Home

Very Large Telescope

1. 0 00 0 50 0 25 0 50 0 50 0 50 1 T F T 7 Ar T T e D E 7 ST SG i AN XS TM J an E ull E e E GL E S EGE EN E J g gt x GE gt ool g 3 af 7 E N b 4 or A ps D l l i l Es l i l 0 02 0 0 02 65 24 63 26 63 28 126 48 126 5 126 52 X mm X mm X mm 0 00 0 25 0 25 0 25 0 50 0 25 x ET Cl A T A A FT r ke NP M O o e o i o o EN a pu E pale 4 SE E 12 E o E c 1 E c E 4 M Si Sen gt a A SH 4 Mp E ME 7 bel e o e l l i i l f i 0 02 0 0 02 63 24 63 26 63 28 126 48 126 5 126 52 X mm X mm X mm 0 00 0 00 0 25 0 00 0 50 0 00 T T T T T T T CH CH CH oL 4 oL 4 oL J o o o E M e E Eo n3 o gt gt gt CH CH CH o o o C3 X of H oj l i l i l l l l i i 0 02 0 0 02 63 24 63 26 63 28 26 48 126 5 126 52 X mm X mm X mm Figure 4 Sample spot diagrams for 500nm light in a quadrant of the OmegaCAM focal plane A 2x2 pixel grid 301m x 30m corresponding to 0 43 is shown as well OmegaCAM User Manual VST MAN OCM 23110 3110 version 95 0 7 and rotation can be tracked accurately The other two auxiliary CCDs are mounted 2mm outside the focal plane one in front one behind and are used for recording defocused star images for on line analysis of
2. 20 19 18 17 4 3 21 1 A N Fits extensions bottom numbers order may change whenever Fiera settings change chip names ESO CCD index above E 10000 5000 0 5000 10000 pointing center X prescan overscan Figure 3 The layout of the 32 science CCDs and the four auxiliary CCDs in the focal plane The science array covers a 1 degree x 1 degree area of sky with 16k x 16k pixels of 0 21 arcsec All CCDs fit inside a circle of diameter 1 4 degrees on the sky Note that this is the layout as seen from below i e the projection of the array onto the sky OmegaCAM data are delivered as multi extension FITS files The CCD name 4265 96 is included in the header of each fits extension CCDs can also be identified by the number of their fits extension but beware this order may change from what is shown in the Figure whenever CCD controller parameters are changed Two of the auxiliary CCDs are mounted out of the focal plane Extra and Intra focal IF amp EF at displacements 2mm in order to enable image analysis and real time active optics corrections Auxiliary CCDs G1 and G2 are used for guiding All 4 auxiliary CCDs are partly vignetted around their respective outermost corner OmegaCAM User Manual VST MAN OCM 23110 3110 version 95 0 6
3. es 19 2 5 3 Differential Guiding e e irisa aa moui e ania a a ok T mr US X Eo 19 2 5 4 Image Analysis with OmegaCAM en 20 2 5 5 Traffic rules between science and auxiliary CCD readouts 20 2 0 Galibration lampsz L5 25h a Ree Rm ORO RO Ls re AL ACRAS RR ELS XE Pd 20 2 7 Photometric Properties ssaa m Aet e a uag Xue em REGN NR Ros 21 2 8 Astrometric Properties 2 0 21 2 9 Flat fielding of OmegaCAM data Illumination correction 21 2 9 1 Performances of the new Baffling system e 22 3 Observing with OmegaCAM 22 3 1 Offsetting Modes and Observing Strategies o a 25 Sel Offsetting Modes sss 2 id st a a ER AA 25 3 1 2 Observing Strategies ice a no hos ced wee EUR xU S Wed hr eee Ee dnd 26 OmegaCAM User Manual VST MAN OCM 23110 3110 version 95 0 iv 3 2 Atmospheric Refraction and Dispersion ls 26 3 2 1 Atmospheric Refraction 27 3 2 2 Atmospheric Dispersion 000000 eee ee 27 3 9 OB pr paration 2 5 a e te ben eed eb IE de es 29 3 94 Target acquisition uw curet ge eere cR E VR een ARMES en 29 3 9 2 Scientific exposures E Vsus d Re a E pM dud 29 3 3 3 Dither and jitter patterns 29 3 4 The Observing templates ens 29 3 4 1 Target Acquisition Templates 29 9 12 Science Templates A8 unom er ARA SS ark dee deas 30 3 4 3 Nighttime Calibration 31 344 Overheads x Gk e ME E A EA AA VER s
4. OmegaCAM User Manual VST MAN OCM 23110 3110 version 95 0 40 200 20 200 200 20 renge AE SEET See 200 e E EJ ver Ss cul o 3 Er ET Gg Sie cul Figure 24 Crosstalk between CCDs 93 96 Each panel shows the counts in a raw frame of a receiver CCD as a function of the counts in another emitter CCD Positive negative slopes correspond to positive negative crosstalk In several cases there is a clear linear relation illustrating crosstalk from unsaturated pixels Saturated pixels also cause crosstalk in many combinations of emitting and receiving CCD their median count level is shown as a star Each row corresponds to a different receiver CCD within a row the three possible emitter CCDs are plotted in numerical order OmegaCAM User Manual VST MAN OCM 23110 3110 version 95 0 41 4 Calibrating and Reducing OmegaCAM data All OmegaCAM data are taken in service mode in the context of a calibration plan RD2 whose function is to maintain the overall calibration of the instrument and atmosphere Thus the aim is not to calibrate individual data sets but rather to calibrate the sky telescope instrument chain Data volumes from OmegaCAM are LARGE A single exposure leads to 0 5 GB of raw uncompressed data at 2 bytes per pixel processed exposures with a 4 byte real number per pixel are over a gigabyte in size Typical data v
5. The focal plane of the telescope is not totally flat such that even with perfect image analysis performed with the OmegaCAM auxiliary chip mild PSF variations across the field of view are unavoidable This applies both to FWHM and ellipticity and is in general more pronounced for redder filters In Fig 18 we show a PSF anisotropy analysis plot obtained for a 240 seconds i band exposure IQ variations of 10 2596 across the FOV can occur depending on outside seeing filter and telescope mirror position Fig 19 shows a histogram of the PSF FWHM measured in the g r i bands between August and Thttp www eso org observing etc bin gen form INS NAME OMEGACAM INS MODE imaging OmegaCAM User Manual VST MAN OCM 23110 3110 version 95 0 34 OmegaCAM IQ distribution August December 2011 each data point is average IQ in one image over full chip array 1 03 i l Dn O7 i VE os 1 D i DEE 0 2 x TM delia 1 16 18 2 24 ADT Geman a TAA te os os te os os Ge 1 omegacam iq at 475nm arcsec Ab omagacam iq at E Mom arcsec AMEI Median IQ ing 0 95 Median IQ in r 0 85 Median IQ in i 0 80 2 o Internal IQ 0 4 0 5 D TA CH all DL 06 08 1 12 1 2 22 24 cenegacam iq at 756r arcsec AMT gt Outside median IQ 0 80 600nm 02 O Julio N
6. 3 3 1 Target acquisition A target acquisition presets the telescope pointing and configures the mirrors corrector and instrument rotator In addition it simultaneously prepares the instrument by loading the appropriate filter in the beam and acquiring guide stars and image analysis stars if required 3 3 2 Scientific exposure The scientific exposures load an appropriate filter lock onto the guide and IA stars perform any offsets clear the CCD array open the shutter for the appropriate time close the shutter and read out the CCD array A science template may command a number of exposures possibly interspersed with offsets Within a template parameter lists are cycled through if the number of values is smaller than the number of exposures specified Several science templates are needed if consecutive observations with different filters are required 3 3 3 Dither and jitter patterns Different templates are used for observations taken in dither or jitter mode These templates imple ment different default offset patterns The patterns are described in the Template Manual RD1 3 4 The Observing templates The detailed use of the templates is described in a separate document RD1 Here the main func tionality of the available templates is described 3 4 1 Target Acquisition Templates Each OB must start with an acquisition template whose execution points the telescope and prepares the instrument 5http www eso org observing e
7. at the shutter location the light from a given star is smeared out over a width of about one tenth of the full shutter aperture This means that the time over which a given pixel receives photons is longer than the actual exposure time by an amount of roughly 0 08 s Seen from an individual pixel the photon rate over time does not have steep edges but is smeared out OmegaCAM User Manual VST MAN OCM 23110 3110 version 95 0 19 2 5 Auto Guiding and Focussing Both the VST and OmegaCAM contain systems for auto guiding and for focussing Focussing happens automatically through real time sensing of the wavefront Filter dependent focus offsets are also automatically applied Traditional through focus sequences are only needed to calibrate these auto focus systems The VST tracks well and for exposures shorter than 60 sec there is no need to auto guide Depending on the user s image quality requirements up to 2 minutes without guiding can be accepted if the target never approaches the zenith to closer than 10 degrees Commissioning and verification tests have shown that tracking drifts are of order 0 1 0 2 per minute but increase when the telescope approaches zenith Whether autoguiding is desired or not can be specified in the acquisition template of an Observation Block 2 5 1 VST Guide Probe The VST contains a guide probe with a pick off mirror which can patrol the field in front of the filters It can be used for guiding the tele
8. 7 1 Update to overheads and PSF Version number aligned with Period number 24 05 2012 Sects 3 4 4 Update to overheads IAstar and telescope offset 15 12 2012 Sect 3 7 5 New Section 3 7 5 about vignetting by wind mast 07 03 2013 Sect 3 4 4 Update to overheads parallelised readout and preset filter set 23 05 2013 Sect 3 4 4 Update to overheads first offset in a template is 5s shorter 23 08 2014 Sect 2 9 1 Performances of the new Baffling system installed in P94 OmegaCAM User Manual VST MAN OCM 23110 3110 version 95 0 iii Contents 1 Introduction 1 Job Credits i s iur tense Se Dd Be beh de bevor em uid E RE Rede 1 E COPE aue de ete exuere IDE ete A T eene a uus 1 13 More Info sta a to e lo o in ti bu Ee ta dd e E A 1 di References E eras we tet he a a A SE 1 1 ACTOS da ete d AAA dE Eur er Ate i 2 2 Instrument characteristics 3 DL COVEIVIEW a a RACE ERR A A sd A Gor du 3 2 2 Detector Mosaic oi a See ee E a de 7 2 2 1 The 268 Million Pixel Science Array o o e 7 2 2 2 Auxiliary CODS Laprida ads 13 2 97 Piers E ARS A A ath E Boss a tata a 13 2 3 1 Note on filter magazines e 15 2 3 2 Providing Private Filters lt lt o 17 2 4 Shutter 22 4 2p dete kc a v de run A a do e dod 18 2 5 Auto Guiding and Focussing es 19 2 5 1 MST Guide Probes eiii se pops ok 9 eedeBobe a o B Pope 19 2 5 2 Auto guiding with OmegaCAM
9. It is the only instrument on this telescope and is operated most nights In principle all observations are carried out in service mode Its main function is to perform large optical imaging surveys which may be used to feed the VLT telescopes with targets With its wide field broad band filter set and image quality matched to Paranal seeing conditions OmegaCAM is well suited to this task This document serves as a user manual for those planning or preparing observations with OmegaCAM It describes the main instrument characteristics and how to observe with it using Observation Blocks and Observing Templates The most recent version of this manual is kept at http www eso org sci facilities paranal instruments omegacam doc 1 3 More Info Further information can be obtained by contacting usd help eso org and consulting the public OmegaCAM webpage http www eso org sci facilities paranal instruments omegacam ESO Quality Control Pages are found at http www eso org observing dfo quality ALL 1 4 References RD1 Template Manual VST MAN OCM 23100 3111 see http www eso org sci facilities paranal instruments omegacam doc RD2 Calibration Plan VST PLA OCM 23100 3090 see http www eso org sci facilities paranal instruments omegacam doc RD3 DFS User Requirements VST SPE OCM 23100 3050 see http www eso org sci facilities paranal instruments omegacam doc RD4 WFI documentation see http www 1s eso org lasill
10. Magnitude zero points for the SDSS filter set Illumination correction functions e Key difference between the dither and jitter modes lees Maximum airmasses without using the AIC ee Overhead times for observations 2 OmegaCAM Photometric Standard Star Fields List of Figures 10 11 12 13 14 15 Schematic drawing of Omega AN Optieallayout ofthe VST e iia RE m uk he e a eel A A Layout of the CCDs in the focal plane Spot Diagrams 4 voee rad Atak as ta MEE SUPE dati Photograph of the detector mosaic ooo Single CCD pixel layout ue Sete hb BP AA A Ee de Seg Schematic of CCD readout e Linearity ofthe CCDs iii RA bho Ree ee E RV ky ee a Filter times CCD throughput An OmegaCAM filter hh hen Shadow cast by the cross in composite filters 0 o e eee Orientation of filter as function of magazine The shutter Wb ta Ze eg 8 om ara er acte ERAI a A Illumination correction derivation for OmegaCAM Sloanr sss Asymmetric straylight in flatfields for OmegaCAM Sloanr 53 13 16 17 21 22 26 28 32 43 OmegaCAM User Manual VST MAN OCM 23110 3110 version 95 0 vi 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 Raw dome flat divided by raw twilight flat for OmegaCAM Gloanr 25 Limit on exposure time due to atmospheric refraction 28 OmegaCAM PSF anisotropies lees 33 OmegaCAM IQ distribution collected betw
