Home

User`s manual

Contents

1. E File System PLACES sf_blanco 3 virtual Desktop ij Trash NETWORK gt Browse Network 1 item Free space 2 9 GB Figure 7 Access to a shared folder from the virtual machine GNU Linux systems Python distribution On debian based GNU Linux distributions such as Ubuntu or Linux Mint the following packages should be installed by using the visual package manager i e Ubuntu Software Center or Synaptic or from a terminal by executing apt get install package_name 8 python 2 7 5 or higher but not 3 x branch python numpy 1 9 2 or higher python scipy 0 16 0 or higher python astropy 1 0 3 or higher python matplotlib 1 4 3 or higher python statsmodel 0 6 1 or higher python pip 1 5 4 or higher cython 0 22 1 or higher gfortran 4 8 4 or higher lockfile 0 10 2 or higher build essential Shttp www ubuntu com ubuntu features ubuntu software centre http wiki debian org Synaptic 8 Administrator rights will be needed DRAFT VERSION ISPEC FRAMEWORK USER S MANUAL Optionally one extra package can be installed by using the python package installer PIP from a terminal only needed if we want iSpec to comunicate with TOPCAT or similars i pip install sampy Listing 1 Extra packages installation 2 22 iSpec Framework Once the system is ready for execution of python applications it is possible to proceed with iSpec installation by decompressing the e
2. 14Tnternational Astronomical Union 12 DRAFT VERSION ISPEC FRAMEWORK USER S MANUAL 3 1 4 Segments Segment region files should be plain text files with tab character as column delimiter Two columns should exists wave_base and wave_top the first line should contain those header names They indicate the beginning and end of each region one per line For instance wave_base wave_top 480 6000 480 6100 3 481 1570 481 1670 491 2240 491 2260 492 5800 492 5990 Listing 10 Fragment of a segments file The values in the column wave base should be always lower than wave top or iSpec will not be able to process the file 3 2 Execution The editor can be initiated by double clicking iSpec command or executing in a terminal located in iSpec s directory 1 iSpec command Listing 11 Command line execution Files Operations Parameters Spectra Help Spectra 1 0 0 8 0 6 x S T e arm o 0 2 0 0 2 0 4 0 6 0 8 1 0 wavelength nm AEIR amp e x 0 943278 y 0 79902 Action Stats Create Modify Remove Element Continuum Lines C Segments 0 Figure 11 Empty editor Once initiated it is possible to load spectra or region files through the File menu Alternatively they can also be specified from command line like for example 13 DRAFT VERSION ISPEC FRAMEWORK USER S MANUAL iSpec command continuum input LUMBA UVES_MRD_sun_cma
3. 3 26 Interoperability with other SAMP applications iSpec is a SAMP ready application and therefore it can send receive spectra to from other astronomical programs i e TOPCAT VOSpec splat running in the same machine To enable the interopability option it is necessary to run a SAMP hub in the system The easiest way to do so is to install and run TOPCAT it starts automatically a SAMP hub 1Snttp www star bris ac uk mbt topcat 16nttp waw sciops esa int index php project ESAVO amp page vospec http star www dur ac uk pdraper splat splat html 31 DRAFT VERSION ISPEC FRAMEWORK USER S MANUAL SAMP Hub File Clients Profiles pRegistration Public ID hub Profile Internal Metadata samp name Hub samp description text Clients org astrogrid samp hub Hub _ Figure 39 TOPCAT SAMP hub with different programs already connected From the menu Edit Send spectrum to the current active spectrum can be sent to any external program connected to the SAMP hub that supports spectra or data tables Send spectrum to extern Application topcat 9 Figure 40 iSpec Send spectrum to option TOPCAT Se alo Elise uE SM 8 Ce 00 Table List 2 tmpiGYNem xml sun s gz Current Table Properties Label tmpiGYNem xml Location SVE tmpiGYNem xml eno TOPCAT 2 Table Browser Name
4. www ivoa net 33 DRAFT VERSION ISPEC FRAMEWORK USER S MANUAL FLT Spectral synthesis and parameter determination In case that the optional functionality of synthetic spectrum generation has been activated compiled it is going to be possible to use SPECTRUM a Stellar Spectral Synthesis Program 3 to generate synthetic spectrum determine astrophysical parameters and calculate chemical element abundances 3 27 1 Synthetic spectrum generation Through Spectra Synthesize spectrum iSpec can generate synthetic spectrum The user can specify the different parameters and select a line list among different options e VALD line list extracted from the VALD database 14 February 2012 with a wavelength range from 300 to 11000 nm e GES linelist they are used in the Gaia ESO Survey GES It covers the wavelength range from 420 to 920 nm e Kurucz linelist covering from 300 to 1100 nm e NIST linelist covering from 300 to 1100 nm e SPECTRUM linelist 300nm 1100nm 3 February 2012 which contains atomic and molecular lines obtained mainly from the NIST Atomic Spectra Database 17 and Kurucz line lists Its wavelength range covers from 300 to 11000 nm On the other hand the solar abundance can be choosen too from different authors and publications e Asplund 2005 2009 e Grevese 1998 2007 e Anders 1989 Additionally iSpec incorporates several different grids of model atmospheres e MARCS GES APOGEE
5. E A A Un IE A yee ga Gaussian standard deviation sigma 0 0055 Gaussian base level mean continuum 1 0000 Cursor on wavelength 595 6058 and flux 0 7933 Figure 19 Continuum and lines fitted Finally the fitted lines can be removed by using the function Operations Clear Fitted lines 3 7 Automatic continuum regions It is possible to find continuum regions automatically by analyzing the spectra slice by slice The slice is selected as continuum if the following conditions are met e The region is at least as big as the size specified by the user e The standard deviation specified by the user is less than a given maximum e The median flux is above or below the continuum fit but not more than a given percentage Depending on the spectrum type the slice size and the standard deviation can be adjusted to find better results This functionality is located in the Operations Find continuum regions menu and can be applied over the whole spectra or only inside the defined segments In both cases it needs a previously fitted continuum After the computation it removes the current continuum regions if there are any and draws the ones that the process has found 19 DRAFT VERSION ISPEC FRAMEWORK USER S MANUAL Properties for finding continuum regions Check for regions of minimum size Maximum standard deviation 0 001 Maximum fitted continuum difference 1 0 Look for continuum regions in The whole
6. Radiation pressure Prad VTURB Microturbulent velocity in meters second e ATLAS9 Kurucz CastelliApogee Kirby models 19 plane parallel Effective temperature Teff 3500 3750 4000 4250 4500 4750 5000 5250 5500 5750 6000 6250 6500 6750 7000 7250 7500 7750 8000 8250 8500 8750 K Gravity Logg 0 0 0 5 1 0 1 5 2 0 2 5 3 0 3 5 4 0 4 5 5 0 dex Metallicity M H 5 00 4 50 4 00 3 50 3 00 2 50 2 00 1 50 1 00 0 50 0 30 0 20 0 10 0 00 0 10 0 20 0 30 0 50 1 00 dex Standard abundances no enhanced Apart from the astrophysical parameters the model atmospheres and linelists Spec ask for other parameters like the rotational velocity Vsin i micro macroturbulence resolution and wavelength range steps Synthetic spectrum generator Model atmosphere IMARCS GES E Solar abundances Grevesse 2007 Line list IGES_atom 475_685nm l Effective temperature K 5777 0 Surface gravity log g 4 44 Metallicity Fe H 0 0 Microturbulence velocity km s 1 0 Macroturbulence velocity km s 0 0 Rotation v sin i km s 2 0 Limb darkening coefficient 0 0 Resolution 300000 Wavelength min nm 470 0 Wavelength max nm 680 0 Wavelength step nm 0 001 Generate spectrum for Custom range defined above C Segments C Line masks OK Cancel Figure 44 Properties for synthetic spectrum generation It is recommended not to use t
7. e Telluric lines It can be used to identify the position of the telluric lines thus these regions can be ignored or to evaluate if a given spectrum has already been corrected by the barycentric velocity if not the output velocity will be zero Additionally in this process iSpec can estimate the resolving power of the instrument as described in section 3 17 e Template Any loaded spectrum or an internal synthetic one can be used for determining the relative radial velocity Velocity profile relative to atomic lines Velocity profile relative to telluric lines Velocity lower limit km s Velocity lower limit km s Velocity upper limit km s 200 velocity upper limit km s 100 Velocity steps km s 1 0 Velocity steps km s 05 Fitting model 2nd order polynomial a Fitting aa land order polynomial a y CCF in Fourier space T CCF E a Mask linelist Atlas Sun 372_926nm v Mal ale Demi 20 Mask size km s 2 0 Minimum depth o1 Minimum depth 0 01 OK Cancel OK Cancel Velocity profile relative to template Velocity lower limit km s 200 Velocity upper limit km s 200 Velocity steps km s 1 0 Fitting model 2nd order polynomial a CCF in Fourier space Cross correlate with Internal template 2 OK Cancel Figure 25 Velocity determination by using atomic telluric lines or a template The generation of the velocity profile is done by an implementation of the cross match correlation
8. e determine_tellurics_shift_with_mask e determine_tellurics_shift_with_template 38 DRAFT VERSION ISPEC FRAMEWORK degrade_resolution smooth_spectrum resample_spectrum coadd_spectra merge_spectra normalize_spectrum_using_continuum_regions normalize_spectrum_in_segments normalize_whole_spectrum_strategy2 normalize_whole_spectrum_strategy1 normalize_whole_spectrum_strategy1_ignoring_prefixed_strong_lines filter_cosmic_rays find_continuum_regions find_continuum_regions_in_segments find _linemasks fit_lines_and_determine_ew calculate_barycentric_velocity estimate_snr_from_flux estimate_snr_from_err estimate_errors_from_snr clean_spectrum clean_telluric_regions adjust_line_masks create_segments_around_linemasks synthesize_spectrum add_noise_to_spectrum generate_new_random_realizations_from_spectrum precompute_synthetic_grid determine_astrophysical_parameters_using_synth_spectra USER S MANUAL determine_astrophysical_parameters_using_synth_spectra_and_precomputed_grid determine_abundances_using_synth_spectra determine_astrophysical_parameters_from_ew determine_abundances_from_ew calculate_theoretical_ew_and_depth paralelize_code 39 DRAFT VERSION ISPEC FRAMEWORK USER S MANUAL e estimate_vmic_from_empirical_relation e estimate_vmac_from_empirical_relation e generate_and_plot_YY_isochrone The easiest way to use the scr
