Home

INTEGRAL Science Data Centre SPI Analysis User Manual

1. Name Name Type Description main script executable spibounds nregions nregions integer Number of regions for which distinct binning algorithms are to be defined example 4 spibounds regions regions string The boundaries of the energy regions example 20 40 100 1000 8000 62 ISDC SPI Analysis User Manual Issue 9 0 spibounds_nbins nbins string The number of bins per region Nega tive numbers indicate logarithmically spaced bins example 10 10 9 7 8 4 2 SPI OBS HIST spi obs hist main purpose is generation of the event spectra that may be used for imaging spectral or timing analysis but in the same time it also derives Good Time Intervals and the Dead time for each pointing and pseudo detector More details can be found in the spi obs hist User Manual which can be downloaded from the following page http isdc unige ch Support spi doc doc html 8 4 3 SPI ADD SIM spt_add_sim allows to add a simulated source into a real observation analysis A source position and flux must be given as input parameters spi_add_sim uses the IRF response to compute the detector counts expected for this source and adds the counts to the data set This program is useful to investigate crowding effects by comparing outputs with inputs for the added source s in the presence of other real sources This program must be used as part of the ISDC pipeline as it needs an OG group with all the
2. existence of file n absence of file Thus the fw type means to test whether the file given as a value of the parameter is writable The parameter mode can be a auto h hidden q query hl hidden learn ql query learn With the a mode the effective mode is set to the value of the parameter named as mode in the parameter file If the mode parameter is not found the effective mode is set to h With the h mode no questions are asked unless the default value is invalid for a given data type whereas with the q mode you are always asked for a parameter With the additional P mode if an application changes a parameter value the new value will be written to the parameter file when the application terminates 66599 664 99 spimodfit makes use of parameter types i r and s Some of the parameters require to enter multiple arguments with the same parameters This is done via a string which is then decomposed into different elements and interpreted A multiple arguments parameter can be either a vector or a list e A vector is an array of values of the same kind i e reals integers etc separated by blanks for example 1100 5 1433 0 1500 0 is a vector of 3 reals This is a extension or the IRAF format coded in the OSA PIL library e A list is a array of values or fields not necessary of the same kind separated by commas
3. SPI Analysis User Manual Issue 9 0 Figure 38 Main GUI window of the spi_science_analysis script 000 3 spi_science_analysis Figure 39 Energy boundaries or ebounds GUI 000 X Energy definition ISDC SPI Analysis User Manual Issue 9 0 47 48 Figure 40 spiros GUI main window X Spiros options M SPIROS General Setup Ok Run SPIROS in Mode SPECTRA y Help m Further Options for Imaging Timing m Selection Parameters energy subset pointing subset detector subset m Other Parameters Background method fi Optimization statistic cruz y Bins for src location FIRST g Close all sub windows by clicking twice Ok then using the Save As button you can save the parameters you have just entered into a file Enter the name grb_analysis par and store it in the current directory PLEASE MAKE SURE NO SPELLING MISTAKE IS INTRODUCED THE GRB SCRIPTS READS THE grb_analysis par FILE LO CATED IN THE CURRENT DIRECTORY TO LOAD THE ANALYSIS PARAME TERS h Once the parameters are saved click on Quit to quite the process We do not want to run the analysis yet This procedure is just a convenient way to enter parameters for the analyses Alternative If you do not want to use the GUI you can specify the above parameters by editing the spi science analysis par file located in your PFILES
4. 6 spiros parameters included into the main script a oa oa a a a e 0000 spi cl ean parameters aa a a qu soy ee E Aa a A A Y spi_spectral_display parameters e spi_dsp_display parameters ISDC SPI Analysis User Manual Issue 9 0 10 14 ix Acronyms and Abbreviations ACS ARF BGO DFEE DOL DPE DS FCFOV FOV FWHM GRB GTI GUI IC IJD IRF Anti Coincidence Shield Ancillary Response Files Bismuth Germanate Digital Front End Electronics Data Object Locator Data Processing Electronics Data Structure Fully Coded Field of View Field of View Full Width at Half Maximum Gamma Ray Burst Good Time Interval Graphic User Interface Instrument Characteristics ISDC Julian Date Image Response File ISDC ISOC OBT OG PHA PI PSA PSAC PSD RMF ScW SWG TBW TM UTC Integral Science Data Center Integral Science Operations Centre On Board Time Observation Group Pulse Height Amplifier Pulse Invariant Pulse Shape Amplifier Plastic Scintillator Anti Coincidence Subassembly Pulse Shape Discriminator System Redistribution Matrix Files Science Window Science Window Group To be written Telemetry Coordinated Universal Time ISDC SPI Analysis User Manual Issue 9 0 Glossary of Terms e ISDC system the complete ground software system devoted to the processing of the INTE GRAL data and running at the ISDC It includes contributions from the ISDC and from the INTEGRA
5. ISDC SPT Analysis User Manual 1 March 2010 9 0 ISDC OSA UM SPI INTEGRAL Science Data Centre SPI ANALYSIS USER MANUAL Reference ISDC OSA UM SPI Issue 9 0 Date 1 March 2010 INTEGRAL Science Data Centre Chemin d cogia 16 CH 1290 Versoix Switzerland http isdc unige ch Authors and Approvals ISDC SPT Analysis User Manual 1 March 2010 9 0 Prepared by P Dubath I Kreykenbohm C Ferrigno Agreed by Re Walters car slats Dal dia ara aaa tava tray aia der Oh eats Approved by To COURVOISIE a esta Roses nd ob h he mafia A ISDC SPI Analysis User Manual Issue 9 0 i Document Status Sheet ISDC SPI Analysis User Manual 2 April 2003 1 0 First Release 19 May 2003 11 Update of the First Release Sections 7 2 7 3 7 5 2 9 8 3 and Table 4 were updated 18 July 2003 2 0 Second Release Sections 6 8 9 10 and the bibliography were updated 8 December 2003 3 0 Third Release Sections 6 10 and the bibliography were updated 19 July 2004 4 0 Forth Release Sections 5 6 8 10 and the bibliography were updated 6 December 2004 4 2 Update of Forth Release Sections 3 3 5 6 10 and the bibliography were updated 31 May 2005 5 0 Major update for OSA 5 0 Release All sections were modified 31 June 2006 6 0 Major update for OSA 6 0 Release Cookbook adapted to the new spi science analysis with added functionality on phase resolve analy
6. flatfield DOL in the panel of Fig 13 an appropriate path is required With this option the detector to detector ratios are kept fixed from that of the empty field but one background coefficient is derived through the imaging reconstruction process In fact one coefficient is derived for each time interval and the length of the intervals is provided as a number of consecutive pointings by the user To use the flat field model select the flat field in the main background GUI and press the button for the flat fields options Provide the number of pointings for which you assume the background to be constant Use a value of 5 in this example und leave the write single background extension unselected Note that in general much longer time interval can be used in the range 20 to 100 pointings The write single background extension must only be selected when using spimodfit instead of spiros see the section on spimodfit Sect 6 7 Another possibility to model the background accurately is to use background templates the second option in the main background GUI The background templates are provided by the SPI instrument team in Toulouse These templates are derived from background tracers but are fine tuned by the instrument specialists Several templates exist based on various background tracers A typical template is the GEDSAT which is based on the count rate of the events saturating the Germanium detector electronic hence the
7. 2 from the pvphase to the first annealing approximately revolution 20 to 38 3 from the first annealing up to the latest revolutions During the first period various modes and setups have been tried and the data are a bit more difficult to analyze During the second one because of the telemetry budget limitation all the information from the individual single events could not be downloaded They were binned on board and transmitted to ground in the form of spectra instead The software can handle these spectra in a rather transparent way although it may be useful to keep this fact in mind in some cases During the third period starting after the first annealing the full information about all events has been downloaded to the ground and SPI has been operated in the standard scientific photon mode It may also be useful to know that at the beginning of this period between the first and the second annealing data were lost at different rates also due to telemetry limitation These losses are taken into account in the analysis and you ISDC SPI Analysis User Manual Issue 9 0 57 may notice unusual ratio between the ontimes which exclude the data loss gaps and the difference Table 5 SPI annealing periods UT IJD Revolutions included start stop start stop first last 1 2003 02 06 09h 2003 02 14 20h 1132 1146 39 42 2 2003 07 15 12h 2003 07 29 18h 1291 1306 92 96 3 20
8. 3 FUNCTION 4 DET GROUP 5 MCM 6 DP modulation default 5 optistat string Enter optimization statistic possible values CHI2 LIKEH default LIKEH solution constr string Image solution constraints possible values NONE POSITIVE default NONE Selection Parameters ISDC SPI Analysis User Manual Issue 9 0 67 energy subset string Enter subset of energy bins a b c d f g default pointing subset string Enter subset of pointing bins a b c d f g default detector subset string Enter subset of detector bins a b c d f g default spiros Timing Mode source timing mode string Enter timing analysis mode possible values QUICKLOOK WINDOW TRANSIENT default WINDOW source timing scale real Enter timing observation integration time days seconds default 0 5 General imaging options nofsources integer Maximum number of sources to be searched for default 3 sigmathres real Enter lower sigma threshold to stop source search default 3 0 srclocbins string Enter count spectrum bins used for source location possible values BLANC SUM FIRST ALL default FIRST reference coord string Enter output image reference coordinate system possible values RADEC GALACTIC default GALACTIC image proj string Enter outp
9. 7 18 Fitting SPI spectra using XSPEC o e 58 7 19 Using a SPIROS output catalogue in input source_res fits to source_cat fits 59 7 20 Adding a new parameter in a spi science analysis analysis 59 7 21 GTIsin SPI data analysiS o e E 60 8 Data analysisisteps magat impure use ae a Se Sap as poda Gh ly ds AE 61 8 1 Catalog source extraction CAT EXTRACT 61 8 2 Energy correction SPI GAIN CORR o o o 61 8 3 Pointing definition SPLOBS POINT LL LL LL LL 62 8 4 Energy bin definition and events binning 0 0 62 8 4 1 SPIBOUNDS 42 ce AE a a a ee RE 62 8 4 2 SEI OBS HIST tos A ES ae Ee deo da do A SS 63 8 4 3 SPIAADD SIM tics Bite dd e EGS 63 8 5 Background model generation SPLOBS BACK 64 8 6 Image reconstruction and spectral extraction SPIROS 65 8 6 1 IMAGING mode 2 2 20 0 A 65 8 6 2 SPECTRAL extraction mod s gt e e o 66 8 6 3 TIMING analysis mode e ai A E 66 8 6 4 Parameter list oia a bl Gatos ee we ee A ed 67 8 7 Spectral Extraction SPMODFIT 0 000 69 8 7 1 Use of sprmod pit o so las as a as Gea ae ne OO Ma ay de ae 70 8 7 2 spimodfit analysis SCTIPb o ee 70 8 7 3 Use and syntax of the parameter fille 71 8 7 4 Suggested settings for point source analysis 72 8 7 5 Known ASS
10. at 50 keV The mask is located 171 cm from the detector plane A picture of the mask pattern is given in Figure 3 3 3 The Detectors The SPI detectors are the only elements of the INTEGRAL spacecraft that are actively cooled down A low temperature is required for an optimum sensitivity and resolution as well as for limiting the radiation damage An active cooling system brings the temperature of the cold plate on which the detectors are mounted down to 85 K using two pairs of cryocoolers In normal operating mode all coolers work simultaneously In case of failure of one of the cryocoolers or of the electronics the instrument can function with only one pair of cryocoolers in a degraded mode with a detector temperature of about 100 K The detector assembly is placed inside a cold box which is kept at approximately 210 K by the passive cooling system All temperatures of the cryostat subsystems are regularly monitored to provide temperature information that can be used for the data processing The detector plane is composed of 19 hexagonal shaped Germanium detectors encapsulated into beryllium capsules and mounted in the most compact configuration on the cold plate The size of a detector is 5 6 cm flat to flat with a height of 7 cm The detectors are mounted with minimum space between them such that the axes of two adjacent detectors are 6 cm apart At the time of writing two of these 19 detectors have ceased to function properly and are
11. o e e Options for the template background e Options for modeling the background using tracers like GEDSAT SPIROS main GU o a ne EE ge a A O a ACA SPIROS imaging mode GUI 00 00 Catalogue extraction GUI cpu 4 4 4005445 YPN ee a AAA Crab spectrum from cookbook example 00 00000008 SPIROS timing mode GUl o A A Crab light curve from the cookbook example o o GUI of spimodfit analysis with the default setting GUI of spimodfit analysis energy definition here 22 equally spaced logaritmic bins have been selected o o o GUI of spimodfit analysis regarding spimodfit oo o GUI of spimodfit_analysis regarding the spimodfit background multipliers Count rate spectrum of the Crab from the cookbook example and the residuals from the bestifitimodel o PA eee yy poe Dk ee eee eee Se AD ASAS Histogram lt options GUI se avai eee ck a sy eee gh A Phase parameter GUP Laura a we tt a ads ek ye AS Orbit parameter GUI 0 00002 ee ee Illustration of the different GRB start stop times Output of og create on the terminal 00000004 SPI Analysis User Manual Issue 9 0 vii viii 32 33 34 35 36 37 38 39 40 Al 42 43 Main GUI window of the spi science analysis script 4
12. spiros_image proj CAR spiros_image fov POINTING ZCFOV spiros_nofsources 3 spiros_sigmathres 6 spiros iteration output NO spiros optistat CHI2 spiros source timing mode QUICKLOOK spiros source timing scale 0 1 ISDC SPI Analysis User Manual Issue 9 0 51 7 2 Repeating and saving analyses As data access is carried out through the observation group OG all the spi_science_analysis script needs to know about the input data is the default name og spi fits All output data files are created with default standard names and they are attached to the OG so that subsequent tasks can access them through the OG It is always possible to repeat an analysis starting from any step Simply select the tasks you want to execute by clicking on the boxes in front of the spi science analysis buttons Be careful however to define a coherent analysis In other words do not leave gaps in the succession of the tasks In such cases the spi science analysis would stop immediately with a Incoherence in the succession of tasks error A concept of analysis level hierarchy has been defined as indicated in Table 3 When the pipeline processing proceeds from one step to another all the data are stamped with the cur rent level A fresh initial OG starts with the DEAD level and then goes trough the POIN BIN_I BKGT and IMA SPE or LC levels successively in a complete a
13. 0 2 2 0 00000 4 1 3 Background Gamma Ray Lines 0 4 2 Instrumental Characterization and Calibration 4 3 Measured Performance 4 3 1 Imaging Resolution 2 2 0 0 2 0 0 02 0000 4 3 2 Spectral Resolution 2 0 0 0 2 e 4 3 3 Sensitivity e sata a Sd a de AE 4 3 4 Dithering Sensitivity Degradation o o 4 3 5 Detection of off axis sources o ee ee 4 3 6 Imaging Capabilities 2 2 en 4 3 7 Timing Capabilities 2 en ISDC SPI Analysis User Manual Issue 9 0 xi iii II Data Analysis 14 iv 5 6 OVEEVICW 3 os has E ana ee AEE Boh NR a PT ie a bbw EEES A 14 Cookbook az 2 a4 pire hes BRE SEBS SS e AAA A Se Bae aa 15 6 1 Setting up the analysis environment 15 6 1 1 Input data organization 2 ee eee 15 6 1 2 Downloading data from the ISDC archive 16 6 1 3 Setting the software environment 0 17 6 2 Observation group OG creation o o 18 6 3 Image reconstruction without input catalogue 18 6 3 1 Data scr enin 4 44 444 ra a a a 28 6 4 Image reconstruction with catalogue information 30 6 4 1 Creating a new source_cat fits fle 30 6 4 2 Using a source cat fits file in the analysis 31 6 5 Spectral Extraction DIA DS ee 32 6 6 Tight curves 64 imc AA a do th aa a ee ida A 33 6
14. 11 50 Figure 31 Output of og create on the terminal 000 xterm gt isdesf7tgrb_analysis_tutorial 123 gt og create SWG index DOL 1 SWG DOL or ASCII list of SWG DOL scw 0066 006600940010 001 sug fits Observation group id used for the path grb030501 Base directory from where the obs branch will be build If null take REP BASE PROD Instruments comma separated list of ALL or any of SPI IBIS JMX1 JMX2 0MC SPI Log 1 OG level created repositories and data structures Log 1 Index scw 0066 006600940010 001 sug fits open Log 1 SWG level created repositories and data structures isdesf7tgrb analysis tutorial 124 gt E A Figure 32 Main GUI window of the spi science analysis script 000 3 IX spi science analysis 42 ISDC SPI Analysis User Manual Issue 9 0 b Make sure that the catalogue extraction is un selected and that no SPIROS Input Catalogue in this imaging example we assume that the GRB position is unknown c Check the List of detectors use 0 18 in most cases and the Coordinate System in that window and click on Energy definition to open the GUI window displays in Fig 33 Figure 33 Energy boundaries or ebounds GUI X Energy definition A Parameters energy binning Ok Number of energy regions 1 F Help Regions energy boundaries 20 400 Numbers of bins in each region 1 d Select your energy
15. Analysis User Manual Issue 9 0 21 22 e spi flat grp 0012 fits revolution 72 to be used for all revolutions up to 140 i e with all 19 detector live e spi flat grp 0013 fits revolution 158 to be used for revolutions 142 to 214 i e with 18 live detector excluding number 2 e spi flat grp 0014 fits revolution 220 to be used for revolution 215 to 281 i e with 17 live detectors excluding 2 and 17 e spi flat grp 0015 fits revolution 343 344 to be used for revolution 282 to 399 i e with 17 live detectors excluding 2 and 17 e spi flat grp 0016 fits revolution 456 457 to be used for revolution 400 to 488 i e with 17 live detectors excluding 2 and 17 e spi flat grp 0017 fits revolution 521 522 to be used for revolution 489 to 587 i e with 17 live detectors excluding 2 and 17 e spi flat grp 0018 fits revolution 653 654 to be used for revolution 588 to 775 i e with 17 live detectors excluding 2 and 17 e spi flat grp 0019 fits revolution 807 810 to be used for revolutions after 776 i e with 16 live detectors excluding 2 5 and 17 These spectra are rebinned in this step to match the bin size of your analysis and the program will select and use the appropriate flat field spectrum automatically However the temporal proximity criterion used in the pipeline might not be the optimal one In case of bad fitting results the user is encouraged to choose another flat field for the
16. However the first bin 0 to 0 03 and the last bin 0 98 to 1 00 will be merged in the program resulting in a phase bin 0 02 to 0 03 or equivalently 0 98 to 1 03 one period later Therefore the above example is a 6 phase bin analysis despite the Number of phase bins 7 The number and boundaries of the phase bins are clearly written in the logfile that you should consult in case of any doubt see below The following parameter Subtract an off phase bin allows to specify an off pulse phase bin number The counts from the off pulse phase bin will be subtracted to the counts of all other bins after being properly scaled to their actual phase bin sizes The phase bin are numbered starting with 0 This option can be used with regularly or irregularly spaced bins although in general an off pulse phase bin will have to be defined using specific bounds Do not forget that in the case the first and last bins are merged see above Below is the log file output for the above case Again the parameter Number of phase bins is 7 in this example 8 boundaries are provided but it is a 6 phase bin case in reality numbered from 0 to 5 In this case Subtract an off phase bin was also selected with a Bin number equal to 4 ISDC SPI Analysis User Manual Issue 9 0 39 Number of phase bin 6 Phase bin No O from 0 000000 to 0 030000 and from 0 980000 to 1 000000 Phase bin No from 030000 to 0 200
17. Sometimes an additional keyword can be put at the end of the list to modify its interpretation For example 1785 1802 1802 1815 1815 1826 keV selects 3 energy ranges in keV This possibility is a local extension of the IRAF standards Unless specified in the parameter description all parameters are required to appear in the param eter file even if their value is not used by the program Blank lines and lines beginning with comments are ignored 8 7 4 Suggested settings for point source analysis The parameters as done in the default parameter file should be set in order to have e an isotropic background component e the contributions from those point sources which are within 20 from the optical axis of the instrument for a given science window The isotropic component is modelled as the product of a common scaling factor and a relative sensitivity for each of the 19 detectors of SPI taking into account the three detector failures during the course of the mission so far The relative detector sensitivity is permitted to vary on 72 ISDC SPI Analysis User Manual Issue 9 0 a timescale of 10 revolutions nearly 30 days as a compromise between assuming no variability at all resulting in a degraded fit and the minimum reasonable variability timescale of one revolution approx 3 days resulting in larger uncertainties due to the increased number of fit parameters Assuming a shorter timescale than 30 days did
