory CIA scd info CSCD143006010001 96082815175532 I K A 2k AC Ek A ak K K FK K SCD CS5CD143006010001_96082815175532 Detector channel LW Entrance wheel HOLE Selection wheel FABRY MIRROR LW Lens 6 00000 Lens wheel position 360 Filter LW2 Filter wheel position 95 Integration time second 2 10007 Integration time camtu 15 Gain 1 Offset 512 Mode IDLE Model Min 25 0000 Model Max 15 0000 Model Mean 3 26709 Model Median 0 000000 192 Model standard deviation CHAPTER 16 DATA STRUCTURE MANIPULATION 4 83574 FKK K K K K K K K KK K K FK K K FK FK FK FK FK FK FK FK K K K ok FK FK ok K K ok ok FK FK oo oko eoe FK FK 2K K K Additional inputs allow for reading of saved structures directly from disk and displaying of selected fields only It is recommended that you refer to the on line help for detailed information on each form of the routine structure info 16 1 10 structure find This function returns a list of structures of type structure which have a field containing a given value Calling the routine is identical for all its flavours the only difference being the fields of the structure Note that arguments are restricted to the numeric or string type ie you can t mix a search for a string va
354. rs Enter the number 30 in the field since you want to select all SCDs with at least 30 frames Hit the Enter or Return key All the wished for SCDs have now been selected e Click on the Redisplay button If you are not yet satisfied with the displayed SCDs and want to unselect more you can use other items from the Advanced slicing menu you just have to remember to unselect all SCDs before because this menu makes the selection You can even unselect the last SCDs by hand using directly the SCDs toggle buttons but do not try to MERGE SCDs at this point Once everything looks good hit the Continue button 12 3 7 Choosing names Once you have clicked on the Continue button a new window will appear asking you to give the name of the SSCD you want You can choose to enter a name for example _obs_lw2 or to use the official name in our example it is 7180014050000 because the data comes from orbit 180 TDT sequence number 014 OSN 05 and CIA s sequence number 0000 In any case an SCD name is made of 31 characters for example an official name would look like CSCD180014050002_96051517052200 and an unofficial user defined name would look like CSCD_obs_lw2__02_96051517052200 The first four characters are always CSCD or cscd for an SCD CSSC or for an SSCD The next 10 may be user defined 12 next for an SSCD or may follow the official convention of the sequence
355. s With one cube input there will be no choice with one raster structure you will have cube mask image rms npix and with two rasters the choice is doubled cubel mask1 imagel rms1 npixl cube2 mask2 image2 rms2 npix2 The last button Scale is the palette of the intensity of the image it can be linear logarithmic or histogram equalized A vertical slider along the image enables to choose the index of the displayed image 14 4 4 1 Advanced Features Two arguments The input can be 2 arguments In this case the image displayed comes from the first argument In the Plot Window the 2 plots are drawn This is useful to compare the same cube before and after processing for example transient correction or deglitching Deglitching This tools enables a manual deglitching First call xcube with the block option Select the button Masking on Then you define a region with the mouse left button a region is one pixel or a rectangle Click on the middle button to label the region bad the right button to label the region good The manual deglitching modifies the mask but not the cube For pds structures the field image can be deglitched as well In this case npix is temporarily set to its opposite value After exiting xcube the masked npix are set to zero 132 CHAPTER 14 IMAGE ANALYSIS AND DISPLAY Mean The selected region in the Frame Window can be a rectangle In this case it is the sum or the mean of the intensity of the pixel
356. s Axis e File You have three choices Save Graph you can save the plot currently displayed as a PostScript encapsulated PostScript IDL saveset or FITS file with or without the color bar Save Frame you can save the frame currently displayed as a PostScript PNG JPG TIFF or FITS file with or without the color bar Quit exit xcube e Window You have tow choices Graph create another window filled by the current plot This is not updated a nd is useful for comparison Pior Fame create another window filled by the current image This is not updated a nd is useful for comparison Color calls the idl routine xloadct to select the color table Zoom This menu selects the Z range of the plot in the next window This menu has a meaning only if Z is time ie the button Temporal Cut of the Frame Window is selected You have four choices Clicking Limits Z range defined by the user click on the first frame drag and release on the last frame see advanced features Block one scd same pointing same wheels same integration time Configuration one configuration All Range everything e Axis This menu selects the x and y scale of the plot in the next window You have four choices Linear Linear Linear Log Log Linear Log Log e Help display an help text The two last menus apply to the Plot Window 3use of cmps_form pro by Craig B M
357. s main window 20 10 Making custom FLATs with flat_builder flat_builder allows you to build a FLAT from EXPOSURES in a raster observation EXPO SURES can be interactively selected and a FLAT created from a median or mean The FLAT can be tested and tailored to produce optimum results An additional feature is that regions of a selected EXPOSURE can be excluded from the FLAT computation so the danger of including source signal in the FLAT can be avoided 20 10 1 Building a simple flat Suppose we have a raster PDS flat_builder accepts the cube of EXPOSUREs in the PDS field IMAGE as an argument Optionally the CAL G FLAT already in CALG FLAT can also be supplied This will allow you to compare your custom FLAT with the CAL G FLAT during a flat builder session flat builder can be called directly or via corr_flat Section 20 2 5 If the later is chosen then upon exiting of flat builder the custom FLAT will be automatically applied to the data If the former is chosen then the custom FLAT must be used as input to corr flat To call flat builder directly CIA built flat flat builder raster pds image calg raster pds calg flat Looking at Figure 20 4 you can see the main flat builder window An additional window not in the figure displays all the EXPOSUREs for reference we will call this the reference window The following steps should guide you through using flat builder 20 10 MAKING CUSTOM FLATS WITH FLAT BUILDER 253 1 Y
358. s of the SCDs are in fact a history of the observation Reading down through the list below you can see the behaviour of CAM through out the AOT In particular look out for which filter the field FLTRWHL and which OP MODE the field MODE is given for each SCD Remember in this observation the filter defines the difference between the two CONFIGURATIONS 12 2 AUTOMATIC DATA SLICING CIA sscd info spd sscd deg LW LW LW LW LW LW LW LW LW LW LW LW LW LW LW LW LW LW LW LW LW LW LW LW LW LW LW LW LW LW LW LW LW LW LW LW LW LW LW LW LW Pe gt ON OOF WN gt Q0 INO IO IO lO FIO FIO l2 RP RB O ON RUN DA ZW O IN 41 SCDs in the SSCD C88C143006010101 98060117121857 Seq channel mode fltrwhl pfov tint gain offset size olelelelelelek IDLE IDLE OBS OBS OBS OBS OBS OBS OBS OBS OBS OBS OBS OBS OBS OBS OBS OBS IDLE IDLE IDLE OBS OBS OBS OBS OBS OBS OBS OBS OBS OBS OBS OBS OBS OBS OBS OBS IDLE FLAT FLAT IDLE LW2 LW2 LW3 LW3 LW3 LW3 LW3 LW3 LW3 LW3 LW3 LW3 LW3 LW3 LW3 LW3 LW3 LW3 LW3 LW3 LW6 LW6 LW6 LW6 LW6 LW6 LW6 LW6 LW6 LW6 LW6 LW6 LW6 LW6 LW6 LW6 LW6 LW DARK LW6 LW6 6 FEF 0000000000000000000000000000000000000 OOo Oo oo ooo
359. s of this region which is plotted versus time in the Plot Window This can be useful when you have an observation with a pixel field of view of 3 arcseconds and some jitter you sum over a region containing the source You can also select a range of indexes the displayed image will be the average over this range 14 4 4 2 Example CIA old raster raster CIA deglitch raster method mm CIA bf xcube raster old raster gt gt click in the Plot Window to select one image gt gt click in the Frame Window on a masked pixel gt gt if needed type new values in the Min Max text widgets of the Plot Window gt gt click on the pixel button gt gt click on the button File then select Quit 14 4 5 Cube analysis with x3d x3d is a widget based program specifically for analysing cubes of images x3d is started by calling it on the CIA command line with an IDL cube or a PDS as input If the input is a PDS then the data in CUBE and MASK are loaded Optionally the CUBE and MASK can be explicitly specified as the first two arguments An example call is CIA gt x3d cvf_pds The x3d window Figure 14 9 displays the current image from the cube being investigated The current image is changed by either clicking on the buttons next frame and previous frame or by moving the slider bar to scroll through the cube A plot window displays either the temporal history of the current pixel a horizontal cut of the row of the curr
360. s pmmmmmmm nnnxxxyy and products pmmmmmmm others where nnnxxxyy refers to the combined TDT and OSN number see Table C 1 for a summary of variables used to identify directories containing data products These lower directories respectively contain data pertaining to a revolution observation data and calibration data C 2 OVERVIEW OF THE CD ROM CONTENTS 321 C 2 3 Where to find nice images of ISO Images of the ISO launch and instruments in JPEG GIF and MPEG can be found in the directory aboutiso images A standard image display package should exist on your system for viewing these images 322 APPENDIX C THE ISO CD ROM Variable Description nnn revolution number XXX TDT sequence number within a revolution yy OSN number nnnxxx TDT number nnnxxxyy combined lt TDT OSN gt pmmmmmmm product set number kkkk data product version number Table C 1 Common variables used in CIA and the CIA User s Manual when identifying data products For example the location of the data products on an ISO CD ROM is prod ucts pmmmmmmm nnnxxxyy Appendix D Guidelines for writing CIA routines D 1 Introduction Users are encouraged to supply their own routines for inclusion into CIA In such cases it is asked that the following be understood e Routines should not duplicate existing code or data structures If in doubt please ask before coding Additional data structures will not be permitted e Supplying a routine to CIA carries with it
361. s the outline of the process e We experiment with the flat field methods to define the best choice of parameters per method e With that choice of parameters method we estimate the long term transient component and subtract it from the cube e In that new cube we make a more refined correction of the flat field In general this is enough to get a significant improvement of the data quality However if needed one can loop on the long term transient variable flat field correction In fact as you will see later see sec 21 5 3 the first step of this process is really a way to determine which flat field method to use in the second step For the rest of this section I will assume that the LW3 and LW2 data are stored in CIA raster structures PDS called 1w3 and 1w2 Remember that to work with SLICE it needs to be initialized once with the command CIA gt cia_slice_init and that to transfer a raster structure to SLICE you have to type CIA raster2slice lw3 and respectively 1w2 for the LW2 data structure SLICE cannot handle more than one structure so you will need to perform all the processing raster per raster although for simplicity s sake we are describing them in parallel in the following sections Once all the processing is done transfer your data back from SLICE to CIA with CIA new_lw3 raster2slice if you were working on the LW3 data modify accordingly for the LW2 data Important warning It may happen that