11. atmospheric dispersion induced spectrum reaches 0 2 0 5 and 1 arcsec for the SDSS filters OmegaCAM User Manual VST MAN OCM 23110 3110 version 95 0 29 3 3 OB preparation Preparing an Observing Block for OmegaCAM is done with ESO s P2PP3 tool RD5 An OB consists of a number of templates each of which represents a standard action taken by the telescope and or instrument Each OB must include one target acquisition template whose main role it is to point the telescope at the target and one or more science templates which perform the actual observation In addition an OB contains a constraint set which specifies under which conditions seeing transparency moon phase timing constraints the OB may be executed Full details are given in the P2PP instructions on the ESO web site RD5 For reasons of flexible scheduling of service mode observations at ESO and to increase the chances that an OB is executed fully within constraints the total execution time of each OB must be at most one hour Longer observations need to be split into separate OBs Unless there are good reasons to do otherwise OBs should only use one filter each If filter changes inside an OB are required it is most efficient to order the exposures in such a way that filters are alternately selected from the two magazines See Table 4 for the filter distribution To help with planning observations one can use the Exposure Time Calculator ETC from the ESO web site
12. automatically as does the computation of aberration coefficients Image analysis is performed at the same wavelength as the autoguiding which in most cases is the same as the science filter see Sect 2 3 Image Analysis in the aquisition typically requires 3 iterations of 40 seconds each During science exposures aberrations are constantly measured by the IA system and corrections are applied to the mirrors during the readout The minimum science integration time that guarantees at least one aberration measurement is about 1 minute OBs with shorter science exposures will not get additional IA corrections after the aquisition Tests have shown that the IQ remains reasonably stable for about half an hour without IA corrections OBs that consist of short integration times should therefore not last more than half an hour Image Analysis in the aquisition can be disabled for concatenations of short OBs that perform a sequence of observations back to back in adjacent fields 2 5 5 Traffic rules between science and auxiliary CCD readouts The OmegaCAM CCDs are controlled by three controllers called FIERA 1 2 3 where FIERAs 1 amp 2 control the science CCDs and FIERA 3 the auxiliary CCDs Fig 3 FIERAI is the master controller Fig 7 The readout rules are such that the science array cannot be read out at the same time than the auxiliary CCDs Otherwise strong interference noise patterns will appear in auxiliary and science CCDs Therefore the
13. mast position Test sky flats taken in November 2012 showed that the VST pupil is vignetted by several percent by the Paranal wind mast when the telescope points at altitude 45 degrees and 162 5 degrees azimuth This is in the north north west Fig 22 shows the respective sequence of sky flats taken at 40 degrees altitude and 4 5 degrees from the mast The linear structures are from the wind mast OmegaCAM User Manual VST MAN OCM 23110 3110 version 95 0 36 Ghost radius surf br remark arcsec Amag arcsec 1 70 16 window filter 2 330 16 6 CCD outer window 3 150 17 1 CCD inner window 4 160 17 6 window filter Figure 20 Example ghost reflection from a g 6 5 magnitude star in a 60 second g band exposure in CCD 91 The ghosts line up along a radius vector through the center of the array which is towards the bottom right in this case Appearances will be different in other locations Ghosts are numbered in order of decreasing surface brightness from top bottom numbers 4 2 1 3 The table on the right gives their radii and surface brightness scaled to a magnitude zero parent star If stars of this brightness or above as well as OmegaCAM observations were uniformly distributed in the sky every fourth to fifth field would contain one such star OmegaCAM User Manual VST MAN OCM 23110 3110 version 95 0 37 Figure 21 Example of reflection of light from a star in this case two stars to the West outside the field S
14. scientific data are analysed by Quality Control in Garching anymore OmegaCAM User Manual VST MAN OCM 23110 3110 version 95 0 45 Measurements in center Separation 2 mm Figure 26 Location of the filter throughput measurements and coordinate system definition The concentric circles have radii of 29 57 5 86 and 115mm additional measurements on the diagonals were made at radii of 144 and 172mm The scale on the sky is 14 3 mm The x and y coordinate axes are indicated the filter moves towards its positive y with the two notches ahead when it is inserted in the instrument The filter coordinate system is identical with the instrument coordinate system if the filter is loaded from magazine B A Filter throughput curves In this section we present the results of measurements of the filter throughput performed in the lab at the Universit ts Sternwarte M nchen A 1 Broad band Sloan filters u gr i z As described in Sect 2 3 the Sloan filters are interference filter sandwiches Because of the manu facturing process and because the beam of the telescope crosses the filters under different angles in different parts of the image see Fig 2 the bandpass of the filter is slightly field dependent The laboratory measurements were designed to mimic the converging tilted beam from the telescope as a function of position in the focal plane Below we provide fitting formulae for the radial variation of central w
15. specially designed system of ribs will be installed in the new baffling mechanism The ribs will allow to remove most of parasitic light coming from the Moon bright stars and the sky A prototype was positively tested on two nights between June 14 15 and Jun 16 17 The final ribs system is expected to be installed by the end of 2014 OmegaCAM users are invited to follow the OmegaCam News page where updated information related to the new ribs installation will be posted 3 Observing with OmegaCAM To carry out observations with OmegaCAM users need to prepare Observing Blocks OBs before the observations and submit those to ESO using the Phase 2 Proposal Preparation P2PP tool More details on how to define OBs are given in Sect 3 3 and in RD5 First the offsetting modes and the effects of atmospheric refraction and dispersion are described 2http www eso org sci facilities paranal instruments omegacam news html OmegaCAM User Manual VST MAN OCM 23110 3110 version 95 0 23 24 95 24 95 24 85 24 85 24 65 24 65 8000 6000 4000 2000 0 2000 4000 6000 8000 0 075 0 025 0 075 8000 6000 4000 2000 0 2000 4000 6000 8000 8000 6000 4000 2000 0 2000 4000 6000 8000 xpos xpos Figure 14 Example of illumination correction derivation for OmegaCAM Sloan r Input data are 32 dither observations of SA113 The magnitude residuals after flatfielding are modeled with one 2D polynomial 32 ZPs as free paramet
16. to wind apply it may not be possible to observe any Landolt equatorial standard star field during an entire night In this case backup standard fields in the southern hemisphere are observed taken from the Stetson catalog Figure 25 shows schematically with what frequencies calibration data are typically taken During the first year of operations regular observations of extended Landolt fields are taken to provide a comprehensive set of secondary standards for the OmegaCAM filter system Shttp www3 cadc ccda hia iha nrc cnrc gc ca community STETSON standards OmegaCAM User Manual VST MAN OCM 23110 3110 version 95 0 42 Monitoring the Photometric Calibration 1 Week Requirement Applied filters Field 1 3Night Night Run Month 1 Year Years 5 6 2 Monitoring EU Polar sk ton x x E 5 63 Zeropoint w g r 7 wr Eq 1 elo k K Eq 2 l x a Eq 3 E ES K i 318 Eq 8 a d 5 64 Zeropoint V Uy Da Eq 1 a Eq 2 Sy Eq 3 Eq 8 5 65 Userkey w gr i Geh Eq n i V Dk D 542 Flat field vert Ge Dome K D V Va VU 54 7 Quick check wir Dome zeck OK gi Figure 25 The frequency with which the various types of calibration data are taken More details are given in the Calibration Plan RD2 The symbol U denotes a user band OmegaCAM User Manual VST MAN OCM 23110 3110
17. 1 L L L 5000 0 5000 am g Se se g E Xpos CRPIX1 pixel Xpos CRPIX1 pixel Ypos CRPIX2 pixel Figure 18 PSF anisotropy analysis plot obtained for an i band observation For the given example outside seeing as indicated by the DIMM around V band was 0 9 and IQ on OmegaCAM chips was 0 65 In general IQ variations of 10 2596 across the FOV can occur depending on outside seeing filter and telescope mirror position 3 6 Selecting exposure times and number of sub exposures ESO provides an Exposure Time Calculator ETC for OmegaCAM It may be used to estimate signal to noise count rates etc for OmegaCAM observations The sky background varies considerably with filter and with moon phase The ETC gives the sky count rates per pixel at various moon phases It is important to make sure that the read noise of the CCDs 5 7 electrons equivalent to the noise of a 30 electrons per pixel background does not add significantly to the sky background noise Exposure times should ideally be such that the sky background is above 250 electrons per pixel Particularly with the u and narrow band filters this requires minimum exposure times of several minutes particularly in dark time 3 7 Special features of the data OmegaCAM data as other wide field CCD imager data have several features that complicate data analysis Data reduction software can cope with these but it is important to be aware of them 3 7 1 Point Spread Function
18. 2 1 No offsets are done To allow removal of cosmic rays by taking the median of images N must be at least 3 2 OFFSET A sequence of N exposures with almost full flexibility of offsets Two limitations are to be considered 1 Large offsets beyond the default dither pattern Sect 3 4 2 increase the overheads due to the need of guide star reaquisition The execution time for offset sequences therefore increases by about 1 minute for each offset that goes beyond the default dither pattern P2PP takes care of this calculation This penalty will not apply if guiding is disabled in the aquisition which is possible for short integrations of 1 2 minutes 2 The maximum of a single offset accepted by the telescope is 1 degree in each of RA and DEC 3 DITHER A sequence of N exposures with offsets that are sufficiently large to bridge the gaps up to 80 between the CCDs in the detector mosaic A DITHER observation results in an image which covers the field continuously without large holes provided N gt 3 Some parts of the sky are seen by several CCDs and all parts of the field are seen in at least N 2 exposures A dither amplitude of 310 is recommended to cover the shadow cast by the cross of the segmented filters B V NB 659 See also Sect 2 3 4 JITTER A sequence of N exposures with offsets that are sufficiently large of order 1 arcsec to shift cosmetic CCD blemishes to different parts of the sky A JITTER observation results in a 3
19. 2 fragment image one fragment per CCD in which the gaps between CCDs remain but in which any part of the sky is seen by at most one CCD OmegaCAM User Manual VST MAN OCM 23110 3110 version 95 0 26 The pattern of offsets between successive exposures on a field distinguishes dither and jitter They represent different trade offs between trying to obtain data with uniform exposure level over the full OmegaCAM field and trying to avoid having areas on the sky where data from different CCDs need to be combined Such overlaps can give rise to discontinuities in the way the PSF varies across the field in the final stacked image but this is unavoidable if the full field is to be exposed The key differences between the two modes are summarised in Table 7 A number of patterns have been pre defined for dither and jitter They are described in more detail in the Template Manual RD1 E 3TTER DITHER Correct what CCD blemishes only also gaps in focal plane Offsets few arcsec 25 100 arcsec Data product fragmented 1 square degree image filled 1 square degree image Advantage no combination of data from different CCDs homogeneous exposure level Penalty gaps between CCDs possible sharp changes in PSF Table 7 Key difference between the dither and jitter modes Programmes that require very accurate PSF measurements should preferably use jitter as it results in a PSF which varies continuously over each CCD except near the very edges