18. always be consulted later A large part of these information provided duplicates those found in the ISOC AO documentation 3 We recommend to start with the overview and cookbook in Sect 5 and 6 respectively The cookbook is meant to be a hands on experience It guides you through a practical example and provided some limited explanatory information Once you have exercised the cookbook example browse the Tips and Tricks section Sect 7 and start your own analysis Come back to the cookbook and the Tips and Tricks sections when you are confronted with problems or difficult decisions These two sections should provide enough information for users of the main SPI analysis script the so called spi_science_analysis and pimodfit analysis Since the analysis of SPI data can be difficult in crowded fields or when there are variable source it is important that the user reads carefully the documents provided by the SPI team 10 in order to understand how these situations should be handled The documents refer to the older pipeline spi science analysis but the general principles are valid also for the alternative method spimodfit delivered with version 8 0 of the OSA package The documents are available at the page http www isdc unige ch integral archive For more expert users who want to develop their own scripts or to execute individual programs independently we advise to continue with Sect 8 and mo
19. coded and both nofsources and kofsources get default values at this stage After that the script copies from the spi science analysis par all spiros parameters and supersede the default values of these parameters In this way the value of spiros nofsources replaces the default value of nofsources However the default value of kofsources Kind OF SOURCES POINT cannot be changed as spiros kofsources is missing in spi science analysis par A nice feature however is that it can easily be added Take your favorite editor copy the kofsources line from spiros par into the spi science analysis par and add spiros_ before kofsources Save the spi science analysis par and launch the GUI make ISDC SPI Analysis User Manual Issue 9 0 59 sure the spi_science_analysis par edited is at the right PFILES location The parameter now ap pears at the bottom of the main GUI or at the bottom of the Hidden GUT if it is a hidden param eter The GUI does not look very nice but you can now modify the value of spiros kofsources and it will be transmitted to spiros as kofsources You can also add spiros_kofsources on a command line run After your run you can check the actual value of kofsources in the spiros logfile 7 21 GTIs in SPI data analysis Current SPI data analyses carried out with the spi science analysis script are not making use of any GTIs
20. directory Now the grb_analysis par file located in the current directory now contains the main parameters you want to use in your burst analysis This file will be used by the script Run spi grb analysis Without any command line argument you get the following help text You know the input parameters from the first point above so you can then launch the spi_grb_analysis script Make sure the grb _start sec_to_avoid_before_grb is not before the pointing start and that grb_stop sec_to_avoid_after_grb is not after the pointing end as there is no corresponding checks in the spi_grb_analysis script In our example spi_grb_analysis 2003 05 01T03 10 10 000 2003 05 01T03 10 30 000 UTC 10 10 You can re run in the same directory spi grb analysis as many times as you wish changing the parameters as explain in above section 5 However when making a new run the outputs of the previous run will be deleted If you want to keep your outputs before making a new run save them by copying the entire directory e g cd cp r grb030501 grb030501 spectra cd grb030501 before continuing After a successful run of spi grb analysis step 7 you obtain a spectrum of the burst called spectrum your GRB ID fits containing a link to the appropriate RMF response ISDC SPI Analysis User Manual Issue 9 0 Figure 41 Outputs of spi_grb_analysis when invoked without any input parameter
21. disabled e Detector 2 died on 6 December 2003 2003 12 06 revolution 140 e Detector 17 died on 18 July 2004 2004 07 18 Revolution 215 4 ISDC SPI Analysis User Manual Issue 9 0 Figure 3 The passive mask of the SPI instrument The bottom picture indicates the direction of the spacecraft Y and Z axes with respect to the mask e Detector 5 died on 19 February 2009 2009 02 19 Revolution 776 The death of the detectors results in a decrease of the effective area of the instrument roughly by 1 19 or 2 19 from revolution 215 on the effective area is about 90 of the original area In order to recover from the radiation damage to the Germanium crystal structure every 6 to 12 months the detectors go through an annealing operation They are heated to 105 C for a few days The instrument is not available for scientific observations during the time needed for the annealing operation and the cooling phase afterwards The signals from the detectors pre amplifiers are sent to the amplification chain which is made up of a Pulse Shape Amplifier and a Pulse Height Amplifier The pulses are amplified in such a way that the performance of the spectrometer is optimized This is done by making a compromise between getting the best signal to noise ratio for the pulses operating in the full 20 keV 8 MeV energy band of the instrument and making the output pulses insensitive to the fluctuations in the detector signal rise ti
22. displaying source spectra produced with spiros The result appears in a separate ROOT display where you have a possibility to zoom write output into a postscript and much more You will see a menu by clicking on the right button of the mouse Note that depending on the mouse position near one of the axes in the middle of the window etc the menus will be different Parameters of the script are given in Table 12 There are no hidden parameters in the script Table 12 spi spectral display parameters Name Type Description obs group string DOL of the observation Group to work with default og spi fits GROUPING spec_dol string DOL of the spectrum to be displayed You have not to set it when an observation group containing spectra is given default ebounds dol string DOL of the energy boundaries You have not to set it when an observation group containing spectra is given default main title string Main title of the plot appears only on the top of the display default Spectral plot perkev boolean Flux units possible values yes for displaying the spectra in counts cm 2 sec keV no for displaying the spectra in counts cm 2 sec default yes loglog integer Defines the scaling mode for the X and Y axis possible values 0 linear linear scaling 1 log linear scaling 2 linear log scaling 3 log log scaling Instead of using this parameter you should
23. gaps i e 20 40keV and 80 100keV in your binning you simply set the number of bins in the region in between to 0 Figure 11 Energy bin definition GUI energy_d efinition Parameters for task spibounds Number of energy regions 1 a Regions energy boundaries 20 40 Numbers of bins in each region E For our example we use a 20 40keV energy region with one bin The histogram option allows to run phase resolved analyses or alternatively to add a simu lated source in the mono energetic analysis For this first example do not change the default values and leave both options un selected They will be described in dedicated sections The background can be modeled using several options a flat fields b templates c tracers In this example we use the default flat field option which is recommended by the SPI team Pre defined flat field spectra are available in the ic spi cal directory of the REP_BASE_PROD They were derived by binning in 0 5 kev bins all events from empty field observation revolu tions i e either the position of a known counterpart from other wavelength catalogues or a position derived from other Gamma ray observations with higher spatial resolutions such as the IBIS ones 20 ISDC SPI Analysis User Manual Issue 9 0 Figure 12 Main background GUI 000 X Background options Figure 13 Options for the flat field background ISDC SPI
24. generation is to provide a model for the time and energy variations in order to limit the number of background parameters in a more realistic imaging solution In the default GEDSAT mode the background is assumed to be proportional to the saturated detector trigger rate Up to five different components may be included into background model The computed model background rates are averaged over the corresponding period of good times These rates are then multiplied by the live time to produce the final background counts stored in the output data structure The resulting background rates detector spectra are formatted as the events detector spectra More details can be found in the spi obs back User Manual which can be downloaded from the following page http isdc unige ch Support spi doc doc html Table 9 spi_obs_back parameters included into the main script Name Type Description Main Script spi obs back nmodel integer Number of background model components possible values 1 5 default 1 spi obs back modelOk string Model type of component k possible values CONST ONTIME DEADTIME DEADLIVE ADJA CENT VETOGATE VETONONSAT VETOSAT GEDRATE GEDSAT default GEDSAT k 42 ADJACENT k 2 spi obs back mpar0k string Model parameters of component k default spi_obs_back_norm0k string Component k normalization type possible
25. in these modes will be biased One possible indication that a source flux change with time is when the analysis residuals remain large even after all cares have been taken in the analysis see next section In these cases we recommend to switch to the SPIROS timing modes imaging timing to obtain images timing spectra to obtain spectra and timing to get light curves and to let some of the suspected sources vary during the observation To tell spiros to let a certain source vary open your source catalogue with fv and edit the SEL FLAG column and set it to 2 normally this column contains 1 one for the sources to be used and 0 for the sources to be ignored by spiros It is probably best to first let vary only one source the brightest of the suspects and to check the residuals again If the residuals are still too large try with a second source and so on But do not forget that as usual the best solution will be the simplest model capable of reproducing the observed counts with statistically reasonable residuals Note that it is a particular bad idea to simply set many or ISDC SPI Analysis User Manual Issue 9 0 55 all sources to variable In such a case spiros will be unable to derive a reliable solution and the results may be arbitrary A detailed discussion with exmaples is provided in two separate documents SPI Compact Source Analysis and SPI Complex Case Analysis both available from the Documentation sec
26. is limited to finding the two brightest sources Remember that SPIROS maximum number of sources parameter is meant in addition to those specified in the catalogue Once you have explored your data set it is a good practice to put all well detected sources in the catalogue possibly with precise positions from the reference catalogue and to ask SPIROS to search for 1 2 additional sources You may then check whether these additional sources are real or not for example by cutting your data set into two independent ones and checking if these additional sources are found at the same position in the two independent analyses 7 10 Spectral extraction In SPECTRAL mode SPIROS simply needs to know the source positions provided through the source_cat file at which to extract spectra The only important parameters in this step are the background method and the optimization statistics and the remarks made before for image reconstruction do apply here as well An RMF response is generated and attached to the derived spectra so that they can directly be used in XSPEC for spectral fitting Extracted spectra are called spectrum sourcelD fits 7 11 Variable sources and timing analysis A strong limitation of the SPIROS imaging and spectral mode lies in the underlying assumption that the source flux is constant over the whole observation period If this assumption is wrong and a relatively bright source does vary significantly the results obtained
27. locations and fluxes before their counts are subtracted from the input counts IROS repeats the iterative procedure above locating each new source approximately always allowing them to float to more optimal locations creating new residue count data and stopping when nothing significant can be found This procedure can be repeated in any number of energy bins to calculate a mean location for each source over the entire spectrum range For the user running spiros in its IMAGING mode the main task is to get the appropriate binned count data output from spi_obs_hist and decide how many sources to look for The user may have an input catalog of known sources to get their values without calculating the position or to find an extra one or two in addition or just to make an image of them 8 6 2 SPECTRAL extraction mode The SPECTRAL mode of spiros is even easier to use as it does not search for sources but gets their locations and width from an input catalog which is located under spi source cat fits in case the default names from the spi science analysis par file have been applied Sources which have a SEL FLAG 1 in this input catalog will be taken into account A different input catalog than the one attached to the Observation Group can also be given via source cat dol parameter spiros finds the flux values of all of the flagged sources simultaneously but independently in each energy bin of the input observation count spectrum This method of
28. low energy range through the following polynomial 1 E co X sa te tc x PHA c3 x PHA while for the high energy range only the following linear relation is used E co c x PHA By default the parameter randomize is set to yes and the executable performs a randomization of event energies over the instrumental channel width using a uniform random number generator If randomise no then the exact energy of the channel center is attributed to all events of a given PHA channel More details can be found in the spi gain corr User Manual which can be downloaded from the following page http isdc unige ch Support spi doc doc html 8 3 Pointing definition SPILOBS_POINT The executable spi obs point defines time intervals with stable attitude during the observation More details can be found in the spi obs point User Manual which can be downloaded from the following page http isdc unige ch Support spi doc doc html 8 4 Energy bin definition and events binning 8 4 1 SPIBOUNDS The executable spibounds is used to construct energy boundaries to be used for the binning of SPI event data Options include uniform linear or logarithmic spacing as well as superimposing sub regions e g of finer spacing over a global binning scheme such that you can investigate more carefully particular spectral regions e g for spectral regions likely to contain line emission Table 7 spibounds parameters included into the main script
29. or spectra are our particular format to store binned events but that these count spectra may comprise only one energy bin in case of imaging analysis The next background step is phase bin independent since the background is eventually scaled to the data in the image reconstruction and consequently it is executed only once Spiros is run independently for each phase bin in turn and one set of results is produced for each bin Again spectra and their index and numbered with introducing a phase_i to names used for phase independent analyses where again i is the phase bin number The same RMF response file spectral response rmf fits is valid for all phased resolved spectra and automatically attached to them and therefore the XSPEC spectral fitting of these spectra is straightforward 40 ISDC SPI Analysis User Manual Issue 9 0 6 8 1 Input ephemeris file Ephemeris is provided to the program through a FITS file because in the case of the Crab one set of ephemeris parameters per month is required to follow the Crab phase over periods much longer than one month The FITS table includes one ephemeris parameter set in each row with a validity time interval specified through MJDSTART and MJDSTOP The program can then automatically select a valid ephemeris parameter set in a given row for each processed science window You just have to make sure that the validity intervals encompass the time span of all data included in
30. possibly the SEL FLAG need to be changed as follow e open the SPIROS output catalogue with fv e click on Table of SPI SRCL RES e check that the column SEL FLAG third but last column is 1 for all sources you want to fix in your analysis source with SEL FLAG 0 will be ignored e click on Header of SPI SRCL RES e change EXTNAME to SPI SRCL CAT e change ISDCLEVL to CAT T e save the changes Alt S 7 20 Adding a new parameter in a spi science analysis analysis The spi science analysis calls a number of individual programs and feed them with the relevant parameters with a certain logic Let us take a concrete example the program spiros for the sake of clarity The spiros parameters are listed in spiros par found at the PFILES location We have selected the most important spiros parameters when building the spi science analysis script and these parameters are named spiros parameter s name in spi science analysis par For example nofsources from spiros par becomes spiros nofsources in spi science analysis par However the spiros parameter kofsources from spiros par is not in spi science analysis par be cause we decided it was not a parameter user should normally change Remember that users can only directly access the parameters of spi science analysis par either from the GUI or from the command line In the spi science analysis code all parameters are first specified i e hard
31. prime results of the package are fitted model parameters with their uncertainty complementing these the fitted models are given also in the forms of skymaps and spectra This program is based on a former model fitting program for SPI spidiffit developped by A Strong MPE Garching The main changes in spimodfit w r t spidiffit are e amore flexible handling of time variability for all kind of sources e time dependent instrument response function IRF e the use of NAG free constrained minimization procedures and the addition of a Levenberg marquadt algorithm e optimized matrix calculations and memory usage e more output files updated catalog of sources background model residues etc spimodfit is mainly dedicated to the study of long duration observations large scale emission focusing on the emission morphology For the details of the fitting methods and the statistical issues we refer the reader to the handbook of spimodfit from which this section of the handbook is extracted ISDC SPI Analysis User Manual Issue 9 0 69 8 7 1 Use of spimodfit From the user point of view the data processing is done in 4 major steps which all have their own options e Definition of the input observation data on input spimodfit expects the different files con stituting a SPI observation events rate poitings dead times good times energy bins and optionally background models The energy bins can be redefined and will b
32. spi science analysis script creates a new log file with a spisa UT date log naming scheme each time it is invoked In our case the first log file contains outputs from the previous analysis without catalogue inputs while the second file includes outputs from the catalogue extraction step In the third file you can find the outputs of the spiros with catalogue inputs run below the usual heading Looking in some details at this log file you will notice that spiros provides the name of the catalogue that is used and that it is explained that one source was provided through the catalogue while two other sources have been detected The results images source res fits log files are almost identical to those of the previous case however you will notice that the Crab position in source res fits as well as im the log file is exactly the same as the catalogue one ISDC SPI Analysis User Manual Issue 9 0 31 6 5 Spectral Extraction Before starting you may want to save your imaging analysis again by copying the current directory cd cp r spi analysis spi imaging with catalogue cd spi analysis The SPIROS program can only extract spectra at fixed known positions Therefore a source cat fits file must be provided in this case We assume that you have created checked a source cat fits as explained in last section Launch spi science analysis de select the catalogue extraction step and provide the inpu
33. the analyzed observation group We are aware that using a FITS file may be inconvenient for most X ray pulsars for which only one set of ephemeris parameters is needed For the time being this set has to be provided through a FITS table with a unique row with a very wide validity interval covering the whole INTEGRAL mission Ephemeris files for the Crab V0332 53 and 3A0535 262 can be found in http isdc unige ch Instrument spi download pulsar ephemeris Crab ephemeris fits http isdc unige ch Instrument spi download pulsar ephemeris V0332 53 ephemeris fits http isdc unige ch Instrument spi download pulsar ephemeris 3 40535 262 ephemeris fits The meaning of the ephemeris parameters are standard information can easily be found trough the WWW for the Crab in particular look at http www jb man ac uk pulsar crab html For other not so fast pulsars some of the parameters such as second derivative can be set to zero 6 9 Gamma Ray Burst imaging In the following cookbook section we explain how to derive images of short bursts such as GRB from SPI data We assume that readers have some knowledge about SPI data analysis with the ISDC spi science analysis script Please follow the first cookbook example if it is not your case 1 The first step is to record the start grb start and stop grb stop times of the period you want to use to derive the image and possibly the time before sec to avoid before grb and after sec