362. s two names This is because you will find an explicit value of the variable in the SCD top level and an integer value in one of its substructures usually the CAL one Note that only the OTF is allowed to vary in an SCD because it is stored in the HK house keeping substructure All other variables have to be constant in an SCD Moreover OTF is only a part of the QLA flag Table 19 1 Slicing variables used in x slicer 224 CHAPTER 19 ADVANCED SLICING 19 6 5 1 Wheels position and Detector related variables The meanings of these variables are straightforward The detector offset variable is no longer useful because there is only one offset per gain of the detector The important point is the difference between channel and selection wheel position 19 6 5 2 Channel versus selection wheel One of the common mistakes is to think that choosing one of these two variables is enough to determine if the images are LW ones or SW ones Think twice it can happen due to wheel motion or dark measurement for examples that the light beam goes through the SW channel while the LW detector is on You will then get LW images but the slicer and you will believe that they are SW ones It may result that the SCD will not be created if you mix the two kinds of images 19 6 5 3 Electronic setup These two variables are rather important e On Board Processing This variable describes the kind of treatment that is applied to your observations by I
363. s write permission Run only one CIA session per directory Some CIA routines create temporary files in the current directory Running more than one CIA session in the same directory can cause a file I O conflict e xv_raster might crash for some North oriented rasters It is recommended that you use ximage instead e ximage will crash if you use it together with ciainfo or widget_olh Using mosaics fit_isopsf and photom_psf will crash for point sources close to the edge of the mosaic e corr flat will fail to flat correct a PDS that contains data with an optical configuration that is not constant e There has been reports from users that MIDAS has problems reading CIA generated FITS files It appears that MIDAS does not implement the full FITS standard and therefore can not read CIA astrometry There is a known problem with the CIA routine sscd del Deleting an SSCD may corrupt those SSCDs which have been derived or extracted from the original For example CIA gt spdtoscd cisp02600506 fits sscd dir cia_vers test nowrite CIA gt cleaned_sscd sscd_clean sscd CIA gt sscd_del sscd 2 4 CIA CAVEATS 15 Any manipulation of the derived SSCDs as named in cleaned sscd will now fail CIA sscd write cleaned sscd 0 ARRAY HANDLE LONG 72 3 Feb 2000 14 14 19 00 IA make array V 1 0 First argument array handle is not an handle E gt etc CIA raster pds get sscdraster cleaned sscd 0 ARRAY
364. scd_info above tells us that this is the third SCD in our SSCD CIA gt scds sscd_elem sscd CIA gt data scd_get model scds 2 CIA gt stab_img reduce_cube data median You can ignore any output from reduce_cube Now supply this image to stabilize CIA gt stabilize sscd method fs stab_img stab_img Clean the SSCD and split into new SSCDs for each CONFIGURATION CIA gt cleaned_sscd sscd_clean sscd Create a raster PDS from the first CONFIGURATION Reduce flat field the data and build the MOSAIC image in the usual way CIA raster_pds get sscdraster cleaned sscd 0 CIA reduce raster pds CIA corr flat raster pds CIA raster scan raster pds This step may be repeated for the remaining CONFIGURATIONS or SSCDs 20 4 RASTER MOSAIC CREATION 243 Figure 20 2 The raster MOSAIC with SSCD calibration and Fouks Schubert transient correc tion Display for comparison with Figure 3 3 6 It might be interesting to compare the results of this calibration and those in Chapter 3 see Figure 20 2 and Figure 3 3 respectively Clearly in Figure 20 2 the background is cleaner and the ghost image to the left of the source is not present CIA tviso raster pds raster 20 4 Raster MOSAIC creation Final step in calibrating a raster PDS is the creation of the raster MOSAIC from the EXPO SUREs in IMAGE Four methods can be used to compute the MOSAIC all of which are handled by raster scan
365. se in the first pass It is also important that you check that your choice of flat field method is compatible with the long term transient determination see sec 21 5 3 21 5 3 Long term transient determination Now that we have seen how to derive the flat field in our data we are going to try and determine the long term transient which is assumed to be a global i e pixel independent additive time drift of the signal It is important to understand that the long term transient determination is made by com paring the data cube to an estimation of the sky Thus a flat fielding method is necessary and among the parameters to set for this step we will find again those which have been encountered in the previous section It is extremely important that a good flat fielding method is chosen otherwise you will see very strong artifacts appearing in the long term transient curves These artifacts are quite characteristic they form an oscillating signal with a number of peaks equal to half the number of raster legs An example of such artifacts is shown in figure 21 4 If that is your case try and tune the flat field parameters better or select another flat field method this is why we do that exercise in sec 21 5 2 284 CHAPTER 21 USING SLICE WITHIN CIA 35 0000 10 0000 25 0000 Figure 21 3 The resulting maps for the first flat field determination with the DivSky Note that the emission gradient produced by the long term tra
366. serving mode OP MODE when CAM is obtaining observation data Also abbreviated as OBS OBS see observing mode official name Refers to the full name of a data product or CIA data structure 309 OFLT Optical FLaT field OLP Off Line Processing also known as the pipeline OP MODE Operational Mode CAM can be operating in one of several modes IDLE CAM is idle OBS performing an astronomical observation DARK obtaining a dark frame FLAT obtaining an internal flat field image CLEAN executed to remove saturation remnants from the detector WAIT CAM is waiting for a good CONFIGURATION An OP MODE may be comprised of one or more STATEs See CONFIGURATION and AOT ORBIT A CAL G file containing information on orbital parameters for all ISO revolutions up to and at the very least including the revolution during which your AOT is performed origin SAD A flavour of SAD containing which contains an EXPOSURE one from each of the CCIM and CMAP data products or similarly CIA processed data Also may contain a glitch list a point source list or jitter information OSN Observation Sequence Number Identifies an AOT within a TDT See also lt TDT OSN gt OTF On Target Flag OTF 1 indicates that CAM is looking at the proposed target When OTF 0 then CAM is off target Each IMAGE has its own OTF value When slicing your data you have a choice of slicing by our OTF or just OTF The former means that you take 143 as the good value of t
367. shows the LW2 data Both data sets are affected by periodic patterns due to bad flat field determination as well as long term transients 2 a soi 2s gogo X oo RR RS Od nox dom RR 278 21 2 The resulting maps for the Perturbed Single Flat Field determination Note that the map orientation has changed as SLICE always produces maps with North up and East left Imprints of the individual raster pointings are still visible 282 21 3 The resulting maps for the first flat held determination with the DivSky Note that the emission gradient produced by the long term transient is much smoother now and pointing imprints are mostly gone ooo o a 284 21 4 An example of artifacts obtained with a incorrect long term transient determi nation the signal oscillates and the number of complete oscillations is roughly equal to half the number of raster legs The dashed line represent the fitted cor rection see text for details In fact these artifacts were generated while using the DivSky flat field method in the long term transient determination for the LW3 image The LW2 image shows similar problems 285 21 5 The long term transient corrections derived by SLICE The continuous curves are the exact corrections and the dashed ones the fitted corrections assuming the long term transient effect is a combination of two exponentials On the left the LW3 case and on the right the LW2 case Some oscillation appear on the LW2 exact curve but
368. slicer assumes that it is on the ISO CD ROM and looks for it in user directory O THERS assuming UNIX notations e x slicer looks for the CSTA file in the user directory e x slicer looks for the orbit file in the user directory OTHERS with the name OR BIT FIT If the orbit file is not found x slicer looks for it in the default orbit directory under the name default orbit file Unless you are working at the ISO data center at Saclay the default data directory is set to on the UNIX system and to arc_dat on the VAX system In UNIX the search for the ERD or STore Data or TDF will therefore start from the directory where x slicer is invoked the VAX VMS system the user can choose x slicer s working directory by typing the following command before entering the CIA session DEFINE ARC DAT SAPIO1 DKA200 DELANEY X SLICER DATA 19 6 1 2 Customizing the default data directory By editing the x_slicer code you can customize the default data directory for all the people working on a node To do this follow these steps e Run a CIA session on the machine where your data are e Enter the following commands l Taken from Aussel H 1996 ISOCAM Data Preparation with X slicer v2 1 Sections 3 amp 4 19 6 ADVANCED SLICING WITH X_SLICER 219 CIA COMMON SESSION PARAMS CIA print NODE The IDL name of your machine is printed out e Open the x slicer pro routine with a text editor e Look for the lines CA
369. something like the following is displayed SCD number no frames M RASTER N RASTER 1 3 3 2 1 1 2 0 0 14 2 T 3 2 Select the SCDs i i 1 and i 2 then press the Merge i and i 1 button and the Merge i 1 and i 2 button The error will be corrected 19 6 6 4 Build only SCDs of SPD flavor If your observation is done using the Normal mode you receive two frames per image a Reset frame and a End Of Integration EOI frame If you do not ask for it the x slicer will create SCDs of ERD flavor i e with EOI end Reset frames If you ask for it it makes the slicer call the erdtospd routine that will translate this format into real images it is only a subtraction LW and little more complicated for the SW 19 6 6 5 DSD files DSDs are Diagnostic Specific Data files These files are interesting only for the ISOCAM experts in case of an instrument problem Fortunately there was no ISOCAM problems during the mission so there was no reason to access these files CIA contains some routines ia_all_status ia_temp_status to perform some checks on the DSD files 19 6 6 6 Saving slicer file The use of this button is important if you work on big datasets see below 19 6 6 7 Entering names and directories At the point of really creating your SCDs after hitting the Continue button you will be asked to enter a name and a directory AGAIN DO NOT FORGET TO HIT THE RETURN KEY otherwise x_slicer will not take into ac