9. 33 0 03 0 0008 0 03 0 0008 0 0259 Line 517 42 517 41 517 43 0 01 0 0009 0 01 0 0010 0 0228 Line 518 30 518 29 518 32 0 02 0 0007 0 02 0 0007 0 0253 Line 518 40 518 39 518 40 0 02 0 0039 0 02 0 0039 0 0442 Line 518 44 518 43 518 44 0 04 0 0074 0 04 0 0074 0 0636 Line 518 54 518 51 518 55 0 01 0 0003 0 01 0 0003 0 0160 teff logg MH vmic vmac vsini Limb R Solution 6174 09 4 08 0 43 1 76 13 52 2 00 0 00 70000 Errors 245 50 0 38 0 16 0 44 1 68 0 00 0 00 0 Calculation time 0 0 53 4 DOF niter nsynthesis chisq tanh rchisq tanh chisq rchisq rms Stats 12795 7 45 15 23 0 0012 16 31 0 0013 0 0357 Return code 2 Figure 46 Output with the determined astrophysical parameters 3 21 3 Abundance determination Abundance determination can be done by synthetic spectra fitting as exposed in section 3 27 2 but alternatively iSpec can measure equivalent widths and use SPECTRUM a Stellar Spectral Synthesis Program 3 to derive the chemical element abundances when the temperature gravity metallicity and microturbulence is already known Abundances determination Modelatmosphere MARCS y Effective temperature K 5777 0 Gravity log g 4 44 Metallicity M H 0 02 Microturbulence velocity km s 2 0 Figure 47 Properties for the chemical abundance determination The following steps are needed for the determination of abundances e Load a spectrum and the line masks to be used in the analysis e Fit the continuum see section 37 DR
10. Consider Custom range defined above Segments Line masks Replace by Completely remove El OK Cancel Figure 33 Wavelength range reduction properties S21 Apply mathematical expression The wavelength fluxes and error values of the active spectrum can be modified by applying a mathe matical expression which can contain any combination of the following functions e sin x Trigonometric sine element wise e cos x Cosine elementwise e tan x Compute tangent element wise e arcsin x Inverse sine element wise e arccos x Trigonometric inverse cosine element wise e arctan x Trigonometric inverse tangent element wise e arctan2 x1 x2 Element wise arc tangent of x1 x2 choosing the quadrant correctly e sinh x Hyperbolic sine element wise e cosh x Hyperbolic cosine element wise e tanh x Compute hyperbolic tangent element wise e arcsinh x Inverse hyperbolic sine elementwise e arccosh x Inverse hyperbolic cosine elementwise e arctanh x Inverse hyperbolic tangent elementwise e around a decimals out Evenly round to the given number of decimals e floor x Return the floor of the input element wise e ceil x Return the ceiling of the input element wise e exp x Calculate the exponential of all elements in the input array e log x Natural logarithm element wise e logl10 x Return the base 10 logarithm of the input array element wise e log2 x Base 2 logarithm of x 28 DRAFT VERS
11. Correlation A Guide for Self Improvement aj 134 1843 1848 November 2007 F Kupka M L Dubernet and VAMDC Collaboration Vamdc as a Resource for Atomic and Molecular Data and the New Release of Vald Baltic Astronomy 20 503 510 2011 J L Bertaux R Lallement S Ferron C Boone and R Bodichon TAPAS a web based service of atmospheric transmission computation for astronomy ArXiv e prints November 2013 S Zucker Cross correlation and maximum likelihood analysis a new approach to combining cross correlation functions mnras 342 1291 1298 July 2003 17 Y Ralchenko NIST atomic spectra database Memorie della Societa Astronomica Italiana Supplementi 8 96 2005 49 DRAFT VERSION ISPEC FRAMEWORK USER S MANUAL 18 B Gustafsson B Edvardsson K Eriksson U G J rgensen A Nordlund and B Plez A grid of MARCS model atmospheres for late type stars Methods and general properties aap 486 951 970 August 2008 19 R L Kurucz ATLAS12 SYNTHE ATLAS9 WIDTHS9 et cetera Memorie della Societa Astro nomica Italiana Supplementi 8 14 2005 20 J A Valenti and N Piskunov Spectroscopy made easy A new tool for fitting observations with synthetic spectra aaps 118 595 603 September 1996 50 DRAFT VERSION
12. Rupprecht G Lo Curto G Avila W Benz J L Bertaux X Bonfils T Dall H Dekker B Delabre W Eckert M Fleury A Gilliotte D Gojak J C Guzman D Kohler J L Lizon A Longinotti C Lovis D Megevand L Pasquini J Reyes J P Sivan D Sosnowska R Soto S Udry A van Kesteren L Weber and U Weilenmann Setting New Standards with HARPS The Messenger 114 20 24 December 2003 J Moultaka S llovaisky P Prugniel and C Soubiran ELODIE SOPHIE Spectroscopic archive In F Combes D Barret T Contini F Meynadier amp L Pagani editor SF2A 2004 Semaine de l Astrophysique Francaise page 547 December 2004 E W Greisen and M R Calabretta Representations of world coordinates in FITS aap 395 1061 1075 December 2002 E W Greisen M R Calabretta F G Valdes and S L Allen Representations of spectral coordinates in FITS aap 446 747 771 February 2006 P Grosbol R H Harten E W Greisen and D C Wells Generalized extensions and blocking factors for FITS aaps 73 359 364 June 1988 P Stumpff Two Self Consistent FORTRAN Subroutines for the Computation of the Earth s Motion aaps 41 1 June 1980 F Pepe M Mayor F Galland D Naef D Queloz N C Santos S Udry and M Burnet The CORALIE survey for southern extra solar planets VII Two short period Saturnian companions to HD 108147 and HD 168746 aap 388 632 638 June 2002 C Allende Prieto Velocities from Cross
13. and line masks The user is responsible for not creating incoherent overlapping regions the program does not perform this kind of validations It is highly recommended to work with nanometer units instead of arm strongs since some operations in iSpec expect this The corresponding files should respect the format exposed in this section 11 DRAFT VERSION ISPEC FRAMEWORK USER S MANUAL 3 1 1 Spectra Spectra should be in plain text files with tab character as column delimiter Three columns should exists wavelength flux and error although in case the error is unknown it can be set all to zero The first line should contain the header names waveobs flux and err such as in the following example waveobs flux err 370 000000000 1 26095742505 1 53596736433 gt 370 001897436 1 22468868618 1 55692475754 370 003794872 1 18323884263 1 47304952231 5 370 005692308 1 16766911881 1 49393329036 Listing 7 Fragment of a spectrum file For all the iSpec operation the error propagation is taken into account To save space the file can be compressed in gzip format Spec can read FITS with the format following the standards of the IAU 8 9 where the spectral coordinates wavelengths are specified in the header via CRVAL1 and CDELT1 keywords The fluxes and errors should be stored respectively in the primary data unit and in an image extension 10 as 1D arrays In the same way the spectrum can be saved i
14. can also be executed from the terminal by invoking Ispec in any directory It is also possible to share a folder from your real computer to the virtual one thus files can be easily accessed from it To activate that option go to Settings Shared folders and add a shared folder by clicking on the plus sign It is important to mark the Auto mount option to have an easier access 5 DRAFT VERSION ISPEC FRAMEWORK USER S MANUAL from the virtual machine 60 Oracle VM VirtualBox Manager ud VY Details acti Gi Snapshots New Start Discard oan tebe General Preview Powered Off Name Ubuntu Operating System Ubuntu 32 bit System Base Memory 1024 MB iSpec_Xubuntu Figure 5 Access to the virtual machine settings iSpec_Xubuntu Shared Folders a H aha Aa 600 Add Share General System Folder Path 4y Users blanco Folders List Folder Name blanco Read onl 2 __ Make Permanent 3 Name Path Machine Fole Transient Fe Figure 6 Add a shared folder between the real and virtual machine The shared folder can be accessed from the virtual machine by double clicking the folder media on the Desktop 6 DRAFT VERSION ISPEC FRAMEWORK USER S MANUAL Zee 2 2 1 iSpec_Xubuntu Running Gi media File Manage om Ty 4 al media File Manager A File Edit View Go Help gt A amp E home virtual Desktop media
15. chosen by the user and if the difference between the theoretical and measured wavelength peak is smaller than a given limit the information will be linked to line region The user can choose also to freely fit the measured wavelength peak instead of using the line mark linked to each line region iSpec can also do an additional verification calculating second derivatives from the observed data to improve the absorption lines limits although generally it is not needed Additionally iSpec will cross match the lines with a telluric line list to dentify the element that produces each absorption line it will be written in the line note and if it may be affected by a telluric line indicated with a symbol in the line note It is worth noting that if a region can not be fitted it will be removed 18 DRAFT VERSION ISPEC FRAMEWORK USER S MANUAL The information related to the fit and the cross match can be visualized if the Stats action is selected and a line region is clicked iSpec lines not saved Files Operations Parameters Spectra Help A narval_sun_norm s gz Tere rrrererrerre gt MEPERETRERELE LETTE 2 CER AER E A AAA 8 E AA 595 2 595 3 5954 5955 595 6 595 7 595 8 wavelength nm A Q O A Bi a x 595 606 y 0 793266 Action Stats Create Modify Remove Element Continuum Lines C Segments 0 Flux standard deviation 0 142118 cote i I E E as o emg amplitude A
16. list of chemical abundances a a 5 8 Flag abundances to be filtered a a a 59 Plot abundances 1 6 4 6 64 4585 em ee teld dood aae a HH ee A ey GO as a eRe Pe RE eA Ee eee eG 5 11 Visualizing the results gt o a o u c s ccs aa k derea rrara rka ra 5 12 Assessment aoaaa 6 Bug reporting 38 40 40 41 42 42 43 43 44 44 45 45 46 47 48 ISPEC FRAMEWORK USER S MANUAL 1 Introduction iSpec is an open source framework for spectral analysis 1 It is suitable for the creation of spectral libraries such as the Gaia FGK Benchmark Stars library 2 and the determination of astrophysical parameters such as effective temperature surface gravity metallicity and individual abundances iSpec works in conjunction with SPECTRUM a Stellar Spectral Synthesis Program 3 for the generation of synthetic spectra and the derivation of abundances from equivalent widths The framework can be used from python scripts in an automatic way by using its API but it also includes a visual interface that is SAMP ready and it can interoperate with other astronomical applications such as TOPCAT VOSpec and splat facilitating a indirect way to access the Virtual Observatory iSpec can be downloaded from http www blancocuaresma com s It Is distributed under the terms of the GNU Affero General Public License open source license except the SPECTRUM code which is also included in the framework thanks to R O Gray Th
17. next to the wavelength step field in order to have a point of reference The interpolation method can be linear bessel 4 points are considered at each interpolated point the result tends to be smoother than the linear interpolation or spline it might help to smooth the 30 DRAFT VERSION ISPEC FRAMEWORK USER S MANUAL spectrum 3 25 Spectra combination All the open spectra can be combine through Edit Combine all spectra by finding the mean median substracting active spectrum minus the rest co adding or dividing active spectrum divided by the rest the flux values It is recommended that before doing so all the spectra should be corrected for its radial barycentric velocity and they should have the same resolution Resample amp combine spectrum Base wavelength Top wavelength 683 26 Wavelength step 0 001 Operation Median Mean C Subtract C Add C Divide Cancel Figure 38 Spectra combination properties The process does the following operations 1 It builds a common wavelength coordinates which will depend on the ranges and wavelength step specified by the user 2 It homogenizes all the spectra by interpolating the flux at each point of the common wavelength coordinates It is worth noting that the visualized spectra is not replaced by the homogenized version they are only used internally by the combine function 3 The homogenized spectra is combined and the result is displayed