20. 2011 05 25 144237 UTC 33 33 g Figure 8 Plot of the linearity of the response of each CCD as a function of illumination level Obtained by exposing the mosaic to a calibration lamp with varying exposure times OmegaCAM User Manual VST MAN OCM 23110 3110 version 95 0 13 Table 2 Properties of all 32 CCDs N at PA 0 degrees CCD number gain e ADU readnoise e E hot pixels i cold pixels dark current ADU hr CAM many of which follow from the properties of the detectors 2 2 2 Auxiliary CCDs In the periphery of the focal plane OmegaCAM contains four auxiliary CCDs Two of these are used for guiding called G1 and G2 both in position and position angle of the field The other two are mounted out of focus one is 2mm in front of the focal plane one 2mm behind Analysis of pairs of defocused images from these CCDs allows optical aberrations defocus coma astigmatism to be measured and the telescope to be kept in focus 2 3 Filters The filter exchange mechanism Fig 10 selects filters from two magazines located either side of the focal plane inside the instrument and slides these into the beam Each magazine can store six filters Filter positioning is repeatable to very high accuracy resulting in less than 0 396 rms flux variations between flat fields taken at different times The footprint of a stellar image is about 1cm in size The available filters are lis
21. 31 3 5 Survey Area Definition Tool 32 3 6 Selecting exposure times and number of sub exposures 33 3 7 Special features of the data 33 3 7 1 Point Spread Function 33 3 7 2 CCD blemishes and particle hits ss 34 3 7 3 Ghosts and reflections 34 3 54 Sky concentration 4 4e kk dr edem yy eR Rx e gd 35 3 7 5 Vignetting close to wind mast position eA 35 3 606 Diffraction Spikes ss x gue A A eV eg EA 38 ST Eringes 0 sk Ae hos eene PX bao dabat e iet a ag 38 3 60 REMANENCE xem xen bon nisi a uc ed Eh e Der tuu xs 38 3 1 9 CCD Crosstalk 4 5 o re ORE OE da atn 39 4 Calibrating and Reducing OmegaCAM data A1 4l Calibration Pl n b ecit uuo eR a e EUR E rur 41 4 2 Data reduction Dipen 43 4 3 Quality Controle 25 5 xe uou ese a RS RUNS a A D RR dece x e 44 A Filter throughput curves 45 A 1 Broad band Sloan filters UgG TEL ees 45 A 2 Segmented Broad band filters Johnson Band Vi 51 A 3 Narrow band filters Str mgren v and Hoi 51 OmegaCAM User Manual VST MAN OCM 23110 3110 version 95 0 A 4 Radial variations in bandpass ee B Detector response curves List of Tables 10 Basic CCD characteristics i a a a a e E A aE E a a ia Properties of all 32 CCDs er bea a ime eeu ae 4 Filters available in OmegaCAM ls Current OmegaCAM filter distribution The opaque filter is required to protect the instrument when it is not in operation e
22. AG and IA loops always stop about 10 seconds before the end of each science exposure Furthermore upon science CCD readout the OmegaCAM shutter is closed such that no light falls on the auxiliary CCDs either AG and IA are therefore always paused from 10s before the end of a science integration until the start of the next science integration 2 6 Calibration lamps Dome flat exposures are taken with calibration lamps mounted at 90 intervals on the top ring of the telescope illuminating a white screen on the inside of the telescope dome The lamps are operated by carefully controlling the electric current and their brightness is stable to about a percent over a timescale of several weeks Dome flat fields are used for monitoring the small scale pixel to pixel variations in sensitivity of OmegaCAM and for overall system throughput monitoring Two sets of four lamps each are mounted side by side One set is used for routine daily calibrations meanwhile the second set is cross calibrated against the first at regular intervals Once the lifetime of the first set of lamps has expired its role is taken over by the second set and the first set is replaced with new lamps This procedure ensures a continuous record of throughput measurements provides dome flat fields and serves as a daytime health check OmegaCAM User Manual VST MAN OCM 23110 3110 version 95 0 21 2 7 Photometric Properties Indicative magnitude zero points the AB magnitud
23. Dither data should be used if the primary aim is to image a field completely 3 1 2 Observing Strategies Observations are also taken under different Observing Strategies which are defined in the observing templates in P2PP Note that these strategy keywords are ignored in ESO s Data Flow System but may help the users in their own data reduction The strategy used is written to the image headers keyword TPL OBSSTRG Strategies extend beyond single OBs The defined strategies are STANDARD FREQ DEEP or MOSAIC They can relate to how fields are laid out during observation preparation and they can also provide specific instructions for data reduction and for scheduling 1 STANDARD Used for a stand alone OB with no particular relation to other observations 2 DEEP Observations intended to generate a deep image built up over several typically many OBs 3 FREQ Observations in which a field is monitored to form a time series Specific time links from minutes to months may be specified for such observations as part of the OB preparation in P2PP Specific time intervals can be specified for each OB complex scheduling constraints e g observe OB 1 five days after OB 2 can be specified in P2PP via time links 4 MOSAIC Observations which will be used to map an area of the sky consisting of several typically many adjacent pointings 3 2 Atmospheric Refraction and Dispersion The atmosphere refracts light as a result of wh
24. Ds the explanation lies either in the CCDs concerned themselves or some manufacturing problem of the detector head electronics There are no current plans to take remedial action as any intervention in the detector head entails a high risk The crosstalk has its origin in bright objects The threshold for triggering crosstalk depends on the CCD and exposure time Mostly it requires saturating the A D converter but also unsaturated sources can lead to electronic ghosts in other CCDs at a level of up to 0 4 of their parent These ghosts look like real sources and cannot be removed by dithering The crosstalk can be both positive bright and negative dark with a range of up to 0 496 it is generally weaker the larger the distance between source and target CCD is The ghost images always occur at the same relative pixel coordinates in the target CCD at which the star in the source CCD is located The crosstalk pattern present in any image consists of several sub patterns see Fig 24 The most common ones are i Negative crosstalk from 96 to 95 This is always present ii Positive crosstalk from 94 to 95 and on to 96 It seems that 95 may sometimes be skipped over iii Positive crosstalk from 96 to 95 to 94 and very rarely to 93 The OmegaCAM consortium is working on dedicated data reduction procedures to flag or remove these artefacts The effect of this crosstalk on the flat fields is still to be investigated in detail
25. ES EUROPEAN SOUTHERN OBSERVATORY Q Organisation Europ ene pour des Recherches Astronomiques dans l H misph re Austral Europ ische Organisation f r astronomische Forschung in der s dlichen Hemisphare ESO European Southern Observatory Karl Schwarzschild Str 2 D 85748 Garching bei M nchen Very Large Telescope Paranal Science Operations OmegaCAM User Manual Doc No VST MAN OCM 23110 3110 Issue 95 0 Date 27 08 2014 G Beccari S Mieske OmegaCAM consortium Prepared A NEE ee Date Signature A Kaufer hunc EE Date Signature C Dumas A Date Signature OmegaCAM User Manual VST MAN OCM 23110 3110 version 95 0 This page was intentionally left blank OmegaCAM User Manual VST MAN OCM 23110 3110 version 95 0 ii Document Change Record Issue Sect Paragr Reasons Remarks affected 1 0 31 03 2004 all first issue PAE 11 13 05 2004 most changes in response to first RIXes DBA RHA 2 0 draft 20 12 2006 small textual changes Appendix Laboratory measurements of filter throughput 2 5 26 06 2011 update based on commissioning 2 6 11 08 2011 2 3 updates on filter naming and characteristics 2 7 05 09 2011 many note 3 2 2 and 3 4 4 Update by ESO PSO after science verification 26 1 14 09 2011 Minor language updates by ESO USD 3 0 20 12 2011 most Update based on commissioning reports and start of survey operations Manual officially taken over by ESO 27 02 2012 Sects 3 4 1 amp 3
26. These OBs need not be specified as part of a science programme OMEGACAM img cal skyflat takes a sequence of five sky flats per filter with typically 2 3 filters that can be observed in evening twilight The skyflat template will gradually increase the exposure times during a sequence to ensure constant flux level for the five flats of order 20000 30000 counts Skyflats are not observed during morning twilight for operational reasons OMEGACAM img cal zp takes an exposure of a standard field through any filter using the two lens corrector Twice per night an equatorial standard field is observed for the five filters u g r i z see Table 10 The first execution is typically done in the evening twilight after skyflats The second execution is done in the middle of the night Furthermore when science observations in user bands are done B V v Narrow Band filters an equatorial standard field is also observed in these respective filters OMEGACAM img cal monit takes an exposure of a polar standard field see Table 10 through the composite u g r i calibration filter using the two lens corrector This is typically done three times per night end of evening twilight middle of the night end of the night 3 4 4 Overheads There are various overheads associated with OmegaCAM observations Telescope preset to a new target typically takes about 2 minutes slew speed better than 1 deg s An additional overhead of PA 2 seconds is added for Posit
27. a sciops 2p2 E2p2M WFI RD5 P2PP instructions see http www eso org sci observing phase2 P2PPSurveys html RD6 OmegaCAM public webpage see http www eso org sci facilities paranal instruments omegacam RD7 ESO QC web page see http www eso org observing dfo quality ALL OmegaCAM User Manual VST MAN OCM 23110 3110 version 95 0 2 1 5 Acronyms ADC Atmospheric dispersion compensator a device which corrects for the chromatic effects of the atmosphere at low elevations CCD Charge couple device a solid state panoramic light detector of high quantum efficiency CTE Charge Transfer Efficiency The efficiency of the charge transport during readout DFS Data Flow System DIMM Differential Image Motion Monitor e2v Manufacturer of the OmegaCAM CCDs ETC Exposure Time Calculator a software tool provided by ESO as an aid in planning observations FOV Field of view FWHM Full Width at Half Maximum IQ Image Quality MPG Max Planck Gesellschaft OB Observation Block a set of instructions to telescope and instrument to carry out a specific observation OmegaCAM The wide field CCD camera for the VST P2PP Phase 2 Proposal Preparation the process at ESO in which the users define the OBs for their approved programmes by means of special software PA Position Angle orientation of the field of view on the sky For VST OmegaCAM PA 0 means that North is in positive y direction and East is in negative x
28. ats at the same rotator angle as the flatfield used in determining the illumination correction The lhttp www eso org observing etc bin gen form INS NAME OMEGACAM INS MODE imaging OmegaCAM User Manual VST MAN OCM 23110 3110 version 95 0 22 Coeff xx Coeff xy Coeff yy DM 1 6153389e 09 1 9320844e 10 1 4650246e 09 oi 1 7019895e 09 3 3104022e 11 1 6704393e 09 r 1 5453463e 09 2 5500341e 11 1 3853747e 09 i 1 5270148e 09 1 0777857e 10 1 1392408e 09 Table 6 This table gives the polynomial coefficients for illumination correction functions as illustrated in Fig 15 x y is in pixel units with x y 0 0 at the mosaic center At distances of 10 000 pixels from mosaic center close to the edge of the FOV the illumination corrections is around 15 OmegaCAM consortium is therefore investigating alternative approaches e g using only domeflats which appear to have an amplitude of 5 or less in the amplitude of the sky concentration 2 9 1 Performances of the new Baffling system The installation of the new baffling system has been completed during P93 The new baffles are designed to reduce the amount of background brightness as well as the amount of stray light component in OmegaCAM images After the installation of the baffling a number of tests on sky have been performed in order to quantify the amount of background and stray light removed In particular the analysis of the ratios between raw dome flatfields taken before and after th