34. the pipeline Depending on the number of science windows the histogram building and image deconvolution can take quite some time Once the pipeline is finished no errors should be apparent you find in your directory amongst other output files the following three kinds of images i e intensity significance and error spiros_image_intensity_result fits spiros image sigma result fits spiros image error result fits Before examining these images convert the output catalogue produced by spiros into a format usable by ds9 cat2ds9 source res fits source res reg ds9 spiros image intensity result fits region source res reg ds9 spiros image sigma result fits region source res reg After adjusting the color tables etc the resulting intensity and significance image should look like this ISDC SPI Analysis User Manual Issue 9 0 27 There are three sources in these images In addition to the obvious strong Crab detection spiros has found two spurious sources with very low significance In most cases spiros does find spurious sources with detection significance in the range 3 to 5 sigmas Great care should be taken to make sure that sources with such detection significance are indeed real and if you only want to see sources detected beyond any doubt raise the spiros lower sigma threshold to higher values 6 is usually a good choice The output file source_res fits provides quantitative results formatted as a FITS t
35. to avoid after grb the burst that should not be included in the period used to derive the background Times can be in UTC or IJD The best way to look at the timing is to display the burst light curve for example from IOSM Of course you can repeat the analysis in as many time bins as you wish e g to derive a set of images and a SPI light curve Figure 30 provides an illustration of the required times on top of a real GRB light curve 2 Set up your analysis environment as indicated in the Setting up the analysis environment section of the SPI data analysis cookbook 3 Identify the pointing encompassing the burst and use og_create to build an observation group only analysis of single pointing are possible in this frame as shown in Fig 31 This produces an observation group with a standard name og_spi fits located in obs grb030501 in our example 4 Move to the observation group location cd obs grb030501 in our example 5 Two alternatives can be used to specify the parameters of your analysis The simplest is to use spi science analysis GUI a Run spi science analysis and specify parameters through the GUI as displayed in Fig 32 ISDC SPI Analysis User Manual Issue 9 0 41 Figure 30 Illustration of the different GRB start stop times 10000 ScW TSTART SCW TSTOP El z O e a Ao 8000 9 9 a lt g 9 e A 3 6000 a 9 ta BACKGROUN i BACKGROUND 03 11 08 14 11 10 11 20 11 30 11 40
36. unige ch Soft download osa osa_sw osa_sw 7 0 osa_inst_guide 2 0 4 pdf 3 SPI observer s manual http astro estec esa nl Integral AO5 AO05_SPI_om pdf 4 The INTEGRAL spectrometer SPI performance of point source data analysis Dubath P Kn delseder J Skinner G K et al 2005 MNRAS 357 420 http isdc unige ch Instrument spi papers mnr8675 420 428 pdf 5 Jiirgen Kn dlseder web page devoted to SPI software http www cesr fr jurgen isdc index html 6 Connell P H Documentation of the SPIROS software package for imaging source location spectral extrac tion and timing analysis using observation data from the INTEGRAL spectrometer SPI http isdc unige ch Instrument spi doc manuals SPIROS_documentation ps A Strong A spiskymax User Manual and Explanatory Supplement http isdc unige ch Instrument spi doc manuals spiskymax ps gz 8 Walter R Favre P Dubath P et al 2003 A amp A 411 L25 9 Beckmann V in the proceedings of 27t Rencontres de Moriond p 417 2002 astro ph 0206506 10 SPI Compact Source Analysis prepared by the SPI Team SPI Complex Case Analysis prepared by the SPI Tea ISDC SPI Analysis User Manual Issue 9 0 79
37. use the functionality which is implemented in the ROOT display i e set loglog 0 and apply logarithmic scaling clicking the right mouse button above the displayed spectrum default 0 chatter integer Chatter level possible values 0 10 O no screen dump information 10 maximal screen dump default 1 76 ISDC SPI Analysis User Manual Issue 9 0 9 2 SPI DSP DISPLAY This script is displaying detector spectra using root BIN I level or higher You can use this on data which are of The maximum number of spectra displayed is 20 As in the case of spi spectral display result appears in a separate ROOT display see discussion in Section 9 1 Pa rameters of the script are given in Table 12 There are no hidden parameters in the script Table 13 spi dsp display parameters Name Type Description obs group string DOL of the observation Group to work with default og spi fits GROUPING dsp dol string DOL of the detector spectrum You have not to set it when an observation group containing spectra is given default ebounds_dol string DOL of the energy boundaries You have not to set it when an observation group containing spectra is given default detector integer number of the detector to display the spectrum possible values O 141 also you can use 1 for all detectors and 2 for their sum default 2 pointing integer pointing number to be dis
38. values NO GLOBAL DETE LINE CONT DFFLINE DFFCONT default NO spi_obs_back_npar0k string Component k normalization parameters default spi obs back scale0k real Component k scaling factor default 1 0 64 ISDC SPI Analysis User Manual Issue 9 0 8 6 Image reconstruction and spectral extraction SPIROS A detailed description of spiros is given in the Spiros User Manual written by the author of spiros P H Connell 6 Here we give just some brief explanations and describe only the parameters used by main script Note that even for the explained parameters we do not give the full range of possible values SPI is a coded mask instrument so direct deconvolution is in principle possible but in practice the response is complex and the data include many pointing directions of the instrument so indirect imaging is essential Indirect imaging implies forward folding for any candidate image we convolve with the instrument response function and compare the result with the observed data This gives the basis for any iterative method which seeks to successively improve the agreement of the predicted with the observed data by adjusting the image One approach is to regard the image as made up of point sources and to adjust their positions and fluxes to give a best fit to the data This is the basic principle of the spiros method The primary purpos
39. wants to run spimodfit another time since the program does not replace existing files In Fig 43 we show the GUI with the standard settings for internal background computation no image fitting and standard background multipliers 8 7 3 Use and syntax of the parameter file spimodfit can be used from the command line after preparing the input data This modality is suggested for advanced use A script spimodfit rmfgen csh is created to call spirmf for the adopted energy binning The script should be run after spimodfit from the command line Thttp www mpe mpg de gamma integral ISDC SPI Analysis User Manual Issue 9 0 71 The parameters of the the program are defined in a file called spimodfit par This file should be readable in one of the directory set in the PFILES environment variable Usually the user updates the parameter file in the working directory so one should set the variable accordingly setenv PFILES The syntax of the file follows the IRAF standards that is the one used by ISDC softwares In this file each parameter is described in the format parameter_name parameter_type parameter_mode default_value minimum_value maximum_value description_prompt 665 99 The parameter_type can be b boolean i integer r real s string or f filename bh The f type can be followed by any combination of r read access w write access e ER
40. will be detected in a given observation In order to avoid specifying wrong positions where no significant sources can be detected it is a good idea to first explore your data set and see which sources can be detected without using any prior information De select the catalogue extraction and leave the SPIROS input catalogue entry empty we explain how to use this step in next Sect 6 4 Select the pointing definition event binning background modeling and image analysis tasks and click on the corresponding buttons to specify task specific parameters 1 For the pointing step you can use a pre defined list of bad pointings to automatically exclude from your analysis pointings with known problems such as high background level or out of nominal instrument characteristics This option is selected by default and we strongly recommend not to un select it or only for testing purposes In the energy definition Fig 11 you have three parameters In the first field you can define a number of energy regions In the second one you specify the actual limits in keV of these regions i e 20 40 60 is for two regions 20 40keV and 40 60keV In the third field you provide the number of bins for each energy region positive numbers provide linear scaling negative numbers logarithmic scaling This allows for a more elaborate definition of the binning than a simple logarithmic or linear binning over the whole range If you want
41. 0 012 4 E 0 01 4 0 008 4 0 006 4 t co A ne ota 1322 65 1322 7 1322 75 1322 8 1322 85 1322 9 TIME d Figure 21 Crab light curve from the cookbook example 6 7 Spectral Extraction using spimodfit First set up the environment and create the observation group as detailed in the sections for spi_science_analysis Then select a suitable catalogue for the observation you are analyzing Using the cookbok example the name is source_cat fits you should leave this name In our example we set the column FLAG to the value 1 for Crab Then start spimodfit_analysis and select the steps POINT BIN_I and FIT Fig 22 The user can skip the background step since spimodfit handles it internally Then click on energy definition and select the number of channels you want in the final spectrum we chose 22 logarithmic bins from 25 and 300kev Fig 23 Close this window and open the window on spimodfit options Fig 24 and choose correctly in the upper left frame the selected bins according to your binning choice You can rebin at this stage with an integer number of bins All the opther parameters can be left to the default values Several files are produced as output one should check the fit quality on the program log spi_sa_DATE log where DATE is the start time of the analysis variable from run to run For each energy range there is the value of the x2 and of the C statistics at optimum These should be around one if a
42. 000 Phase bin No 200000 to 0 380000 Phase bin No from 380000 to 0 500000 O from O O Phase bin No from 0 500000 to 0 800000 O 4 PONE Phase bin No 5 from 0 800000 to 0 980000 Counts from phase bin 4 will be subtracted from all spectra Another example if you select Equal bin width the parameters are less confusing and with a Number of phase bin 5 the log file contains the following outputs Number of phase bin 5 Phase bin No O from 0 000000 to 0 200000 Phase bin No from 0 200000 to 0 400000 Phase bin No from 0 400000 to 0 600000 Phase bin No from 0 600000 to 0 800000 Phase bin No from 0 800000 to 1 000000 PONE Finally an orbit can be specified in the case of a pulsar with a companion X ray binary by entering orbit parameters in the orbit window Fig 29 Leave zeros if there is no orbit Figure 29 Orbit parameter GUI TAAA gt IX Orbit m Orbital parameters Ok Asini 1 E Help Orbital period days oy 5 Al Orbit epoch T90 parameter oy Orbit eccentricity 1 F a ES Orbit omega_d parameter 0 2 a ES Orbit pporb parameter 0 E The phase resolved analysis is then executed after you have clicked three times on Ok buttons The binning step produces one set of detector spectra for each phase bin called evts det spec phase i fits where i is the phase bin number Note that these files named spec
43. 03 11 11 12h 2003 11 26 20h 1411 1426 132 136 4 2004 06 17 09h 2004 07 02 18h 1629 1645 205 209 5 2005 01 19 09h 2005 02 05 06h 1132 1146 277 282 6 2005 06 14 00h 2005 06 30 00h 1991 2008 326 330 7 2006 01 09 00h 2006 01 26 00h 2200 2217 395 400 8 2006 06 08 00h 2006 06 27 00h 2350 2369 446 452 9 2006 12 04 10h 2006 12 22 15h 2529 2547 506 511 10 2007 05 29 12h 2007 06 17 17h 2705 2724 565 571 11 2008 01 12 04h 2008 01 31 07h 2933 2952 641 647 12 2008 08 17 04h 2008 09 05 19h 3151 3171 714 720 13 2009 04 20 08h 2009 05 11 23h 3397 3418 796 802 14 2009 10 19 18h 2009 11 09 9h 3579 3600 857 863 15 2010 03 30 9h 2010 04 17 1h 3741 3759 911 916 between the GTI stops and starts 7 17 Crowded fields When the number of sources in the field of view are larger than a few 3 4 the analysis of SPI data becomes increasingly difficult especially when sources are separated with angles comparable to the SPI resolution of about 2 degrees SPIROS can get confused and attribute the counts due to one source to another A good test is to add a mock source or a few in the middle of the field and to compare the flux extracted by SPIROS with the input one It is possible to derive the contribution of a mo
44. 2 Energy boundaries or ebounds GUI 00 020000 43 Background GUL ss wy ate hs RS E Bo GE ES 43 spiros GUI main Window 44 spitos GUI imaging Window 44 Outputs of spi grb analysis when invoked without any input parameters 45 Main GUI window of the spi science analysis script 47 Energy boundaries or ebounds GUI e eee 47 spiros GUI main Window 48 Outputs of spi grb analysis when invoked without any input parameters 49 GUI with the standard settings for a flat field background and logarithmically in creasing energy binning 2 2 2 a 71 GUI with the standard settings for internal background computation no image fitting and standard background multipliers 00 0 71 ISDC SPI Analysis User Manual Issue 9 0 List of Tables 10 11 12 13 Main characteristics of the SPI instrument 000 511 keV background line strength with and without PSAC Overview of the successive steps of the spi science analysis script SPI detector failures sos uso same aaa e a we AD E SPL annealing periods s sm js as anana tote te ata HL ae BS OH GS SL cat_extract parameters included in the main script spibounds parameters included into the main script spi_add_sim parameters included into the main script 0 spi_obs_back parameters included into the main script
45. 7 102 26 2198 8 30 9 18 0 2 16 1 72 0 00 8 102 27 2196 8 36 9 18 0 3 15 2 05 0 00 9 102 28 2194 8 22 7 18 0 0 79 1 26 0 00 10 102 29 2189 8 16 6 18 0 0 24 0 92 0 00 Ptg Rev Exp ONTIME CHI2 ML Expected Diff Reduced Data no no no secs value value STD CHI2 ML excl 28 ISDC SPI Analysis User Manual Issue 9 0 1 102 20 2155 2 19 1 16 3 0 49 1 17 0 00 2 102 21 1932 7 15 7 16 3 0 11 0 96 0 00 3 102 22 2200 8 43 5 16 3 4 76 2 67 0 00 4 102 23 2198 8 31 5 16 3 2 66 1 93 0 00 5 102 24 2196 8 19 3 16 3 0 53 1 19 0 00 6 102 25 2143 8 18 5 16 3 0 38 1 13 0 00 7 102 26 2198 8 18 3 16 3 0 36 1 12 0 00 8 102 27 2196 8 19 8 16 3 0 61 1 21 0 00 9 102 28 2194 8 30 9 16 3 2 55 1 89 0 00 10 102 29 2189 8 9 7 16 3 1 16 0 59 0 00 The most important information to consider is Diff STD which gives the difference expressed in term of standard deviation between the CHI2 or ML resulting values and the expected ones In our case these residuals appear reasonable for all pointing with the exception of number 3 with 5 24 sigma When using your own data you may find some pointings with very large residual values DO NOT TRUST THE RESULTS if some of the involved pointings have residuals much larger than 3 standard deviations although see Sect 7 for possible exception to this rule You must get rid of these residuals The first thing to do is to use the AUTO filter of spiros Launch spi science analysis again de select all tasks except spiros a
46. 7 Spectral Extraction using spimodfit o oo 35 6 7 1 Furthersettines tn trap Mette ted ducted A ATA de e di de de do a 36 6 8 Phase Resolved Analysis LL LL LL a 37 6 8 1 Input ephemeris file LL LL LL LL e 41 6 9 Gamma Ray Burst imaging e 41 6 10 Gamma Ray Burst spectroscopy e 46 Tipsiand Tricks sf a ri a E T o CR ist te Ca aunt he a A 50 7 1 Example of command line run 0 00000 eee 50 Ed Repeating and saving analyses e ee ee 52 7 3 Pseudo detectors and detector list o oo ooa ee 52 7 4 Catalogue extraction e 53 7 5 Event energy correction e 53 7 6 Event binning GTI and dead times o a 53 7 7 Background modeling ssa sa ee ee ala e RR b 54 7 8 Image reconstruction ecos Fae ee aaa aca a a Ra ee a 54 7 9 Biases from bright sources just outside the image FOV 55 7 10 Spectral extraction e 55 7 11 Variable sources and timing analysis LL LL LL LL a 55 ISDC SPI Analysis User Manual Issue 9 0 7 12 Converging on a good solution checking residuals 56 7 13 High time resolution with SPT o e 56 7 14 SPI failed detectors ao spa a a ee we AA 56 oo Annealing phases Tatie ag A es ee do do do a ES 57 7 16 Instrument modes and SPI data o 57 TAT Crowded fields GA ssa Be ee de de da ee Ae Beate oe die E a AE DS A 58
47. C Finally after every detector annealing a thorough check is done of the instrument imaging and spectroscopic response since these may change as a result of the annealing process 4 3 Measured Performance 4 3 1 Imaging Resolution The design of the instrument is such that the angular resolution for isolated point sources is about 2 5 FWHM The location of point sources can be done with an accuracy much better than this however depending on the strength of the source As explained above Section 3 10 dithering is required for imaging more complex regions 10 ISDC SPI Analysis User Manual Issue 9 0 4 3 2 Spectral Resolution The spectral resolution has been measured in the laboratory with detectors that are representative of the flight units and afterwards with flight model detectors and pre amplifiers An example spectrum obtained in this way is shown in Figure 6 The measured energy resolution as a function of energy for an individual detector is given in Figure 7 capture Fe n gamma 600 coublet 7631 1 7645 5 keV sop aop ES E D ami 300 200 100 a iy 0 L al L 4 L 6600 6800 7000 7200 7400 7600 7800 Energy keV Figure 6 Example spectrum taken with laboratory detector units representative of the flight units 4 3 3 Sensitivity The continuum and line sensitivities of the SPI instrument are given in Figure 8 and
48. Figure 9 The continuum sensitivities are for AE E 2 and are calculated from the narrow line sensitivity by dividing those by VR AE where R is the instrument resolution for lines The line sensitivities are fluxes in photons cm 7 s 71 the continuum sensitivities are fluxes in photons cm 7 s 7 keV 7 The line sensitivities are for narrow lines The 511 keV sensitivity is worse than the surrounding continuum due to the strong 511 keV background line originating in the instrument 4 3 4 Dithering Sensitivity Degradation The instrumental sensitivities given in Figures 8 9 are for a source on axis and do not take into account dithering As stated earlier observations with SPI should not be done in staring mode since this makes the identification and removal of the background impossible Dithering on the other hand has the disadvantage that the source is not observed for the full integration time in the center of the fully coded field of view The SPI response falls off towards the edge of the field of view and therefore dithering will degrade the sensitivity of the instrument somewhat The hexagonal dither a central pointing with six surrounding pointings in hexagonal pattern all 2 apart only samples the central part of the SPI fully coded field of view Therefore no reduction in the sensitivity is noticeable The 5 by 5 dither however samples closer to the edge of the fully coded field of view In this case the sensitivity is degraded by
49. In order to eliminate bad data only entire pointing are excluded from the analysis see Sect 6 3 1 In addition the GTIs require to analysis short bursts such as GRBs can be automatically by the GRB analysis script In some cases you may need to create a GTI anyway and this can be done following the below procedure e Create an empty GTI structure dal create obj_name usergti fits template GNRL USER GTL tpl e fill some keywords if you prefer GUIs use fv fparkey value 0 0 fitsfile usergti fits 1 keyword TSTART fparkey value 10000 0 fitsfile usergti fits 1 keyword TSTOP fparkey value G00D fitsfile usergti fits 1 keyword GTI _TYPE TSTART and TSTOP define the start and end validity of your selection not the selection itself The GTI_TYPE can be GOOD use the time specified in the user GTI file or BAD exclude the time specified in the user GTI file e open the fits file with fv fv usergti fits e click on Table e click on Edit e select Insert and then row e insert as many rows as you want to have good time intervals e fillin the row either the UTC or the IJD or the OBT columns corresponding to your good bad time you want to define e Save the fits file 60 ISDC SPI Analysis User Manual Issue 9 0 8 Data analysis steps As explained in the Overview Section the main script calls a number of executables which are launched in a given order In the present ch
50. In this case however it cannot be guaranteed that the cleaning will work properly 74 ISDC SPI Analysis User Manual Issue 9 0 Table 11 spi_clean parameters Name Type Description obsgroup endlevel chatter force clean_swg string string integer boolean boolean DOL of observation group default og spi fits GROUPING output level of the OG possible values PRP CAT I POIN BIN IT BKG_I IMA SPE default PRP verbosity level possible values 0 5 Chatter 5 is for debugging default 3 Force removal of files possible values no spi_clean will neither work on corrupt OGs nor delete any files using the system command rm yes spi clean will also work on broken OGs and delete any detached file using rm Furthermore spi clean will also try to repair a broken OG in the sense that it checks that links in the OG are ok Corrupt links are removed The level of the OG is set to the highest level of the children of the OG Be careful The results when working on broken OGs can be quite unpredictable default yes Determines whether science window group should also be cleaned not recommended default no ISDC SPI Analysis User Manual Issue 9 0 75 9 Additional tools In this section we describe additional tools which may help you to analyze the results of your analysis 9 1 SPISPECTRAL DISPLAY This script is