370. source detection A well deglitched CUBE will have few or no spiky pixels Glitches can be manually removed by clicking on the button glitch and then clicking on a pixel in the image with the middle mouse button The selected pixel will be masked and the MASK will be automatically updated note that the MASK should not be supplied to x3d as an expression if the update is to be saved See Section 20 2 2 for alternative manual deglitching methods within CIA Note that there are two other manual deglitchers man mask and sl viewcube in the contrib directory of CIA If mask 1 complex then all instances of non zero mask are indicated AII of this depends of course on such factors as the integration time the number of frames per state the response of detector to the source and the intensity of the glitch 14 4 CUBE ANALYSIS 135 EXE Es MS ESL CE EE EE Plot type Multi Q Single Display Quit Figure 14 10 xv_temp window 14 4 7 xv_temp xv temp is yet another tool for examining cubes It does however offer different functionality from x3d and xcube It can be invoked simply as CIA xv_temp any pds cube to any_pds to Take a look at Figure 14 10 Note the following points e The image displayed is the mean of all the IMAGEs in any pds cube A white box sur rounds a feature in this image we will call the pixels enclosed within this box the region e The position of the region can be changed by clic
371. sscd1 16 1 1 2 scd init This function initialises an SCD in memory An example using all the available options and following from the example of Section 16 1 1 1 follows CIA scdi scd init 143006010101 10 ssid sscd1 ERD deid 0 ack ok CIA help scdi ok SCD1 STRING C8CD143006010101 96080512231512 OK INT 1 The first argument is the combined lt TDT OSN CN STATE gt number that is used to name the returned SCD scd1 that is created by scd init see Section 15 2 1 second argument specifies the number of planes in 1 The keyword ssid specifies the name of the SSCD we have initialised in Section 16 1 1 1 and to which scd1 now belongs If the keyword ERD is set scd1 is an SCD of ERD flavour see Section 15 2 2 otherwise it defaults to an SCD of SPD flavour The keyword deid is either 0 LW detector or 1 SW detector and defaults to LW detector 16 1 1 3 ssad init To initialise a new SSAD in memory we can do the following CIA ssadi ssad init 7143006010101 ack ok CIA help ssadi SSAD1 STRING 088A143006010101 96080512201019 OK INT 1 The first argument is the combined lt TDT OSN CN STATE gt number that is used to name the returned SSAD ssad1 that is created by ssad_init see Section 15 2 1 A keyword argument not used here is old ssad which can be used to pass an already existing SSAD to ssad init so as to copy its parameters to ssad1 16 1 CIA DATA STRUCTURE INTERF
372. ster dat verb CIA corr dark 1w6 raster CIA deglitch lw6 raster CIA stabilize 1w6 raster 2 We can look at the results of our processing so far by viewing the the corrected IMAGEs in Iw6 raster cube with x3d CIA x3d lw6 raster Clicking on the button mask indicates the pixels that have been masked as unstable By looking at the temporal history of the pixels you can immediately see if good stabilisation and deglitching have been achieved A vertical profile of some background pixels will tell you if a good dark correction has been applied Now can proceed with flat fielding Firstly all the IMAGES per CAM pointing SCD must be averaged to EXPOSUREs the routine reduce does just that Then the flat fielding is performed on each EXPOSURE with corr flat CIA reduce 1w6 raster CIA corr flat 1w6 raster Now we have completed the final processing step with the core calibration routines 13 2 CALIBRATING THE PDS 101 13 2 2 Raster MOSAIC creation The final step in processing raster observation data is creation of the raster MOSAIC from the corrected EXPOSUREs The routine raster_scan will project all the EXPOSUREs on to the raster field of view and place the resultant raster MOSAIC in the PDS field raster CIA gt raster_scan lw6_raster If you wish you can also convert the MOSAIC pixels to milli janskys mJy CIA gt conv_flux bs_pds raster By looking at the MOSAIC you can really get an idea as
373. t OLP has been scientifically validated these problems should no longer exist A brief history of OLP e Up to version 2 41 OLP did not specify the instrument for which the IIPH was computed e Version 2 43 and later specified the instrument for which the IIPH was computed e Version 4 1 and later filtered the RA DEC and ROLL values e Version 6 1 and later provides improved values of RA DEC and ROLL namely CRA CDEC and CROLL Moreover it is in the IIPH that x slicer reads the crucial information about how the obser vation went was it OK or was the Target not acquired etc If you have the misfortune to see on your screen a message like the one of Figure 19 2 two possible reasons are 19 6 ADVANCED SLICING WITH X_SLICER 221 It appears that your observation did not go well X Slicer will continue to proceed through the loading but it is very unlikely that you will get great things from this file Refer to the X Slicer manual to get information on this case your ATTERROR number is 1 ARGHHHH Figure 19 2 x slicer s dreaded message e ATTERROR 1 The target was not acquired Your observation is lost e ATTERROR 2 Due to a break in the telemetry down link system the IIPH file will contain uncorrected data from the star trackers All your RA DEC and Roll will be false but your images should be OK For normal observations both these problems should be spotted during the quality checking process
374. t gain offset size ra dec 0 LW IDLE LW5 1 5 2 10 2 512 13 223 797 53 680 1 LW OBS LW7 1 5 2 10 2 512 8 223 796 53 680 2 LW OBS LW7 1 5 2 10 2 512 101 223 796 53 680 3 LW IDLE LW7 1 5 2 10 2 512 23 223 796 53 680 Compared to other observations there are very few SCDs STATEs This is typical of a staring observation 31 32 CHAPTER 4 STARING OBSERVATION 01 Remove unwanted SCDs with sscd clean CIA cleaned sscd sscd clean sscd Out of 4 SCDs 2 are rejected due to mode is rejected due to csh flag 3 are rejected due to qla flag In total 1 is accepted We are left with only one SSCD and one corresponding SCD Now we must place the contents of the SSCD into a PDS For a staring observation we use a general PDS This is created with get sscdstruct CIA staring pds get sscdstruct cleaned sscd You might be interested to see the data at this stage x3d can be used to do this CIA x3d staring pds Now we can proceed with the calibration We will perform the standard calibration steps on the cube i e staring pds CUBE In this observation the data does not need stabilization correction however feel free to experiment with this CIA corr dark staring pds CIA deglitch staring pds Now we have a nicely calibrated PDS You might want to check this with x3d Use the same calling sequence as above This time you can click on the button mask to see which pixel have been masked by t
375. t the SCD Number and the corresponding buttons The bottom of the x handle slice window is made of a set of buttons on the left and a pull down menu named Advanced Slicing on the right The functions of the buttons are e Redisplay Redisplay the x_handle_slice window showing only the selected SCDs Note that you must redisplay before slicing e Continue When satisfied with the slicing go into the SSCD SCD and eventually DSD creation e Unselect All SCDs self explanatory 12 8 DATA SLICING WITH X SLICER 91 Figure 12 2 x handle slice window title staring SC St La N Ent Whl Sel Whl LW Lns LW Fit SW Lns SW Fit Ch Beam Switching A O T Ob T Off Integ Ti G Obs OTF R M M R P N R P S N P M CHAPTER 12 DATA SLICING description SCD Number starting readout of the SCD last readout of the SCD maximum number of images that can be built entrance wheel position selection wheel position LW lens wheel position LW filter wheel position SW lens wheel position SW filter wheel position channel which detector is on Beam Switching AOT variable observation type detector offset integration time detector gain observation mode on target flag raster mode M raster position N raster position sequence number parallel mode flag displayed always always always always if selected if selected if selected if selected if selected if selected if sele
376. ted routine called deglitch ksig clip PDS side effects Glitched pixels are flagged in MASK 7 method manu method Deglitch manu The IMAGEs in CUBE can be manually deglitched with this method Upon calling instructions for use are displayed in the CIA command window and a graphics window displaying a single IMAGE at a time from CUBE is opened The mouse is used to select suspected glitched pixels and those pixels are replaced with the median of their neighboring pixels Note that there are two other manual deglitchers man mask and sl_viewcube in the contrib directory of CIA In addition x3d Section 14 4 5 now has manual deglitching capability An example call is CIA deglitch pds method manu This method is arduous and is probably best used to get rid of persistent glitches that other deglitching methods have failed to find routine deglitch man PDS side effects Glitched pixels are removed from IMAGEs in CUBE i e CUBE is modified Glitched pixels are also flagged in MASK reference ISOCAM Handbook Chapter Data processing methods Section Manual deglitch 20 2 3 Stabilization There are several methods that may be used to achieve stabilization all of which are implemented by stabilize and individual low level routines As with the other core calibration routines the keyword method selects the required method All the available methods are described below but before reading on it is worth noting th
377. tem gt telnet ssd jpl nasa gov 6775 JPL Horizons vers SUN v3 0 Type for brief intro for more details System news updated Jan 15 2002 Horizons gt tempel tuttle gt EXACT lt name search SPACE sensitive NAME TEMPEL TUTTLE Continue lt cr gt yes n no y JPL DASTCOM3 Small body Index Search Results 2002 Feb 26 08 44 47 Comet amp asteroid index search NAME TEMPEL TUTTLE Matching small bodies 20 7 ANALYSIS OF SOLAR SYSTEM OBJECTS 247 Record Epoch yr Primary Desig gt NAME lt 201399 1998 55P Tempel Tuttle 201400 1998 55P Tempel Tuttle 201401 1998 55P Tempel Tuttle 201402 1998 55P Tempel Tuttle 201403 1998 55P Tempel Tuttle 201404 1998 55P Tempel Tuttle 6 matches To SELECT enter record integer followed by semi colon FOO OOOO IG I I I I I I ak a 2k a 2k a 2k ak Select Fltp 11 Rledisplay cr 201403 ROO RRR RRR ROR RRR RII IRI I A KK aK a A 2k 24 2k JPL HORIZONS 55P Tempel Tuttle 2002 Feb 26 08 45 17 Rec 201400 COV Soln date 2002 Jan 03_16 05 39 obs 392 1865 1998 FK5 J2000 0 helio ecliptic osc elements AU DAYS DEG period Julian yrs EPOCH 2451040 5 1998 Aug 15 0000000 CT Residua