29. avarrete ESO Chile Figure 19 OmegaCAM IQ distribution collected between August and December 2011 December 2011 Median IQ on the detectors is 0 8 for i and 0 95 for g Given the internal IQ of 0 4 0 5 the measurements are consistent with a median Paranal seeing of 0 8 at 600nm 3 7 2 CCD blemishes and particle hits The OmegaCAM CCDs contain a number of cosmetic defects hot pixels cold pixels and traps see Table 1 The data on these pixels must be masked out before the data can be analysed if such gaps in the data are unacceptable the observations need to be dithered or at least jittered Hot pixels and traps can result in entire columns of data being lost so jitter dither offsets must not be parallel to the CCD columns Some dust particles are also visible on the detectors mostly these cover patches no more than about five pixels 1 arcsec in size At this point it is not clear whether how often the dust particles move around In addition particle hits on the CCDs release charge and obliterate small parts of the image If these are a concern their number grows with the exposure time then several exposures are needed 3 7 3 Ghosts and reflections Unwanted but at some level unavoidable reflections at air glass and air silicon interfaces result in some scattered light in the image Some of these reflections are reasonably well focused resulting in ghost reflections spurious images from bright sources appearing i
30. avelength blue edge and red edge of each filter as well as plots of the filter profile at the center of the filter Apart from the wavelength shift the shape of the filter response is rather constant with field position The filter response curves have all been multiplied by the average quantum efficiency of the OmegaCAM CCDs In the case of two filters D and r the bandpass variation is not quite axisymmetric about the physical centre of the filter For these a best fit center was determined the location of the center is reported in Sect A 4 OmegaCAM User Manual VST MAN OCM 23110 3110 version 95 0 46 u filter x1 Ch z 4 400 600 800 1000 SN SAA D N 386 384 382 380 378 332 330 328 326 324 322 0 50 100 150 0 50 100 150 Radius mm Radius mm Red Blue Figure 27 Top panels throughput of the u filter x average CCD response curve including a blow up by a factor of 1000 to show the out of band blocking Bottom left variation of central wavelength with radius on the focal plane dots measurements line polynomial fit Bottom right corresponding variation of wavelength at half maximum throughput of the blue and red edges of the filters The edge of the CCD array is reached at a radius of 180mm OmegaCAM User Manual VST MAN OCM 23110 3110 version 95 0 47 g filter x1 x1000 lt 472 E 470 0 50 100 150 0 50 100 150 Radius mm Radius mm Figure 28 Top pa
31. band The fringes are caused by the emission lines in the night sky spectrum and can be highly time variable As the equivalent width of these lines varies in the sky spectrum so do the fringe amplitudes and to some extent the patterns This additive effect is mapped by comparing twilight sky exposures with relatively weak sky emission lines with dark sky exposures Appropriately scaling this fringe map to individual exposures then allows the fringe pattern to be subtracted Fig 23 shows excerpts of a fringe map for the Sloan A and z bands with intensity contrasts in the image at 5 3 7 8 Remanence Exposure to a very bright star can saturate the CCD and can leave a surplus of charge on the surface of the detector This can take some time to diffuse away and meanwhile any subsequent readouts will show a faint residual of the saturated star Such remanence can take well over half an hour to decay away Note that remanence on your service mode data may indeed be caused by a previous image obtained for a different program Overall if naked eye stars as well as OmegaCAM observations were uniformly distributed in the sky every fourth to fifth field would contain one such star that could leave notable remanence OmegaCAM User Manual VST MAN OCM 23110 3110 version 95 0 39 3 7 9 CCD Crosstalk CCDs 93 96 are known to suffer some electronic crosstalk Since crosstalk has not been found down to very low levels in any of the other CC
32. ble but not simple to use private filters with OmegaCAM There are a number of constraints 1 The filters need to be mounted in special frames These frames grab the filters by the baseplate which needs to be approx 5mm thick Spare frames exist 2 Overall optical thickness should be equivalent to 15mm fused silica 3 The costs of these large filters are high 50 100 kEuro and manufacturing times long up to a year or more People interested in using private filters with OmegaCAM should contact the ESO User Support Department usd help eso org OmegaCAM User Manual VST MAN OCM 23110 3110 version 95 0 18 Figure 13 Picture of the partly disassembled shutter unit Bonn shutter The two blades as well as the central aperture are visible in the lower part of the unit 2 4 Shutter The shutter of OmegaCAM is located on top of the instrument in front of the filters When closed it covers all 36 CCDs science auxiliary It is a low acceleration twin blade photometric shutter of aperture 370x292mm The leading blade exposes the CCDs the trailing blade covers them again By ensuring that both blades travel with the same acceleration profile the exposure is kept homogeneous over the full beam For very short exposures below 0 8 s the trailing blade starts its motion before the leading blade has finished moving Over the science CCDs the blades move at constant speed Even for exposures as short as 0 1 seconds the homogeneity o
33. calibration file The OmegaCAM pipeline performs the following main steps using calibration files derived by the ESO Data Flow Operations Group in Garching 1 Each CCD frame is debiased 2 Bad pixels are flagged 3 Cosmic rays and satellite tracks are detected and flagged 4 Data are divided by a normalised flat field image 5 If appropriate fringes are removed 6 An astrometric solution is computed for each CCD based on a search of USNO stars in the image 7 Images are background subtracted corrected for sky concentration and resampled to a common linear world coordinate system 8 Different exposures in the same template are combined into one single image per CCD taking into accounts weights and bad pixel masks The USNO astrometric star catalogues contain over 400 million stars over the whole sky an average of 10 000 per square degree or 300 per OmegaCAM CCD down to 20th magnitude in R Typical astrometric uncertainty for each star is 0 3 rms More information on the catalogue can be found at http archive eso org skycat usno html OmegaCAM User Manual VST MAN OCM 23110 3110 version 95 0 44 It must be stressed that ESO does not provide any reduced science images Furthermore and like for VIRCAM the ESO version of the pipeline is mainly designed for reduction of calibration frames and it is not released to outsider users The user should contact the consortium for the pipeline science grade products 4 3 Q
34. direction PSF Point Spread Function RMS Root mean square SADT Survey Area Definition Tool SDSS Sloan Digital Sky Survey a large imaging and spectroscopy survey of mostly the Northern hemisphere OmegaCAM s science filter set include the five SDSS filters word z USM Universitatssternwarte M nchen USNO US Naval Observatory Their all sky catalogue is used for astrometric calibration of Omega CAM data VLT Very Large Telescope four 8m telescopes on ESO s Paranal Observatory VST VLT Survey Telescope a 2 6m telescope at ESO s Paranal Observatory dedicated to optical surveys and host to OmegaCAM WFI Wide Field Imager an 8 CCD mosaic camera on the ESO MPG 2 2 telescope on La Silla with similar pixel scale and sensitivity to OmegaCAM OmegaCAM User Manual VST MAN OCM 23110 3110 version 95 0 3 lt 1 5 Meter Instrument Attachment Flange gt Filter Storage Exposure shutter lle change Filter Storage mJ a e ZECIL 31035 BS Filter Optics MO Ort E EDO D H tee GA 0H 00 SSS EL o QJ 16k x 132 CCD chips r1 u W 0 0 D D oh Magazine A Detector Head Magagine B Eh 9 f YR oo uo J l 320 0 O Gg SC 1 ER Ek Cooling a x System aN LN2 Reservoir Figure 1 Schematic view of the instrument Key components are labelled 2 Instrument character
35. e coordinates in the focal plane for the default instrument rotator position angle of 0 X points East and Y points North The CCDs labelled IF and EF are the intra and extrafocal CCDs for image analysis these are mounted 2mm above and below the rest of the CCDs respectively G1 and G2 are guider CCDs
36. e new baffling installation shows that 696 to 896 of light distribution has been removed in all bands A confirmation is offered by the same analysis performed using the twilights flatfields Moreover the ratios of raw twilights flatfields are consistent with M2 chimney and M1 plug together removing foremost an oval circular stray light distribution but removing slightly less stray light near CCD edges Some residual stray light is still visible after the study of twilights flatfields and with test observations performed in the vicinity of bright stars and the Moon Fig 16 shows the ratio of raw r band dome flat divided by raw twilight flat taken at the same absolute rotator angles 137 degrees before and after the installation of the new baffling system left and right panel respectively It is visible that dome to twilight ratio images are consistent with the presence of residual stray light in the twilight flatfileds compared to dome flatfields Moreover the residual stray light component shows an irregular distribution that does not simply rotate with the change of rotator angle but changes morphology as well As a consequence it remains the case that the domeflats appear ideal to model pixel sensitivities of science images thanks to having 1 less stray light and 2 stray light in a rather simple circular distribution The use of twilight flatfields is instead recommended in order to proper model the stray light in science images A
37. e of a star that generates one electron per second are given in Table 5 These numbers are consistent with the on line Exposure Time Calculator ETC which is based on the throughputs and reflectivities of the individual optical components Because the angle of light incidence varies with distance from the center the OmegaCAM filters show some bandpass variation across the field This variation is mostly axisymmetric and it introduces a radially dependent colour term Measurements will be included once they are available Mag zero pt AB Table 5 Magnitude zero points for the SDSS and Johnson filter sets These are average values chip to chip scatter is of order 0 1 mag 2 8 Astrometric Properties By design OmegaCAM and the VST deliver a quite constant plate scale and a rather uniform PSF The radial plate scale is constant over the field to better than 1 part in 1000 and independent of the filter used The pixel scale with ADC in is 0 215 pix With the ADC out hence two lens corrector in it is 0 213 pix 2 9 Flat fielding of OmegaCAM data Illumination correction Both twilight and to lesser extent dome flats have a straylight component This straylight component is centrally concentrated sky concentration and makes the flatfield exposures overexposed in the center of the FOV Applying the flatfield exposures without any correction would result in an erroneous apparent trend of photometric ZP with distance fro
38. e time are specified No offsets are made OMEGACAM img obs jitter performs a jittered exposure see Sect 3 1 1 The number of exposures the exposure times and the step size for the jitter pattern can be specified A step size of 1 arcsec suffices to correct for CCD blemishes and most dust particles on the detectors If it is required that a given source is seen by a completely different set of pixels in each sub exposure the step size needs to be increased to above that of the sources This is particularly true if accurate real time fringe maps need to be derived from the observations themselves In those cases dithering may be more appropriate The jitter template allows step sizes up to 100 arcsec but such large offsets are not recommended OMEGACAM img obs dither performs a dithered exposure see Sect 3 1 1 The number of exposures the exposure times and the step size for the dither pattern can be specified The dither patterns are designed to make sure the full field is observed without gaps Two step sizes can be given one in X along CCD rows and one in Y along columns It is recommended that these are set to 25 and 85 arcsec respectively which are slighty larger than the largest gaps in the mosaic Exceptions are the segmented filters B V and NB 659 for which the step sizes should be 310 in both X and Y to dither out the shadowing by the filter cross Sect 2 3 Two dither patters are provided The diag pattern sim