51. L instrument teams e Pointing Period during which the spacecraft axis pointing direction remains stable Because of the INTEGRAL dithering strategy the nominal pointing duration is of order of 30 minutes e Slew Period during which the spacecraft is manoeuvered from one stable position to another i e from one pointing to another e Science Window ScW For the operations ISDC defines atomic bits of INTEGRAL opera tions as either a pointing or a slew and calls them ScWs A set of data produced during a ScW is a basic piece of INTEGRAL data in the ISDC system e Observation Any group of ScW used in the data analysis The observation defined from ISOC in relation with the proposal is only one example of possible ISDC observations Other combinations of Science Windows 1 e of observations are used for example for the Quick Look Analysis or for Off Line Scientific Analysis ISDC SPI Analysis User Manual Issue 9 0 xi 1 Introduction The present SPI Analysis User Manual includes two major parts 1 a description of the INTE GRAL SPI instrument and 2 information about the SPI data analysis system provided by the ISDC A more general overview on the INTEGRAL Data Analysis can be found in the Introduction to the INTEGRAL Data Analysis 1 People starting with SPI data analysis are expected to skip the first part of the document as they should already have some basic knowledge about the instrument This part can
52. SA software you must also set your software environment correctly If not already set by default by your system administrator you should set some environment variables typing setenv ISDC_ENV directory_of_OSA_sw_installation setenv ISDC REF CAT REP_BASE_PROD cat hec gnrl_refr_cat_0031 fits 1 source ISDC ENV bin isdc init env csh 3Note that the naming scheme is different for revision 1 and revision 2 data For the revision 1 data the name of the prepared Science Window Group is swg_prp fits instead of swg fits ISDC SPI Analysis User Manual Issue 9 0 17 in order to e set ISDC_ENV to the location where OSA is installed e set ISDC_REF_CAT to the DOL of the ISDC Reference Catalog e run the OSA set up script isdc init env csh which initializes further environment vari ables relative to ISDC_ENV The SPI pipeline creates a log file with all relevant information automatically each time it is executed If you however want to log not only the output from the pipeline but also from all your other OSA activities set the environment COMMONLOGFILE setenv COMMONLOGFILE name log where name log is the name of the log file You can use any name you want but do not forget the in front otherwise no information will be written to the screen any more As your level of expertise with the software increases you may wish not to have the Graphical User Interface GUI pop up when you launch your analysis In this case th
53. UES do s ecce cuana a Baal aS 73 8 8 Response interpolation SPIRME 73 8 9 Image reconstruction SPISKYMAX 02000 74 8 10 Cleaning Tool SPLCLEAN o e e 74 ISDC SPI Analysis User Manual Issue 9 0 v vi 10 Additional tools pumba a ass oe ES eee ee Se a ee ee 4 76 9 1 SPISPECTRAL DISPLAY a 76 9 2 SPIE DSP DISPLAY goto de p eh PR Se SE ES 77 Known limitations gt sic cee Balasore oe EE Ree ee eee teehee amp 4 78 ISDC SPI Analysis User Manual Issue 9 0 List 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 ISDC of Figures A cut away view of the SPI instrument a Numbering of SPI detectors 2 LL LL 0 000 2b ee eee ee ee The mask of the SPI instrument 0 0 0 0 2 2 00000 Dithering patterns and instrument field of views o o The SPI continuum background components 000000 Spectrum example suas ag do IA ar os bol dn ee a a a Epnerew resolution o a la A EAS IS gra Tee A A AS AD Continuum sensitivity a oe a de BS Oh cas ear Narrow line sensitivity suas a em a RE hee gea aa spi science analysis main GU l o 0 000 002 0000 0000 Energy bin definition GUI 2 2 0 00 0 000 200 0020 00008 Main background GU A ida Sadek BABA ele SN A De a Options for the flat field background o o
54. a factor 0 84 ISDC SPI Analysis User Manual Issue 9 0 11 T ree T proto 0 resolution E 35104 E L710 11 FWHM keV 0 1000 2000 3000 4000 5000 6000 7000 8000 9000 Energy key Figure 7 The measured energy resolution of an individual SPI detector This was measured using lab oratory detectors The resolution of the full instrument with all 19 detectors is slightly lower than this ph em a Lev log Fluz do 1 2 3 4 log Energy keV Figure 8 The continuum sensitivity of the SPI instrument for a 3 sigma detection in 10 seconds on axis The fluxes are for E AE 2 The dashed lines indicate extrapolations from the X rays using an powerlaw with photon index 2 for 1 10 and 100 mCrab 12 ISDC SPI Analysis User Manual Issue 9 0 45 ph cmz a th log Fluz a tn 1 2 3 4 log Energy keV Figure 9 The narrow line sensitivity of the SPI instrument for a 3 sigma detection in 10 seconds Note that the 511 keV line is not shown in this figure 4 3 5 Detection of off axis sources The wide field of view of SPI allows the detection of off axis sources However it also means that off axis sources will create a shadowgram on the detector that increases the background photons for the prime target To remove this background a proper mapping of the source and the surroundings is necessary This is the main reason why hexagonal dithering should only be used for isolate
55. a gov docs heasarc ofwg docs ofwg_recomm r11 html ISDC SPI Analysis User Manual Issue 9 0 53 7 7 Background modeling In the most general case the SPI image reconstruction is an intractable problem as it would require solving for one background value for each combination of detector pointing and energy bin in addition to solving for the source parameters This problem has more unknowns than measurements Clearly some assumptions on the background behavior have to be introduced For all point source analyses we recommend to use the default flat field option In this case pre defined flat field spectra are used They are rebinned to match the bin size of your analysis and the program will select and use the appropriate flat field spectrum for your analysis automatically With this option the detector to detector ratios are kept fixed from that of the empty field but one background coefficient is derived through the imaging reconstruction process In fact one coefficient is derived for each time interval and the length of the intervals is provided as a number of consecutive pointing through the next parameter We use a value of 5 in the cook book example The background is then assumed to be constant within these intervals of 5 pointings Note that in general much longer time interval should be used in the range 20 to 100 pointings A good background modeling alternative is provided by either the GEDSAT template or the GED SAT mod
56. able Open this file with the ftool fv to look at the results The resulting spiros Crab position is RA 83 617 DEC 22 019 while the catalogue position is RA 83 633 DEC 22 014 The difference is of the order of 1 arcmin As expected for very high S N sources this value is much smaller than the SPI spatial resolution The Crab flux in this 20 40 keV energy band is 0 175 0 001 ph sec cm corresponding to a detection significance of 159 You can also see that the detection significance of the spurious sources is 5 7 and 3 8 All these results can also be viewed perhaps more conveniently in the end of the log file called spi sa_ UT date log below the following title where you find the most important information about the analysis such as information about the sources found at each iteration and about the resulting images 6 3 1 Data screening Near the end of the spisa UT date log log file you can find a table with chi square values per pointing looking as follow with a slightly different formatting however Contributions to CHI2 parameter by pointing exposure Ptg Rev Exp ONTIME CHI2 ML Expected Diff Reduced Data no no no secs value value STD CHI2 ML excl 1 102 20 2155 2 21 5 18 0 0 58 1 19 0 00 2 102 21 1932 7 19 6 18 0 0 27 1 09 0 00 3 102 22 2200 8 49 4 18 0 5 24 2 75 0 00 4 102 23 2198 8 40 0 18 0 3 67 2 22 0 00 5 102 24 2196 8 19 2 18 0 0 20 1 07 0 00 6 102 25 2143 8 24 9 18 0 1 15 1 38 0 00
57. able containing the results of the fit By default this file contains the different sky models original map and fitted images and a binary table containing the fitted parameters with associated configuration and estimated errors Optionally it can also add an exposure map in this FITS file create a background model for the observation write residues of the model light curves spectra and an updated point source catalog into other files 8 7 2 spimodfit analysis script A set of standard parameters based on the ones used for the SPI quick look analysis available at http www mpe mpg de gamma instruments integral spi www public_data index_srcspe html is implemented in the script spimodfit_analaysis that is delivered for a more user friendly uti lization of spimodfit The first steps before the background calculation are the same of spi science analysis The energy channels of the input data and the set of detectors which are used must be coherently set between the binning and spectral extraction steps Regarding the background the default option is the calculation internal to spimodfit The time variability is set by default to account for variability on science window time scale and at each annealing or detector failure A feature which regards the background multipliers is introduced to account for different efficiency of the pixels The relative default time variability accounts for the annealings and detector failures and a mon
58. al values which you have to remove from your analysis to obtain reliable results To exclude such bad pointings from your analysis note the pointing numbers of the bad pointings Then launch spi science analysis again de select all tasks except spiros Open the Spiros options GUI Enter all good pointings in the pointing subset entry field Le if pointing 7 and 8 of a total of 20 pointings are bad enter 1 6 9 20 as pointing subset Now rerun the analysis You will find that this time the resulting images are much better If you check the log file you ll see that the chi square values of all pointings have improved significantly If you still encounter some pointings with bad chi square values simply repeat the procedure until all chi square values are reasonable But beware it is easy to obtain results with very small residuals by removing all bad pointings from the analysis However you must be very careful when removing pointings all really bad pointings are removed automatically ISDC SPI Analysis User Manual Issue 9 0 29 by the pipeline during the pointing step and by the AUTO filter of spiros Very often when you still encounter bad pointings these are due to an insufficient sky model i e you may have a strong variable source or a transient in the FoV 6 4 Image reconstruction with catalogue information Once you have seen which sources are detected by spiros you can obtain better results by fixing the source positi
59. analysis pipeline uses these files by default for on the fly energy correction in the event binning step program spi_obs_hist This standard energy correction is excellent for most cases even for line analysis only line analysis in revolutions with large temperature changes may require better correction You can however if you want apply your own set of gain coefficients by providing the name of an alternative index of gain coefficient files in the Gain Coefficient file name entry field of the spi_science_analysis GUI click on the upper right hidden button to find this parameter 7 6 Event binning GTI and dead times In addition to the actual event binning this step includes the energy bin definition through the spibounds software the derivation of the GTI and of the dead time GTI and DEAD tasks So far no real GTI filter has been implemented at this stage and the complete time intervals of all pointings are considered as good remember that bad pointing can be eliminated during the pointing step using a list of known bad pointings The actual dead times are provided by the SPI electronic Dead time ratio and live time are defined by LIVETIME DEADC ONTIME where the ontime is the net observing time excluding any gaps in the data taking period such as gaps due to slews or data loses and the LIVETIME is the time corrected from the dead time These definitions are borrowed from http legacy gsfc nas
60. and without the PSAC are given in Table 2 component with PSAC without PSAC continuum 4 9 x 107 5 4 x 107 passive material 2 5 x 1074 2 5 x 1074 shield leakage 8 6 x 107 8 6 x 1078 mask 1 2 x 1075 2 1 x 1074 BGO shield 1 4 x 1076 1 4 x 1076 blocking time Total 3 2 x 1074 5 2 x 1074 Table 2 511 keV background line strength with and without PSAC Calculations are for 5 cm BGO shield 80 keV shield threshold without PSD Fluxes in cts cm s 1 4 1 3 Background Gamma Ray Lines Background gamma ray lines are emitted in passive materials close to the detectors and in the detector material itself Primary and secondary cosmic ray particles protons and neutrons induce excited nuclei in nuclear reactions with nuclei of the passive material The prompt or delayed radioactive de excitation of these nuclei leads to gamma ray lines which can be detected by the germanium detectors Calculations show that lines originating in the mask should not pose a problem for SPI 4 2 Instrumental Characterization and Calibration The SPI instrument has been fully tested and calibrated on ground before the launch Soon after the launch the observations of the Crab nebula and Cygnus X 1 were done for further calibrations The detector gains thresholds and resolution versus energy are determined from normal event data and ACS off spectra for consistency checks in the routine monitoring task of ISD
61. apter we describe some executables in more details in order to explain how the main script works and what parameters you have to enter for a proper analysis Describing the executables we also discuss the parameters used by spi_science_analysis To learn more about a particular executable type its name with the h option Non hidden parameters of the main script are marked with bold font Usually names of the main script parameters are derived from the corresponding name of the parameters of the executables by adding the name of the executable in the beginning e g parameter pointing dol of the executable spi obs point has a name spi obs point pointing dol in the main script In the rare case when this rule is broken we give both names in the description of the executable A detailed description of the products of each step can be found in the Appendix section where also the description of the raw data is given 8 1 Catalog source extraction CAT EXTRACT The executable cat extract performs the source selection from a reference catalog The output catalog consists of the sources that are in the FOV of the pointings included in the Observation Group You can also put additional criteria on the sources that would be selected by setting the desirable position with the parameters radiusMin and radiusMax which specifies a surface obtained by merging the areas lying between radiusMin and radiusMax from the central point of each pointing and the de
62. back model idx fits 1 to use the background computed in the background step The following entry regards this case so that the user can allow for multiple background components this is an advanced feature that should be used with a deep knowledge of spimodfit The entry with the fitted detector ID range allows the user to select different ranges of the detectors to fit the background the choice 00 18 is suggested so that the background at this stage is handled as a constant The entry time variability definition allows the user to define time variability nodes and intervals The default values regard the SPI annealing times and the detector failures plus a variability on science window time scale 0 084 days The button on background multipliers opens a window Fig 25 through which the user can set up the background multipliers This is a features that allows the program to compute a background pattern The relevant parameters are the detector ranges and the Variability coefficients The former is by default the list of all detectors so that each detector can be treated separately The latter by defaults covers the SPI annealing instants and a monthly variability This is the best choice for point sources We suggest not to touch the other parameters of the window unless a specific analysis is pursued 6 8 Phase Resolved Analysis Starting with OSA 6 0 the spi science analysis script offers phased resolved analysis capabilities The main switch
63. bins in principle a single wide band for a first image click Ok to close this window e Click on Background options and un select the flat field option and select the Use tracers to model background option as shown in Fig 34 and click Ok to close this window Figure 34 Background GUI Flateield model detector variation model gt ok Use empty field observations to model background I Help if yes number of pointings with const back F Time variation background models from tracers Number of Models to use 1 f Parameters Modei Parameters Model 2 Model GEDSAT hd Model GEDSAT v Model Parameters Model Parameter Normalisation NO Normalisation No Nom Parameters Nom Parameter Scale 1 Y scale 1 4 Parameters Model 3 e Parameters Model 4 Model GEDSAT Z Model GEDSAT y Model Parameters SSS Model Parameters SSS Normalisation NO Normalisation NO Norm Parameters S Norm Parameters 70 Scale 1 3 Scale 1 Back in the parent window click spiros to open the GUI window shown in Fig 35 g Make sure the imaging mode is specified do not introduce any selection parameters and specify the Background method and the Optimization statistic Background method 1 means that the background derived from the periods before and after the burst and rescaled to the burst duration will be subtracted from t
64. calculating source flux values independently in each energy bin is required as a base line mode of spiros to output raw or photo peak spectra which are then to be used by the XSPEC spectral modeling software to extract their true spectra this is described in the Cookbook Section 6 8 6 3 TIMING analysis mode The TIMING mode of spiros is again easy to use as it does not search for sources but gets their locations from an input catalog as in SPECTRAL mode The idea is that the user will bin count data over some observation period in as many energy bins as desired and that spiros will calculate source flux values at a sequence of MJD times over some integration time interval to be given by the user Note that the data at this level are already grouped into pointings thus no timing analysis can be performed on time scales shorter than the pointing duration At present TIMING mode has three forms of analysis e QUICKLOOK This computes the flux for all sources and background in small groups of pointing exposures covering the integration time interval in parameter source timing scale and in each count spectrum energy bin 66 ISDC SPI Analysis User Manual Issue 9 0 e WINDOW This mode assumes a mean count modulation model MCM or known background variations and computes source flux values simultaneously in a sequence of integration time intervals spanning the period of observation For a large number of integration intervals and ene
65. ck source with the program spi_add sim The detector counts expected from a source at given position and intensity are computed using the instrumental response and added into the binned event data set This step should be disabled for most standard analysis The program only works for single energy bin analysis The user should check the documents provided by the SPI team for further reference 10 7 18 Fitting SPI spectra using XSPEC e Extract a spectrum with the spi_science_analysis script following the instructions provided in Sect 6 5 The spi science analysis script sets all relevant parameters in the resulting spectra creates an appropriate response file and links it to the spectrum e Below is a simple example of an XSPEC spectral fitting session with a Crab spectrum spec trum_Crab fits xspec cpd xs data spectrum_Crab fits setpl energy model power 58 ISDC SPI Analysis User Manual Issue 9 0 2 0 8 0 fit 1000 flux 20 200 plot ldata delchi 7 19 Using a SPIROS output catalogue in input source res fits to source_cat fit One of the SPIROS outputs is a catalogue source res fits containing source positions and fluxes After a SPIROS run you may want to fix the positions of some of the detected sources and run SPIROS again see Sect 6 4 and for that you need to turn an output catalogue source res fits into an input one source_cat fits Their structures are very similar therefore only two keywords and
66. courage the reader to consult the specific handbook of spimodfit to understand how this is handled We suggest the general user not to exploit this functionality unless he she knows well the instrument and the program 36 ISDC SPI Analysis User Manual Issue 9 0 e oe Energy definition Parameters energy binning Ok Number of energy regions 13 Help Regions energy boundaries 25 300 Numbers of bins in each region 22 Figure 23 GUI of spimodfit_analysis energy definition here 22 equally spaced logaritmic bins have been selected 00e X Spimodfit options m Energy Setup Image Options Ok First selected bin 1 f Init images form file No 0 ves 1 0 y Help Last selected bin 22 Filename Number of bins per rebinned energy Image Parameters m Source Variability Parameter Time variability definition d ays p pointings Odi Background Parameters Background file empty back model idx fits 1 viaximum number of background loaded components 14 Background multipliers Fitted detector ID ranges 00 18 Time variability definition d ays p pointings 1144 3 d n 1305 2 d n 1426 0 d n 1435 0 d n 1660 0 dn 1 Figure 24 GUI of spimodfit_analysis regarding spimodfit The bottom frame regards the background settings The entry for the background file can be left empty to allow the internal handling this is our suggested choice or can assume the value