18. spectra C Only inside segments OK Cancel Figure 20 Properties for the automatic mechanism of finding continuum regions 3 8 Automatic line masks iSpec can find line regions automatically by applying the following steps Properties for Finding line masks Minimum depth of the continuum Maximum depth of the continuum 1 0 Select elements comma separated Fel Fe 2 Resolution 80000 Velocity respect to telluric lines km s 0 0 Discard affected by tellurics Check derivatives before fitting Line list VALD_atom 300_1 100nm y Maximum atomic wavelength difference 0 005 Look for line masks in The whole spectra C Only inside segments OK Cancel Figure 21 Properties for the automatic mechanism of finding line masks e Search local maximum and minimum points of a spectrum that has been smooth by using two times the resolution e Select those line candidates that have a minimum depth 1 0 represents 100 of depth with respect to the continuum e Fit the line candidates with a Gaussian model and discard those with bad fit e Cross match the remaining lines with an atomic line list considering the velocity specified by the user e Select the lines that correspond to the elements specified by the user comma separated or blank to avoid this filter e Cross match again with a telluric line list considering the velocity specified by the user e Discard lines that may be affected by tellu
19. step recommended to be half of the mask size This implies that the distance between the elements in the mask is variable in wavelength but constant in velocity depends on the mask size specified by the user line list 1 Build the mask with a Mask size Avelocity 5 given size in lt gt terms of Resampled VEICA mask Step Avelocity a Resampled avec 7 2 Resample A A A Spectrum AS gt uniformly in terms of Spectrum rie 7 velocity Figure 26 Mask for cross correlation iSpec permits to choose the mask size in velocity and the minimum depth The following formula is used for determining the wavelength ranges 23 DRAFT VERSION ISPEC FRAMEWORK USER S MANUAL i velocity M41 A As 1 eee 2 14 ve oct y After the mask construction the spectrum is re sampled to have the same points as the mask and the cross correlation algorithm can be easily applied by shifting the mask values to the left right each shift represents a constant increment decrement in velocity This approach allows to reduce the computation time needed for calculating the velocity respect to the atomic telluric mask The results are presented in a new window where the velocity profile is shown The mean velocity is calculated by fitting a second order polynomial near the peak and additionally a Gaussian Voigt is fitted with fixed mean velocity to determine other complementary paramete
20. tmpiGYNem xml Rows 81 879 Dx Columns 3 _ Table Browser for 2 tmpiGYNem xml Sort Order 4 m waveobs flux err 470 0 0616 470 05775 470 05361 470 04756 470 03673 470 03244 470 03763 470 04794 470 0543 470 05702 470 06091 470 06109 470 06141 470 0598 470 05853 0 061 Row Subset All 4 SAMP Messages O OnN OM l WN e O0000000000o0oososSs 0 0 950 500 550 wavelength n A rRegistration HOO eB Public ID hub p E Profile Internal Action Stats Create _ Modify Remove Elements Property Maliin Metadata samp name Hub Subscriptions arose Figure 41 TOPCAT with data sent from iSpec Cursor on wavelength 653 5767 and flux 0 0565 32 DRAFT VERSION ISPEC FRAMEWORK Starlink SPLA USER S MANUAL T VO A Spectral Analysis Tool BIO RRC Ia ELIS EAE IGS Global list of spectra narval_sun s gz Short name Full name 4 s 5 gt Properties of current spectra A narval_sun s gz narval_sun s gz file localhost var folders zc f79yb4h1l4cv3qdpd11sss5tC TABLE Coordinates Data Errors waveobs flux Jl Save Reset Starlink SPLAT VO lt plot0 gt Format Columns Colour 600 R Bajaj Displaying Wavelength 501 255 X scale 1 0 w A G Fa 4 t Gal ell VA
21. AFT VERSION ISPEC FRAMEWORK e Fit the lines see section USER S MANUAL A gaussian voigt profile will be fitted to determine the equivalent width central wavelength etc By using the fitted paramters a cross match process with the internal atomic data will be performed to determine the elements and other fundamental information for each line e Determine abundances with fitted lines It is necessary to know the astrophysical parameters of the star Teff log g metallicity and microturbulence Chemical abundances Abundances abundance dex 3 959 500 550 600 650 700 wavelength nm sy OB Property Value Fe 1 abundance 0 00224 Fe 1 median abundance in log N Ntot dex 4 65 Fe 1 mean ahuindance inloa N Ntot dex 4 66 Recalculate again Figure 48 Chemical abundance determination 4 Automatic usage The iSpec s functions described in this document and some additional ones that cannot be accessed through the visual interface can be called from a python script An example script named exam ple py is included into the Spec distribution files The example includes a list of functions for different generally useful actions e read_write_spectrum e convert_air_to_vacuum e plot e cut_spectrum_from_range e cut_spectrum_from_segments e determine_radial_velocity_with_mask e determine_radial_velocity with_template e correct_radial_velocity
22. E REE Ee HE 21 3 12 Velocity determination and correction a a a 21 3 13 Spectroscopic binaries identification a a a e e 25 3 14 Signal to Noise Ratio SNR estimation o o 20000000 25 3 15 Errors estimation a 26 1 DRAFT VERSION LO AMOS 2 cae bh bob ES eR be eee Gh A 3 17 Resolving power estimation e a a e a a a 3 18 Resolution degradation o oo aa e a 3 19 Continuum normalization a s sss cis saky ra Ra ee we 3 20 Wavelength range reduction e a a e a 3 21 Apply mathematical expression oaa a a a a a 3 22 Fluxes and errors cleaning 3 23 Clean telluric regions aa a 3 24 Spectrum resampling ooa a a a a 3 25 Spectra combination oaoa a bn e a ERED ODER HERE RE ES 3 26 Interoperability with other SAMP applications 0 4 3 27 Spectral synthesis and parameter determination 2 048 3 27 1 Synthetic spectrum generation ooo e ee 3 27 2 Parameters determination a a a a 3 27 3 Abundance determination 0 0 a a a a a a a Automatic usage Pipeline 5 1 Generate a small grid of synthetic spectra 2 ee 5 2 Preprocessing cee ee eR Ree eee eb ERAS owe eRe oS 5 3 Generate list of pre processed spectra 1 e 5 4 Determine atmospheric parameters and normalize 204 5 5 Generate list of atmospheric parameters a a a a a a 5 6 Determine chemical abundances a 5 7 Generate
23. EWORK USER S MANUAL 3 17 Resolving power estimation iSpec can try to estimate the resolving power of the instrument that has observed the spectrum based on the FWHM of the telluric lines This estimation can be found in the velocity determination function relative to telluric lines section 3 12 and it is obtained by the following equation C Ra l FW H Meetturic FW H Mtneoretical where c is the speed of light in the vacuum km s FW H M etluric km s correspond to the telluric lines observed in the spectrum and FW H Mtneoretical km s to the theoretical telluric lines The resolving power is also estimated when iSpec determines the velocity profile relative to the atomic data however this is a bad estimator since it uses directly the FWHM measured and there are several factors that contribute to the broadening of the absorption lines i e star s rotation 3 18 Resolution degradation The active spectrum resolution can be degraded to a lower one by selecting the menu Operations Degrade resolution and selecting the original and target resolution If the user has previously generated a velocity profile function sections 3 12 and 3 17 the current resolution has been already estimated and it will be shown in the degradation dialog although the user can modify it Degrade spectrum resolution If initial resolution is set to zero spectrum will be just smoothed by a gaussian determined by the final resolution F
24. ION ISPEC FRAMEWORK USER S MANUAL e sqrt x Return the positive square root of an array element wise e absolute x Compute the absolute values elementwise e add x1 x2 Add arguments element wise e multiply x1 x2 Multiply arguments element wise e divide x1 x2 Divide arguments element wise e power x1 x2 First array elements raised to powers from second array element wise e subtract x1 x2 Subtract arguments element wise e mod x1 x2 Return element wise remainder of division Apply mathematical expres Wavelengths Fluxes flux Errors err OK Cancel Figure 34 Apply a mathematical expression to the active spectrum The functionality can be found in Operations Apply mathematical expression and the current values can be refered as waveobs flux and err Some examples of the utility of this functionality e Convert the wavelengths from Armstrong to nanometers by dividing the wavelengths by 10 Waves waveobs 10 e Change the scale of the fluxes Fluxes power flux 2 e Modify the errors by using a percentage relative to the flux Errors flux 0 05 e Assign a constant value to the error values Errors err 0 0 1 0 The result of the mathematical operation should always be an array with the same number of elements as the original spectrum 3 22 Fluxes and errors cleaning In order to filter out bad measurements the current active spectrum can be cleaned by se
25. T A 17 AB narval_sun s gz Y limits automatic _ V hair log phot flux 0 087155 _ log Track free v Y scale 1 0 Action Stats J Property Cursor on wavelength 4 Phot flux versus Wavelength 500 520 540 560 580 Wavelength nm 600 640 660 680 Figure 42 Splat with data sent from iSpec VOSpec VOSpec File Edit View Operations Plastic SAMP Help elslelele x al a ale E micr Flux Unit VOSpec Spectral Analysis Tool yy w T T T T T T 7 RedShift 000 21 E E 80 De reddening S 6 0 22V 0 00 s Y axis error M 2 X axis error Y Graphic Mode i L E Pa t m 4 80 5 40 0 5 80 6 00 6 40 80 Wavelength micron logarithmic x10 x at ae Y Y Y Y Y Y AY YY O He in lla la 18 ln Yo Ol x ol x ol x ol x ol x ol x ky Spectra List gt M Local Data View RETRIEVE UnmarkAll Reset e f e Figure 43 VOSpec with data sent from iSpec It is also possible to send spectra from external programs to iSpec in this case it is required that the data is structured in three columns in the following order wavelengths fluxes and errors It is highly recommended to name the columns with the following labels waveobs flux err By using this functionality iSpec can receive data from the Virtual Observatory in a indirect way through external programs such as TOPCAT VOSpec and splat 1Shttp