39. een August and December 2011 34 Example ghost reflection 2 22 ee 36 Reflection from a star outside the field 37 Vignettine by wind Master da deas 37 Fringe maps in and band 38 Crosstalk between CCDs 93 96 ees 40 Frequency of calibration observations eA 42 Filter measurement location and coordinate system llle 45 u filter througlip tes cs Loud eh ae E ROO eX Re RR E po 46 g filter throughput 44 A Se amp ES E A PIE eias AT r filter throughput 22 525 odor eb deg Rp ERREUR RE A AE dei 48 1 filter throughput xou ee A NOROE UO Ee OX E Re poU EC edes 49 2 alter throughput L5 o eibi Seno edge ex Sue kin sechs ee 50 CCD Quantum efficiency CUIVES a 53 OmegaCAM User Manual VST MAN OCM 23110 3110 version 95 0 1 1 Introduction 1 1 Credits This manual is largely based on the OmegaCAM user manual drafted and maintained by the Omega CAM consortium Konrad Kuijken Bernard Muschielok Andrea Baruffolo between 2004 and 2011 until end of OmegaCAM commissioning It also includes content from the OmegaCAM DFS Com missioning Reports and input from the OmegaCAM IOT It will be updated continously by Paranal Science Operations as survey operations evolve and more experience is gathered 1 2 Scope OmegaCAM is the wide field imager for the Cassegrain focus of the VLT Survey Telescope VST on Paranal a 2 6m modified Richie Chretien alt az telescope designed specifically for wide field imaging
40. el is 15 u The readout seqence is illustrated in Fig 7 The science mosaic is read out by two FIERA systems one for each half East West of the array Total readout time is 29sec Combined with other overheads principally the time needed to wipe the array before a new exposure this leads to a total minimum shutter closed time between successive exposures of 40 seconds Generally the CCDs in the array are similar but they do differ in detail A number of properties of the CCDs are tabulated in Table 2 in a format that corresponds to their layout in the science array top view At the default sky position angle of zero degrees this layout corresponds to N at the top and East to the right Complete QE curves for each of the CCDs are given in Appendix B All CCDs but three have parallel and serial CTE above 0 999995 the exceptions being e CCD 65 and 96 which have serial CTE 0 999994 e CCD 72 which has serial CTE 0 999993 With a gain of ca 2 6 electrons per ADU the analog to digital conversion of the amplifier output of each CCD saturates before the CCD full well capacity of over 200 000 electrons Below this saturation the response of the CCDs is linear to within 196 as can be seen from Fig 8 The flat fields of the CCDs show different behaviour dependent on the wavelength In the u band the flat fields show a diagonal diamond like pattern which is caused by the thinning process A few of the CCDs 65 66 86 87 also s
41. entation of the mosaic of the pointing center in the top or bottom row of CCDs The reflections are linear narrow features perpendicular to the edge of the mosaic An example is shown in Fig 21 Stars outside the top and bottom edges also cause linear reflections but less frequently They require a star 0 56 0 565 degrees N or S of the pointing center These reflections are weaker and less focussed than the previous ones Their cause is also still under investigation 3 7 4 Sky concentration Multiple reflections also generate a very defocused image of the field which is added to the sky background in the image The effect usually referred to as sky concentration is a smooth additive component to each image The additive term is also present in flat fields be it from the sky or from an exposure of the dome screen If this additive term is not subtracted from the flat fields first the result of flat fielding is an image with a flat background but a varying photometric zero point Determining such a photometric flat field is best done by observing large standard star fields Note that one way to mitigate the effect is to calibrate each exposure CCD by CCD since the pattern is expected to be dominated by scales larger than an individual chip Sky concentration has been shown in commissioning to be for Sloan band observations up to 20 in twilight and 5 or less in dome flats See also Sect 2 9 3 7 5 Vignetting close to wind
42. ers SDSS DR7 is used as standard star catalog Top left magnitude residuals after flatfielding ZP per chip are not subtracted Top right model Bottom left External magnitude residuals after illumination model has been applied Bottom right model with uniform ZP that is with indivudal ZPs per chip subtracted The illumination correction reaches 0 2 mag at the edges of the FOV OmegaCAM User Manual VST MAN OCM 23110 3110 version 95 0 24 Figure 15 Raw r band dome flat divided by raw twilight flat for four different absolute rotator angles From top left to bottom right absolute rotator position angles are 107 6 18 3 163 1 and 71 2 degrees The asymmetric patterns has an amplitude of 5 896 OmegaCAM User Manual VST MAN OCM 23110 3110 version 95 0 25 Figure 16 Raw r band dome flat divided by raw twilight flat taken at the same absolute rotator angles before and after the installation of the new baffling system left and right panel respectively 3 1 Offsetting Modes and Observing Strategies 3 1 1 Offsetting Modes OmegaCAM observations can be taken with a number of Offsetting Modes STARE OFFSET DITHER or JITTER They differ in the kind of offsets that are used between exposures and in the final data product 1 STARE A sequence of N exposures of exactly the same part of the sky except for the effect of differential atmosperic refraction see Sect 3
43. full well is at least 2 times the pixel full well and the summing full well is at least 4 times the pixel full well These full well values Full Well Linearity correspond to the maximum charge that can be handled while attaining the CTE specified Amplifier full well is around 200 000 e Amplifier full well refers to the limit of linearity i e the maximum amount of charge that can be amplified while producing output that matches a linear response to within 1 lt 2 e pixel hour at operating temperature of 120 C There is cross talk at the level of 0 496 between CCDs 93 96 CCDs 87 and 88 shows day to day gain variations of 1 2 RMS Optical Coating Single layer anti reflection coating astro BB OES tees apama insten tm OOOO Package design Four side buttable fourth side with larger gap Invar package with integrated PT100 temperature sensor Table 1 Basic characteristics of the OmegaCAM CCDs See also Table 2 and Fig 8 OmegaCAM User Manual VST MAN OCM 23110 3110 version 95 0 10 Invar Silicon EF Pixels on Jed qc s sis EN Pixel Positions d Nj E E 00 AU d262 eevds ey Updated 21112011 GHe B i S D Z 77 66 6 si PA Yy 64 28 Invar zi A 64 18 Nom Silicon N b 4102x15mu 61 53mm 31 72 Invar 31 62 Nom Silicon Figure 6 The light sensitive area of a single CCD Units are in mm size of a single pix
44. gether with the P2PP for Surveys to generate automatically large numbers of OBs A large survey consists of many tiles or pointing centers Observations of any type stare jitter dither offset can be executed at a tile position The SADT provides the tiles and their central coordinates that may be used to cover any given area of sky Alternative ways of tiling the sky e g the Astro WISE PlateSystem developed for the KiDS survey may also be used but are not supported by the SADT 6For more information about SADT see http www eso org sci observing phase2 SMGuidelines SADT html OmegaCAM User Manual VST MAN OCM 23110 3110 version 95 0 33 SF Anisotropy for frames with DATE OBS 2011 09 03 02 29 07 filter Sloan 32 frame s target KIDS 339 0 31 Alt 62 385 Az 290 349 Rot 96 55557 Seeing 0 92 WindDir 200 0 WinsSp 2 69 Bars eratis ELLIPTICITY EBIN angle w r t x axis THETA Magnitude vs half light radius Radius vs Ellipticity T T T T T T papa T P bs M mn c Mem y w t T y 5000 MAG_ISO U U m S R T T T kamen d f Ellipticity d bl i 123 45 67 8 9 10 9 D 9 x9 9 oa 0 oP 95 FLUX_RADIUS pixel ZE g ge e Distance from optical axis pixel FWHM vs x position FWHM vs y position T T T T T T Ypos CRPIX2 pixel 5000 WS FWHM IMAGE arcsec FWHM IMAGE arcsec E TT 1 1 1 eI _ Blipticity 20 0 i 1 h
45. how an ink pattern areas of very high UV sensitivity the origin of this feature is unknown At intermediate wavelengths g r the flat fields are reasonably featureless At longer wavelengths i and particularly 2 fringing occurs leading to swirling patterns reminiscent of a thin film of oil on water All artifacts can be removed in data reduction The efficiency of the filters folded with the detector response is plotted in Fig 9 Detailed analysis of the filter throughput including bandpass variations across the fields is deferred to Appendix A See Sect 3 7 for a description of particular features that occur in scientific data obtained with Omega OmegaCAM User Manual VST MAN OCM 23110 3110 version 95 0 IO TELESCOPE AREA MEN _ Master FIERA Slave FIERA OmegaCAM Scientific CCDs Mosaic data transfer storage display readout time partial file from slave Instr WS partial file Linux fram master merging into final file done in parallel with 1 next exposure Figure 7 Data flow diagram for readout of the OmegaCAM science array Normalised flux ADU s OmegaCAM User Manual VST MAN OCM 23110 3110 version 95 0 12 3 1026 1015 1 630 1 005 D 0 945 ome 0 985 0 986 1 020 1015 1010 1 0 995 EE 0 185 H 1029 1015 1010 D ass ome KE o SES SEES CIS CGE effet Lee ESSE Exposure level ADU Created
46. ich sources appear slightly closer to the zenith than they would be without the atmosphere The effect is slightly wavelength dependent This has two consequences atmospheric refraction and atmospheric dispersion 3Note that even though the CCD mosaic is extremely flat significant jumps in PSF can still occur in coadded dithered images if the seeing varies from one exposure to the next OmegaCAM User Manual VST MAN OCM 23110 3110 version 95 0 27 3 2 1 Atmospheric Refraction A star s extra displacement r towards the zenith due to atmospheric refraction is given by r RtanZ rad 2 where Z is the zenith distance of the star and R is the refractive index of air minus 1 To a good approximation R larcmin 3 R is slightly wavelength dependent see Sect 3 2 2 Over the field of OmegaCAM r takes slightly different values from the center to the edge Z differs by Z 0 5 and so r varies by r Rsec Z Z c 0 5 sec Z 4 Thus the size of the field changes with zenith distance by an amount 26r 1 0 sec Z This is not a concern since the astrometric standard stars in the field suffer the same refraction and so this effect is taken out in standard data reduction However atmospheric refraction does affect long exposures at lower elevations If r changes signifi cantly during an exposure then it becomes impossible to track objects in the center and outer parts of the mosaic at the same time guiding with two stars w
47. ield see example spot diagrams in Fig 4 The field distortion is very low so that the image scale is virtually constant over the whole field There are narrow gaps between the CCDs the overall geometric filling factor of the array is 91 4 In addition to the 32 CCDs making up the science array OmegaCAM also contains four auxiliary CCDs around the edges of the field Two of these are used for autoguiding so that both field position OmegaCAM User Manual VST MAN OCM 23110 3110 version 95 0 4 OVERALL VIEW 1000 200 100 100 200 700 600 500 400 300 200 100 0 Figure 2 Optical layout and selected light paths through the VST The configuration with the two lens corrector is shown The bottom panel shows a zoom onto the corrector optics the incoming light passes through two spherical lenses a planar filter and the spherical dewar entrance window before reaching the CCDs on the right Scales are in mm OmegaCAM User Manual VST MAN OCM 23110 3110 version 95 0 5 10000 5000 p 5000 10000 Mosaic Layout for OMEGACAM configuration 0 Fiera 2 Fiera 1 96 95 94 93 92 91 90 89 32 131 30 29 16 15 14 13 Gl 88 87 86 85 84 83 82 81 EF 28 27 26 25 1 12 11 10 9 IF 80 79 78 77 76 75 74 73 G2 l24123122121 18 171 61 5 Fiera 3 Fiera 3 4 72 71 70 69 68 67 66 65
48. ill favour the center As a result image quality in the outer parts of the field will be degraded The way to prevent this is to keep the exposures short and so avoid a large change of r during an exposure For r not to change by more than 0 1 arcsec over an exposure requires 0 5 sec Z lt 0 1 or Z lt 6 cotZ cog Z 5 At Z 45 therefore Z must not change by more than 3 degrees per exposure for a target near the celestial equator this means no more than 12 min exposure at Z 60 the limit is below a degree change allowed 4 min exposure In poor seeing or if image quality constraints are not so stringent this limit can be relaxed and presents no practical concern Figure 17 shows a plot of the limit given by eq 5 3 2 2 Atmospheric Dispersion Because refraction is wavelength dependent more strongly so in blue light the atmosphere turns ob jects into little spectra with the blue end pointed at zenith Within the bandpass of a broad band filter this can cause image elongation and degradation A model for the amount of atmospheric dispersion on Paranal has been calculated by E Marchetti see http eso org gen fac pubs astclim lasilla diffrefr html Based on these data Table 8 shows the airmass at which atmospheric dispersion causes a smearing of the light over a 0 2 arcsec 1 pixel 0 5 and 1 arcsec length It is the responsibility of the user to specify the airmass constraints of their observati