378. ter 12 13 1 Creating a PDS from an SSCD At this stage it is assumed that the data has been prepared sliced correctly and placed in a set of SPD SCDs see Figure 11 1 To progress to the calibration process you need to create a PDS from this set of SPD SCDs The PDS can conveniently hold all the data we need for further processing There are four flavours of PDS to accommodate some differences in the nature of the data from different CAM AOTs and observations the raster PDS the CVF PDS the BS PDS and the general PDS 13 1 1 PDS caveats Before we proceed there are some important points about PDSs which should be understood The PDS must only contain data from STATEs where CAM is in OP MODE OBS Other STATEs such as IDLE will crash CIA processing routines Also the PDS must be perfectly filled for calibration to work telemetry drops missing raster points gain change due to satu ration can all contaminate a PDS A PDS is not suitable for some kinds of data e g data from a polarization observation or calibration methods where contiguous data is required as input For instructions on how to calibrate an SSCD see Section 20 3 In addition Chapter 8 contains an example of calibrating SSCDs containing data from a polarization observation If you have used sscd_clean to slice your SSCD then telemetry drops will have been patched with fake SCDs and the subsequently derived PDS should be fine Alternatively you can continue the data anal
379. ter at the lower left corner The number of points in the M direction of the raster are held in the raster PDS field RASTERCOL which takes its value from the FITS keyword ATTRNPTS The raster pds field M STEPCOL or the FITS keyword ATTRDPTS gives the distance between adjacent columns in arcsecs Likewise the number of points in the N direction of the raster are held in the in the raster PDS field RASTERLINE which takes its value from the FITS keyword ATTRNLNS The raster_pds field N STEPLINE or the FITS keyword ATTRDLNS gives the distance between adjacent lines in arcsecs E 1 DEFINITIONS 329 North N direction of the raster Y M M direction of the raster Z 07 Figure E 2 Schematic of a Y axis raster Crosses x mark successive positions of the raster The start and end points are also indicated The LW detector with its 24th column marked are shown on one position of the raster The trajectory of ISO during the raster is shown by the broken line This is a diagram of a M 7 N 5 raster 330 APPENDIX E ISOCAM ASTROMETRY ANGLES AND COORDINATES By construction The M axis is identical to the Y axis while the N axis is in opposition to the Z axis e a is the roll of the camera as in Figure E 1 In the raster PDS a is found in the fields ANGLE RASTER and INFO ROLL In the the SCD and SSCD it is simply the field ROLL e 3 is the so called SSCD and raster PDS field RASTER ROTATION It is the pos
380. th UTC at beginning of data day UTC at beginning of data hour UTC at beginning of data minute UTC at beginning of data second year of creation month of creation day of creation hour of creation minute of creation second of creation 0 01 second of creation pe 2 AOT number Type integer This field can have the following values value description Cv 4A CS C5 CUS use only CAM OI staring raster micro scan or tracking CAM02 not used historic reasons only 3 beam switch cvf 5 polarization 164 10 11 12 13 14 CHAPTER 15 CIA DATA STRUCTURE HIGH LEVEL ARCHITECTURE TARGET Target name as it appears on original proposal Type string e g HARO 3 OBSERVER Observer s name as it appears on the original proposal Type string e g LMETCALF TDT TDT number Type integer OSN OSN number Type integer F_RASTER Indicates the type of CAMOI Set to UNDEFINED for AOTs other then CAMOI Type string The table below lists the possible values of _ value description RASTER raster observation STARING staring observation MICRO_SCAN micro scan observation TRACKING tracking observation UNDEFINED observation is not CAMOI CHANNEL CAM channel used in the observation Possible values are LW or SW Type string MODE String indicating
381. th raster images of the same object radec is set to orient them in the standard astronomical fashion The keyword shift will make isocont attempt to match the images not only by astrometry but by correlation also This helps to compensate for possible pointing inaccuracies CIA isocont lw6 raster lw3 raster radec shift 14 6 3 x isocont x_isocont is a widget interface to isocont see Section 14 6 2 It has been designed for easy experimentation of overlays rather then specify keywords you can point and click With x_isocont you can easily set contour levels titles produce hardcopies etc x_isocont is started in the same way as isocont but without the keywords e g CIA isocont l1w6 raster haro3 fits Looking at Figure 14 17 you can see that it is split into several panels and drop down menus Click on info help for an online help description of the functions of the drop down menus The functions of the panels are described below with the corresponding isocont keyword given for each Input Images Displays names of files or data structures The data structure name is taken from the field SAD NAME Check buttons can be set to switch on off display of images The button FOV is equivalent to the keyword scan Astrometry Parameters Menus for setting astrometry parameters These parameters are equivalent to keywords epoch ref magnify Contour Levels Set number of contour levels minimum and maximum values pro
382. that dark correction de glitching and transient correction are performed in CIA You will generally start the SLICE processing after these three steps ie where in a more standard reduction session you would have made the flat field correction To start SLICE the first step is to initialize its common blocks as well as place its directories in your IDL path This is simply done with CIA gt cia_slice_init This creates variables that will be used to store your data any variable with the same name is therefore erased Their names and content are listed in Table 21 1 You need now to transfer your data into these variables This is done with the CIA routine raster2slice therefore type CIA raster2slice data The next section will describe the processing that you can do in SLICE For completeness reasons let us see here also how to transfer data back from SLICE into a CIA structure This is also very simple as you just have to type CIA new raster slice2raster This function requires no argument as all the required data are already in the SLICE commons You will see that the content of the output raster structure is different from that of the input one In particular the raster field that contains the final map is generally larger This is due to the SLICE convention of only building maps with North up and East left the astronomical convention This in general results in larger raster maps Note however that this new raster s
383. the EOI and RESET FRAMES from a single STATE and in addition parameters describing that STATE e g coordinates lens filter etc The SPD SCD differs primarily in that it holds IMAGEs which have been computed from the EOI and RESET FRAMEs These IMAGES are either directly taken from the CISP data product or computed by CIA from the FRAMES in the ERD SCD Set of Science CAM Data SSCD The SSCD is primarily designed to catalogue a set of SCDs either ERD SCDs or SPD SCDs but not both together belonging to the same CONFIGURATION and variables which describe that CONFIGURATION However it may be used to catalogue any number of S TATEs from all the STATEs in AOT down to a single STATE For an indirect way of treating a CIA structure as a regular IDL structure see Section 16 1 2 66 CHAPTER 9 THE DATA PRODUCTS AND CIA DATA STRUCTURES Science Analysed Data SAD AAR Level Contains two EXPOSUREs one calibrated in detector coordinates from the CCIM data product and the other calibrated in celestial coordinates from the CMAP data product The SAD is also used to hold EXPOSUREs from the CMOS data product In addition to AAR data products the SAD may be used to hold CIA calibrated EXPOSUREs and MOSAICs Set of Science Analysed Data SSAD Contains a catalogue of a set of SADs Its function is analogous to that of the SSCD In general best use of the SSAD is made when it catalogues SADs that belong to a single CONFIGURATION However
384. the SAD substructure CSSP Click on Update SAD and when you exit cvf display you will find CSSP filled 1See Section 15 5 3 for a description of the CVF PDS data structure and Section 13 4 2 for an introduction to calibrating a CVF observation 118 CHAPTER 14 IMAGE ANALYSIS AND DISPLAY Lem 55 er Figure 14 2 cvf display window CIA cvf_cssp sad_get cssp cvf pds sad name CIA help cvf_cssp str Structure SPECTRUM_STRUC_1 11 tags length 4152 NAME STRING spec000000000000 00 NSPEC LONG 0 RA FLOAT 288 132 DEC FLOAT 67 6615 N LONG 76 WAVELENG FLOAT Array 200 BANDWIDT FLOAT Array 200 STAT FLOAT Array 200 FLUX FLOAT Array 200 DFLUX FLOAT Array 200 MASK BYTE Array 4 32 5 If you have lost the SAD maybe by saving the CVF PDS only and restoring without the SAD in another CIA session then don t worry The keyword flux returns the last spectrum displayed before quitting cvf_display Wavelength information can be found in the CVF PDS fields CVF INCR WAVELENGTH START and WAVELENGTH_END 6 xloadct can be invoked by clicking on Color 7 The button Axis allows you to change either or both axes to a logarithmic scale 14 2 2 xcvf xcvf is a visualization tool for CVF data and a front end for the lower level routine conv cvf2isap xcvf only takes CVF PDSs as input while conv_cvf2isap works with both CVF PDSs and AAR products Since this chapter is concerned mostly
385. the angle between the M and Y axis measured in the direct trigonometrical sense from the M axis to the Y axis Note that in that case and cannot be deduced from one another Given that the rasters are reconstructed with the M axis as horizontal axis and the N axis as the vertical axis prior to project an individual position in that image the data have to be rotated by y Looking at Figure E 3 one has y a f 90 or ROLL RASTER ROTATION 90 E 1 DEFINITIONS 331 Figure E 3 Schematic of a M 4 N 3 raster oriented with reference to the North axis Crosses x mark successive positions of the raster Here the raster position angle given in PGA was approximately 65 see text Note however that the actual RASTER ROTATION is in fact 245 as the location of start and end points shows The camera its 24 column and pixel 0 0 are once again displayed for clarity purposes Angles 3 and y are explained in the text 332 APPENDIX E ISOCAM ASTROMETRY ANGLES AND COORDINATES E 2 Trouble shooting astrometry in CIA structures You ve read the previous section you think you understand ISO s angles and you have been told that the current version of CIA complies with these definitions Thus you confidently reconstruct your raster and it fails If the astrometry is seriously wrong then most likely you are using data that is not scientifically validated data from OLP prior to version 4 0 is not scientifical
386. the previous SCD and users can later manually unmask this invalid data for inclusion in the finally MOSAIC image 17 1 4 SCDs from ERD erdtoscd With the exception that erdtoscd creates ERD SCDs from ERD data products everything in the description of spdtoscd in Section 17 1 3 also applies here An example call might be CIA erdtoscd cier14300601 fits sscd dir product_dir CIA gt scd_dat scd_dir ack ack 17 1 5 CDSs from CAL G files calg2cds To convert CDS data products from files on the CD ROM to CDSs in memory or on disk we use the function calg2cds Using the assigned variables of Section 17 1 1 an example call is Remember that since the CDSs in CIA are more recent then the distributed CAL G files Section 9 4 3 it is unlikely that you will frequently use this routine 17 2 IMPORTING FITS TO REGULAR IDL DATA STRUCTURES 203 CIA cds calg2cds ccglwdark fits dir product dir ack ok CIA help cds ok CDS STRING CCGLWDARK_96080515324098 OK INT 1 calg2cds returns the name of the formed CDS in memory Nothing is written to disk by this function Use the CIA procedure cds_write to save the CDS data structure to disk see Section 16 1 5 17 2 Importing FITS to regular IDL data structures This section describes CIA routines that can be used to read any extended FITS data products As such they can read the ISO data products into a regular IDL structure The IDL ASTROLIB also contains