67. ction to the INTEGRAL Data Analysis 1 Table 1 Main characteristics of the SPI instrument Energy range 20 keV 8 MeV Energy resolution FWHM 2 35 keV at 1 33 MeV Detector area 500cm Field of View 16 x 16 fully coded 35 x 35 zero coded Angular resolution 2 8 for point sources Point source positioning lt 1 3 for point sources depending on point source intensity Narrow line Sensitivity 5 x107 ph cm s at 1 MeV Timing accuracy 160 us 2 ISDC SPI Analysis User Manual Issue 9 0 3 Instrument Description 3 1 The Overall Design The SPI instrument is a coded mask spectrometer designed with a hexagonal geometry to maximize its compactness An overall cut out view of the instrument is given in Figure 1 Figure 1 A cut away view of the SPI instrument The mask plastic scintillator camera and ACS sub systems are highlighted The detector plane of the instrument consists of 19 cooled hexagonal shaped high purity Germa nium detectors providing a total area of about 500 cm In Figure 2 the detector numbering is shown An Anti Coincidence Shield ACS defines the instrument Field Of View FOV by pro viding veto pulses for photons and particles coming from the sides A plastic scintillator PSAC placed underneath the coded mask detects charged particles coming through the FOV and also provides veto signal The sensitivity of the instrument is lim
68. d e Click on the More Options button at the top or at the bottom of the web page e De select the All check box at the top of the Catalog table and select the SCW Science Window Data one e Change the Search Radius from Default to 10 degrees for our case e Press the Specify Additional Parameters button at the bottom of the web page e De select the View All check box press twice on it at the top of the Query table e Select scw_id and put the value 0102 without the quotes to specify all ScWs from Revolution 102 e Select scw type and put the value pointing without the quotes or simply po to get only pointings e Select the scw_id for the Sort option roundish button next to scw_id e Press the Start Search button at the bottom of the web page 16 ISDC SPI Analysis User Manual Issue 9 0 At this point you should be at the Query Results page with all the ScWs available for revolution 102 Sort the Scw_id column by clicking on the left arrow below the column Name You can then select the ScWs we are interested in i e 010200200010 010200210010 010200220010 010200230010 010200240010 010200250010 010200260010 010200270010 010200280010 010200290010 Press the Save SCW list for the creation of Observation Groups button at the bottom of that table and save the file with the name spi lst in your analysis c
69. d point sources where no significant contribution is expected from other sources within about 20 4 3 6 Imaging Capabilities A thorough presentation of the SPI imaging capabilities can be found in the paper The INTE GRAL spectrometer SPI performance of point source data analysis published in MNRAS 4 4 3 7 Timing Capabilities In the SPI standard operation mode each photon data set includes timing information given by a 100us clock signal This clock is synchronized to the on board clock and thus to the UTC The timing error budget for SPI is derived from e the accuracy of the on board clock and the synchronization e the conversion between on board time and UTC e the conversion between UTC arrival time at the spacecraft and the arrival time at the solar system barycentre when appropriate The resulting SPI timing accuracy calculated in this way is 129s 30 accuracy and a 90 confi dence accuracy of 94s ISDC SPI Analysis User Manual Issue 9 0 13 Part II Data Analysis 5 Overview The first task when starting a SPI data analysis is to identify and collect the input data There are two main cases 1 guest observers receiving their private data and 2 users of the public data browsing and retrieving data from the INTEGRAL archive Basically a list of pointings see Glossary have to be collected into an ASCII file This ASCII file can then be fed into og create to produce what is called an observation
70. dead detector will increase sharply at the time of the failure while the other detector backgrounds will remain approximately constant As a consequence it will no longer be possible to assume that the background follow the same time evolution in all detectors as you do when using the saturated Germanium event count rate as a time model for example It is no longer necessary to specifically exclude dead detectors from the analysis The software does that automatically 7 15 Annealing phases About every 6 months the SPI instrument goes through a so called annealing phase During these phases the detectors are heated up to about 100C for a few days so that their crystal structure can re arrange and correct itself for most of the accumulated radiation damages During these phases listed in Table 5 SPI does not produce any scientific data At the end of an annealing phase the detectors are cooled down again and they are switched on before the temperature reached the nominal value During this period the standard energy correction is not so good and special care has to be taken for accurate spectral analysis These periods with unstable temperature are included in the ranges of revolution provided in Table 5 Data from revolutions outside of these ranges should be fine in this respect 7 16 Instrument modes and SPI data From the SPI operation point of view three main different phases can be identified 1 the first 2 months pvphase
71. dft analysis Y Spi Figure 42 GUI with the standard settings for a flat field background and logarithmically increasing energy binning ok Energy Setup og ep Fires Detector iD Ranges Rangs Range oe ooa Defintion o oos Range oa ooma Demonas oos Pange ts ooma O Deron o oos Input Model ages men ooo Joss Eq OEE soe Y Backgrou ES mes turer or componere 8 m RETA intromer 08 sie 503 Figure 43 GUI with the standard settings for internal background computation no image fitting and standard background multipliers appropriate parameters For this option and the introduction of images for the diffuse emission models we refer to the spimodfit handbook However we warn the reader that the full exploitation of these and more advanced features should be carried out in touch with the experts at MPE In Fig 42 we show the GUI with the standard settings for a flat field background and logarithmically increasing energy binning Finally the spimodfit final products are not attached to the observation group decoupling the spectral extraction step from the previous ones The program can then be used separately once the data are binned and the optional background is created without the need of running any DAL tool to clean the observation group However the side effect is that the user should delete the output files by hand if he she
72. e of spiros is to use SPI data to locate point or point like sources in the observation field of view and to output a catalog of their parameters along with images their spectra and any flux variability in time spiros has three basic operating modes to choose from e IMAGING mode Here the aim is only to locate point or point like sources in the observation field of view and to output a catalog of their locations and width parameters This step also produces images e IMAGING TIMING This step produces images in the case that variable bright sources are in the field of view e SPECTRAL and SPECTRAL TIMING mode In this mode spiros reconstructs spectra of sources in the SPI field of view Source positions are either taken from the source location process IMAGING mode or from a catalog The output spectra can be used for model fitting in XSPEC The SPECTRAL TIMING mode is no different from the normal SPECTRAL mode except that it conside variable sources as in the timing mode e TIMING mode Asin SPECTRAL mode a catalog of known sources is used to create a light curve of sources and background in several energy bins one flux value per pointing 8 6 1 IMAGING mode Given its main task of locating point sources and creating images of them in the observation field of view the first question is how spiros goes about locating them A fast and simple method results by assuming some prior knowledge of the sources expected in the field of view nam
73. e processed independently It is also possible to select a subset of the detectors in the input data This step are usually performed via the standard OSA processing from GTI to Background levels but ready to use data are also available at http www mpe mpg de gamma instruments integral spi public service data These data are binned at high spectral resolution therefore the user interested in point sources should rebin the energy channels using the parameters first energy bin last energy bin and m energy rebin in spimodfit The background models can be collected into one single model setting the flag spi flatfield single to yes or limited to a maximum number only the first n background loaded ones will be loaded In addition to these files the Instrument response function files IRFs file has to be specified More than one IRFs files can be given to handle a time dependent response the default values are appropriately inserted in the standard parameter file e Definition of the fitted parameters spimodfit can fit extended emission images point sources sources and background parameters For each of these components the time and detector variability should be given as well as some other fitting or normalization parameters explained below e Fitting options as the minimization routines and its initialization MCMC options are cur rently not implemented e Definition of the output files spimodfit generates a FITS file t
74. e program spiskymaz regards the image as a pixelated skymap and the aim is to obtain the intensity in each pixel The most important idea is that the data constrain the image within some limits in an N dimensional space where N is the number of pixels Hence there is no unique best image and we have to make some choice out of all the possible images within the constrained region Maximum Entropy method as implemented here is one way of doing this and of quantifying the uncertainty of the result Since the number of pixels is usually large e g 104 10 the techniques involved are rather different from those of model fitting spiskymaz is adapted to the particular needs of SPI The background temporal variations are treated via a template prepared by spiback and the coefficients are fitted during the imaging process The input count spectra contain data for many energy bins and spiskymax analyses a subset or all of these energies as specified by the parameters energy range min energy range maz The number of iterations can be specified since the automatic stopping criterion is not always appropriate and in any case may not be reached in the CPU time available Sources to be analyzed are defined by their positions in the input source catalog SPI SRCL CAT only those with the SEL FLAG flag 1 are analyzed and their fluxes and lo errors written to the output catalog After the image is produced the flux and its error for a number of us
75. e variable COMMON SCRIPT must be defined setenv COMMONSCRIPT 1 To revert to having the GUI unset the variable unsetenv COMMONSCRIPT 6 2 Observation group OG creation The first step of the analysis is to create the Observation Group for more information on og_create see 1 The easiest is to launch the first program from the analysis top level directory i e REP BASE PROD so that the parameter baseDir can be set to the current directory og_create idxSwg spi lst ogid spi analysis baseDir instrument SPI spi lst is the file containing the list of pointing DOLs created as explained in above Sect 6 1 2 Note that if this file is located in another directory the full path should be specified together with its name spi_analysis is a user chosen parameter used to name the subdirectory where all analysis data will be written instrument should never be changed for SPI data analysis The result is an observation group named og spi fits located in obs spi_analysis This later location is where the following of your analysis should be executed and where all further analysis results will be written Depending on the number of science windows the speed of your machine and your data server this can take from a few seconds for just one science window to more than an hour for hundreds of science windows and of the order of a few seconds for this example 6 3 Image reconstruction without inpu
76. ecause of the additional parameters of the splines Crashes have been reported when using a large gt 1000 files dataset their origin is still unclear 8 8 Response interpolation SPIRMF spirmf performs a 2 dimensional interpolation of an existing RMF onto a desired grid in input energy x output energy channel space This accommodates spectral analyses invoking essentially arbitrary binning of the data The current limitations are that 1 the input and output matrices must be square i e number of detector channels equals number of input energy bins 2 the input matrices cannot be compressed compression here referring to the OGIP condensed matrix file format although the output matrices are automatically compressed and 3 the output energy grid must fall within the span of the input grid or very nearly so spirmf can be used for either XSPEC 12 compatible component RMFs see appendix B 0 3 or for single RMFs for use with ISDC SPI Analysis User Manual Issue 9 0 73 spiros extracted spectra Elements above the diagonal are treated as zero spirmf can optionally update a PHA II data file to facilitate ease of use with XSPEC Both the input and output RMF s will be in the condensed file format with zero valued elements ignored 8 9 Image reconstruction SPISKYMAX The detailed description of spiskymaz is given in the document written by its author A Strong MPE Garching 7 Here we give only a brief overview of th
77. ectory of ic files installation idx ln s directory of cat installation cat ln s directory of local archive scw ln s directory of local archive aux The different directory of can be all the same location For example if you are logged on the ISDC network and have a direct access to the revision 2 ISDC archive these links become ln sf isdc arc rev 2 ic ln sf isdc arc rev 2 idx ln sf isdc arc rev 2 cat ln sf isdc arc rev 2 scw ln sf isdc arc rev 2 aux Alternatively if you do not have any of the above data on your local system or if you do not have a local archive with the scw and the aux branch available follow the next section instructions to download data from the ISDC WWW site If you do have already the data available on your system you can skip next section 6 1 2 you need however to build a list of DOLs for your analysis that you can either produced with the archive browser or by hand using shell command or and an editor 6 1 2 Downloading data from the ISDC archive The below example is based on observations of the Crab from Revolution 102 To retrieve the required analysis data from the archive go to the following URL http isdc unige ch index cgi Data browse You reach the W3Browse web page which will allow you to build a list of Science Windows SCWs needed to create your observation group for OSA e Type the name of the object Crab in the Object Name Or Coordinates fiel
78. ectral_response rmf fits 1 data set is in use XSPEC gt cpd xs XSPEC gt setp en XSPEC gt plot ld XSPEC gt ign 30 ignoring channels Jas 11 in dataset 1 XSPEC gt model power Model powerlaw lt 1 gt Input parameter value delta min bot top and max values for 1 0 01 3 2 9 10 1 powerlaw Pholndex gt 1 0 01 0 0 1E 24 1E 24 2 powerlaw norm gt 32 ISDC SPI Analysis User Manual Issue 9 0 Model powerlaw lt 1 gt Model Fit Model Component Parameter Unit Value par par comp 1 1 1 powerlaw PhoIndex 1 00000 0 00000 2 2 1 powerlaw norm 1 00000 0 00000 Chi Squared 4010090 using 89 PHA bins Reduced chi squared 46092 99 for 87 degrees of freedom Null hypothesis probability 0 00 XSPEC gt fit 50 lt snip gt Model powerlaw lt 1 gt Model Fit Model Component Parameter Unit Value par par comp 1 1 1 powerlaw PhoIndex 2 16210 0 123967E 01 2 2 1 powerlaw norm 13 3916 0 626640 Chi Squared 115 7470 using 89 PHA bins Reduced chi squared 1 330425 for 87 degrees of freedom Null hypothesis probability 2 137E 02 XSPEC gt plot ld de The final plot ld de produces the following spectral display Note that the remaining deviations are indeed really small especially considering that we are using only 10 science windows in this example which is a very small number for SPI 6 6 Light curves SPIROS can only produce light curves for sources provided through an input catalogue The ea
79. el which are based on the count rate of the events saturating the Germanium detector electronic In this case un select the flat field option and select the corresponding background option For the GEDSAT model set the Number of Models to use to 1 and select the GEDSAT model In the image reconstruction process we will assume that the real detector background varies in the same way as this model which will be scaled to the actual data to determine the real background GEDSAT is a good model for most point source analyses It is recommended to read further advises from the user manuals of the spi obs back and SPIROS programs before thinking about using more sophisticated background models See also the docu ments provided by the SPI team 10 for difficult cases 7 8 Image reconstruction Image reconstruction is carried out using SPIROS in IMAGING mode It is recommended to first use background method 3 in conjunction with the flat field background model The CHI2 optimization statistic should first be used while the LIKEH can be tried in a second run In principle both optimization statistics should produce compatible results However with the CHI2 statistic all the bins with zero counts are ignored and this can lead to biases hard tail at high energy or problem with pseudo detectors at low energy where the number of double triple can be null On the other hand CHI2 statistic seems to be more robust and faster in most cases SPIROS
80. ely if they are point or point like and that there are only a small finite number of them tens but not hundreds or thousands On this assumption the sources can be located using the Iterative Removal of Sources TROS method which literally attempts to locate each source in the field of view one after the other from the strongest to the weakest This method works as follows e A fast but blurred location image is created It shows the most probable location or region of the greatest source emission This image may be called a pixel independent correlation ISDC SPI Analysis User Manual Issue 9 0 65 map as it is constructed by scanning the imaging field of view with a source probe and calculating the flux by assuming all input counts are due to that source and any background events alone This typically produces a convolved image with a global maximum at a good first guess for a probable source location e With this approximate location find the strength probably upper limit of the source and then the detector counts expected from it e Subtract source counts from the input count data to create residue count data more or less due to other sources not yet located e With these residue counts the procedure above is repeated creating a new location image searching it for any new source and adding it to a growing list Each time a new source is added to the list all sources are allowed to float to more precise
81. er defined sources can be optionally determined Each source is specified in terms of a position and ON and OFF radii The ON region is the circle centered on the source position with radius ON and the OFF region is the annulus between the ON and OFF radii The source flux is defined as X intensity in ON region X intensity in OFF region solid angle ON solid angle OFF The error is based on a Bayesian analysis which marginalizes over all the unwanted degrees of free dom so the significance of a source by this method may not correspond to a classical significance level In general significances increase for smaller skymaps since the degrees of freedom decrease 8 10 Cleaning Tool SPI CLEAN Run the SPI cleaning tool spi clean to remove all previous pipeline products spi clean obs group og spi fits 1 endlevel PRP Note that spi clean is much faster than og clean but it only works on SPI data It is launched automatically by the by the pipeline when necessary spi_clean checks all entries in the OG og spi fits All entries that have a higher level than the requested end level are detached from the OG and deleted If in force mode spi_clean checks whether the detaching was successful and if the file still exists uses the UNIX rm command to delete file In force mode spi_clean is also able to work with corrupted OGs i e if one member of an OG has been manually deleted or renamed and it will try to fix the OG
82. fluxes and you can provide an optional reference catalogue to use If you want the resulting source list to contain only sources which have been de tected with SPI before for most observations this is a good idea enter SISDC REF CAT SPI FLAG 1 in the reference catalog field provided that the environment variable ISDC REF CAT is set which should be the case When you are done with your changes click OK and run the pipeline When the pipeline has finished analyze the contents of the catalog 30 ISDC SPI Analysis User Manual Issue 9 0 fv source_cat fits Review the catalogue carefully If you don t like what you see e g the sources you are interested in are missing quit fv restart spi_science_analysis and change the options for the catalogue extraction step i e lowering the flux threshold if you re source is weaker than the threshold For minor modifications or removing entries use fv to modify the catalogue according to your needs In our case depending on the catalogue version used one or two sources H 0614 091 and or X per may be selected in addition to Crab We do not want to use these sources in the analysis Using the fv menu find the SEL_FLAG column one before the last one and enter SEL_FLAG 0 for these sources Source with a selection flag of zero will be ignored by spiros alternatively you can also removed the un wanted rows Save the catalogue and quit fv See Sect 7 fo