26. VORITES E All My Files m AirDrop J Applications iSpec_Xubuntu vbo Spec_Xubuntu vdi Logs E blanco x x prev E Desktop a Documents O Downloads Figure 2 Open iSpec Xubuntu vbox with VirtualBox to add it to the list of virtual machines The new virtual machine will be listed in VirtualBox now it can be run by selecting it and clicking Start 4 DRAFT VERSION ISPEC FRAMEWORK USER S MANUAL 600 Oracle VM VirtualBox Manager qe TE gt VY Details _ Details Snapshots New Settings Start Discard ee tebe General Preview Powered Off Name Ubuntu Operating System Ubuntu 32 bit System Base Memory 1024 MB iSpec_Xu buntu Figure 3 Start the virtual machine By default the virtual machine will show the Xubuntu Desktop with two icons that allow to execute iSpec normal execution and test Double clicking the test will launch iSpec with the example spectra iSpec_Xubuntu Running qm ty y A Tras Files Operations Parameters Spectra Help fe fe An 550 600 wavelength nm 4 O O Gis Action Stats Create Modify Remove Element Continuum Lines Cursor on wavelength 459 0653 and flux 0 1941 a y PR Ole 7 Figure 4 Virtual machine running iSpec with a solar spectrum Spec is installed on home virtual shared iSpec in the virtual machine check section 2 2 2 for more details about the contents and the application
27. WHM wavelengths final resolution Initial resolution 25813 Final resolution 12906 OK Cancel Figure 32 Resolution degradation properties For each flux value the process will 1 Define a window based on the FWHM size which depends on the original and target resolution 2 Build a gaussian using the sigma value and the wavelength values of the spectra window 3 Convolve the spectra window with the gaussian and save the convolved value In case that the user specifies an initial resolution of zero the spectrum will be just smoothed by a gaussian with variable FWHM determined by each wavelength and the final resolution wavelength FWHM 7 final_resolution 3 19 Continuum normalization The continuum normalization operation can be found in Operations Continuum normalization and it divides all the fluxes of the active spectrum by the fitted continuum Errors in continuum placement will be propagated in this operation if the continuum was fitted taken into account the errors from the spectrum 27 DRAFT VERSION ISPEC FRAMEWORK USER S MANUAL 3 20 Wavelength range reduction The current active spectrum can be cut by selecting Operations Wavelength range reduction and specifying a base and top wavelength or using directly the defined segments line regions The user can also choose to replace the removed fluxed by zeros Wavelength range reduction Base wavelength Top wavelength 680 0
28. _AP txt Generate Grid Joutput processed Pre processing output abundances Determine APs output list_abundances_differential txt Determine abundances Figure 55 Generate list of abundances Input in dark blue output in light blue output list_abundances_reference txt output list_abundances_summary_ID txt 5 8 Flag abundances to be filtered Some of the atmospheric parameters and abundances might not be of good quality i e too hot cold stars outliers stars not members of their cluster etc thus this step tries to identify those abundances that should be discarded There are several parameters in setup py that are going to control how the discarding process is going to be done python 007 _identify_abundances_to_filter py output list_abundances_summary_ID txt output list_abundances_summary_ID_flag_F txt Listing 21 Execution AA DRAFT VERSION ISPEC FRAMEWORK USER S MANUAL Pre processing Determine APs output list_abundances_summary_1 D txt Generate Grid Determine abundances Figure 56 Flag Input in dark blue output in light blue 5 9 Plot abundances Taking into account the previous results several plots are generated in this step in order to visually inspect the results python 009_plot_abundances py output list_abundances_summary_ID_flag_F txt output plots abundances Listing 22 Execution output list_abundances_summary_ D_
29. algorithm 12 which sum up the spectrum s fluxes f multipled by a template function p C v Y_ 5 v pia v flux piz 1 lines pix 1 Calculate C v where p pix v is varied from a lower to an upper velocity in fixed steps 2 Normalize C v The p function represents the fraction of the line of a template spectrum which depends on the spectral type of the star that falls on a given pixel at a given velocity The cross correlation can be computed in Fourier space taking advantage of the correlation theorem 13 although when the spectrum has a large wavelength the computation can take more time compare to the normal cross correlation Spec includes an internal template it can be found in the directory input spectra synthetic which corresponds to a synthetic spectrum of a star with Teff 5777 0 gravity logg 4 44 and metallicity 0 02 solar type which has been generated by using SPECTRUM 3 atomic data extracted from the VALD database 14 350 to 11000 nm and MARCS model atmospheres In case that another ground base observed spectrum is used as template it is recommended to clean the regions that may be affected by telluric lines as explained in the section 3 23 In the case of selecting the option of using atomic lines a mask is used instead of a template which means that the p function has values only on the peak of the line and the value corresponds to the depth of the line The possible masks line list
30. ars correspond all to the Sun The rest of stars default are marked as not reference In both cases there will be a maximum of 6 iterations to determine the parameters teff 1 logg 71 MH l ymic ymac e The regions used for the determination of atmospheric parameters and abundances are specified in the file e Different parameters related to how stars abundances are going to be filtered are also present Generate Grid Custom read_spectrum function Telluric mask Zeropoint template RV Wavelength ranges Resolution Determine APs Model atmospheres Atomic linelist Solar abundance Configuration for AP determination Reference spectra differential analysis Regions for AP determination Regions for abundance determination Filtering options Pre processing Determine abundances Figure 58 Setup 5 11 Visualizing the results To validate the results it is possible to manually inspect the results with iSpec The spectra used for the determination of the atmospheric parameters and the final solution synthetic spectra can be easily visualized by executing the following commands python visualize_ap py output list txt output processed M67_No1054 515 525 python visualize_ap py output list txt output processed M67_No1054 515 525 Fe 1 gt python visualize_ap py output list txt output processed M67_No1054 515 525 None Listing 23 Execution The commands
31. ative to atomic lines IA rel intensity s 2 80 Sigma km s 4 93 7 Velocity lower limit km s 200 0 Velocity upper limit km s 200 0 Velocity steps km s 1 0 Fitting model 2nd order polynomial a y CCF in Fourier space Mask linelist Atlas Sun 372_926nm x gt Mask size km s 2 0 al B0 7 9 80 Minimum depth 0 1 Velocity km s OK Cancel Figure 28 Cross match correlation determined by ELODIE and iSpec for HD005516A spectro scopic binary Spec incorporates into the input spectra binaries directory the spectrum of HD005516A observed with ELODIE 7 e Date 4 10 1996 RA 00 57 12 40 DEC 23 25 03 54 e ELODIE RV component 1 25 49 km s RV component 2 4 72 km s e iSpec results with parameters by default Barycentric vel 5 15 km s RV component 1 30 16 km s RV component 2 0 04 km s RV corrected component 1 25 01 km s RV corrected component 2 5 11 km s iSpec will try to automatically detect outliers peaks in the velocity profile in order to detect spectroscopic binaries and fit more than one Gaussian Voigt 3 14 Signal to Noise Ratio SNR estimation Spec provides two slightly different approaches to estimate the spectrum s SNR 25 DRAFT VERSION ISPEC FRAMEWORK USER S MANUAL e From errors The SNR is calculated by using the flux divided by the reported errors in the spectrum This is the best way to calculate the SNR if the er
32. ces txt python 007 _identify_abundances_to_filter py output list_abundances_summary_ID txt output list_abundances_summary_ID_flag_F txt python 009_plot_abundances py output list_abundances_summary_ID_flag_F txt output plots abundances Listing 13 Complete list of commands to execute The commands that end with a number are the ones that can be executed in parallel For instance if we want to execute one of them using 16 cores then it is sufficient to change the last number to 16 The key files that affect the behaviour of the pipeline and that the user needs to adapt to his her needs are the initial list of input spectra input spectra_list txt and the configuration file setup py 5 1 Generate a small grid of synthetic spectra The first step is going to compute a grid of synthetic spectra that is going to help to accelerate the determination of atmospheric parameters This step should be done just once For instance it is not necessary to re run it in case we add new input spectra to analyze python 000_generate_grid py input grid_ranges txt output grid 1 Listing 14 Execution AO DRAFT VERSION ISPEC FRAMEWORK USER S MANUAL input grid_ranges txt Teff logg MH Generate Grid Pre processing 3500 1 0 2 0 3500 1 0 1 0 4500 1 0 1 0 Determine APs Determine abundances Figure 49 Generate grid Input in dark blue output in light blue 5 2 Pre processing The pre processin
33. click and Left click to Lines region and its empty space drag modifies remove the Segments statistical and a new the left edge region information will region will Right click and be visible in the appear If it is drag modifies bottom part of a line region it the right edge the window will ask for an optional note Line marks Left click on a Left click on a Left click on a line region to line region to line region to add a note modify the peak remove the mark note Right click on a line region to add modify a note Table 2 Combination of actions and elements It is important to remark that Zoom Pan mode should be disabled in order to be able to execute the above actions On the other hand the user is responsible for not creating incoherent overlapping regions the program does not perform this kind of validations 3 5 Continuum fitting The continuum of the star can be fitted by going to the menu Operations Fit continuum 16 DRAFT VERSION ISPEC FRAMEWORK USER S MANUAL Properties for fitting continuum Suggested number of splines based on the wavelength range 199 Number of splines Degree 2 Resolution 0 0 Filtering order median max Wavelength step for median selection 0 01 Wavelength step for max selection 1 Use spectrum s errors as weights for the fitting process Y Automatically find and ignore strong lines Strong line probability threshold 0 5 Consider only continuum regi