387. their orientation A description of the FITS convention and how it relates to CAM angles can be found in Section E 3 E 1 Definitions Ambiguous definitions sometimes intentional of angles of importance in ISOCAM observations are spreading confusion in minds and in routines This appendix sums things up to the best of our current knowledge for the LW part of CAM the SW is a priori and simply deducible from the LW case Before the drawings some points to note e We are using only the astronomical convention for projections In this convention when the celestial North is pointing upward the celestial East is pointing leftward In this representation 6 increases upward and a increases leftward e ISOCAM rasters are reconstructed along axes such that the M scanning direction goes from left to right and the N scanning direction from bottom to top Therefore the North axis is generally not pointing upward in these images However a simple rotation is enough to restore the astronomical convention i e no mirroring of the image is needed e The sign after a letter is a convention to specify an oriented axis e The Y and Z axes are defined by the spacecraft and are identical for all instruments though they project differently in their various focal planes The X axis never appears because it is pointing to the target Note that the XYZ referential is a direct one in 3D space but that due to the astronomical convention used the YZ
388. these are not obviously related to the raster scan period 286 LIST OF FIGURES xvii 21 6 21 2 21 3 21 6 22 1 22 2 E 1 E 2 E 3 4 5 The results of the long term transient correction and variable flat field determina tion Variable flat field was performed using the DivSky method with parameter setup as indicated in Table 21 4 LW3 is on the left and LW2 on the right Com pare with Fig 21 1 to measure the improvement 287 The resulting maps for the Perturbed Single Flat Field determination Note that the map orientation has changed as SLICE always produces maps with North up and East left Imprints of the individual raster pointings are still visible 290 The resulting maps for the first flat field determination with the DivSky Note that the emission gradient produced by the long term transient is much smoother now and pointing imprints are mostly 290 The raster maps using a standard CIA procedure see text for details Left panel shows the LW3 data while the right panel shows the LW2 data Both data sets are affected by periodic patterns due to bad flat field determination as well as long term transients cc fag bee hee de ee Gee oe OR et 291 The results of the long term transient correction and variable flat field determina tion Variable flat field was performed using the DivSky method with parameter setup as indicated in Table 21 4 LW3 is
389. through the slow process of regenerating them again ssad write is dedicated to saving an SSAD and all its catalogued SADs CIA ssad write cssa000014300601_96092310115335 dir sads where dir is the destination directory You may repeat the command for the other SSAD 16 3 Saving and restoring PDSs Since a PDS is a regular IDL data structure it can be saved with the IDL commands SAVE and RESTORE It is recommended that you use the keyword xdr so as to avoid compatibility problems when sharing data across platforms If you have completed the calibration of your data and you want to save the results then you can remove some unnecessary data from your PDS prune pds removes the fields MASK 16 4 MANIPULATING THE MASK 195 and CUBE these fields contains the vast bulk of the data in a PDS so removing them greatly decreases the PDS size and saves disk and memory space CIA raster pds prune pds raster pds CIA save file raster_pds dat raster pds verb SAVE Portable XDR SAVE RESTORE file SAVE Saved variable raster pds raster_pds is a raster PDS which is originally more than 2 MB in size prune pds reduces it to about 0 5 MB CIA 1s la raster pds dat rw r r 1 mdelaney ssamr 500568 Oct 14 13 17 raster pds dat 16 4 Manipulating the MASK This section describes CIA MASK manipulation routines For more details on the MASK in CIA data structures and a list of the possible MASK
390. tial coordinates from the CMAP data product The SAD is also used to hold EXPOSUREs from the CMOS data product In addition to AAR data products the SAD may be used to hold CIA calibrated EXPOSUREs and MOSAICs See also Section 15 2 4 Set of Science Analysed Data SSAD Contains a catalogue of a set of SADs Its function is analogous to that of the SSCD In general best use of the SSAD is made when it catalogues SADs that belong to a single CONFIGURATION However it may catalogue any subset of SADs from an AOT See also Section 15 2 5 Diagnostic Specific Data DSD Contains physical parameters of the camera tempera tures voltages wheel positions etc It is directly used only by instrument experts for in depth investigation of ISOCAM behaviour Such data that is of interest to the normal user is also held in the SCD Documentation of the DSD structure is beyond the scope of the CIA User s Manual 15 2 1 Standard fields of observation data structures There is some similarity between observation data structures in their architecture that is to say standard fields exist in all Generally they contain information which is relevant to each data structure e g it is necessary to know the PFOV during an observation whether you have raw FRAMES in a ERD SCD or calibrated IMAGEs in SAD but it is not necessary to keep RESET FRAMEs in the SAD This section describes these standard fields Later sections describe fields particular to each dat
391. ting data product types to filenames You can consider the data products to be organised into five groups Raw Data Standard Processed Data Automatic Analysis Results Calibration Data and Auxiliary Data Further 1 Also a good reference is Harten R H 1988 The FITS tables extension A amp A Suppl Ser 73 365 372 61 62 CHAPTER 9 THE DATA PRODUCTS AND CIA DATA STRUCTURES explanation of these data types and the files where actual data may be found follows This may not be an exhaustive review of the data products but should suffice at this stage 9 3 1 Raw data products The ISOCAM raw data products consist of CAM Compact Status Data and CAM Edited Raw Data Compact Status Data CSTA Data on the status of ISOCAM during the observation It is delivered in the CSTA file ISO CD ROM users can find this file in the directory products pmmmmmmm nnnzrxyy Though used by some routines these data are not essential CAM Edited Raw Data CIER ERD refers to raw CAM data These are all the data from and associated with an observation including internal calibration and housekeeping data These data are delivered to you in the CIER filet ISO CD ROM users can find this file in the directory products pmmmmmmmy nnnaxryy 9 3 2 Standard Processed Data SPD SPD are data which have had some processing Housekeeping data has been removed and IMAGEs are computed from the RESET and End Of Integration EOI FRAMES that are present in CIER Thes
392. tion CAMO5 dedicated CAM99 8 1 Description of the observation The data used here is from a CAM polarization observation of the object HIC085371 This observation is basically a 2 x 2 raster observation though at each raster pointing the entrance wheel cycles through the 3 polarizors 3 times Hence there are many SCDs for this observations 2 x 2 3 3 in addition to several IDLE and other uninteresting SCDs This chapter provides a good example of the technique of calibrating an SSCD 8 2 Data analysis It is assumed in this section that you have read Chapter 3 Generally concepts described in that section will not be re described here In this example it is necessary to calibrate the SSCD and not the PDS This is because the polarization data will no longer be contiguous when it is frozen in a PDS see Section 13 1 1 for details of the limitations of a PDS This will become more obvious later in the chapter when we get to the SSCD cleaning stage For more on calibration of SSCDs see Section 20 3 8 2 1 Overview of calibration steps The steps involved in the calibration of data from this polarization observation may be summa rized as 1 Slice in the usual way using one of the slicing routines e g spdtoscd or x slicer Perform dark deglitching and transient correction on the SSCD using the routines corr dark deglitch and stabilize 2 After the core calibration steps are complete we manually clean and combine SSCDs and SCDs be
393. tion from the wheel description see Section 16 5 for an example To access these files directly follow the procedure below 1 Firstly let s have a look at the hk wheels txt files In unix these files can be found in CIA hk dir vers tables and in VMS in CIA hk dir CIA TABLES Now list the contents of hk dir CIA cd hk dir CIA dir hk wheel txt hk wheel 1 txt hk wheel 3 txt hk wheel 5 txt hk wheel 2 txt hk wheel 4 txt hk wheel 6 txt 178 CHAPTER 15 CIA DATA STRUCTURE HIGH LEVEL ARCHITECTURE Each file corresponds to a different CAM wheel file hk wheel 1 txt hk wheel 2 txt hk wheel 3 txt hk wheel 4 txt hk wheel 5 txt hk wheel 6 txt wheel number description WHEEL 1 entrance wheel WHEEL 2 selection wheel WHEEL 3 LW lens wheel WHEEL 4 LW filter wheel WHEEL 5 SW lens wheel WHEEL 6 SW filter WHEEL 2 Now take a look in hk wheel 1 txt for entrance wheel positions CIA more hk wheel 1 txt C WHEEL 1 Entrance wheel R Gastaud 10 Jan 1995 CREATOR C R Gastaud DAPNIA CEA SACLAY CALIBRAT C Interactive Analysis version 1 0 VERSION C Individual file version number 1 0 DATE C Date of file creation 10 01 95 TIME C Time of file creation 10 00 00 TELESCOP C ISU INSTRUME C CAM END 360 0 UNKNOWN 1 UNKNOWN 2 UNKNOWN 3 UNKNOWN 4 UNKNOWN 5 UNKNOWN 6 UNKNOWN 7 polarizor 2 etc So entrance wheel position 7 corresponds to polarizor 2 15
394. to 0 e Very advanced users can configure CIA logging and debugging verbosity with the system variable CIAERR CIA help ciaerr str Structure 749788 5 tags length 12 refs 2 SCREEN BYTE 2 LOG BYTE 3 XWARN BYTE 0 DEBUG BYTE 0 PRINT STRING PRINT PRINT Defines the routine used by the CIA s cia print to print information Can be used to redirect output of cia print DEBUG Flag indicating debugging level Set to 0 for silent output and 1 for verbose output For example CIA ciaerr debug 1 will cause CIA to output debugging information 12 CHAPTER 2 ABOUT CIA LOG Log file reporting verbosity Ranging from 1 to 3 for increasing verbosity SCREEN Screen reporting verbosity Ranging from 1 to 3 for increasing verbosity XWARN CIA uses IDL s keyword inheritance Setting this flag will alert you if a supplied keyword parameter is not known by the called routine If a warning appears you have either deliberately specified a keyword used by a low level routine or mistyped the keyword in which case it will be ignored For example CIA ciaerr xwarn 1 CIA corr dark pds goodbye hello CORR DARK Undefined keywords GOODBYE HELLO Dark correction with model and no scaling inclusion of contributed routines to CIA s path Only for VMS under Unix these routines are included by default setting the level of the on line help see section 2 3 2 Either the complete help for all routines Programmer