83. from the first point above so you can then launch the spi_grb_analysis script Make sure the grb start sec_to_avoid_before_grb is not before the pointing start and that grb_stop sec_to_avoid_after_grb is not after the pointing end as there is no corresponding checks in the spi grb analysis script In our example spi_grb_analysis 2003 05 01T03 10 10 000 2003 05 01T03 10 30 000 UTC 10 10 You can re run spi grb analysis in the same directory as many times as you wish changing the parameters as explained in above section 5 However when making a new run the outputs of the previous run will be deleted If you want to keep your outputs before making a new run save them by copying the entire directory e g cd cp r grb030501 grb030501 imaging cd grb030501 before continuing 8 spi grb analysis is calling a number of executables and spi science_analysis several times Since spi science analysis produces its own log file the log file situation after running spi grb analysis is somewhat complicated The resulting spi grb analysis log is the high level log file containing logs from the executables called directly by spi grb analysis In ad dition you can find three log files spi_pointing log spi binning log and spi spiros log The first contains the outputs of spi obs point the second includes the logs from the binning to background steps and the third contai
84. g ISDC SPI Analysis User Manual Issue 9 0 Figure 14 Options for the template background gt ax aeos Hele E ISDC SPI Analysis User Manual Issue 9 0 23 O00 24 Figure 15 Options for modeling the background using tracers like GEDSAT Background Model options 13 GEDSAT v GEDSAT v ISDC SPI Analysis User Manual Issue 9 0 5 The last step of the SPI pipeline is the image deconvolution by spiros Click on the spiros button to get the spiros GUI Spiros performs image deconvolution in order to create images spectra or light curves Since we want to create an image first we set the spiros mode to IMAGING Figure 16 SPIROS main GUI spires TE SPIROS General Setup ok Run SPIROS in Mode IMAGING y Help Further Options for imaging Selection Parameters timing energy subset pointing subset detector subset Other Parameters Background method 2 y Optimization statistic CHI2 x Image solution constraint NONE y Bins for src location FIRST In the Selection Parameters fields of the spiros GUI you can specify three subsets of energy bins pointings or detectors respectively If you leave the fields empty spiros will use all data otherwise only the specified energy bins pointings and detectors will be taken into account in the analysis You can automatically ignore obviously bad da
85. g photon This contributes to raise the full energy peak and to reduce the Compton continuum Unfortunately it also degrades spatial information as it is not possible in general to determine which detector was hit first All signals produced from the detection plane are also directed towards the Pulse Shape Discrimi nator PSD System see more details in Section 3 5 All events detected by SPI are classified in accordance with how many detectors respond to an incoming photon The event detected by only one detector is called a single event Multiple events are detected by several detectors and the value of multiplicity corresponds to the number of the detectors hit 3 5 Pulse Shape Discriminator PSD The photons interacting in a given detector can either deposit all their energy at a single location the so called localized events or in two or more sites the un localized events Pre launch simula tions predicted that the instrumental background would be dominated by localized 5 decays in the energy range 200 keV 1 MeV The PSD electronic sub system was design to determine whether interactions are localized or not Unfortunately the fraction of 87 decays observed in flight data turns out not to be large enough for a significant background reduction Despite important efforts from SPI experts it has not been possible to improve significantly the data analysis by using the information provided by the PSD sub system 3 6 Anti Coinc
86. group OG The OG includes links to all input and later output data and all software read write data through this group Once you have created an OG an image reconstruction or a spectral extraction can be achieved by running a single script spi science analysis This script analyzes the data going through a number of different steps briefly described in Table 3 These steps correspond to so called Levels identified with 3 5 capital letter acronyms Their meaning is common to all INTEGRAL instruments A very important difference with the other instruments is however that almost always the complete set of pointings are analyzed together simultaneously Apart for GRB analysis single pointings are never analyzed independently Table 3 Overview of the successive steps of the spi_science_analysis script Level Program Short description of the step CATI cat_extract Extracts from the INTEGRAL reference catalogue information position and flux about known sources lying in the field of view POIN spi obs_point Extracts the start stop times and the SPI attitude for all point ings included in the observation and stores the results in the pointing fits file On request this program can access a list of bad pointings usually of order of 1 2 of the total and exclude them for the current analysis BINI spibounds Defines the energy bins creating the energy_boundaries fits file spi_obs_hist On the fly energy correction even
87. hbor detec tor all multiple event are affected similarly Currently we have three independent instrumental responses 1 one for all 19 detectors for the period preceding the first failure 2 one for 18 56 ISDC SPI Analysis User Manual Issue 9 0 detectors for the period between the first two failures and 3 one for 17 detectors for the period after the second failure and before the third failure The introductin of a fourth 4 respoonse valid after the failure of the third detector will be released in the next OSA release after the appropriate testing is performed Table 4 SPI detector failures Det ID UT IJD Pointing Revolution 2 2003 12 06T09 58 30 1435 41635 014000040010 140 17 2004 07 17T11 00 00 1659 46000 021400600010 021500000012 214 215 5 2009 02 19T12 00 00 3337 50000 077500610010 077600000012 775 776 these times are approximation as the failure occurred at some unknown time during the perigee passage Our software cannot currently handle a time dependent response and the easiest way is to an alyze the three possible cases independently and to combine the results later on It is possible however to analyze different mixtures of data together using one of the three responses as they are not so different Great care should be taken in this case anyway For example because of the effect mentioned in last paragraph the single event background in the neighbors of a
88. he No of sources parameter and that there is an additional source in the FOV In this case SPIROS cannot take this source into account and its counts are attributed to the other sources and or to the background When all the above checks are performed and if there are still large residuals it may be time to let varying some sources SPIROS in imaging and spectra modes assumes that the source fluxes are constant throughout the whole observation period A good solution can therefore not be obtained if some of the sources are variable see Sect 7 11 Only if bad pointings remain despite all efforts you can remove them from the analysis Note all bad pointings and enter the GOOD pointings in the pointings subset entry field see also Sect 6 3 1 7 13 High time resolution with SPI The highest time resolution possible for light curves is one science window i e about 30 minutes A higher time resolution is not possible for spiros However it is possible to analyze the raw SPI event data which have a very high time resolution but no further tools are provided however plenty of tools e g the FTOOLS exist to work on event data 7 14 SPI failed detectors Three of the 19 SPI Germanium detector have ceased to function properly and they are now dis abled The failed detectors also affect the response of their neighbors as the double events involving before a neighbor and the to die detector are now seen as single events in the neig
89. he burst data as a fixed background With background method 3 the background will be scaled through the image deconvolution process solving for a single background scaling coefficient Do not use other options For the Optimization statistic Chi2 is recommended while LIKEH can be tried in a second run Click on imaging to check that the default parameters shown in Fig 36 are correctly entered and update them if necessary h Close all sub windows by clicking twice Ok then using the Save As button you can save the parameters you have just entered into a file Enter the name grb_analysis par and store it in the current directory PLEASE MAKE SURE NO SPELLING MISTAKE 5Note that although this file has a par extension it is not a standard parameter file It is a csh script containing all parameter values The value of all parameters are then later extracted from this file in the main grb analysis script ISDC SPI Analysis User Manual Issue 9 0 43 44 Figure 35 spiros GUI main window 000 IX Spiros options Figure 36 spiros GUI imaging window 000 Imaging CAR y POINTING ZCFOY y ISDC SPI Analysis User Manual Issue 9 0 IS INTRODUCED THE GRB SCRIPTS READS THE grb analysis par FILE LO CATED IN THE CURRENT DIRECTORY TO LOAD THE ANALYSIS PARAME TERS i Once the parameters are saved click on Quit to quite the process We do not
90. idence Subassembly ACS The main function of the Anti Coincidence Subassembly ACS is to shield the Germanium detec tors against background photons and particles from sources outside the field of view The ACS system consists of 91 Bismuth Germanate scintillator crystals in combination with photo multiplier tubes The scintillator crystals are used to convert all incoming events into photons in the 480 nm region visible light Photo multiplier tubes are used to detect these photons and convert them into electrical pulses which are sorted normalized and summed up by the ACS electronics Each photon registered by ACS induces a time tagged veto signal directed to the Digital Front End Electronics DFEE which reject any simultaneous detector events 3 7 The Plastic Scintillator Anti Coincidence Subassembly PSAC The purpose of the plastic scintillator subassembly PSAC is to reduce the 511 keV background due to particle emission by the passive mask The detector consists of a plastic scintillator inside a light tight box located just below the passive mask It has a good gamma ray transparency and actively detects particles which deposit energies in excess of 300 keV The light flashes that are produced by the impacts of these particles are detected by four photo multiplier tubes located around the light tight box and converted into electrical pulses which are processed by the PSAC 6 ISDC SPI Analysis User Manual Issue 9 0 electr
91. idual components are identified The total background spectrum is indicated with the black line 4 1 2 511 keV Background The 511 keV background can be split into five components e the continuum background under the 511 keV line e 511 keV photons from passive material due to 3 decays of unstable nuclei in the materials These are nuclei hit by protons and neutrons produced by interactions between cosmic ray particles and material from the detectors shield or cryostat The unstable nuclei decay through 67 decay The annihilation of the positron leads to the emission of two 511 keV photons in opposite directions If one is absorbed by the detector and the other escapes a 511 keV background event is produced ISDC SPI Analysis User Manual Issue 9 0 9 e 511 keV photons originating from interactions of cosmic rays with passive spacecraft mate rials that are not rejected by the BGO shield shield leakage e 511 keV photons originating from cosmic ray interactions with the mask material The main source is pair creation by cosmic ray proton interactions with W nuclei This component is significantly reduced with the Plastic Scintillator e BGO shield blocking time component 511 keV photons produced by 8 decays in the BGO shield when the ACS electronics is blocked by a large energy deposit and the veto is on All these components were reproduced with a Monte Carlo method The resulting line strengths for the 511 keV line with
92. ir uncertainty This tool is oriented towards large scale surveys which combine a large set of individual pointings of the spacecraft It concentrates on fitting spatial as opposed to spectral models although through using the different energy channels of the measurement it will be an important method to generate spectra of diffuse emission The principal use will be fitting to maps of physically based tracers of y ray emission Examples are fitting the 1809 keV 2 A line to free free or IR emission maps fitting diffuse continuum y ray s to gas HI CO maps It will also allow fitting spatial functions e g Gaussian exponential where there is no physical model available Fitting components is the best method to get the spectra of diffuse emission when a good spatial but a poor spectral model is available The analysis of measurements with this tool is complementary to deriving results from deconvolved images e g from methods specifically designed for point sources e g spiros However it can be fruitfully used to extract the spectra of point sources The algorithm performs fitting of raw data binned counts for many observations to multi component models using the full instrument response information In addition to parameter estimation a Bayesian statistical analysis is used to include information on model characteris tics and other prior information so that the uncertainties of the fitted parameters are properly assessed The
93. is in the Histogram options of the top level GUI Clicking on this button opens a window shown in Fig 27 This feature is not implemented in spimodfit analysis ISDC SPI Analysis User Manual Issue 9 0 37 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 1046 3121 d n 1144 3 d n 1305 2 d n 1345 3845 d n 1426 0 d n 143 Energy keV A9x Sjunoo X SV Figure 26 Count rate spectrum of the Crab from the cookbook example and the residuals from the best fit model 38 ISDC SPI Analysis User Manual Issue 9 0 Figure 27 Histogram options GUI 000 IX Histogram options Figure 28 Phase parameter GUI 000 Phase Parameters ephemeris fits Open the Phase Parameters window Fig 28 by selecting Phase resolved analysis and clicking on Phase Parameters The pulsar ephemeris has to be provided through a FITS file the format of which is described in the next sub section Sect 6 8 1 You can then specify the number of phase bins Equally spaced bins will be used by default but if you un select Equal bin width you can provide specific phase bounds Be careful when entering this parameter as it may seem confusing Here is an example for the Crab of a phase bound vector with Number of phase bins ES Phase Bounds 0 00 0 03 0 20 0 38 0 50 0 80 0 98 1 00 Phase Bounds must include values between 0 and 1 and it must start with 0 and end with 1
94. is mode should only be used for isolated point sources and is not really suitable for imaging The side lobes are still present but significantly less with the 5 by 5 dither pattern about 50 of the hexagonal case To remove these side lobes which will cause artifacts in reconstructed images the only possibility is to enlarge the imaged area by using many more pointings i e many dithering patterns ISDC SPI Analysis User Manual Issue 9 0 7 Schematic view of possible dithering patterns is shown in Figure 4 A different dithering strategy can be adopted depending on the circumstances For SPI a 5 by 5 dithering pattern with 2 spacing around the target source are almost always preferable In the case there is only a bright isolated source in the field of view an hexagonal dithering pattern with one pointing centered on the source and surrounded by six pointings with distances of 2 is also acceptable SPI FCFOV O Target position o 7 point hexagonal pattern o 25 point rectangular pattern Figure 4 Schematic view of the dithering patterns and of the instrument Fully Coded Fields of Views Both dithering patterns use a dwell time of the order of 30 minutes per point 4 Performance of the Instrument 4 1 Components and Sources of Instrumental Background As we have already mentioned above the SPI instrument is background limited The sensitivity of the instrument is therefore largely dependent on the background and
95. ited by the background due to the primary and secondary cosmic ray particles and the cosmic background radiation Each photon that is absorbed in a Germanium detector will give a pulse that is sent to the electronics The electronics analyze the incoming pulses reject those simultaneous with veto signals and tags good photons with the energy the time and the type of event see Section 3 4 for the discussion on the possible event types These data are then sent to the ground All photons detected by the detectors are summed into background spectra and are sent to the ground every 30 minutes ISDC SPI Analysis User Manual Issue 9 0 3 Z Sun O X pointing Figure 2 Numbering of SPI detectors 3 2 The Passive Mask The passive mask is located on top of the SPI instrument above the plastic scintillator The pur pose of the mask is to code the incident gamma rays in the field of view providing the instrument imaging capabilities The mask consists of a pattern made of hexagonal tungsten blocks with a 120 rotation symmetry The mask is made of 127 elements arrange in a 78 cm diameter circle Of these elements 63 are opaque and 64 are transparent Each opaque element is 30 mm thick and 60 mm flat to flat in size The tungsten elements stop the gamma ray radiation in the range 20 keV to 8 MeV with an absorption efficiency greater than 95 at 1MeV The holes in the mask have a gamma ray transparency of 60 at 20keV and 80
96. ky image is reconstructed over a user defined field of view FOV it may happen that a bright source located in the instrument FOV is not included in the image FOV For example if the image FOV is specified as POINTING FCFOV i e pointing center SPI fully coded FOV there might be a bright source in a corner outside of the fully coded FOV but inside the zero coded FOV A large fraction of the detector events can then come from this source and SPIROS searches for a source location to best reproduce these counts SPIROS however cannot find the real source as the search is restricted to the specified sky image It will be forced to find one or several spurious positions at some coding or random noise peak inside the image FOV and it will attribute the counts to it This will lead to completely biased results In order to avoid such biases first make sure all detectable sources in and close to the FOV of SPI fall inside the specified image FOV In general it is enough to use the POINTING ZCFOV pointing zero coded FOV option in the spi science analysis FOV entry field click spiros first in the main GUI and imaging in the spiros GUI If really necessary you can defined your own custom image FOV see the SPIROS user manual in such a case Second check that SPIROS maximum number of sources is not too restrictive You will also obtain biased results if there are 3 bright sources in your sky image and that SPIROS
97. ltiple responses which are appropriately used during the data processing The final response accounts for the modified responses accordingly The spiros imaging software is quite a complex tool with many different options and param eters Not all possible cases have been fully scientifically validated The best tested modes include imaging and spectral extraction and the more robust background method is number 2 combined with a time variation model proportional to the Germanium saturated events Other modes such as timing and spectral timing and other background methods are being further tested and validated The spimodfit software is an on going project which has now reached a stable configuration but not all the features have been scientifically tested At least in one case a long staring pointing which is split up into several science windows in the ISDC system is not handled correctly in the SPI data analysis It concerns ScWs 008200220010 001 008200220020 001 008200220030 001 Only the first pointing is properly included in the analysis while the subsequent ones are ignored Please report if you find any other such cases ISDC SPI Analysis User Manual Issue 9 0 References 1 ISDC OSA INTRO Introduction to the INTEGRAL Data Analysis http isdc unige ch Soft download osa osa_doc prod doc_tree html 2 ISDC OSA INST GUIDE Installation Guide for the INTEGRAL Off line Scientific Analysis http isdc
98. lysis before running spiros Instead of this program the script calls the program spimodfit which allows for a different approach in the simultaneous fit of background point sources and diffuse emission using the SPi data This will be described in Sect 6 7 of the cookbook and with more detail in Sect 8 7 For analysis of Gamma ray bursts a dedicated script spi_grb_analysis is available This script is documented on line at http isdc unige ch index cgi Support spi 6 Cookbook We assume that you have already successfully installed the ISDC Off line Scientific Analysis OSA Software The directory in which OSA is installed is referred later as the ISDC_ENV directory If it is not the case look at the Installation Guide for the INTEGRAL Data Analysis System 2 for detailed help 6 1 Setting up the analysis environment 6 1 1 Input data organization The commands below apply to the csh family of shells i e csh and tcsh and should be adapted for other families of shells In order to set up a proper environment you first have to create an analysis directory e g spi data rep and cd into it mkdir spi data rep cd spi data rep setenv REP BASE PROD PWD This working directory will be referred to as the REP BASE PROD directory in the following All the data required in your analysis should then be available from this top directory and they should be organized as follow scw data produced by the instrument