34. ct resolution of the input spectra so that the convolution to a homogeneous resolution is done properly e f the average signal to noise ratio is known it is recommended to specify it Otherwise just set it to zero and it will be estimated e If the spectra is already normalized or the telluric lines are already removed or do not exist space observations it should be specified Y for yes N for no 1 python 001_preprocessing py input spectra_list txt input spectra output preprocessed 1 Listing 15 Execution 41 DRAFT VERSION ISPEC FRAMEWORK USER S MANUAL Generate Grid input spectra_list txt None or cluster name Pre processing Star Star name Origin ee Instrument Operation none coadd merge ID Unique identifier Filename Input filename Determine APs R Resolution Snr SNR or 0 Generate list Normalized Y N tellurics_removed Y N input spectra Output filenames based on ID Files separate in directories based on origin Figure 50 Pre processing Input in dark blue output in light blue Determine abundances 5 3 Generate list of pre processed spectra After the pre processing is completed a list should be automatically generated with all the information needed for the next steps Optionally we can fill the file input clusters txt with the clusters names ages radial velocities and metallicities This information can be us
35. ditor s file Spec_v20150728 tar gz the version number may vary with time 1 tar zxvf iSpec_v20150728 tar gz 2 cd iSpec_v20150728 Listing 2 iSpec installation OPTIONAL In case we want to have the functionality of generating synthetic spectrum and deter mining astrophysical parameters an additional step should be executed in order to compile the needed libraries 1 make Listing 3 Compilation of SPECTRUM a Stellar Spectral Synthesis Program 3 Finally it is possible to test iSpec by double clicking test command or executing test command in a terminal and selecting a star by entering its number 600 G blanco macpblanco obs u bordeaux1 fr bash 80x24 pr Last login Fri Sep 21 14 12 89 on ttys0B2 blanco macpblanco Users blanco Downloads sve test command exit Resolving power per instrument Narval 78 000 98 000 ESPaDOnS 68 000 81 000 HARPS 115 888 1 Procyon Metal Rich Dwarf Teff 6545 logg 3 99 M H 0 02 HARPS 2 Sun Metal Rich Dwarf Teff 5777 logg 4 44 M H 0 00 Narwal 3 Mu Cas A Metal Poor Dwarf Teff 5308 logg 4 41 MZH 0 89 Narwal 4 Arcturus Metal Poor Giant Teff 4247 logg 1 59 M H 0 54 Narwal 5 Mu Leo Metal Rich Giant Teff 4433 logg 2 50 M H 0 29 ESPaDOnS 6 ALL of them simultaneously 1 to 5 7 Sun with example region masks B Exit 7 Figure 8 Test execution script which allows to view 5 different sta
36. e gravity 6 Bug reporting Bugs can be reported to the author by providing all the information needed to reproduce it 1 Python script or detailed step list if the bug is related to the visual interface 2 Data used 3 Output printed on the terminal 48 DRAFT VERSION ISPEC FRAMEWORK USER S MANUAL References 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 S Blanco Cuaresma C Soubiran U Heiter and P Jofr Determining stellar atmospheric pa rameters and chemical abundances of FGK stars with iSpec aap 569 A111 September 2014 S Blanco Cuaresma C Soubiran P Jofr and U Heiter The Gaia FGK benchmark stars High resolution spectral library aap 566 A98 June 2014 R O Gray and C J Corbally The calibration of MK spectral classes using spectral synthesis 1 The effective temperature calibration of dwarf stars aj 107 742 746 February 1994 M Auri re Stellar Polarimetry with NARVAL In J Arnaud and N Meunier editors EAS Publications Series volume 9 of EAS Publications Series page 105 2003 J F Donati ESPaDOnS An Echelle SpectroPolarimetric Device for the Observation of Stars at CFHT In J Trujillo Bueno and J Sanchez Almeida editors Astronomical Society of the Pacific Conference Series volume 307 of Astronomical Society of the Pacific Conference Series page 41 2003 M Mayor F Pepe D Queloz F Bouchy G
37. e latest SPECTRUM version can also be obtained from http www appstate edu grayro spectrum spectrum html This document presents the installation steps for the Spec framework together with a general overview of the functionalities that can be accessed through the visual interface and from python scripts Zi Installation 2 1 Virtual Machine The fastest way to experiment with Spec on any platform Mac Windows Linux and Solaris is to use a ready to use the virtual machine with Spec and all its dependencies already included i e python packages and compilers To run it the only necessary step is to install the sofware for running virtual machines called VirtualBox free software http www star bris ac uk mbt topcat 2http www sciops esa int index php project ESAVO page vospec 3http star www dur ac uk pdraper splat splat html 4nttp www ivoa net Shttps www gnu org licenses agp1 3 0 html 3 DRAFT VERSION ISPEC FRAMEWORK CAN About Screenshots Downloads Documentation End user docs ve Technical docs Contribute Community USER S MANUAL 2 https www virtualbox org wiki Downloads search Login Preferences Download VirtualBox Here you will find links to VirtualBox binaries and its source code VirtualBox binaries By downloading you agree to the terms and conditions of the respective license VirtualBox platform packages The binaries are released under th
38. e terms of the GPL VirtualBox 4 3 10 for Windows hosts gt x86 amd64 J 1 VirtualBox 4 3 10 for OS X hosts gt x86 amd64 VirtualBox 4 3 10 for Linux hosts VirtualBox 4 3 10 for Solaris hosts x86 amd64 VirtualBox 4 3 10 Oracle VM VirtualBox Extension da LETRA 2 Support for USB 2 0 devices VirtualBox RDP and PXE boot for Intel cards See this chapter from the User Manual for an introduction to this Extension Pack The Extension Pack binaries are released under the VirtualBox Personal Use and Evaluation License PUEL Please install the extension pack with the same version as your installed version of VirtualBox If you are using VirtualBox 4 2 24 please download the extension pack gt here If you are using VirtualBox 4 1 32 please download the extension pack gt here If you are using VirtualBox 4 0 24 please download the extension pack gt here Figure 1 Oracle VirtualBox website download section 2014 03 31 1 Go to the Download section of http www virtualbox org download the VirtualBox pack age for your platform and install it A reboot might be necessary 2 Download the VirtualBox Extension Pack too platform independent executed it and it will be automatically integrated into your installation of VirtualBox Once installed the iSpec virtual machine can be decompressed and recognized by VirtualBox by opening iSpec_Xubuntu vbox 600 7 iSpec_Xubuntu_v20140330 lt i gt EX a FA
39. e to discard stars in later stages of the pipeline python 002 _generate_list py input clusters txt output preprocessed input spectra_list txt output list txt Listing 16 Execution input clusters txt Generate Grid Cluster None or cluster name Age_ref Cluster s age RV_ref Radial velocity M H ref Metallicity Determine APs Joutput preprocessed Pre processing Determine abundances input spectra_list txt Figure 51 Generate list of spectra Input in dark blue output in light blue 5 4 Determine atmospheric parameters and normalize The determination of atmospheric parameters depends on several parameters that can be modified in setup py For instance the normalization can be done using a first fast estimation of the atmospheric parameters of the star That is why the normalization takes place in this step python 003_determine_ap py output list txt output preprocessed output grid output processed 1 Listing 17 Execution A2 DRAFT VERSION ISPEC FRAMEWORK USER S MANUAL output list txt Generate Grid output preprocessed Pre processing output grid Determine APs Determine abundances Figure 52 Determine AP Input in dark blue output in light blue 5 5 Generate list of atmospheric parameters Once the previous step has finished a list should be automatically generated compiling all the atmo spheric parameters in a single file py
40. e top of the window by default it depends on the wavelength range and it proposes 1 spline every 1 nm There is a third fitting model named fixed value which is useful when the spectrum is already normalized and it is not necessary to re normalize 17 DRAFT VERSION ISPEC FRAMEWORK USER S MANUAL 0 8 0 6 0 4 05 561 0 561 5 562 0 562 5 563 0 563 5 561 0 561 5 562 0 562 5 563 0 563 5 0 8 0 6 0 4 05 561 0 561 5 562 0 562 5 563 0 563 5 560 5 561 0 561 5 562 0 562 5 563 0 563 5 Figure 17 Continuum fitting algorithm Once the continuum fit is executed related information can be visualized if the Stats action is selected and a region is clicked On the other hand the fitted continuum can be removed by using the function Operations Clear Fitted continuum 3 6 Line fitting For all the line masks a Gaussian fit can be done by using the Operations Fit lines menu option It requires a previously fitted continuum and the velocities respect to atomic telluric lines This fields will be automatically filled if the velocity determination process has been executed previously Velocity respect to telluric lines km s 0 0 Line list VALD_atom 300_1100nm a Maximum atomic wavelength difference 0 005 Y Allow peak position adjustment Check derivatives before fitting OK Cancel Figure 18 Properties for fitting lines Fitted lines will be cross matched with an atomic line list
41. flag_F txt Generate Grid Pre processing Determine APs Determine abundances X Fe 1 Figure 57 Plot Input in dark blue output in light blue 5 10 Setup In the setup py file several parameters can be modified and they are going to affect the behaviour of the pipeline e A custom read_spectrum function can be defined here in case we have input spectra in a format that is not directly understood by iSpec It is important to remember that the wavelength should be in nanometers e The telluric mask and the template to determine radial velocities is specified here the default values are usually good enough e For the homogenization process it is going to be taken into account the wavelength range that we want to use in our analysis and the common resolution 45 DRAFT VERSION ISPEC FRAMEWORK USER S MANUAL e The determination of atmospheric parameters and abundances will use the model atmosphere atomic line list and solar abundances indicated in this file e A configuration section is present to specify how is going to be treated different groups of stars For instance in the example All the stars with origin REF_HARPS REF_NARVAL and REF_UVES share the same configuration they are going to be considered as the reference star for the differential analysis i e config abundance_reference True If we check the input spectra list we will see that those st
42. g is going to read the input spectra perform different operations depending on what is specified in input spectra_list txt convolve to a common resolution specified in setup py and remove telluric regions The possible operations are e None The input spectrum is used as it is e Coadd All the spectra with the same value in the fields cluster star and origin will be co added e Merge All the spectra with the same value in the fields cluster star and origin will be merged useful when we have spectra corresponding to different wavelength regions In the input list of spectra is very important to e Correctly specify the fields cluster star and origin and make sure they are coherent with the operation that we want to perform e g coadd or merge Cluster usually filled with the number of the cluster to which the star belongs to If it is a field star we can specify just field or any other name we prefer Star name of the star if we have several spectra of the same star it is necessary to always use the same name independently if we want to coadd merge or not Origin typically filled with the name of the instrument of origin e g UVES HARPS NARVAL or used for grouping special group of stars such as the solar spectra used as a reference for the differential analysis e Ensure that all the spectra have a unique ID field e Specify the corre