395. to how good the calibration is We can display the MOSAIC most easily with tviso CIA gt tviso lw6_raster raster 13 2 3 Staring analysis Actually there is no real dedicated staring analysis After the creation of the corrected EXPO SUREs as in Section 13 2 1 the analysis of staring observation data is almost complete the final formality is the conversion of the EXPOSURE pixels to mJy CIA gt conv_flux staring_pds image CIA gt print staring_pds image_unit mJy pix To check the quality of your analysis you can use tviso to display individual EXPOSUREs CIA tviso staring pds image 0 Or you can use x3d though only if you have more than one EXPOSURE CIA x3d staring pds image 13 2 4 Beam switch MOSAIC creation The final step in calibrating the BS PDS is the creation of the beam switch MOSAIC i e the difference between the on source field EXPOSUREs and the reference field EXPOSUREs The routine reduce bs will perform this operation CIA reduce bs bs pds For historical reasons the beam switch MOSAIC is stored in the PDS field raster even though beam switching has nothing to do with rasters Now you can convert the MOSAIC pixels to mJy CIA conv flux bs pds raster As for the raster PDS you can check the results of the analysis with tviso CIA tviso bs pds raster 102 CHAPTER 13 DATA CALIBRATION 13 2 5 CVF analysis Unlike the raster PDS or BS PDS there is no MOSAIC to create in a C
396. to this automatically upon login Before any modifications to your CIA set up can be made you must do the following 1 Make a personal copy of the generic CIA startup script user_init pro this should be found in the CIA installation directory In your LOGIN COM file change the logical idl_startup to point to your personal user_init pro Them make sure that your defi nition of the logical idl_startup is used For example include the following lines in your login com CIA_ENV DEFINE IDL STARTUP SYS LOGIN USER_INIT PRO If you want to modify CIA s logging directory then you also have to create your personal CIA VMS ENV COM which has to be executed instead of the general one To specify the directory to which CIA log files are to written in your CIA VMS ENV COM change the logical logfile dir For example DEFINE LOGFILE DIR SAPIO1 DKA200 DELANEY CIA LOGFILES However for most users switching off CIA s logging is the preferred option You can do so by placing the following command in your CIA VMS ENV COM file DEFINE NOLOG LOGFILE DIR NL To use your CIA VMS ENV COM instead of the general one include the following lines in your login com SYS LOGIN CIA_VMS_ENV COM If you also wish to have your private user init pro then change the logical idl startup within your CIA VMS ENV COM file as described above 2 3 4 2 Unix You can find out what the command cia is by typing gt which cia 14 CH
397. tre of the reference pixel counting from 1 E g If the reference pixel is indexed in IDL as 9 9 then CRPIX1 10 5 and CRPIX2 10 5 166 12 13 14 15 16 1T 18 CHAPTER 15 CIA DATA STRUCTURE HIGH LEVEL ARCHITECTURE SIZE Number of IMAGEs in the SPD SCD and the number of FRAMEs in the ERD SCD Type integer DATA Cube containing IMAGEs in the SPD SCD Does not exist for ERD SCD Type integer IDL cube Unit ADU EOI Cube containing EOI FRAMEs Only exists for ERD SCD Type integer IDL cube Unit ADU RESET Cube containing RESET FRAMES present in both ERD SCD and SPD SCD Type integer IDL cube Unit ADU MODEL Cube available for holding processed images MODEL has the same dimensions as DATA or RESET and is only present in an SPD SCD Type float IDL cube EOI MODEL Similar to MODEL though used for processed EOI frames MODEL is only present in ERD SCD Type float IDL cube RESET MODEL Similar to MODEL though used for processed RESET frames RE SET_MODEL is only present in ERD SCD Type float IDL cube MASK A cube of the same dimensions as DATA EOI or RESET Each pixel in the MASK corresponds to a pixel in DATA SPD SCD or to a pair of pixels in EOI RESET ERD SCD and records the status of that pixel Masked pixels have one of two possible byte representations in the MASK This is determined by the value of the system variable IMASK Setting MASK to 0 selects the simple MAS
398. tructure put fills a field of this structure e structure get returns a field of this structure e structure write writes this structure to a file e structure read reads this structure from a file structure list returns the names of all existing structures of this type structure del deletes a structure structure find returns names of structures with a field equal to a given value e structure info returns information about a structure s fields structure elem returns the names of the SADs SCDs contained in an SSAD SSCD Ap plies to SSADs SSCDs only 187 188 CHAPTER 16 DATA STRUCTURE MANIPULATION 16 1 1 structure init This function is used to initialise a data structure in memory That is to say it will create an empty structure for you to fill with your own data Due to differences in the nature of the data structures each flavour of structure INIT are slightly different and so will be treated separately 16 1 1 1 sscd init To initialise a new SSCD in memory we can do the following CIA sscdi sscd init 7143006010101 ack ok CIA help sscdi ok SSCD1 STRING CSSC143006010101_96080512201019 OK INT 1 The first argument is the combined lt TDT OSN CN STATE gt number that is used to name the returned SSCD sscd1 that is created by sscd_init see Section 15 2 1 A keyword argument not used here is source which can be used to pass an already existing SSCD to sscd_init so as to copy its parameters to
399. tructure is fully compatible with CIA In particular it can be examined with isocont 21 8 ORGANIZATION OF DATA IN SLICE 275 Table 21 1 The SLICE variables and their content Pe M the data cube i e all readouts of the raster E lat the flat field cube one plane per readouts mapO initial version of the raster map created at the long term transient correction stage if im param does not exists not used otherwise map the raster map as created by the current flat field correction mask a mask associated to the raster map filled by the source action indicates the location of detected sources in the field map before the raster map as created by the previous flat field correction useful to compare the result of different operations mapi a copy of map created either by the make map action for the Single Flat flat field method or by the bad pixels action map2 a copy of map created by the ghost action map3 a copy of map created by the make map action for all flat field methods except Single Flat map4 equivalent to map but with the bad pixels masked created by the bad pixels action map5 equivalent to map but showing only the sources detected at by the source action map mjy the calibrated map hmjy header corresponding to map_mjy redun the redundancy map coverage factor for sky pixels error map the error map associated to map obs param a structure containing configuration parameters act the structure describing
400. ts sscds if they have less than min scd states was introduced sscd clean now acts properly on trailing CAM parallel readouts which still contain the raster point ID of the last raster point e Upgrade of point source photometry routines fit isopsf doesn t modify the input parameters xin and yin more As option the output remainder a fltarr containing the PSF subtracted input image was added The option confused improves fit isopsf s behaviour for confused regions Additionally it also returns a statistical measure of the positional error the new routine photom psf was introduced as an user friendly way to do PSF photometry the new routine photom aper was introduced as an user friendly way to do aperture photometry xcvf now has the ability to read write apertures This way users can simply load a previously defined aperture to work on several different cvf datasets The file I O is now done via IDL s dialog pickfile routine e It is now possible to recover previously unusable data suffering from the mis use of the raster point with the program repair rpid It patches a 0 0 raster point id contained in GPSCRPID of an ISOCAM ERD or SPD fits file with the best guess rasterpoint ID contained in GPSCFILL Calling syntax is CIA gt repair_rpid file e The slicers were improved use corrected coordinates from the pointing files by default the number of read outs accumulated or sampled on bo
401. uation can occur when CIA is running on remote machine and a PC X Windows server is used for display The IDL Astronomy User s Library routine getrot returns the ROLL angle and not the rotation angle as specified See Appendix E CIA creates log files when a session is initiated Information about your session and any errors which may occur are recorded in these files You can use the CIA routine error level to set the level of verbosity of error reporting both to the screen and the to log files See the on line help or cia help Section 2 3 2 for usage CIA uses IDL s READ and RESTORE to save CIA data structures to file Such a data file cannot be restored by a version of IDL pre dating the version which saved the file However the converse is not true IDL can always restore data saved by a preceding version As is usual with IDL in the event of a crash you may not automatically return to the main IDL level and so your variables will seem to have disappeared Generally you can recover by typing RETALL on the command line Following a widget crash you may find that all subsequently called widgets appear dead on your screen This is a problem with the IDL widget manager XMANAGER Usually invoking it manually type XMANAGER on the CIA command line will reactivate your widget Please be aware that the use of netscape or another X windows resource hog might get your CIA session stuck when using widget routines like xdisp or xv raster
402. ucture e g CIA struct scd extract CSCD143006010110_97092611592803 Now follow the following procedure 1 Create a variable to hold our MASK CIA gt our_mask bytarr 32 32 19 2 set the pixels we wanted masked to 1 Again we use the example that the first column of all our CAM IMAGEs are blind CIA our mask 0 1 3 Call ia put mask to put our_mask into struct converting the pixels of value of 1 to the appropriate value for blind pixels CIA ia put mask blind struct our mask 16 5 CDS DATA EXTRACTION 197 16 5 CDS data extraction This section describes ways in which you can examine the contents of a CDS remember that a CDS is a CIA data structure which holds CAM calibration data see Section 15 3 The different types of calibration data i e CAL G files and the CDS name for each is given in Table 15 3 2 on page 173 Let s assume you want to find out without using find best psf see Section 20 12 2 what PSFs we have for an observation configured with a 3 PFOV and the LW2 filter 1 Firstly you have to read in the LW PSF CDS CIA psf cds cds read lwpsf 2 Secondly you have to know the LW2 filter wheel position and the LW lens wheel position for the 3PFOV You can find this information either in the ISOCAM documentation e g ISOCAM User s Manual in the CIA files hk wheels n txt see Section 15 4 3 or with the function convert wheel back CIA lw2 position