99. me 3 4 Event Types The photon which entered the telescope can be detected due to its interaction with the absorbing material of the detector Three major types of interactions play a dominant role photoelectric absorption Compton scattering and pair production In the photoelectric absorption process a photon undergoes an interaction with an absorber atom in which the photon completely disappears In its place an energetic photo electron is ejected by the atom The photo electron carries off most of the original photon energy The Compton scattering takes place between the incident gamma ray photon and an electron in the absorbing material The incoming photon is deflected and it ISDC SPI Analysis User Manual Issue 9 0 5 transfers a portion of its energy to the electron In the pair production process the gamma ray photon disappears and is replaced by an electron positron pair The positron will annihilate in the absorbing medium and two annihilation photons are normally produced as secondary products of the interactions Depending on the size of the detector and on the energy of the incoming photon a photon scattered in a Compton interaction can escape the detector or undergo a second interaction The pairs of 511 keV photons produced by the annihilation of the positrons resulting from pair creation can also produce other interactions or escape the detector In SPI the detector array is used to recover the total energy of an incomin
100. n through a distant slow connection or 3 ROOT has not been installed and thus the GUI is not available An easy way to produce a command line script for spi_science_analysis is to use the GUI to enter the parameters and then to use the Save As button This produces a few line sh scripts that you can execute directly or copy into another more sophisticated one Below we also give an example of a command line run corresponding to the cookbook example of Sect 6 3 It includes all spi science analysis parameters The example can be copied and pasted into a terminal or a shell script Note that this example includes all possible parameters and that many of them are actually not used in this particular case but they are still listed such that you can modify them easily spi science analysis obs group og spi fits IC Group idx ic ic master file fits 1 IC_Alias 0SA coeff_DOL MX IRF_DOL RMF_DOL _ catalog _ clobber yes log_File spi_sa log run_cat_extract no run_pointing yes run_binning yes run_background yes run_simulation no run_spiros yes run_phase_analysis no run_gaincorrection no run_fullcheck no detectors 0 18 spiros source cat dol coordinates RADEC N cat extract fluxMin 0 001 cat extract fluxMax 1000 use pointing filter yes spibounds nregions 1 spibounds regions 20 400 spibounds nbins 1 spi phase hist ephemDOL ephemeris fit
101. nalysis When spi science analysis is launched any data with a stamped level higher than the new starting level is deleted before the actual processing starts again As a consequence the outputs from a previous analysis are always overwritten The easiest way to save the results of a previous analysis is to copy the complete directory structure before starting again cd cp r analysis_current_dir analysis_saving_dir cd analysis_current_dir As all analysis data are part of the OG you cannot use rm to delete some of them This would break the OG and prevent any further re processing If you use rm by mistake the safest is to start again at the very beginning with og create If you want to delete some data from the OG without using the spi_science_analysis script you can use the spi_clean cleaning tool described in Sect 8 10 7 3 Pseudo detectors and detector list Incoming photons which hit more than one SPI detector are called multiple events A set of pseudo detectors are defined to incorporate double and triple events into the analysis Simula tions showed that events hitting two detectors originate from a relatively small area smaller than that of a detector located near the border of the two detectors The same is true for triple events they occur very near the common corner of the three involved detectors Hence the idea of con sidering these small areas located between detectors as p
102. name of the template Other templates are based on the detector count rate the non saturated events and many more The user is encouraged to try the various models to see which work best for his specific science and dataset For typical point source analysis a good starting point is the GEDSAT template To use this model un select in the main background GUI the flat field option and select the background templates Press the button for the template options and select the template to use Another way to model the background is the direct use of background tracers the third option in the main background GUI The GEDSAT model which is also based on the count rate of the events saturating the Germanium detector electronic is a good alternative to model the background In this case un select the flat field option and select the tracers option in the main background GUI Press the Background model options button to set the options Set the Number of Models to use to 1 and select the GEDSAT model For more complicated setups you can use more components The GUI offers the possibility to use up to four different model components simultaneously In the image reconstruction process we will assume that the real detector background varies in the same way as this model which will be scaled to the actual data to determine the real background GEDSAT is a good model for most point source analyses see Sect 7 for more information about background modelin
103. nce analysis to run the catalogue extraction step the spi science analysis main GUI is shown in Fig 32 Select catalogue extraction and unselect all other tasks click on catalog to enter the catalogue task parameters Click Ok to close the catalogue window and on Run in the main frame to execute spi science analysis The source_cat fits catalogue produced contains irrelevant data It will probably includes a number of catalogue sources that may be detected on longer observation and certainly not the GRB position at which we want to extract a spectrum The catalogue file should include a single row with the GRB identification and position Edit this file with fv and enter the correct identification and position deleting extra rows if any and save the result Alternatively the position from the SPI image analysis can be used see Sect 7 19 for how to transform a source res fits into a source cat fits although in most cases a more accurate position should be available from IBIS or other sources If you use an existing source cat fits file make sure with fv that it contains the correct identification and position of the GRB and that the ISDCLEVL keyword of this file is CATI 2 As in the previous GRB imaging analysis case two alternatives can be used to specify the parameters of your analysis The simplest is to use spi science analysis GUI a Run spi science analysis and
104. nd enter AUTO in the pointing subset under spiros options and run the pipeline again The AUTO filter automatically removes all data on a per pointing and per detector base that have completely unreasonable values For most datasets this significantly improves the results but not in our case The next thing to do is to work on the background You should use different background models to improve the modelling We will use the MCM method in the next step launch spi science analysis again open the spiros options and select background method 5 and run the pipeline The results are now much better Ptg Rev Exp ONTIME CHI2 ML Expected Diff Reduced Data no no no secs value value STD CHI2 ML excl 1 102 20 2155 2 17 3 15 3 0 36 1 13 0 00 2 102 21 1932 7 16 7 15 3 0 26 1 09 0 00 3 102 22 2200 8 20 3 15 3 0 90 1 32 0 00 4 102 23 2198 8 26 8 15 3 2 09 1 75 0 00 5 102 24 2196 8 24 7 15 3 1 69 1 61 0 00 6 102 25 2143 8 16 3 15 3 0 18 1 07 0 00 7 102 26 2198 8 22 6 15 3 1 31 1 47 0 00 8 102 27 2196 8 15 4 15 3 0 01 1 00 0 00 9 102 28 2194 8 19 2 15 3 0 71 1 26 0 00 10 102 29 2189 8 1133 15 3 0 73 0 74 0 00 All residuals are now below three sigma and we do not need to further work on the extraction However in other datasets you may still have pointings with significant residuals In this case you may have bright variable sources in the FoV See Sect 7 on how to deal with such cases Nevertheless you will encounter pointings with high residu
105. nd the maximum number of sources or when the sigma threshold has been reached In our case we ask spiros to search for up to 3 sources with a detection significant larger than 3 sigma enter 3 instead of the default 6 We use the CAR Cartesian projection type for almost all cases With the CAR projection the derived images have angular coordinates with pixels regularly spaced by a constant angular step Since SPI is a coded mask instrument you have to define the image field of view FOV i e the sky region that spiros should reconstruct An easy way to specify the image FOV is to use the POINTING option The output image will then range from the minimum to the maximum of the central latitude longitude values of all pointings included in your observation Using POINTING FCFOV or POINTING ZCFOV will add half of the SPI fully coded or zero coded FOV on each side of the image defined by the pointing centers POINTING or POINTING FCFOV produces the nicest intensity images but be aware that strong sources just outside your image FOV can completely bias your analysis see Sect 7 and is therefore only recommended when NO sources are close to FoV If there sources close to the FoV POINTING ZCFOV is a good choice for most observations however the resulting intensity images need careful tuning in ds9 When the setup is done click twice Ok once on imaging and once on spiros then Run
106. ns the results from spiros 9 After a successful execution of spi grb analysis it is also possible to run spiros again with the spi science analysis script Launch spi science analysis make sure you select only the spiros step if you select any other step the burst background produced by spi grb analysis will be deleted and you will have to start it all again change the required spiros parameters making a LIKEH analysis rather than a Chi2 one for example and run the show again ISDC SPI Analysis User Manual Issue 9 0 45 10 Please consult the cookbook for information about the analysis results and for more infor mation about the analysis parameters 6 10 Gamma Ray Burst spectroscopy In the following cookbook section we explain how to derive spectra of short bursts such as GRB from SPI data We assume that readers have some knowledge about SPI data analysis with the ISDC spi _science_analysis script Please follow the first cookbook example if it is not your case Please follow points 1 to 4 from the previous section 6 9 on GRB imaging Once this is achieved you have an observation group in the current directory and you can continue with the following steps 1 The burst position at which you want to extract a spectrum must now be specified This is done by providing a source_cat fits file containing the GRB position You can either use an existing file or create a new one using the spi scie
107. on the correct identification of background photons The background can be divided into the following main components e continuum radiation e 511 keV line radiation 8 ISDC SPI Analysis User Manual Issue 9 0 e gamma ray lines 4 1 1 Continuum Background The continuum background can be split into several components depending on their origin First the radiation coming from outside the instrument This can be the cosmic diffuse gamma ray flux that comes in through the instrument aperture or leakage through the BGO shield of cosmic diffuse gamma ray radiation and gamma continuum radiation from the spacecraft induced by energetic cosmic ray particles Secondly scattering in the germanium detectors of neutrons that were produced in the spacecraft or other parts of the instrument Thirdly background components produced inside the spectrometer detectors These consist of localized 67 decays non localized B decays and 3 decays The continuum emission from the mask and the BGO emission when the veto electronics are blacked out veto dead time are negligible The individual components and the total continuum background emission are illustrated in Figure 5 10 erture flux elastic neutron shield leakage B localized 10 aaa Pi non localized mask emission Total t 10 count s em Mev e A 10 Figure 5 The continuum background components for SPI The indiv
108. onics assembly The electronics produced a veto signal associated with the detected events which are later merged with the ACS veto signal by the ACS front end electronics 3 8 Electronics The SPI on board electronics is in charge of the real time acquisition assembly time stamping and intermediate storage of the various pieces of information coming from the detectors the PSD the ACS etc Events are subdivided into classes depending on their origin in the instrument detector electronics detectors PSD veto shield and are used within the overall event energies and system monitoring statistics dead time signal counts etc The detected events are time tagged with a 20 MHz local clock which provides the timing resolution The reset timing reference is done with the 8 Hz satellite clock 3 9 Operating Modes The SPI instrument has only one mode for normal observations All scientific observations with SPI are done in so called photon by photon mode with a high temporal resolution In this mode scientific data are collected and transmitted to the ground for each photon In case the SPI telemetry is continuously overflowing due to background radiation that is higher than expected or due to a strong solar flare the instrument can be operated in a degraded science mode TM emergency mode In this case the single detector events are binned on board and downloaded in the form of spectra Thus the time information of each single photon is los
109. only change the results within their statistical errors The common background scaling factor was permitted to vary on a short timescale of 2h corresponding to typically four science windows This was the conservative choice determined by the actually observed changes in the overall event rate which is strongly dominated by background and the aim to minimise the number of model parameters in order to improve sensitivity A default energy binning can be chosen based on considerations of SPI energy resolution and flux sensitivity It consists of 20 energy bins of logarithmically increasing width between 25 and 8000 keV with a narrower bin around the 511 keV pair annihilation line Many of our sources of interest are too weak to be detected significantly above 200keV in individual bins of even this coarse compared to the SPI energy resolution energy binning So a wider binning consisting of 12 bins can be introduced for weaker sources The narrow and the wide binning should have the first three bins up to 48keV in common then the wider binning roughly combines every two narrow bins up to 500 keV and after that every 3 narrow bins The bin boundaries of the narrow binning could for instance be 25 31 39 48 60 75 93 116 144 179 223 278 346 502 520 668 832 1036 1290 1606 2000 and 8000 keV The bin boundaries of the wider binning could be 25 31 39 48 70 103 150 219 320 502 520 1000 and 8000 keV When the model is construc
110. ons in the image reconstruction For example you may want to fix the position of the bright well known sources to see whether the detection significance of a fainter source improves as it should in case of a real detection First you may want to save the previous results The easiest way is to make a full copy of your current working directory cd cp r spi analysis spi imaging without catalogue cd spi analysis The source positions have to be provided to spiros through a source cat fits file You can either use an existing file or create a new one with the catalogue extraction task middle section of the main GUI Here we explain how to produce a new source cat fits file including sources found in the ISDC reference catalogue in the observation field of view 6 4 1 Creating a new source cat fits file Launch spi science analysis and select the catalogue extraction We strongly advise to stop the script after this step and to check the results before going again further into the analysis This is done by de selecting all tasks but catalogue extraction Figure 18 Catalogue extraction GUI catalog Catalog extraction parameters Minimum radius fo Maximum radius 20 Column for flux fi Minimum flux 0 001 Maximum Flux 1000 Reference Catalog ISDC REF CAT SPI FLAG 1 To change the catalogue extraction options just click the catalog button and the catalog GUI will appear Options are the maximum and minimum
111. other time the script automaticcaly deletes the previous output files with the command rm f spimodfit therefore if the user wants to compare the results of subsequent runs he she should copy the relevant files elsewhere At odds with the spi science analysis pipeline the results of spimodfit are not attached to the og group therefore the deletion of these files does not produce any harm to the og tree 6 7 1 Further settings The central frame allows the user to define a source variability time scale This is applied to all sources in the local catalog and is a delicate feature since it increases the number of parameters that are fitted The default is non variability but different choices can be done for highly variable sources especially in the crowded fields The format of the variability defines nodes IJD d n where IJD is the INTEGRAL Julian day and intervals DD di where DD is the variability timescale in days The variability should be applied only to energy bins with high enough S N typically at the lower energy range for SPI It is not currently possible to separate within the pipeline different spectral regions therefore subsequent runs of spimodfit should be done to allow spectral variability only in some bands and the results combined by the user using his her own tools On the upper right frame of the GUI Fig 23 we provide the functionality for simultaneous fitting of diffuse emission and point sources but we en
112. parallel processing of the single detector events in order to provide additional infor mation about their pulse shape The PSD information was intended to help reducing the background Unfortunately the in flight background conditions are such that even the best experts have failed to achieve significant improvements with the PSD Consequently all the PSD related processing is currently disabled in the analysis pipeline PSD events are simply used as standard single events The basic user choice is then to analyze only single PSD events specifying detector list of 0 18 in the analysis or to consider double and triple detector interaction with detectors 19 84 Three instrumental responses are now included in our system characterizing SPI before between and after the detector 2 and 17 failures Another will be available in the next OSA release to cope with the failure of detector 5 The spi science analysis pipeline cannot currently handle a time variable response The easiest is to analysis the three possible cases independently our software then selects automatically the appropriate response and to combine the results later on It is possible however to analysis different mixtures of different data together using one of the three responses as they are not too different Great care should be taken in this case anyway to avoid possible biases see the Tips and Tricks section of our documentation The spimodfit_analysis can instead handle mu
113. pecific analysis you can either 1 use an existing source_cat fits or 2 perform an extraction cat_extract CAT I first step of the analysis from the ISDC reference catalogue Since the outputs of cat_extract are sometime hard to predict it is strongly advisable to run only cat_extract first and to check the result with the ftool fv before going further in the analysis If there are sources that should not be taken into the analysis either set the SEL FLAG column one before the last one to zero or removed the un wanted rows Alternatively make sure SEL FLAG is set to one for all sources that should be used in the analysis In general it is difficult to know in advance which of the known sources will actually be detected by INTEGRAL It is therefore a good practice to make an analysis without specifying any source positions to look at the results possibly comparing them with the outputs of cat extract before evaluating which source positions should be fixed in the analysis 7 5 Event energy correction Since the start of the second revision of the ISDC processing sets of gain coefficient files are part of the analysis input data For each revolution a dedicated ISDC pipeline 1 builds spectra from all available events 2 derives the positions of a number of background lines and 3 compute gain coefficients by fitting a polynomial to the line position energy relation see Sect 8 2 for more details The spi science
114. played possible values 1 for all pointings and 2 for their sum default 2 main title string Main title of the plot appears only on the top of the display default Detector spectrum perkev boolean Flux units possible values yes for displaying the spectra in counts keV no for displaying the spectra in counts parameters perkev and persec can be used in any combination default yes persec boolean Flux units possible values yes for displaying the spectra in counts sec no for displaying the spectra in counts parameters perkev and persec can be used in any combination default yes chatter integer Chatter level possible values 0 10 O no screen dump information 10 maximal screen dump default 1 ISDC SPI Analysis User Manual Issue 9 0 77 10 Known limitations 78 SPI is a complex Gamma ray instrument almost always dominated by background contribu tions The scientific validation of the SPI data analysis going on at ISDC and at different instrument team sites is as of today far from complete Users are encouraged to look crit ically at any result obtained with the ISDC software and to use external comparisons and simulations when possible Spurious results can be derived for example when using a wrong set of parameters and or an incorrect background modeling The SPI instrument is equipped with a Pulse Shape Analysis PSD electronic which carries out a