49. imes lt 60s For requested seeing gt 1 2 and zenith distances gt 10 degrees up to 2 minute integrations without guiding are probably fine e Concatenate OBs of the same field if not already contained in a group or time link container e No IA is necessary in the aquisition of concatenated OBs during the first 30 minutes of a concatenation If possible avoid filter changes within an OB The overhead for filter change in a preset is zero since preset and filter exchange are done in parallel and preset takes normally longer than filter exchange In contrast filter change between templates costs time while since P92 this happens in parallel to the readout it still takes longer than readout and thus adds overhead e Avoid if possible sequences of large offsets beyond the default dither pattern with long inte gration times This will require reaquisition of guide stars at each offset P2PP will charge you one minute extra overhead for each large offset provided that AG is enabled in the aquisition 3 5 Survey Area Definition Tool To help users in planning how to cover their target survey area a specific tool has been developed the Survey Area Definition Tool SADT It is a mandatory tool for preparing survey observations with the Visible and Infrared Survey Telescope for Astronomy VISTA but it may also be used for preparing VST OmegaCAM survey observations The output of the SADT survey area file in XML format can be used to
50. in half degree steps and or rotating the instrument through multiples of 90 degrees the full field of view can be exposed in all four bandpasses Which CCD was exposed with which filter segment depends on the magazine the filter is loaded in this is recorded in the image FITS header see below Note on filter magazines The auxiliary CCDs used for guiding and image analysis are covered by separate filters Usually these have nearly the same bandpass as the main science filter but in some cases the narrow band filters and the u band filter they are different in order to boost sensitivity and increase the number of guide stars that can be used The Calibration filter is used to observe standard star fields in four bands simultaneously and hence to monitor variations in the atmospheric extinction throughout the night For this purpose a specially calibrated standard star field near the South celestial pole has been established see Sect 4 1 2 3 1 Note on filter magazines Because the filter frames can only mate with the filter insertion mechanism on one side the orientation of a filter with respect to the CCD mosaic depends on the magazine in which it was stored see Fig 12 During observations the filter exchange mechanism always returns a filter to the magazine and slot within that magazine it came from However it may happen that a filter needs to be moved manually from one magazine to the other during daytime operations This fli
51. in r mm ui 107 x107fr 10 59 Central 353 2 7 24 8 29 2 66 Blue edge 327 3 5 71 6 01 1 89 Red edge 383 4 9 15 11 27 3 70 g x10 r x10 7 x10 95 Central 472 5 28 72 37 43 11 91 Blue edge 413 0 22 29 29 10 9 30 Red edge 552 0 38 51 50 15 15 90 r SAO TA x107fr ALOT Spe Central 621 4 4 72 3 55 1 05 Blue edge 560 7 4 07 2 74 1 20 Red edge 696 9 5 42 4 57 0 81 Note Radius cf position 7 9 2 3 i x10 7r x10 7 x10 95 Central 7533 3 64 0 93 1 04 Blue edge 683 1 2 62 0 35 0 93 Red edge 839 5 5 29 2 14 1 33 Note Radius cf position 4 8 14 8 z x107 r xd10 5y 0759 Central 879 4 0 58 0 16 0 22 Blue edge 841 8 0 81 0 61 0 66 Red edge 920 6 0 52 0 04 0 16 Where the radius is referred to an offset centre its location is given in the x y coordinates shown in Fig 26 OmegaCAM User Manual B Detector response curves VST MAN OCM 23110 3110 GZ EF version 95 0 X 53 Figure 32 The QE curves for all CCDs in the mosaic laid out as the mosaic itself top view The solid curve in each panel is the measured QE for that device the dotted curve is the average for all 32 science CCDs The wavelength range from 320 to 1100nm is plotted dotted lines indicate 400 600 800 and 1000nm X and Y ar
52. ion Angle PA on sky different from 0 in the aquisition This overhead occurs because to reach any PA angle the telescope will first preset to the new position at PA 0 and then reposition the rotator with a speed of about 2 degrees per second PA 180 or PA gt 180 is interpreted as modulo 360 degrees For example PA 270 implies a motion to PA 90 and hence 45 seconds overhead Filter exchange in the instrument is done in parallel to the preset it takes between 65 and 115 seconds the precise value depends on whether a filter is being swapped for another one from the same magazine or whether they belong to different magazines Acquisition of guide stars takes about 1 minute A full TA sequence in the aquisition takes on average 3 minutes Successive bias frames can be read out every 40 sec which covers the time needed to wipe the array read it out and transfer it to disk ordering the data into a FITS file takes place in parallel to taking the subsequent exposure For concatenations an additional overhead applies if a PA different from 0 is chosen in subsequent OBs In a given concatenation the amount of this additional overhead equals TEL PRESET ROT CONCAT abs PAop lt n gt abs PAoBp lt n 1 gt Rotatorspeed 6 Rotator speed is about 2 deg s For concatenations where PA is equal for all OBs the associated overhead per OB is hence the PA angle in seconds PA 90 would imply an overhead of 90s per OB Table 9 collects the cu
53. istics 2 1 Overview OmegaCAM see Fig 1 is the wide field imager for the Cassegrain focus of the VLT Survey Telescope VST on Paranal a 2 6m modified Richie Chretien alt az telescope designed specifically for wide field imaging It is the only instrument on this telescope and is operated most nights In principle all observations are carried out in service mode The VST OmegaCAM system is designed for good seeing limited image quality over a wide field The telescope has an actively controlled meniscus primary mirror an active secondary and an image anal ysis system It contains two interchangeable correctors one is a high throughput two lens corrector which works in u z bands the other contains an Atmospheric Dispersion Compensator ADC for work at lower elevations The ADC has almost no throughput in the u band The optical layout of the VST is illustrated in Fig 2 The VST provides a 1 degree unvignetted field of view which OmegaCAM samples with a 32 CCD 16k x 16k detector mosaic Fig 3 at 0 21 arcsec per pixel 0 213 for the two lens corrector 0 215 for the ADC configuration Each CCD is of 2k x 4k format and subtends 7 3 x 14 6 on the sky The CCDs are thinned blue sensitive 3 edge buttable CCD44 82 devices from e2v of high but not perfect cosmetic quality Image quality is specified such that in the absence of seeing 80 of the energy from a point source should fall within a 2x2 pixel area over the full f
54. line polynomial fit Bottom right corresponding variation of wavelength at half maximum throughput of the blue and red edges of the filters Note that for this filter the bandpass variation is most symmetric about the an offset center see Sect A 4 OmegaCAM User Manual VST MAN OCM 23110 3110 version 95 0 50 z filter x1 Ex o o x 400 600 800 1000 Alam 884 882 d lt 880 T T 878 876 5 m 874 0 50 100 150 0 50 100 150 Radius mm Radius mm Figure 31 Top panels throughput of the z filter x average CCD response curve including a blow up by a factor of 1000 to show the out of band blocking Bottom left variation of central wavelength with radius on the focal plane dots measurements line polynomial fit Bottom right corresponding variation of wavelength at hal maximum throughput of the blue and red edges of the filters OmegaCAM User Manual VST MAN OCM 23110 3110 version 95 0 51 A 2 Segmented Broad band filters Johnson B and V See transmission curve data available from the Tools Section of the OmegaCAM webpage RD6 AA Narrow band filters Str mgren v and Ho See transmission curve data available from the Tools Section of the OmegaCAM webpage RD6 OmegaCAM User Manual VST MAN OCM 23110 3110 A 4 Radial variations in bandpass The radial variation of central and half power wavelengths expressed in nm can be described as the sum of the following polynomial terms
55. m FOV center The sky concentration hence needs to be corrected for after flatfielding the data The corresponding procedure is called Illumination Correction see e g Section 2 8 4 of McFarland et al 2011 arXiv 1110 2509 For OmegaCAM it consists of taking 32 dither observations of a photometric standard field such that the same group of standard stars is consecutively observed in each CCD These data is flatfielded with a flatfield that is a combination of the small scale structure in the domeflat and large scale structure of the twilight flat The ZP residuals after flatfielding are then modeled with a 2D polynomial and 32 ZPs as free parameters The relevant steps are illustrated in Fig 14 for the r band The illumination correction reaches 15 2096 at the very edges of the FOV for all Sloan filters Indicative illumination correction fitting coefficients as determined in the commissioning are shown in Table 6 for the w g r i filters An important note must be made in this context twilight flats have in addition to the rotationally symmetric straylight component an irregular contribution see Fig 15 These patterns occur most likely due to the incident angle criterion of the Fabry Perot throughput of those interference filters The non symmetric component is of order 5 896 of the average flux and as such pose a limitation to accurate flat fielding with sky flats Moreoever it requires that the science data are flatfielded using twilights fl
56. n different parts of the field In OmegaCAM most of these ghosts are very defocused and hence very diffuse but the most focused OmegaCAM User Manual VST MAN OCM 23110 3110 version 95 0 35 one projects to radii of a few mm scale is 14 3 mm and so is visible if there is a bright star in the field It arises as a reflection off the entrance surface of the dewar window followed by a reflection off the nearest filter surface Another ghost arises as a reflection off the CCDs followed by reflection at the entrance or exit surfaces of the dewar window These reflections largely disappear when the star falls on a gap between CCDs By way of example Fig 20 shows the four main reflections from a bright star on CCD 91 The associated table lists typical size and surface brightness relative to the bright star of these ghosts Note that the sizes and positions of the ghosts depend on the location of the parent star in the image and their brightnesses on the filter in use Ray tracing calculations predict that a bright star at distance R degrees from the center of the field will have a ghost that is displaced radially outwards by 21 R 70 pee 7 and that this ghost has a diameter of 23 73 Re 8 In some cases reflections from stars outside the field are seen Their origin is still under investigation Initial results are that these are caused by bright stars which lie between 0 5 and 0 53 degrees E or W in the default ori
57. nels throughput of the g filter x average CCD response curve including a blow up by a factor of 1000 to show the out of band blocking Bottom left variation of central wavelength with radius on the focal plane dots measurements line polynomial fit Bottom right corresponding variation of wavelength at hal maximum throughput of the blue and red edges of the filters OmegaCAM User Manual VST MAN OCM 23110 3110 version 95 0 48 r filter x o o x 622 wc y d lt 620 t 618 O 616 3 614 S 0 50 100 150 0 50 100 150 Radius mm Radius mm Figure 29 Top panels throughput of the r filter x average CCD response curve including a blow up by a factor of 1000 to show the out of band blocking Bottom left variation of central wavelength with radius on the focal plane dots measurements line polynomial fit Bottom right corresponding variation of wavelength at hal maximum throughput of the blue and red edges of the filters Note that for this filter the bandpass variation is most symmetric about the an offset center see Sect A A OmegaCAM User Manual VST MAN OCM 23110 3110 version 95 0 49 x1 x1000 0 50 100 150 0 50 100 150 Radius mm Radius mm Figure 30 Top panels throughput of the 7 filter x average CCD response curve including a blow up by a factor of 1000 to show the out of band blocking Bottom left variation of central wavelength with radius on the focal plane dots measurements