403. uctures Many CIA users probably also use non IDL based analysis packages such as IRAF The most convenient way to share data with such external packages is by using the FITS format Due to the inability of many external packages to handle extended FITS files we need to use both non extended and extended FITS It is impossible to export all data from a CIA structure without using FITS extensions so for archiving purposes FITS extensions are unavoidable With these problems in mind two kinds of export have been developed i export for external analysis using non extended FITS and ii export for archiving using FITS extensions A small exception to this rule has been made with the routine conv cvf2isap 18 1 Export to the spectral analysis package ISAP conv_cvf2isap allows export of ISOCAM CVF data to the spectral analysis package ISAP CIA conv cvf2isap cvf pds cvf isap fits As you may guess bf conv cvf2isap accepts a CVF PDS as input As output it writes a FITS file with extensions that can be understood by ISAP The keyword option pix accepts a two element integer array specifying an ISOCAM pixel and a spectrum is extracted a this pixel position bf conv cvf2isap also works with ISOCAM CVF AAR products in this case it accepts the name of a CMAP file as input 18 2 Export to external packages This section describes CIA routines for export of data to non extended FITS files for analysis by external packages raster2fits writes t
404. ue in a field of a structure This can be a useful routine when a data structure field is undefined or incorrectly defined The fields for the relevant structure are given in Chapter 15 Note that you should be careful when using structure put Changing a field value unnecessarily may disrupt your data On no account change the field NAME it will confuse CIA memory management Also read Section 16 4 before attempting to manipulate the MASK with scd put As an example the string EARTH is placed in the field NAME of the specified SCD CIA scd put target EARTH CSCD143006010105 96080110071423 ack ack 190 CHAPTER 16 DATA STRUCTURE MANIPULATION 16 1 4 structure get Function to return a value of a specified field of a structure Following from the example in Section 16 1 3 the value of the field NAME is extracted CIA help scd get name CSCD143006010105_96080110071423 ack Expression STRING EARTH ACK INT 1 See Section 2 4 for restrictions of use 16 1 5 structure write Procedure to write a specified structure to a file The filename will be the name of the structure with cub appended When using sscd write and ssad write the entire set of SCDs or SADs are written to disk In the example below a single SCD is written to disk in the directory product dir see Section 17 1 1 CIA scd write CSCD143006010105 96080110071423 dir scd dir ack ack CIA help ack ACK INT 1 See
405. undesirable SCDs STATEs that are IDLE and STATEs where the filter wheel was not set to LW10 should be discarded CIA scds sscd elem sscd CIA scd del scds 0 4 CIA scd del scds 22 23 Now we must place the contents of the SSCD into a PDS For a beam switch observation we use BS PDS This is created with get sscdbs CIA bs pds get sscdbs sscd Note that there is one irregularity which sometimes arises in beam switch observations some observers have programmed their observations in reverse Section 19 3 tells you how get sscdbs can be used to deal with this problem Now we can proceed with the calibration We will perform the standard calibration steps on the cube i e bs_pds cube CIA x3d bs pds 6 2 DATA ANALYSIS 43 When you are satisfied we can perform the first calibration steps of DARK correction and deglitching As in Section 4 2 the data does not need stabilization correction if you are not convinced you can check this with x3d before progressing CIA corr dark bs pds CIA deglitch bs pds Now the PDS contains a nicely calibrated cube Again you might want to check this with x3d see Figure 6 1 Use the same calling sequence as above 7 Now we can create the EXPOSUREs CIA reduce bs pds and perform flat field correction CIA corr flat bs pds 8 You may be curious as to which EXPOSUREs are source and which are reference CIA print bs pds src image 1 5
406. urce field EXPOSUREs index to ref EXPOSUREs number of beam switch cycles 15 5 6 CAL G PDS substructure reference or type Section 15 2 3 13 Section 15 2 3 14 Section 15 2 3 15 string Section 15 2 3 7 integer integer integer integer integer integer fltarr nx_raster ny_raster fltarr nx_raster ny_raster fltarr nx_raster ny_raster string intarr nscd fltarr fltarr integer Here we describe the CAL G data substructure which is found in all flavours of PDS It is used to contain data taken from the CAL G files distributed with CIA see Section 9 3 5 It is automatically filled when the PDS is created subfield type description of contents CALG DARK fltarr 32 32 CAL G DARK CALG DARK_NAME string DARK name CALG FLAT fltarr 32 32 CAL G FLAT CALG OFLAT NAME string CAL G OFLT name CALG DFLAT NAME string CAL G DFLT name CALG PSF fltarr 32 32 CAL G PSF CALG PSF_NAME string PSF name 15 5 7 INFO PDS substructure This substructure preserves parameters from the individual SCDs used to create the PDS The table lists the fields of the INFO substructure along with a reference to the SCD field 15 5 PREPARED DATA STRUCTURE PDS subfield I I 1 I 1 I I I I NFO RA i NFO DEC i NFO ROLL i NFO CRPIX1 NFO CRPIX2 NFO GAIN i NFO OFFSET i NFO TINT i NFO ENTWHL i NFO SELWHL i NFO PFOV i NFO N_ACCU i NFO SCD_NAME i NFO FLTRWHL i NF
407. ure 2 2 typical ciainfo display for VMS 2 8 GETTING STARTED Supergroup Craphic dispiny imago processing Asiron Bor put output to im and export data with dilierent format all he rou iners 1o keap Erick Ele 05 di Graphic display For your eyes only image processi p create and images Calibration to create dark flat fiped arsi all the calibration tal i input Output Calilxration Coneral_Tools Baia management Find Quit Help Ovarian find display _ NAME CDs INFO _ STRUCT ZEAD PURPOSE SAD_INFO Compute a CVF spectra Select interactively a region in an image SSAD_INFD process and derive the spectra SED INFO CATEGORY DEGLITCH VILSPA 11 2 USER SOURCES 11 GRAPHICS displays plots high level routines DEL PATTERN CALLING SEQUENCE FIT 5RC i cl display CF scil scil T PSODDNT PARAM ni INPUTS none MHTTER P5F KEVED INPUTS PSF AVERAGE SCD display all the rame nested of he reduced images Inpix 32 by 32 image the Image to display Irtead of the mean FLUX image pn dee ri le af Lhe axis axis airing style of T lA CAM STATUS T e ME CALIB_C F o axis FIT 20GAUSS 1_ intar absciss log ordinates FIT DISPLAY b log lear ordinaat cri else see i FIT P5 t nom string coreplementary title the
408. ut file You must follow any changes to this name with the return enter key for the new name to take effect The two display units below the top row show the current spectrum in the source aperture solid line the sky aperture dashed line and the difference between the two thick solid line marked with symbols The left panel shows the full spectrum The right panel shows an expanded zoomed view The zoomed region is marked by two vertical dotted lines in the left panel The set of buttons immediately beneath the plotting windows allow the user to select how the spectrum is computed Choices are average median or the sum of all spectra currently in the aperture The axis selection allows one to plot the spectrum versus the wavelength or the frame number The arrows lt and gt allow the user to reposition the two vertical lines marking the boundaries of the expanded view CHAPTER 14 IMAGE ANALYSIS AND DISPLAY Export Options Spectrum raw thin line background dotted line source thick line wish eA Image Scaling Parameters Beam Selection sse Pixel Information CVF image frame 12 Figure 14 3 The xcvf window Expanded view of the spectrum Hot pixels show current Aperture 14 2 CVF IMAGE ANALYSIS 121 The section of the GUI beneath the plotting region displays frames from the reduced CVF cube The left most part of this region displays astrometry information and allows th
409. uted EXPO SUREs in detector coordinates It is also a field in an SAD used to hold data from the CCIM data product or CIA processed data CGLL CAM Glitch List An AAR level data product containing a list of AA detected glitches CIA user friendly format A term referring to the manner in which CAM parameters are presented in the CIA data structures CIDT CAM Instrument Dedicated Team Located in VilSpa its main tasks were to monitor and calibrate ISOCAM CIER CAM Instrument Edited Raw data i e edited telemetry data An ERD level data product containing EOI and RESET frames CISP or CIER can be the input data for CIA analysis CISP CAM Instrument Standard Processed data A SPD level data product containing IM AGES computed by AA from CIER data CISP or CIER data can be the input data for CIA analysis CIST CAM Instrument Support Team Located in Saclay and Orsay France it supported the CIDT during the mission CJAM CAM Jitter And Memory Also contains stabilization information This is an AAR level data product CLEAN See OP MODE 1 the ISOCAM Handbook for more information on CAL G data products 307 CMAP CAM MAP A data product containing AA computed EXPOSUREs in astronomical coordinates It is also a field in an SAD used to hold data from the CMAP data product the CMOS data product or CIA processed data CMOS CAM MoOsSaic An AAR level data product containing MOSAICS constructed from EXPOSURESs contained in the CM
410. ux error CIA plot est wave est flux 1 RASTER is not a likely name for a beam switch MOSAIC It was chosen early in the development of the beam switch analysis routines in order to make the BS PDS compatible with the raster PDS 246 CHAPTER 20 ADVANCED DATA CALIBRATION dir has to point to a local directory containing the theoretical PSFs see Section 15 4 1 The output of the program is e sort wave wavelength sorted in ascending order e flux The flux of the point source one per executed CVF step e est flux The flux of the point source one per observed wavelength In case a wave length was observed twice est flux is the mean of both observations e est flux error The error of the point source flux one per observed wavelength This error is computed on the fly by Monte Carlo simulations 20 7 Analysis of solar system objects Basic data reduction steps are performed as in chapter 5 2 The exposures of the general PDScreated with get sscdstruct are dark corrected deglitched transient corrected and also flat fielded CIA gt corr_flat sso_pds cube As next step you have to get the ISO centred ephemerids of the solar system object CIA jd convert_time sso_pds utk UTK MJD 24000001 CIA print min jd max jd format d13 4 2450849 0361 2450849 0380 jd contains the time in Julian Date when these data were taken Then you have to access the HORIZONS On Line Ephemeris Sys
411. values see Section 15 2 2 18 Also take a look at Section 2 3 4 for how to configure the MASK handling e ia put mask routine to modify the MASK in a CIA data structure or IDL array To work with ordinary IDL arrays you can use put mask e ia get mask routine to extract MASK from a CIA data structure 16 4 1 Extracting the MASK from CIA data structures Follow the procedure below to extract and examine a MASK from a CIA data structure 1 Suppose we have a CIA structure The first step is to extract the entire MASK from the structure e If our structure was an SCD this can be done as CIA mask scd get mask CSCD143006010110_97092611592803 e If it is a PDS CIA mask pds mask 2 Now to look at the dead pixels use ia get mask to extract and tviso to display CIA dead mask ia get mask dead mask CIA tviso dead mask 0 If you look at the values in dead_mask you will see that they are either 0 or 1 However the values in original mask can be 0 1 2 4 6 8 16 32 64 128 each value having a separate meaning see the table in Section 15 2 2 18 ia get mask translates these values to a simple 0 or 1 In our example above we asked for the pixels masked as dead to be translated so all the pixels with a value of 2 in original mask became 1 in dead mask 196 CHAPTER 16 DATA STRUCTURE MANIPULATION Alternatively you can extract the dead pixels from an SCD directly CIA dead mask scd get