43. iSpec Integrated Spectroscopic Framework USER S MANUAL Author Sergi BLANCO CUARESMA July 28 2015 ISPEC FRAMEWORK USER S MANUAL Contents 1 Introduction 3 2 Installation 3 21 Virtual MACHINE 2 66 5 6064 46 64 2 24 ELOY eR wee ERR a 3 2 2 GNU Linux systems T 2 2 1 Python distribution a a 7 Pee TOPCO TIGER cosmos od eee eee RED RDS Se OE A Sw eX 8 23 WSCOSASIGI S cc eee hw oa 10 205k Pihon AISERBUON essa esa BS ew a we eee a A 10 2 3 2 i9pe c framework 24 462425 6H G8 ee BESO ERG eee EERE DEH ES 11 3 Interactive usage 11 ou Mpu tand output Tes 52424 e583 eda wb Re Rew A oO 11 EEL EMO error ea pee eG EE ARE ee eK 12 la COnN MU ocurra ar Che EE eH HEHE SE See e ss 12 See UG ias o eee eR Hehe dae ee Bee EE 12 3 1 4 Segments aoaaa ee a 13 Gee WA a eee eee ea eRe ERE RRS A 13 33 Basic interaction lt s ss escabassa ew RARER HOEK He eee A HS 14 3 3 1 Opening selecting and closing spectra files 0 2 004 14 3 3 2 Opening and closing region files 2 oa a a a ee o 15 3 3 3 Saving images spectra and regions 2 2 ee 15 3 3 4 Visualization a 15 3 4 Regions 16 35 CONDAL MNE ociosa Be eee Gh ERD Bw ew oe ee ow eS 16 Ore SIUC R sarengan EREE A 18 3 7 Automatic continuum regions 1 1 a 19 3 8 Automatic line masks 20 3 9 Adjust line masks 20 3 10 Create segments around line masks a a 21 3 11 Barycentric velocity calculation a aooaa 4H bE ER
44. ipt is to duplicate it and erase all the functions except the ones that fit the user s needs The variable ispec_dir inside the script should point to the right directory where Spec is installed and the script can be executing by writing on a terminal python example py 5 Pipeline iSpec provides an example of a complete pipeline to process non normalized 1D spectra and derive atmospheric parameters and individual chemical abundances The pipeline consists on several steps that should be run sequentially because the output of one step is used as an input for the following one Nevertheless some steps can be executed in parallel if they are run in a multi core machine The chain of execution is as follows see file runjobs python 000_generate_grid py input grid_ranges txt output grid 1 python 001_preprocessing py input spectra_list txt input spectra output preprocessed 1 python 002 _generate_list py input clusters txt output preprocessed input spectra_list txt output list txt 1 python 003_determine_ap py output list txt output preprocessed output grid output processed 1 python 004_generate_ap_list py output list txt output processed output list_AP txt python 005_determine_abundances py output list_AP txt output processed output abundances 1 python 006_generate_abundances_list py output list_AP txt output processed output abundances output list_abundan
45. lecting Operations Clean fluxes and errors and specifying a base and top flux and error Clean fluxes and errors iv Filter by flux Base fx Top flux 1 2 Filter by error Base error 0 0 Top error 12 Filter cosmics Resampling step 0 001 Window size 15 Variation limit 0 01 Replace by Zeros a OK Cancel Figure 35 Filter out all the measurements that are not in these range limits 29 DRAFT VERSION ISPEC FRAMEWORK USER S MANUAL Additionally the cleaning can be done by considering the mean and standard deviation of a limited region between a top and base flux level This option can be useful to remove cosmics residuals although it should be used carefully since it would remove also emission lines Finally the user can chose to completely remove the fluxes or replace them by the continuum instead of zeros 3 23 Clean telluric regions Through the menu Operations Clean telluric regions one can clean the regions of the spectrum that may be affected by telluric lines For that purpose it is needed the radial velocity of the spectrum the minimum depth of the telluric lines to consider previously measured by iSpec from a synthetic model and the region around those lines to be cleaned in km s The user can chose to completely remove the fluxes or replace them by the continuum instead of zeros By default a margin of 30 km s is suggested since it represents the typical maximum velocity range
46. models 18 plane parallel spherical Effective temperature Teff 2500 2600 2700 2800 2900 3000 3100 3200 3300 3400 3500 3600 3700 3800 3900 4000 4250 4500 4750 5000 5250 5500 5750 6000 6250 6500 6750 7000 7250 7500 7750 8000 K Gravity Logg 0 0 0 5 1 0 1 5 2 0 2 5 3 0 3 5 4 0 4 5 5 0 dex Metallicity M H 5 00 4 00 3 00 2 50 2 00 1 50 1 00 0 70 0 50 0 20 0 00 0 20 0 50 0 70 1 00 dex Standard abundance composition 30 of all models Fe H a Fe C Fe N Fe O Fe 1 00 to 0 00 0 00 0 00 0 00 0 00 0 25 0 10 0 00 0 00 0 10 0 50 0 20 0 00 0 00 0 20 0 75 0 30 0 00 0 00 0 30 1 00 to 5 00 0 40 0 00 0 00 0 40 Table 3 MARCS Standard abundance composition 19nttp marcs astro uu se 20SPECTRUM is a plane parallel synthesizer but some tests has shown that the MARCS spherical models can be used with reasonable results Additionally the transition from MARCS plane parallel to spherical is smooth enough to avoid major impacts 34 DRAFT VERSION ISPEC FRAMEWORK USER S MANUAL The MARCS mod files have been transformed to the format needed by SPECTRUM Kurucz format L T temperature K P gas pressure Pgas Electron density 1 cc XNE PAMAROS NOOB WH RHOX mass depth 1 380658e 16 T ABROSS Rosseland mean absorption coefficient KappaRoss ACCRAD
47. n FITS format if fits extension is specified in the file name 3 12 Continuum Continuum region files should be plain text files with tab character as column delimiter Two columns should exists wave_base and wave_top the first line should contain those header names They indicate the beginning and end of each region one per line For instance wave_base wave_top 480 6000 480 6100 3 481 1570 481 1670 491 2240 491 2260 492 5800 492 5990 Listing 8 Fragment of a continuum region file Additionally wave base should be always lower than wave top or iSpec will not be able to process the file 3 1 3 Lines Line region files should be plain text files with tab character as column delimiter Four columns should exists wave_peak wave_base wave_top and note the first line should contain those header names They indicate the peak of the line beginning and end of each region one per line and a note it can be any string comment For example 1 wave_peak wave_base wave_top 480 8148 480 7970 480 8330 31 496 2572 496 2400 496 2820 499 2785 499 2610 499 2950 5 505 8498 505 8348 505 8660 Listing 9 Fragment of a line regions file The note can be blank but the previous tab character should exists anyway Regarding the wavelenghts wave base should be always lower than wave top and wave peak should be in between or Spec will not be able to process the file
48. ne by selecting the Pan mode left clicking and dragging the mouse Additionally in this mode it is possible also to zoom in out by right clicking and dragging the mouse 15 DRAFT VERSION ISPEC FRAMEWORK USER S MANUAL While Zoom or Pan modes are active no actions stats create modify remove can be performed on any element see section 3 4 3 4 Regions As exposed in section 3 1 there are three different types of regions e Continuum they can be used for fitting the star s continuum instead of using the whole spec trum see 3 5 Continuum region candidates can also be automatically identified by using the functionality described in section 3 7 e Line masks its goal is to isolate lines of interest and they are used for gaussian fitting see section 3 27 3 and astrophysicial parameters abundance determination see section 3 27 e Segments regions that generally include one or more line masks and one or more continuum regions For creating modifying or removing regions an action and an element should be selected in iSpec Action Stats Create Modify Remove Element Continuum Lines C Segments Figure 15 Actions and elements The possibilities with each combination are explained in the following table it is worth noting that the mouse position in wavelength flux can be found in the status bar of the editor s window Stats Create Modify Remove Continuum Left click on a Left click on an Left
49. ns input regions directory Example of continuum regions 9 DRAFT VERSION ISPEC FRAMEWORK USER S MANUAL Line masks for Fe 1 1 Line masks for the wings of H Alpha H Beta and Magnesium triplet Segments for the previous lines Star Type Teff K Log g dex M H Radial Vel km s Procyon Metal Rich Dwarf 6545 3 99 0 02 3 2 Sun Metal Rich Dwarf 5777 4 44 0 00 0 00 u Cas A Metal Poor Dwarf 5308 4 41 0 89 96 06 Arcturus Metal Poor Giant 4247 1 59 0 54 5 18 u Leo Metal Rich Giant 4433 2 50 0 29 14 25 Table 1 Astrophysical paramters for the 5 very well known stars included 2 3 MacOSX systems 2 3 1 Python distribution One of the easiest ways to install a Python distribution on MacOSX with most of the needed libraries for the Spec i e matplotlib is by using MacPorts However it is important to know that the installation can take time 1 hour or more depending on the computer because depending on your operating system version it might compile all the python libraries from source code On the other hand MacPorts is completely open source without limiting licenses and it can be used to install other free software apart from python 1 It should be taken into account that MacPorts requires X11 application for Mac OS X Xcode and the Command Line Tools 2 Download the MacPorts binary version that correspond to the Mac OS X version install it by double clicking and follo
50. o port selfupdate Listing 5 Update MacPorts 6 Install python framework and several required extra libraries it can take several hours because it compiles everything from the sources sudo s Execute everything with administrative privileges root 2 port selfupdate 3 port install python27 4 port select set python python27 5 port install py27 readline 6 port install py27 tkinter 7 port install py27 numpy s port install py2 7 scipy o port install py27 matplotlib io port install py27 astropy ii port install py27 cython 12 port select set cython cython27 13 port install py27 statsmodels 14 port install py27 lockfile 15 port install py27 pip 16 In sf opt local bin pip 2 7 opt local bin pip i7 pip install sampy is exit Stop having administrative privileges root Listing 6 Python installation The user password may be required to gain administrative rights for some commands 2 3 2 iSpec framework Check section 2 2 2 for specific instructions about the installation of iSpec 3 Interactive usage 3 1 Input and output files iSpec can read write spectra and region definition for e Continuum regions spectrum regions where there is only continuum no absorption emission lines e Line masks Individual absorption lines where the mask covers from base point to base point and specifies where is the peak e Segments Group of continuum regions