58. olumes from a single night are 50 100 GB of raw science data another 20 50 GB of raw calibration data are also taken every day OmegaCAM data are processed using a pipeline see RD3 for a full description which is run by the ESO Data Flow Operations group and is used for quality control purposes RD7 On Paranal a special version of the pipeline is running for real time QCO checks of image quality FWHM ellipticity IQ variation and sky transparency Only raw data is made available to the general user All data from OmegaCAM are delivered as multi extension FITS files The first extension contains the observing parameters and extensions 2 33 contain the data from the individual CCDs Note that the name of the CCD as identified in Fig 3 is included in the header of the corresponding FITS extension The order in which extensions appear in the FITS file may vary and should not be relied upon to identify CCDs 4 1 Calibration Plan The full details of the original calibration plan are given in a separate document RD2 We summarise the important features of its Paranal implementation here noting that all calibrations mentioned in the following are free of charge for the user Bias dark and flat field exposures are taken regularly on a daily to weekly basis Biases are taken daily dome flats every 3 5 days and darks weekly In clear conditions sky flats are taken every night in 2 3 filters One can hence expect each key band to have sk
59. ons as a function of the filter used and the image quality requirements To compensate the atmospheric dispersion effect an Atmospheric Dispersion Compensator ADC was designed for the VST It consists of two sets of counter rotating prisms with which one can introduce given amounts of dispersion and thus cancel the atmosperic dispersion The ADC has good throughput longward of 360nm wavelength unfortunately this does not include the u bandt Important Note The ADC has not been fully characterised and commissioned yet P90 Therefore it is not offered 4Note that the atmospheric extinction in u is high so observations in this band should in any case be taken as close to zenith as possible OmegaCAM User Manual VST MAN OCM 23110 3110 version 95 0 28 v e 80 KA on 60 40 20 Max exposure time minutes Maximum change in Z degrees o o 20 40 60 Zenith Distance Z degrees Figure 17 The maximum change AZ in zenith distance degrees that is allowed during an exposure before the image quality is degraded by 0 1 arcsec due to atmospheric refraction The right hand axis approximately translates this limit into an exposure time T minutes for a field on the equator as seen from the Paranal latitude of 24 T AAZ cos 24 Airmass at which dispersion reaches Filter 0 2 arcsec 0 5 arcsec 1 arcsec u g e i Table 8 The airmass sec Z at which the length of the
60. osures If an offset is commanded the auto guiding will simply keep the stars on the pixels where they are found in this new exposure i e the system does not calculate to which pixels the telescope offset should have sent the stars but rather assumes that the offset was applied correctly Telescope offsets that are sufficiently small to keep the guide star on its CCD are accurate to at worst 1 arcsec 2 5 3 Differential Guiding For observations where non siderial pointing is required e g to follow or search for solar system objects the telescope can be given a differential tracking rate It is currently not possible to guide in this mode either with the VST guide probe or with OmegaCAM Differential Guiding has not been fully commissioned and is not offered for the observing periods P88 and later OmegaCAM User Manual VST MAN OCM 23110 3110 version 95 0 20 2 5 4 Image Analysis with OmegaCAM The other two auxiliary CCDs are mounted out of focus see Figs 3 and 5 They intersect the stellar light beams in two planes one in front of and one behind the focal plane This gives information on the deviation of the beams from perfect cones with apex in the focal plane Different kinds of aberration result in different shapes for these defocused stars and this can be used to calculate corrections to be applied to the VST s active primary and secondary mirrors Acquisition of the two image analysis stars one per CCD is designed to happen
61. over the gaps between the CCDs they prevent reflection of light off the bright bond wires of each detector Slight vignetting is observed near these strips and on the long outer sides of the four corner CCDs OmegaCAM User Manual VST MAN OCM 23110 3110 version 95 0 9 CCD type Marconi e2v CCD 44 82 1 A57 PSP tin denies tam Format Active Pixel Area 2048 x 4102 pixels 2048x4100 used p 50 horizontal pre and 50 overscan pixels 2 e at 50 kpix sec R O N 4 e at 500 kpix sec not including system noise 6 e at 1000 kpix sec Full R O N including system noise at default read speed 5 7 e Gain e ADU 2 identical output amplifiers with integrated post chip amplifiers Output Amplifiers Charge can in principle be read through either one completely or through both of them simultaneously In practise only the respective left one is used Device Layout No frame transfer option 3 phase parallel register non split pue 3 phase serial register split Dump Drain A dump drain exists parallel to the serial readout register merece cd so that lines can be discarded as a whole A global defect budget exists specifying the overall defects Cosmetics for all of the 40 science devices Cosmetically the final device cosmetics are much better than devices used for the WFT gt 0 999995 per parallel or serial shift Horizontal Transfer Frequency 2 Mpixels sec Vertical Transfer Frequency 15 20 klines sec gt 200 000 e Serial
62. pixels surrounded by a border whose width is 0 16mm on the short side opposite the readout register 0 5mm along the long sides of the CCD and 5mm on the side of the readout register Unavoidably therefore there are small dead zone gaps in the science array between the light sensitive areas of neighbouring CCDs How the CCDs are mounted together in the array is indicated in Figs 3 and 5 The CCDs are mounted as closely as possible The resulting average gap sizes are e between the long sides of the CCDs 1 5mm 100 pix 21 5 e central gap along the short sides 0 82mm 55 pix 11 8 e wide gap along short sides 5 64mm 376 pix 80 5 e between science array and auxiliary CCDs 15 9mm 1060 pix All CCDs are read out at 280kpix s through one amplifier port each in all cases situated at the left end of the readout register The resulting image comes with 48 columns of pre and overscan and 100 rows of vertical overscan Because of their construction and the way they are arranged in the mosaic all CCDs are read out along their long side in a direction away from the center of the mosaic In the lower half of the CCD array the readout register is at the bottom southern edge of the CCDs whereas in the upper northern half the CCDs are rotated through 180 degrees OmegaCAM User Manual VST MAN OCM 23110 3110 version 95 0 8 A Figure 5 Picture of the 36 CCDs mounted behind the dewar window Note the black strips that c
63. ply offsets in X and Y by the step sizes specified resulting in a diagonal pattern of pointing centers It results in an optimally homogeneous coverage of the field by the mosaic for a given number of exposures The starext pattern is more complex It is designed to keep a small area of the field containing a bright star off the CCD array in order to reduce reflections The size of this box area can be specified recommended values are 8 arcsec x 8 arcsec Subject to that constraint this pattern yields the most homogeneous coverage of the field It works best if the number of exposures is a multiple of 5 The move to gap acquisition template should be used with this template The dither and jitter patterns are illustrated in the Template Manual RD1 OMEGACAM img obs offset allows full flexibility in the definition of offset patterns The specified number of exposures are taken and offsets are taken from a list Offset data are processed in the same way as OMEGACAM img obs dither data Large offsets beyond the default dither pattern increase the overhead due to the need to reaquire a guide star See Sect 3 1 1 and the Template Manual RD1 OmegaCAM User Manual VST MAN OCM 23110 3110 version 95 0 31 3 4 3 Nighttime Calibration A number of templates are provided for photometric zero point and atmospheric extinction measure ments Usually these will be used by Observatory staff executing the calibration plan RD2 see Sect 4 1
64. ps the orientation of the filter by 180 degrees OmegaCAM User Manual VST MAN OCM 23110 3110 version 95 0 16 H 50 100 150 200 250 300 350 400 4 Figure 11 The shadow cast by the cross that separates the different segments of a composite filter here the V filter Table 3 EXC in IEEE POPP name Ha VPHASY 658 6 659 3 Har 659 0 666 0 672 6 679 1 H_ALPHA Ho z 0 3 851 9 861 4 869 0 877 7 NB 852 861 869 878 NB3 453 6 494 3 533 5 575 6 NB 454 494 533 575 NB4 616 1 710 2 755 1 816 4 NB 617 710 755 817 ws wort Mens IF 06 40 Central wavelength and FWHM of filter throughput x mean CCD QE curve IF Interference Filter CG Coloured Glass filter M Monolithic S Segmented 4Q 4 quadrants with dif ferent passband This is a private filter for the VPHAS survey Note the slightly different central wavelength of one of the segments Ha in four slightly overlapping redshift ranges z 0 0 01 0 02 0 03 fThese are private filters for use by Munich University Observatory for the first five years of OmegaCAM operations Note that these are not currently mounted Because of the different manufacturing processes the Calib filter bandpasses differ in detail from the mono lithic filters For the filters consisting of different quadrants the layout of the quadrants on the sky is as shown below when the filter is inserted in magazine A for the default orientation of the instr
65. rrently known overhead values best estimates As of P92 some overhead reduction is achieved due to parallelising readout and preset filter setup Last readout of an OB is effectively 0 seconds and filter changes between templates are 30s faster First offset in a template is now charged with 15s instead of 20s Some guidelines to avoid unnecessary overheads OmegaCAM User Manual VST MAN OCM 23110 3110 version 95 0 32 Value seconds READOUT_TIME 40 Readout Data written to disk READOUT TIME LAST IMG 0 Readout for last image of an OB FILTER SETUP 65 or 115 Filter exchange between aquisition amp 1st science template Depending on filter position see text FILTER SETUP TPL 35 or 85 Filter exchange between science templates TEL PRESET 120 PA 2 2 minutes Preset plus rotator repositioning TEL PRESET CONCAT 30 x 1 5 x is distance to new target in degrees TEL PRESET ROT CONCAT 0 180 For PA 0 in a concatenation there is a rotator motion overhead of PA 1 deg s TEL_OFFSET 15 Telescope offset at the beginning of a science template INS_GUIDESTAR 60 Total time for GS at the OB beginning INS_IASTAR 180 Time for IA at the OB beginning INS_GUIDESTAR_NEW 60 Acquire new GS after offset INS_GUIDESTAR_RE 5 Automatically reacquire same guide star after offset PICK_TIME 45 Time for pick object for MoveToGap acquisition Table 9 The various overhead times associated with OmegaCAM observations e No guiding is necessary for integration t
66. scope and for its active optics system When required the probe can be used in between scientific exposures to measure the optical image quality by means of a Shack Hartmann analysis This information can then be used to configure the telescope s active optics system The probe can in principle also be used for autoguiding and image analysis during scientific exposures However unless a suitable guide star exists very close to the edge of the field the probe vignets the field of view Use of the OmegaCAM guide and image analysis sensors is thus the default way of operating OmegaCAM and VST It also avoids the need to move the VST probe in and out of the field which takes several minutes 2 5 2 Auto guiding with OmegaCAM Two guider CCDs are mounted diametrically opposite to each other near the edges of the field see Fig 3 Auto guiding is done simultaneously on two stars one on each guider CCD Guiding with two stars allows field center and rotation to be tracked during an exposure When the shutter closes the guider CCDs are not exposed to the sky and guiding cannot take place Acquisition of the two guide stars one per CCD is designed to happen automatically If after a readout a second exposure is taken of the same field the guide stars are re acquired once the shutter is reopened If there was no offset commanded for this new exposure then the auto guiding will attempt to return the guide stars to the same pixels as in the previous exp