412. ve described the data products or files on the CD ROM and and how the data products relate to the data product types Here we shall discuss how the data product types relate to the data structures employed by CIA 9 4 1 What is a CIA Data Structure The CIA Data Structures have been specially designed to manage ISOCAM data within a CIA session There are several types of structures each of which are used to hold a specific type of data product or data from different stages of CIA processing 9 4 RELATING DATA PRODUCT TYPES TO CIA DATA STRUCTURES 65 The CIA data structures are quite complex structures containing many data These include images FRAMEs EXPOSUREs and MOSAICs and CAM parameters CVF position RA and DEC of images etc Where possible the data are presented in fields of the structure in a user friendly format telemetry coded parameters have been converted to string information by CIA With the exception of PDSs see Section 9 4 4 these structures are not regular IDL structures Unfortunately due to IDL memory management restrictions these structures do not have a user friendly user interface and may well be the most difficult hurdles for CIA novices You can t do something like CIA help cia struct str However they do offer great flexibility They can handle data from the most exotic of observations observations with telemetry drop outs etc and allow the advanced CIA user the possibility to perfo
413. ves the cursor to the mouse position and echoes the intensity and the coordinates in the terminal of the new cursor position If the input is a raster then additional information about the data cube are printed right click moves the image so that the position of the cursor is in the middle of the displayed image 14 4 CUBE ANALYSIS 139 z Iz ISOCAM CIA Cea Saclav Image Display Banner File Scale Tools Zoom 2 Min 16 5546 amp Max 112 103 ct Y thumbnails Fixed Reset V 4 Data mod za J Raster Statistics Mouse Mode amode C Title Title N66CSMC Lw3 6 39 39 76 8959 00 59 6 65 72 10 24 7 J2000 Intensity amp coords Main Image amp color table Figure 14 11 The main window of ximage 140 CHAPTER 14 IMAGE ANALYSIS AND DISPLAY 14 4 9 Raster visualization This functionality only applies for a raster PDS see Sections 15 5 and 15 5 4 If Data mode is selected a click on the button raster raises a cube analysis window The main image is a mosaic of several sky views which overlap Each sky view or SCD contains several readouts usually from 10 to 30 This new window contains a plot of the history of all the camera pixels that have contributed to the final flux of the pixel selected by the cursor in the main image This camera pixels will b
414. vf sscd dir path CIA scd dat scd dir ack ack The SPD SCDs have now been created and the name of their SSCD is held in cvf sscd CIA print cvf_sscd C38C203056040001 96091918462427 4 To get an overview of the SCDs use sscd info 12 2 AUTOMATIC DATA SLICING 85 CIA sscd info cvf_sscd 77 SCDs in the SSCD CSSC203056040001_96091918462427 Seq channel mode fltrwhl pfov tint gain offset size ra dec 0 LW OBS LW CVF2 3 0 2 10 1 512 99 191233 30 673941 59 1 LW OBS LW CVF2 3 0 2 10 1 512 21 191233 30 673941 59 2 LW OBS LW CVF2 3 0 2 10 1 512 20 191233 30 673941 56 3 LW OBS LW CVF2 3 0 2 10 1 512 20 191233 32 673941 59 4 LW OBS LW CVF2 3 0 2 10 1 512 21 191233 30 673941 64 5 LW OBS LW CVF2 3 0 2 10 1 512 20 191233 31 673941 62 72 LW OBS LW CVF2 3 0 2 10 1 512 20 191233 30 673941 62 72 LW OBS LW CVF2 3 0 2 10 1 512 20 191233 30 673941 62 73 LW OBS LW CVF2 3 0 2 10 1 512 20 191233 30 673941 59 74 LW OBS LW CVF2 3 0 2 10 1 512 21 191233 31 673941 64 75 LW OBS LW CVF2 3 0 2 10 1 512 21 191233 31 673941 62 76 LW IDLE LW CVF2 3 0 2 10 1 512 4 191233 32 673941 51 5 So from the above you can see that there are 76 STATEs of CAM under OP MODE OBS and the final STATE under OP MODE IDLE the list has been abbreviated for neatness This STATE we can discard This is easily done by putting all the SCDs in an IDL string array cvf_scds and then deleting the guilty IDLE STATE CIA cvf_scds sscd_elem cvf sscd CIA scd del
415. vide your own custom levels etc Contour colours can also be set These parameters are equivalent to keywords nlevels min value max value levels black c_style c_color Optional Overlays Allows some extra features to be added to the plot These parameters are equivalent to keywords grid nonorth putinfo title star Register Parameters Set shift configuration isocont can perform manual shifting of the images i e you provide the number of pixels to shift by in x and y directions or automatic shifting by image correlation These parameters are equivalent to keywords shift offset Display Parameters Set parameters equivalent to keywords nosample rect missing Window Parameters Font Parameters 14 6 IMAGE COMPARISON AND OVERLAYING 149 Figure 14 17 x isocont window 150 CHAPTER 14 IMAGE ANALYSIS AND DISPLAY 14 6 4 xcorr astro xcorr astro is a graphical tool which displays two images side by side and allows the user to 1 Cross correlate two source catalogs 2 Determining shifts between two images by identifying stars common to both images 3 Determine PSF fits and centroids for sources on the images The ultimate goal of xcorr astro is to correct for the astrometric shift induced the wheel jitter offsets The astrometry information within the ISO structure is modified to reflect these corrections How to use xcorr astro Calling syntax CIA xcorr astro cia pds reference fits Issuing the c
416. with image visualization the description 14 2 CVF IMAGE ANALYSIS 119 conv_cvf2isap may be found later in Section 18 1 Both these routines output the CVF data to a file that is readable by ISAP The calling sequence for xcvf is CIA xcvf cvf_pds xcvf also accepts the keywords iz Exposure number to display zl Lower bound on the range of intensities to display z2 Upper bound on the range of intensities to display outfile Name of the default output file help If set then xcvf will display a short help screen On startup the user is greeted with a summary of mouse commands and explanation of the plot symbols and lines 14 2 2 1 The display The row of buttons and fields at the top of GUI perform the following operations QUIT Exit xcvf HELP Re print the short summary listed on startup RESET ALL Reset all apertures as well as the lower and upper bounds on the intensity ranges to display COLOR Bring up IDL s xoolor widget for interactive control of color tables and stretches Write ASCII File This will write the current spectrum as an ASCII file with column headers and a short summary of apertures and combination option used to produce the spectrum Export Current Spectrum Call conv_cvf2isap and export the current spectrum for use in ISAP The current spectrum is the one shown by the thick solid line and the symbols in the plotting window see below Output File The name given by conv_cvf2isap to its outp
417. y an optical image with contours from a CAM image 147 14 17 xisocont windOW sx ee ye Poem de OR ea Ee Ga 149 14 18xcorr_astro window 152 15 1 eds display window 175 19 1 Spectrum from an up and down LW CVF 216 19 2 x_slicer s dreaded message 221 20 1 Processing of observations using the small Fabry mirror 241 20 2 The raster MOSAIC with SSCD calibration and Fouks Schubert transient correc CODY e tts ao hg oh pt eaa OG olde eG a s it aala a ee 243 20 3 Comparison of standard vs improved SSO processing 249 20 4 flat builder s main window 252 20 5 bkg_builder s main window 255 20 6 Comparison of standard projection vs weighted projection 261 20 7 Original pixel histories of the same 263 20 8 Back projected pixel histories of the same 264 20 9 Distortion correction of staring 265 20 10 The RMS image that correspond to Figure 3 3 268 20 11 The weight image that correspond to Figure 3 3 268 21 1 The raster maps using a standard CIA procedure see text for details Left panel shows the LW3 data while the right panel
418. ysis process until the image level by reducing the SSCD to aSSAD This later option is unavoidable if the SSCD is comprised of data with non constant tint or PFOV See Section 15 5 for a description of the architecture of a PDS 95 96 CHAPTER 13 DATA CALIBRATION 13 1 2 raster PDS The raster PDS is designed to handle sliced data from the following observation types e raster AOT 1 e micro scan observation AOT 1 To create the raster PDS from the sliced MxN SPD SCDs where N and M are the raster dimensions use get_sscdraster To continue the example of Section 12 2 we assume that the IDL variables Iw6_sscd and Iw6_scd_dir contain strings with the name of an SSCD cataloging the sliced SPD SCDs and the name of the directory path where it is has been saved respectively To avoid confusion and save on memory it is a good idea to delete all SSCDs SCDs that may be have been left over from the slicing process using scd del or sscd del reload the sliced SSCD you want to calibrate from disk and then proceed to build your PDS CIA lw6_sscd sscd_read cssc143006010002_98060117274484 cub dir scd dir CIA 1lw6 raster get sscdraster lw6_sscd CCGLWDARK 97031713382678 not exact matching for GAIN 1 lt gt 0 CCGLWOFLT_98050815090326 not exact matching for TINT 15 lt gt 36 lw6 raster is a raster PDS containing all your prepared data Don t worry too much about any unusual looking messages like the two appeari
419. ytes integers strings etc and may include structures An array is a simple example of a data structure DFLT Detector FLaT field DSD Diagnostic Specific Data CIA data structure containing house keeping data EOI End Of Integration This refers to CCD read out i e FRAME after integration is complete ERD level Edited Raw Data ERD level refers to the level of pipeline processing that pro duces ERD For CAM ERD is synonymous with CIER EXPOSURE EXPOSURE when written in upper case in the CIA User s Manual refers to a single image computed by averaging IMAGEs over a STATE see also MOSAIC and FRAME 308 APPENDIX A GLOSSARY FITS data products The data products on the CD ROM are in the form of FITS files the actual CAM image being stored in a binary extension of the FITS file FLAT A flat field image For CAM this is the product of an OFLT and a DFLT see glossary entries for both of these terms Also an OP MODE when CAM is performing an internal flat field measurement FRAME FRAME when written in upper case in the CIA User s Manual refers to a single CAM CCD read out There are two possible FRAMEs EOI or RESET See also MOSAIC and EXPOSURE future SAD A flavour of SAD which may contain a MOSAIC either taken from the CMOS data product or computed with CIA from EXPOSUREs It may also contain spectra IDA ISO Data Archive Archive of ISO data products at Villafranca which is accessible over the internet

ISOCAM Interactive Analysis User`s Manual Version 5.0

Contents

Download Pdf Manuals

Related Search

Related Contents