51. ols represent lower surface gravity 4T DRAFT VERSION ISPEC FRAMEWORK 2 0 iSpec log g Reference log g dex Dm o m DO yA 00 MO o onm E QU c MMR lt vp On ONEROLGHS5O_SSSOLADO CH 0 EA ANN 50 OS AS OTE OS SSITON OS LOS ELA GI DPI RASTAS neo o DOCO2275 0000820002 OGOODATIATEVYDOOTOVUOLOITIOCE m wn O_o o OUST O YO Vuo cw 50u 3 Soe 5050 ARO HH 00 LAO A PA Y O MA PM ON PALA 00 00 Dd ALO NR MN HH ODO MID OMA INDOOATOOONRADGININN GOD GEN E SOAAHHAHA ANNAN MMM mm Reference surface gravity dex O t TANA Y IFR 910 t T Reference Fe 1 H E USER S MANUAL Figure 60 Differences in surface gravity between the reference the Gaia FGK benchmark stars and the derived value by iSpec pipeline Stars are sorted by surface gravity the color represents the metallicity and larger symbols represent lower surface gravity Spec Fe I H Reference Fe I H NLTE dex mm o oO WONG DO y mM N AP ETE C 00 os NFFJENONS SUA LL20 ES Sox HDL NOUN SU OT OBO A NOOO CGO FO of pAn Gavo Lo 5 TIOCOVaTET ot Lolo sol ovado Ans Reference Fe I H NLTE dex 6400 6000 5600 5200 Y 4800 F 4400 4000 3600 Figure 61 Differences in neutral iron abundances between the reference the Gaia FGK bench mark stars and the derived value by iSpec pipeline Stars are sorted by metallicity the color represents the temperature and the size is linked to the surfac
52. ons Ignore line regions Treat each segment independently Fitting model Splines B Fixed value 1 0 OK Cancel Figure 16 Properties for the fitting of the continuum The process applies a median filter and a maximum filters recommended sense but the inverse order can be chosen too with windows of given sizes specified by the user It is frequently useful to fine tune those values depending on the spectral type and signal to noise ratio because it will affect the continuum placement The steps are visually shown in Fig 17 and the details are better described in 2 The spectral resolution can be specified to optimize the computation but it is completely optional it can be set to zero If the spectrum contains errors they can be propagated thus they will be used as weights during the fitting process Regarding absorption lines ispec implements a probabilistic mechanism to automatically detect them which can also be adjusted The process can consider only the defined continuum regions and or ignore line regions if the corre sponding options are selected if not it will use the whole spectra Additionally it can treat each region independently fitting the continuum without considering the rest of the regions Once the spectra is filtered a model will be fitted which can be several splines recommended degree is 2 or one polynomial model The suggested number of splines degree for the polynomial is shown on th
53. oo narrow wavelength ranges and wavelength steps not smaller of 0 01 being 0 001 the optimum value 3 27 2 Parameters determination iSpec can be used to determine astrophysical parameters by comparing an observed spectrum with synthetic ones generated on the fly in a similar way as the Spectroscopy Made Easy SME tool 20 works A least square algorithm tries to minimize the solution in order to converge towards the most similar synthetic spectrum 21http kurucz harvard edu grids html 39 DRAFT VERSION ISPEC FRAMEWORK USER S MANUAL To do so iSpec should have one active spectrum with one or more segments and one or more line masks defined The synthetic spectrum will be generated only within the segments and only the measurements that fall into the line masks will be used for comparison and minimization In Parameters Determine astrophysical parameters it is possible to specify what parameters should be free and what are their initial values Determine parameters Model atmosphere MARCS GES El Solar abundances Grevesse 2007 El Line list IGES_atom 475_685nm v Effective temperature K Free Teff Surface gravity log g 4 44 Free Log g Metallicity Fe H l o 0 0 M Free Fe H Microturbulence velocity km s 1 0 Y Free Vmic Macroturbulence velocity km s 0 0 Mv Free Vmac Rotation v sin i km s 2 0 Free vsin i Limb darkening coefficient 0 0 Free limb dark coeff Resolu
54. ra harps_procyon_norm s gz A narval_sun_norm s gz flux 540 66 540 67 540 68 540 69 540 70 540 71 540 72 wavelength nm Figure 13 Show errors in plot 5 352 Opening and closing region files Any kind of region files continuum line masks or segments with the format specified on section 3 1 can be opened by Spec through the menu File Once loaded they can also be completely removed from the plot through the menu Operations Clear Continuum Line masks Segments In case some regions have been modified and not saved Spec will ask for confirmation before clearing them 3 3 3 Saving images spectra and regions Through the File menu it is possible to save a PNG image of the current plot the active spectrum or the definition of the different regions It is worth noting that in the title of the editor s window will appear segments lines continuum or name_of_spectrum to indicate that some of these elements have been modified but not saved 3 3 4 Visualization iSpec provides zoom capabilities by activating the zoom mode Once done left clicking and dragging defines the zone to be augmented The home icon reverts the zoom and the back forward arrows permits going back to the previous zoomed region O O Sif Figure 14 From left to right Home button back forward arrows Pan mode and Zoom mode On the other hand it is possible to move the current visualization zo
55. require the name of a star or its ID e g M67_No1054 a wavelength range of interest e g 515 525 and what line masks we want to show e g Fe 1 for neutral iron or None for not showing any line mask Regarding the individual abundances the command is different but the argument logic is the same 1 python visualize_abundances py output list txt output abundances M67_No1054 550 560 O Listing 24 Execution 46 DRAFT VERSION ISPEC FRAMEWORK 5 12 Assessment USER S MANUAL The pipeline has been assessed by analyzing the Gaia FGK Benchmark stars library 2 The median differences with respect to the reference values are e Effective temperature 27 68 K e Surface gravity 0 15 0 14 dex e Metallicity M H 0 16 0 06 dex e Iron abundance Fe I H 0 06 0 05 dex It is up to the user to do a correction to the results obtained with this pipeline For instance a simple approach would be to substract the previous biases to the parameters and abundances that come out of the pipeline Spec T y Reference T K HD220009 Arcturus 1000 0 4 aD ADO periei teniendo aes ASA n 0 8 re 0 1 2 A 200 16 8 400 E 2 0 0 600 ac 2 4 Figure 59 Differences in effective temperature between the reference the Gaia FGK benchmark stars and the derived value by iSpec pipeline Stars are sorted by temperature the color represents the metallicity and larger symb
56. ric lines if the user has indicated so This functionality is located in the Operations Find line masks menu and can be applied over the whole spectra or only inside the defined segments In both cases it needs a previously fitted continuum After the computation it removes the current line regions if there are any and draws the ones that the process has found 3 9 Adjust line masks Line masks automatically found or manually defined for a given type of stars may not fit well enough the shape of the line iSpec can adjust automatically previous defined line masks to match a better 20 DRAFT VERSION ISPEC FRAMEWORK USER S MANUAL beginning end of the masks to the particular form of the active spectrum Adjusting line masks Resolution 100000 OK Cancel Figure 22 Properties for the automatic mechanism of finding line masks To do so iSpec will check were is the optimal limit of the line by looking at a window of a given margin in nanometers around the line center The algorithm will search for the local maximum in each side of the line that is closer to the line center 3 10 Create segments around line masks The user that has a group of line masks already defined could be interested in creating segments around them for instance in order to compute synthetic spectra iSpec can do that automatically the user should specify how many nanometers he wants around each line masks and iSpec will group those line
57. rors are present e From fluxes The whole spectrum is checked resampling to ensure homogeneous steps and taking 10 by 10 measures although this value can be modified by the user calculating the SNR equation 5 for each one and finally selecting the mean SNR as the global SNR Properties for estimating SNR Number of points and wavelength step is only used if SNR is estimated from fluxes not errors Number of points Wavelength step resampling 0 001 Estimate SNR Directly from reported errors C From fluxes in blocks of N points OK Cancel Figure 29 Properties for the global SNR estimation The Signal to Noise Ratio SNR can be defined as the ratio of mean to standard deviation of a measurement SNR 5 QI where u is the mean value and o the standard deviation 3 15 Errors estimation Given a SNR iSpec can estimate the errors associated to each flux measurement This functionality is found in Operations Estimate errors based on SNR Calculate spectrum errors SNR Signalto Noise Ratio 10 0 OK Cancel Figure 30 Properties for the global SNR estimation 3 16 Add noise Poisson Gaussian noise can be artificially added by using the function Operations Add noise to spectrum fluxes Add noise to spectrum SNR Signal to Noise Ratio Distribution Poisson C Gaussian OK Cancel Figure 31 Add Poisson noise given a SNR 26 DRAFT VERSION ISPEC FRAM
58. rs S DRAFT VERSION ISPEC FRAMEWORK USER S MANUAL iSpec Files Operations Parameters Spectra Help Spectra espadons_mu _leo s gz narval_arcturus s gz harps_procyon s gz narval_mu_cas s gz A narval_sun s gz A AA til A 550 600 wavelength nm O cal El a x 611 363 y 0 988235 Action Stats Create Modify Remove Element Continuum Lines C C Segments 0 bs Cursor on wavelength 611 3628 and flux 0 9882 Figure 9 Spec showing several spectra The Spec compressed file contains the program and some additional data in the input directory e Observed spectra for 5 very well known stars Procyon Sun y Cas A Arcturus y Leo in put spectra example directory Filenames NARVAL_ s gz NARVAL 4 spectra resolving power 70 000 90 000 ESPaDOnS_ s gz ESPaDOnS 5 spectra resolving power 68 000 81 000 x HARPS_ s gz HARPS 6 spectra resolving power 115 000 Characteristics Not normalized Radial velocity not corrected Barycentric correction applied Resolution depends on the instrument Wavelength range from 480 to 680 nm e ELODIE 7 spectra for binarity tests input spectra binaries directory see section 3 12 Filenames x elodie_hd005516A_spectroscopic_binary s gz x elodie_hd085503_single_star s gz Characteristics Not normalized Resolution 42 000 x Wavelength range from 400 to 680 nm e Regio