67. tc bin gen form INS NAME 0MEGACAM INS MODE imaging OmegaCAM User Manual VST MAN OCM 23110 3110 version 95 0 30 Two acquisition templates are provided OMEGACAM_img_acq and OMEGACAM_img_acq_movetogap Each acquisition template allows specification of the object position instrument filter for the first exposure ADC or two lens corrector settings COORD and NONE respectively and Cassegrain rotator offset angle It is also possible to specify whether or not autoguiding is required Based on experience so far unguided tracking for 60 seconds does not degrade image quality measurably Finally for OBs in concatenation it is possible to disable the start of Image Analysis in the aquisition The difference between OMEGACAM img acq and OMEGACAM img acq movetogap is that the former ac quires the field blind while the latter takes a short exposure which the operator can display and use to move the telescope so that a selected bright star falls in a gap between CCDs in order to reduce reflections and remanence 3 4 2 Science Templates Science templates perform the actual scientific exposure For each of these templates the user must specify the exposure time and the filter to be employed and optionally the Observing Strategy used see Sect 3 1 2 The templates differ in the number and pattern of offsets that they use OMEGACAM img obs stare takes a number of exposures through the same filter The number of expo sures and the exposur
68. ted in Table 3 Transmission curves as measured in the lab are included in Appendix A The filters are sandwiches of glass substrates ca 14mm thick Each filter has a baseplate usually made of BK7 which provides the mechanical support and mounting to the filter frame Further layers of glass are glued onto this base plate and provide the filter bandpass either through coating or by means of coloured glass For the segmented filters B V and NB 659 these layers consist of four quadrants and the interface of the quadrants casts a slight shadow in the form of a cross onto the OmegaCAM User Manual VST MAN OCM 23110 3110 version 95 0 14 JAN 0 6 0 4 Throughput filter xCCD 0 2 A 400 600 800 1000 A nm Figure 9 Throughput of the OmegaCAM Sloan filters times the average quantum efficiency of the CCDs For comparison standard normalised Johnson Cousins UBVRI filters are also shown OmegaCAM User Manual VST MAN OCM 23110 3110 version 95 0 15 Figure 10 A single OmegaCAM filter ready to be loaded into the instrument Note the separate filters covering the auxiliary CCDs image plane Fig 11 This central vignetting cross covers 310 in X and Y direction In a science observation Sect 3 1 1 it can therefore be dithered out with a dither amplitude of 310 in both X and Y Most of the narrow band filters consist of quadrants where each quadrant has a different bandpass By offsetting the telescope
69. the image quality The CCD layout in the focal plane is shown in Fig 3 A photograph of the detector mosaic is shown in Fig 5 OmegaCAM contains a 12 slot filter exchange mechanism subdivided into two cabinets of six slots each Currently the available filters include the Sloan u g ri z set Johnson B and V filters several narrow band filter mosaics a Str mgren v filter and a special calibration filter see Table 3 OmegaCAM data are taken in the context of a calibration plan RD2 that ensures that all data can be photometrically and astrometrically calibrated to 0 05 magnitudes and 0 1 arcsec rms precision respectively See Sect 4 1 Compared to the wide field imager WFI RD4 on the ESO MPG La Silla 2 2m telescope Omega CAM on the VST offers 1 a four times larger field 2 better cosmetic quality CCDs 3 a primary mirror of 1 4 times the area 4 year round operation in service mode 5 better image quality due to active telescope optics astro climate 6 the Sloan filter set but only very few narrow band filters 2 2 Detector Mosaic 2 2 1 The 268 Million Pixel Science Array The central 32 CCDs of OmegaCAM form the science array the heart of the instrument Basic characteristics of the CCDs are found in Table 1 Consult also the OmegaCAM QC webpages RD for up to date information The CCD type is e2v CCD44 82 The packaging of a single CCD is illustrated in Fig 6 The light sensitive area consists of 4100x2048
70. uality Control ESO performs real time quality control during the observations using QC parameters derived by the data reduction pipeline The main parameters used in real time QC named QCO on Paranal are FWHM ellipticity of point sources in the science images and photometric ZPs derived from standard star exposures For each exposure and a representative subset of 8 of the 32 CCDs the above QC parameters are provided as pipeline output and queried with scripts by the nighttime operators Image Quality average is contrasted with the user requirements to determine whether an executed OB is within constraints Accepted tolerance is 10 in average FWHM wrt user requirements Also ellipticity and image quality variation across the FOV are monitored OBs whose images have more than 0 15 average ellipticity or more than 25 3096 in FWHM variation across the FOV are typically rescheduled All OmegaCAM data are transfered to Garching via USB disks In the medium term future it is envisaged to transport the data via a fiber link between Paranal and Antofagasta EVALSO to Europe such that the delay will only be a few hours at most Health checks of the calibration data are performed by the Quality Control Group in Garching to monitor the status of the instrument RD 7 A scoring scheme is applied to those QC parameters Also the Paranal pipeline provides health check parameters which can be scored to monitor the instrument status Since November 2011 no
71. uch features can occur on each of the four corner CCDs LII UT Em agi T d R Figure 22 Vignetting of the VST FOV by the Paranal wind mast Each image shows a normalised sky flat taken at a telescope altitude of 40 degrees and at azimuth angles between 157 and 167 degrees from left to right Normalisation is done wrt the mean of the leftmost and rightmost image Greyscale is from 10 of the mean flux The sequence crosses the position of the Paranal wind mast at around 162 degrees azimuth north north west Amplitude of the linear structures caused by the wind mast is several percent OmegaCAM User Manual VST MAN OCM 23110 3110 version 95 0 38 SA Rss ST Figure 23 excerpts of a fringe map for the Sloan d left and z band right Intensity contrasts in the image are at 5 The cuts show that fringe amplitudes are a few percent in 2 and up to 10 in z 3 7 6 Diffraction Spikes The spider that holds the secondary mirror in place causes diffraction spikes around bright stars Because the VST is an alt az telescope the camera rotates with respect to the spider during an exposure causing this spike pattern to rotate as well 3 7 7 Fringes Internal reflections in the CCD and the consequent interference give rise to fringes in the sky back ground These effects show up in the i and z bands and to a low extent in r and can have amplitudes up to 10 in z
72. ument with PA 0 North is up East to the left If the filter is mounted in magazine B the orientation is rotated by 180 degrees Ha VPHAS Ha z 0 3 NB3 NB4 Calib Q v 7 659 666 OmegaCAM User Manual VST MAN OCM 23110 3110 version 95 0 17 Table 4 Current OmegaCAM filter distribution The opaque filter is required to protect the instru ment when it is not in operation Manual filter manipulations should always be followed by the taking of a new set of flat fields but if for some reason old flats need to be used it is important to check that these were taken with the same filter orientation Which magazine a filter was loaded from is logged in the image headers keyword INS FILT1 NO magazine A contains filter positions 1 6 magazine B positions 7 12 Changing from one filter to the next is faster when the filter is in the opposite magazine since this allows to perform parallel motions of both filters inserting one removing the other one The typical time for filter exchange is one minute for filters between different magazines and two minutes for filters in the same magazine See also Table 9 which lists the OmegaCAM observing overheads The current order in which the filters are stored in the instrument is shown in Table 4 This order tries to anticipate the typical sequence of filter changes within OBs It is currently optimised for the sequence u g r i z 2 3 2 Providing Private Filters It is possi
73. ver the field is better than 196 However it is important to realize that exposures are not simultaneous there is a lag of about 0 5 second between the exposures at the leading and trailing edge of the field For high timing accuracy observations the delay time tdelay 0 39 0 00174X seconds 1 needs to be added to the absolute time of the exposure start and end Here X is in mm with respect to the center of the detector mosaic 1mm 14 3 arcsec and runs along the short sides of the CCDs in the direction of motion of the blades The signs of the header keywords DET SHUT TMCLOS and DET SHUT TMOPEN record from which direction the shutter was opened After the initialisation reset of the shutter it is always the same blade that covers the mosaic no matter what the previous state was This blade is defined as BLADE 1 and the blade outside the aperture as BLADE 2 BLADE 1 is at the x side of the instrument For the first exposure in a new sequence BLADE 1 will move outside the aperture and after the exposure time BLADE 2 will move into the aperture For the next exposure BLADE 2 will uncover the aperture and BLADE 1 moves in again after the exposure is over The sign of the header keywords are therefore related to the shutter motion in the following way Blade 1 Blade 2 implies a positive value of DET SHUT TM keywords Blade 2 Blade 1 implies a negative value of DET SHUT TM keywords Note that the shutter blades are quite far from focus
74. version 95 0 43 Field Name SA92 00 55 12 0 00 55 58 0 SA95 03 53 49 0 00 02 33 0 SA98 06 52 12 0 00 19 17 0 SA101 09 56 19 0 00 26 27 0 SA104 12 42 12 0 00 30 16 0 SA107 15 39 03 0 00 13 52 0 SA110 18 41 50 0 00 23 09 0 SA113 21 41 54 0 00 29 22 0 Polar Field 03 25 43 0 89 02 33 0 Table 10 OmegaCAM Photometric Standard Star Fields Astrometric calibration is performed by reference to the many USNO catalogue stars that are present on each CCD No special calibration observations are needed With the broadband old zf filters and exposure times between 10 seconds and 10 minutes the USNO stars should be unsaturated Short u band observations less than a minute may not provide sufficient stars for an accurate astrometric solution More specialised calibration files related e g to sky concentration or fringing from night sky lines are derived from the observations themselves or from less frequent dedicated measurements The calibration plan is specified to deliver photometric zero points accurate to 0 05 magnitudes and astrometry with rms error 0 1 arcsec in a fully automatic pipeline reduction mode 4 2 Data reduction Pipeline Data delivered to the ESO archive consist of the raw exposures as well as the calibration data The header of the image contains the nominal astrometric world coordinates they are not corrected for pointing errors etc The photometric zeropoint is provided as a separate
75. y flat every 3 days Master flat fields are derived from a combination of twilight and dome flats but note the impact of straylight gradients in sky flats Sect 2 9 Fringe maps are derived by comparing these flat fields to night sky exposures The photometric zero point is determined as follows The basic approach is that the atmospheric transparency and instrument efficiency are monitored several times per night in the five SDSS bands u g r i z the so called key bands using the two lens corrector whereas other filters so called user bands are cross calibrated against these All filters including u g r i z are considered as user bands when they are used in combination with the Atmospheric Dispersion Corrector ADC OmegaCAM employs a 4 quadrant calibration filter which is used to monitor the sky transparency and the instrument zero point in u g r bands simultaneously Three times a night a standard star field on the celestial pole is observed with this filter to measure atmospheric extinction In evening twilight and around midnight and for clear conditions a high elevation Landolt standard star field is also observed in the five bands u g r i z see Table 10 with the monolithic science filters from the SDSS system Finally if science observations in user bands are made an equatorial standard for the those particular bands is also observed close in time to the science observation When telescope pointing restrictions due

Very Large Telescope

Contents

Download Pdf Manuals

Related Search

Related Contents