59. rs Velocity profile relative to atomic lines Profile relative intensity o ul o Dm ul 150 100 Q O al T a x 31 6235 y 0 877621 ft Mean km s 21 98 Min error km s 0 5000 Baseline 1 00 A rel intensity 3 32 Sigma km s ance Velocity lower limit km s 200 0 Velocity upper limit km s 200 0 Velocity steps km s 1 0 Fitting model 2nd order polynomial a w CCF in Fourier space Mask linelist Atlas Sun 372_926nm 7 Mask size km s 2 0 Minimum depth 0 1 OK Cancel Figure 27 Velocity profile The error in the radial velocity is calculated following 16 2 fC Cm 17 NET Now CA 8 where N is the number of bins in the spectrum C is the cross correlation function and C is its second derivative Finally the spectrum and the regions can be shifted considering the determined velocity or indicating a custom one by using the option Operations Correct velocity The following formula is applied pa velocity C Acorvected A T velocity 4 C 24 DRAFT VERSION ISPEC FRAMEWORK USER S MANUAL where c Is the speed of light in vacuum and A the original wavelengths 3 13 Spectroscopic binaries identification The velocity determination function relative to atomic data section 3 12 can be used to identify spectroscopic binaries In those cases the resulting profile would have two different peaks Velocity profile rel
60. s that are close enough under the same segment Properties for finding segments OK Cancel Figure 23 Properties for the automatic mechanism of finding line masks 3 11 Barycentric velocity calculation Spec incorporates an option for calculating the earth s velocity towards the earth Parameters Calculate barycentric velocity menu algorithm based on Stumpff 1980 11 so that the spectra can be corrected and transformed to the solar barycentric reference frame Operations Correct Barycentric velocity menu For the determination it is necessary to know the date time of the observation and the star s coordinates RA Right Ascension in hours DEC Declination in degrees in epoch J2000 0 Barycentric velocity determination radial 15 02 2012 Velocity correction Time HH MM SS 00 00 00 Epoch J2000 0 Velocity relative to telluric lines km s 18 11 Right ascension HH MM SS 19 50 46 99 Apply correction on Spectra Declination DD MM SS 08 52 5 96 C Regions OK Cancel OK Cancel Figure 24 Barycentric velocity determination and correction 3 12 Velocity determination and correction The velocity profile can be determined relative to three different references e Atomic data Useful for determining the radial velocity of a star considering that the barycentric velocity due to the earth orbit has been already corrected 21 DRAFT VERSION ISPEC FRAMEWORK USER S MANUAL
61. s where the telluric lines may be For instance considering a fictitious star that has zero radial velocity respect to the sun when it is observed from earth a radial velocity might be measured due to the rotation of the earth around the sun This radial velocity respect to the each will typically be between 30 and 30 km s for this fictitious star depending on the region of the sky that it is observed This functionality is mainly useful when the spectrum is going to be used as a template for measuring the radial velocity of another spectrum see section 3 12 Cleaning all this fluxes from the template reduces the impact of potential telluric lines in the spectrum to be measured Clean telluric regions Velocity relative to tellurics Minimum velocity 30 0 Maximum velocity 30 0 Minimum tellurics depth 0 02 Replace by Zeros El OK Cancel Figure 36 Filter out all the measurements potentially affected by telluric lines 3 24 Spectrum resampling By selecting Operations Resample spectrum the current active spectrum can be uniformly resampled from a given wavelength range increment by linear or spline interpolation Resample spectrum Base wavelength 476 61 Top wavelength 683 26 Wavelength step 0 001 Current wavelength step min 0 0001 max 0 005 median 0 003 mean 0 003 Method Bessel Y OK Cancel Figure 37 Properties for spectrum resampling Some basic statistics of the active spectrum are printed
62. s are 22 DRAFT VERSION ISPEC FRAMEWORK USER S MANUAL e Narval Sun lines and depths detected from observed asteroids by the NARVAL spectrograph with a wavelength range from 370 to 1048 nm e Solar and Arcuturs atlas lines and depths detected in the solar and arcturus atlas with a wave length range from 372 to 926 nm e HARPS SOPHIE AO FO G2 KO K5 M5 mask specially prepared to be used by HARPS SO PHIE with a wavelength range from around 375 to 680 nm e Synthetic Sun lines and depths detected from a synthetic spectrum generated with SPECTRUM VALD linelist and MARCS model atmosphere with a wavelength range from 350 to 1100 nm e VALD mask based on a line list extracted from the VALD database 14 with a wavelength range from 300 to 1100 nm The depth of each line correspond to a star with Teff 5770 0 and gravity logg 4 40 solar type For the telluric lines the mask has been generated from the analysis of a synthetic spectra of the typical telluric lines obtained from TAPAS 15 an on line service that provides simulated atmospheric transmission spectra for specific observing conditions Internally for the cross correlation process iSpec creates a mask from the given line list with a user defined size If one or more mask lines fall into one range the max depth is assigned as the mask value otherwise it will be zero Afterwards it re samples the mask uniformly in terms of velocity by using the specified velocity
63. sk txt lines input LUMBA UVES_MRD sun_Fe linelist txt segments input LUMBA UVES_MRD_sun_segments txt input LUMBA UVES_MRD _sun_official s gz Listing 12 Command line execution with all the possible arguments 3 3 Basic interaction 3 3 1 Opening selecting and closing spectra files iSpec can open multiple spectra files simultaneously with the format specified on section 3 1 through the menu File Open spectra At every moment only one spectrum is active and its marked with a A symbol in the legend box narval_mu_cas_norm s gz narval_sun_norm s gz harps_procyon_norm s gz narval_arcturus_norm s gz A espadons mu _leo _norm s gz Figure 12 Multiple spectra the active one is marked with an A in the legend Operations such as continuum fitting or radial velocity determination are perform using only the active spectrum This can be change through the menu Spectra Name of the spectrum or directly closed through Spectra Close spectrum In case the spectrum has been modified and not saved iSpec will ask for confirmation before closing it Additionally the errors associated to each spectrum s point can be plotted by activating Spectra Show errors in plot This will plot two discontinous line above and below the spectrum representing the flux the errors 14 DRAFT VERSION ISPEC FRAMEWORK USER S MANUAL Files Operations Parameters Spectra Help Spect
64. thon 004_generate_ap list py output list txt output processed output list_AP txt Listing 18 Execution output list txt Generate Grid Joutput processed Pre processing Determine APs output list_AP_cluster_summary txt Determine abundances output list_AP_star_summary txt output list_AP_history_synth txt output list_AP_history_synth_summary txt Figure 53 Generate list of AP Input in dark blue output in light blue 5 6 Determine chemical abundances This step proceeds with the determination of individual chemical abundances python 005_determine_abundances py output list_AP txt output processed output abundances 1 Listing 19 Execution A3 DRAFT VERSION ISPEC FRAMEWORK USER S MANUAL output list_AP txt Generate Grid output processed Pre processing Determine APs Determine abundances Figure 54 Determine abundances Input in dark blue output in light blue 5 7 Generate list of chemical abundances Once the previous step has finished a list should be automatically generated compiling all the abun dances in a single file and calculating the weighted average values for the reference star and the rest of the sample The differential calculations are performed by this process too python 006_generate_abundances_list py output list_AP txt output processed output abundances output list_abundances txt Listing 20 Execution output list
65. tion 7 100000 Free resolution Individual abundance Free individual abundance Maximum number of iterations 20 OK Cancel Figure 45 Properties for determining astrophysical parameters The linelist solar abundances and model atmospheres are the same as for the synthetic generation option see section 3 27 1 The results are shown directly in the terminal 36 DRAFT VERSION ISPEC FRAMEWORK USER S MANUAL Generating 6271 52 4 09 0 43 1 76 13 52 2 00 0 00 70000 1 00 Generating 6171 52 4 19 0 43 1 76 13 52 2 00 0 00 70000 1 00 Generating 6171 52 4 09 0 48 1 76 13 52 2 00 0 00 70000 1 00 Generating 6171 52 4 09 0 43 2 26 13 52 2 00 0 00 70000 1 00 Cache 6171 52 4 09 0 43 1 76 15 52 2 00 0 00 70000 1 00 Generating 6174 09 4 08 0 43 1 76 13 52 2 00 0 00 70000 1 00 E Iter 7 CHI SQUARE 15 23354606 DOF 12795 teff 6174 089416 logg 4 084982924 MH 0 4253733734 Vmic 1 759699773 Vmac 13 51913615 Vsini 2 limb_darkening_coeff 0 R 70000 continuum_correction 1 Generating 6176 25 4 08 0 42 1 76 13 51 2 00 0 00 70000 1 00 Generating 6175 42 4 08 0 42 1 76 13 52 2 00 0 00 70000 1 00 Cache 6174 09 4 08 0 43 1 76 13 52 2 00 0 00 70000 1 00 Cache 6174 09 4 08 0 43 1 76 13 52 2 00 0 00 70000 1 00 wave_peak wave_base wave_top chisq tanh rchisq tanh chisq rchisq rms Line 480 29 480 27 480 32 0 05 0 0011 0 05 0 0011 0 0314 Line 480 81 480 80 480 83 0 05 0 0026 0 05 0 0026 0 0451 Line 517 30 517 29 517
66. wing the instructions all options should be left by default w Install MacPorts 2 0 3 MacPorts 2 0 3 pkg Welcome to the MacPorts 2 0 3 Installer Welcome to the MacPorts for Mac OS X Installer Introduction MacPorts provides the infrastructure that allows easy installation and management of freely available software on Mac OS X 10 7 systems Read Me http www macports org License l This installer guides you through the steps necessary to install MacPorts Destination Select 2 0 3 for Mac OS X To get started click Continue installation Type Installation Summary Figure 10 MacPorts installation 3 OPTIONAL After the installation if the user s default shell is bash edit the file SHOME bash_profile and verify that the installer has added the following lines PATH variable includes the MacPorts directories nttp matplotlib sourceforge net 1Onttp www macports org lMhttp guide macports org chunked installing htmltinstalling x11 12http guide macports org chunked installing xcode html IShttp www macports org install php 10 DRAFT VERSION ISPEC FRAMEWORK USER S MANUAL 1 export PATH opt local bin opt local sbin PATH Listing 4 bash profile 4 Close all the terminals and open a new one lf the user s default shell is bash you can instead execute source SHOME bash_profile in order to load the new configuration 5 Update the local ports tree with the ports repository sud

Download Pdf Manuals

Related Search

Related Contents

KX-TD7680 Operating Instructions  Model R5 Service Manual  Home Decorators Collection MPBNVT4922 Instructions / Assembly  Neri-Motori_Catalogo-Generale  Audiovox CTX5000 Cell Phone User Manual  KTM Prime  Kramer Electronics VS-66H3 video switch  PAH_programme animations printemps  REKA App  

Copyright © All rights reserved. Failed to retrieve file