115. ps and the third contains the results from spiros 7 After a successful execution of spi grb analysis it is also possible to run spiros again with the spi science analysis script Launch spi science analysis make sure you select only the spiros step if you select any other step the burst background produced by spi grb analysis will be deleted and you will have to start it all again change the required spiros parameters making a LIKEH analysis rather than a Chi2 one for example and run the show again 8 Please consult the cookbook for information about the analysis results and for more infor mation about the analysis parameters ISDC SPI Analysis User Manual Issue 9 0 49 7 Tips and Tricks This section gathers a number of important information and advises about the SPI data analysis We recommend to browse through this section after having exercised the Cookbook Sect 6 and to come back to one point or another when you are confronted with a particular issue in your own analysis 7 1 Example of command line run The usage of the spi science analysis GUI is strongly recommended It allows to enter all important analysis parameters see Sect 7 20 for how to modify the list of important parameters and then to run the script However command line run are required in some cases because either 1 spi science analysis is called from another higher level script or 2 spi science analysis is ru
116. r advices on how to use catalogue information with spiros 6 4 2 Using a source cat fits file in the analysis In order to use fixed source positions in an analysis launch spi science analysis de select the catalogue extraction step and provide the name of your catalogue file in the SPIROS input catalogue entry field We advise to use the standard name source_cat fits for input catalogues At this stage the complete analysis can be run again and previous results will be overwritten However the catalogues information is only used in spiros and if there is no need to change other analysis parameters such as the energy binning only this last step can be repeated Select Image analysis click on spiros and then on imaging You can then change the maximum No of sources to 2 instead of 3 This change is not really critical but for a better comparison with the previous analysis we would like to restrict spiros to a 3 source analysis Since the Crab source is now provided through the catalogue we specify a maximum of 2 additional sources Click twice on Ok to close the spiros GUIs and Run on the main GUI The script will execute spiros and stop The resulting images can be viewed with ds9 and the source res fits file with fv as explained before The log file situation is now a bit more complicated If you type ls rtl spi_sa log you will notice three log files In fact the
117. relevant information such as the pointing pattern the detector list etc It also does not create any new data structure but it add counts to the detector spectra data structure SPI OBS DSP updating this file At the moment only one source can be added at one time However several sources can be added just by running the program several times The program works only if the image analysis is carried out in one energy bin only If there is more than one energy bin spi add sim does not do anything and the detector spectra will not be changed Table 8 spi_add_sim parameters included into the main script Name Name Type Description Main script executable spi_add_sim_CoordSys CoordSys string Coordinate system possible values CELESTIAL GALACTIC default CELESTIAL spi add sim SrcLong SrcLong real Longitude of simulated source degree default 80 spi add sim SrcLat SrcLat real Latitude of simulated source degree default 19 spi add sim FluxScale FluxScale real Simulated source flux ph cm sec default 0 01 ISDC SPI Analysis User Manual Issue 9 0 63 8 5 Background model generation SPILOBS_BACK spi_obs_back produces background models for imaging timing and spectral analysis The simplest case is to assume a background that does not change with time The background then has only one free parameter per detector in the image analysis The goal of the background
118. rgy bins this might take some time to calculate e TRANSIENT This upgraded mode is similar to WINDOW mode but allows the removal of the discontin uous nature of the HAT like time bin functions to return an implicitly smoothed light curve The integration time intervals spanning the period of observation are used here to create a sequence of splining nodes or knot points which will define a sequence of piecewise or smoothly continuous B spline functions Select either TRANSIENT HAT TRANSIENT LINEAR TRANSIENT QUADR or TRANSIENT CUBIC If no X XXX function is appended a simple HAT type is selected and the results should be the same as in WINDOW mode These functions are actually created by taking a HAT function covering each time interval and smoothing it with a HAT filter several times over to create LINEAR QUADRATIC or CUBIC B spline functions a sequence which will tend towards a Gaussian shape with each further smoothing 8 6 4 Parameter list The spiros parameters included into the main script are given in the Table 10 Table 10 spiros parameters included into the main script Name Type Description mode string Select processing mode possible values IMAGING IMAGING TIMING SPECTRA SPECTRAL TIMING TIMING DIFFUSE default IMAGING background method integer Handling method of background model response values possible values 0 ZERO 1 FIXED 2 FLOATING
119. s spi phase hist phaseBinNum 20 N 50 ISDC SPI Analysis User Manual Issue 9 0 spi_phase_hist_phaseSameWidthBin yes spi phase hist phaseBounds spi phase hist phaseSubtractOff no spi phase hist phase0ffNum 0 spi phase hist orbit no N spi phase hist asini 0 spi phase hist porb 0 spi phase hist T90epoch 0 spi phase hist ecc 0 spi phase hist omega d 0 spi phase hist pporb 0 spi add sim SrcLong 80 spi add sim SrcLat 19 spi add sim FluxScale 0 01 N use background flatfields yes use background templates no use background models no N spi flatfield ptsNbConstBack 5 spi flatfield single no N spi templates type GEDSAT spi templates scaling 1 N spi obs back nmodel 1 spi obs back model01 GEDSAT spi obs back mpar01 spi obs back norm01 NO spi obs back npar01 spi obs back scale01 1 spi obs back model02 GEDSAT spi obs back mpar02 spi obs back norm02 NO spi obs back npar02 spi obs back scale02 1 spi obs back model03 GEDSAT N spi obs back mpar03 spi obs back norm03 NO spi obs back npar03 spi obs back scale03 1 spi obs back model04 GEDSAT N spi obs back mpar04 spi obs back norm04 NO spi obs back npar04 spi obs back scale04 1 spiros mode IMAGING spiros energy subset spiros_pointing subset MX spiros detector subset spiros background method 3 spiros srclocbins FIRST
120. s E E 000 X xterm isdcsf grb030501 313 gt spi_grb_analysis E a a a a a a oo EE EE Ee dd dd dd dd dd oo A O a a spi_orb_analysis analysis of a single pointing only Version 1 3 Date 1 September 2006 Author Pierre Dubath oe t oe oe eo oo oe ee oe dd dd oo oe oo oo RS Usaget spi_grb_analysis lt grb_start gt lt grb_stop gt UTC IJD 4 lt sec_to_avoid_before_grb gt lt sec_to_avoid_after_grb gt the two last parameters are optional defaulted to 20 and 120 seconds respectively UTC format for lt grb start gt lt grb stop is like 2003 05 01703 10 10 000 gt Check your inputs and make sure your observation group include only one pointing ul isdesf tgrb030501 314 gt E You can directly enter XSPEC to display and fit your resulting spectrum see Sect 7 18 for more information 6 spigrb analysis is calling a number of executables and spi science analysis several times Since spi science analysis produces its own log file the log file situation after running spi grb analysis is somewhat complicated The resulting spi grb analysis log is the high level log file containing logs from the executables called directly by spi grb analysis In ad dition you can find three log files spi_pointing log spi binning log and spi spiros log The first contains the outputs of spi obs point the second includes the logs from the binning to background ste
121. s e g event tables cut and stored by ScWs aux auxiliary data provided by the ground segments e g time correlations cat ISDC reference catalogue ic Instrument Characteristics IC such as calibration data and instrument responses idx set of indices used by the software to select appropriate IC data and all data resulting from your analysis will reside in an obs branch in that same directory which will be created by the first analysis step og_create see below Part of the above input data may already be available on your system You can either copy the data to the relevant directory or better create soft links as follow 1f the setenv command fails with a message like setenv command not found or setenv not found then you are probably using the sh family In that case please replace the command setenv my_variable my_value by the following command sequence my_variable my_value export my variable In the same manner replace the command source my script by the following command my script the is not a typo The Instrument Characteristics files OSA IC package and the Reference Catalogue OSA CAT package are part of the OSA software distribution They should be installed following the Installation Guide for the INTEGRAL Data Analysis System 2 ISDC SPI Analysis User Manual Issue 9 0 15 ln s directory_of_ic_files_installation__ ic ln s dir
122. searches for the brightest source first it then removes its contribution to the original binned event data set and continues iteratively with increasingly fainter sources The maximum number of sources SPIROS should search for must be specified as well as the minimum acceptable sigma for source detection The number of sources is meant to be in addition to those that may be specified in a source cat file For the source number try first with the number of sources that should be detected in your INTEGRAL data set given the field of view plus 1 or 2 for possible unexpected sources For the sigma threshold use 5 or 6 to be safe Below that level you may get spurious detections as SPIROS is a bit optimistic at the lower S N end Since SPIROS reconstruct a sky image the field of view FOV has to be specified A convenient way to do that is to use the FOV parameter With POINTING ZCFOV SPIROS will always image a square angular region defined by the extent of the centers of all pointings augmented by the SPI zero coded FOV see below Sect 7 9 for a possible strong bias Sky images are produced at each iteration The final images produced after the last iteration are spiros image intensity result fits and spiros image sigma result fits The source positions and fluxes are output in source_res fits 54 ISDC SPI Analysis User Manual Issue 9 0 7 9 Biases from bright sources just outside the image FOV Because the output s
123. ses and selection of good pointing 10 September 7 0 Update for OSA 7 0 Release Cookbook adapted to the new 2007 features of spi science analysis 1 June 2009 8 0 Update for OSA 8 0 Release Cookbook updated for the introduction of spimodfit analysis 1 March 2010 9 0 Update for OSA 9 0 Release Revision of flat field options 28 APR 2010 Printed ii ISDC SPI Analysis User Manual Issue 9 0 Contents Acronyms and Abbreviations aooaa a a RR RR RR Glossary of Terms 1 Introduction I Instrument Definition 2 Scientific Performance Summary saaa LL RR RR 3 Instrument Description ee ee ee 3 1 TheOverall Designs tt coe ah ete ad AA DA A 3 2 The Passive Mask ir fico ik ee oe EA 3 3 The Detectors a aura de gear arara hoe ole Soe ee ee BAR Ewes a 3 4 Event Types gas 2 4 2 Fada ee tee haw ee Lk ek ae A A a a als 3 5 Pulse Shape Discriminator PSD o oo oo o 3 6 Anti Coincidence Subassembly ACS o o oo oo o 3 7 The Plastic Scintillator Anti Coincidence Subassembly PSAC 3 8 Electronics s 4 4 ARA A E it AAA dd ye 3 9 Operating Mod s ara 4 4 re a RO EG ee A E S 3 10 Dithering Strategy a ae saa eae BA ee Re ED Sa 4 Performance of the Instrument 0 0 0 0 20000000 4 1 Components and Sources of Instrumental Background 4 1 1 Continuum Background 2 2 0000 4 1 2 511 keV Background LL LL LL
124. seudo detectors The response of these pseudo detectors is computed from GEANT simulations and they are included in the analysis as if they were real detectors Events hitting more than 3 detectors are increasingly dominated by background events They are not normally used in scientific analysis In general for analysis up to a few hundreds of keV the inclusion of the double and triple events does not significantly improve the results In other words for point sources most analysis can be restricted to real detectors i e a detector list 0 18 It is probably always a good idea to start an analysis first using only real detectors anyway For high energy analysis double pseudo detectors 19 60 and triple pseudo detectors 61 84 can be added later into the analysis using an extended detector list Note however that the current RMF response available for XSPEC spectral response is valid for analysis with real detectors 0 18 only Sor an alternative OG name if you do not follow the standards 52 ISDC SPI Analysis User Manual Issue 9 0 7 4 Catalogue extraction Although it is possible to carry out an image reconstruction without any prior information better results will be obtained if the positions of sources known to lie in the field of view are specified beforehand In any case source positions have to be provided for spectral extraction Source positions are provided through a source_cat fits file For a s
125. siest starting point in this cook book is to copy the data set used for imaging with a catalogue for which we used a single energy bin from 20 40 keV The command below are to save the results of spectral analysis and to start again with the imaging data set Note that the below naming is somewhat arbitrary working directly in a spi timing directory could also be meaningful cd mv spi_analysis spi_spectra cp r spi imaging with catalogue spi analysis cd spi analysis Launch spi science analysis check that an input catalogue is provided and de select all tasks except spiros Open the Spiros options GUI and set the spiros mode to TIMING Click on the timing button to get the timing GUI Set the mode to QUICKLOOK and the Timing Scale to 0 The Timing Scale is the time resolution of the resulting light curve in days Setting it to O will give you one data point per science window ISDC SPI Analysis User Manual Issue 9 0 33 the highest time resolution possible At the moment it is not possible to create light curves with a resolution smaller than one science window Click Ok twice and then Run The resulting light curve is written to spiros timing data fits This file contains one extension for each source plus the background first data extension This light curve can be displayed with e g fv to produce the output presented in Fig 21 soiros timing data fits FLUX 1 10 0 ss asc
126. sirable flux with the parameters fluxMin and fluxMax Table 6 cat extract parameters included in the main script Name Name Type Description main script executable cat extract radiusMin radiusMin string Lower limit for the position selection default 0 cat extract radiusMax radiusMax string Upper limit for the position selection default 20 cat extract fluxDef fluxDef string Column used for flux selection possible values 1 15 300 keV 2 0 3 8 MeV default 1 cat_extract_fluxMin fluxMin string Low limit for flux selection ph cm sec e g Crab catalogue value is 0 4458 15 300 keV 0 0167 0 3 8 MeV default 0 001 cat extract fluxMax fluxMax string High limit for flux selection ph cm sec default 1000 8 2 Energy correction SPI GAIN CORR When running the spi science analysis script event energy calculation is carried out on the fly within the spi obs hist program On the other hand event energies can be computed off line with the program spi gain corr In the two cases the corrections are computed in the same way deriving the energies in physical units keV from the instrumental channel PHA using a set of ISDC SPI Analysis User Manual Issue 9 0 61 gain coefficients The correction can be evaluated with the help of up to five gain coefficients c However currently four coefficients are used for the
127. specify parameters through the GUI as displayed in Fig 38 b Check the List of detectors use 0 18 in most cases and the Coordinate System in that window de select catalogue extraction enter the name of your file source cat fits in the SPIROS Input Catalog entry field and click on Energy definition to open the GUI window displays in Fig 39 c Select your energy bins in this case 6 bins ranging from 20 to 200 keV logarithmically spaced as the bin number is negative click Ok This will close this window d Click on Background options and un select the flat field defaults option as shown in Fig 34 and click Ok to close this window e Back in the parent window click spiros to open the GUI window shown in Fig 40 Make sure the SPECTRA mode is specified do not introduced any selection param eters and specify the Background method and the Optimization statistic Back ground method 1 means that the background derived from the periods before and after the burst and rescaled to the burst duration will be subtracted from the burst data as a fixed background With background method 3 the background will be scaled through the image deconvolution process solving for a single background scaling coeffi cient Do not use other options For the Optimization statistic Chi2 is recommended while LIKEH can be tried in a second run 46 ISDC
128. st important to look at the individual program user manuals These user manuals can be downloaded from the following WWW page http isdc unige ch Support spi doc doc html A thorough scientific validation of an important part of the ISDC software can be found in the paper The INTEGRAL spectrometer SPI performance of point source data analysis published in MNRAS 4 which also contains a number of useful informations and recommendations For analysis of Gamma ray bursts a dedicated script spi_grb_analysis is available This script is documented on line at http isdc unige ch index cgi Support spi ISDC SPI Analysis User Manual Issue 9 0 1 Part I Instrument Definition 2 Scientific Performance Summary The SPI telescope has been designed to provide at the same time a medium angular resolution and an excellent energy resolution in the range 20 keV 8 MeV The main characteristics of the instrument are given in Table 1 The imaging capabilities are limited by the fact that SPI consists of only 19 detectors pixels and that the data are strongly background dominated To help distinguishing between source and background photons the number of measurements is increased by dithering the whole spacecraft around the target of interest A typical observation is thus split into a number of pointings whose attitude differ by about 2 degrees see Section Strategies of scientific Observations in the Introdu
129. t but the analysis software is still able to handle the data in order to produce spectra and images as described in this document The maximum data generation rate in this mode will be about half the rate for normal photon by photon mode 3 10 Dithering Strategy The SPI data are background dominated The background in each of the 19 independent detectors can vary in time in a different way Several types of background variations can be anticipated e short term variations due to solar activity e variations over the orbital period related to the position of INTEGRAL in the orbit e long term variations over the mission duration In order to reconstruct the image on the detectors for all pixels in the field of view 25 degree field with 2 degree resolution for a single pointing a set of 19 equations with 156 unknowns would need to be solved This is impossible and the only way to increase the number of equations and make the system solvable is to use more pointings Thus in order to solve the problem of background determination an appropriate dithering strategy has to be adopted for every observation The imaging performance of SPI depends on the dithering pattern that is used In general the greater the number of pointings the better the imaging It was shown that when using the hexagonal dither pattern the reconstructed point source response function shows very strong side lobes at distances of 10 to 20 from the center Therefore th
130. t binning and derivation of the dead times Outputs three files 1 gti fits containing always good GTI with ontimes 2 dead_time fits containing dead time ratios and livetimes 3 evts_det_spec fits containing the binned events formatted as spectra spi phase hist Optional event binning in phase intervals computed from pro vided ephemeris and taking into account possible orbital motion Outputs one file per phase bin named evts_det_spec_i fits where i is the phase bin number BKG spiobs back Computes background models from different indicators Two main recommended options are 1 deriving the background from pre defined flat field files and 2 using the Germanium satu rated events as a background tracer Outputs are stored into back_model fits IMA spiros Carries out image reconstruction IMA spectral extraction or SPE or timing analysis LCR depending on the chosen mode SPE This is the most difficult and critical step If the phase binning as or been selected all phase bins are processed independently in turn LCR spirmf Invoked in spectral extraction only creates the appropriate RMF response matrix required for XSPEC spectral fitting 14 ISDC SPI Analysis User Manual Issue 9 0 Since the release of OSA 8 another script is available for the spectral extraction and the contempo rary fit of diffuse emission spimodfit_analysis This script performs all the steps of spi_science_ana
131. t catalogue When og_create has finished without any errors move from the top level directory SREP_BASE_ PROD into the analysis directory obs spi analysis and launch the spi_science_analysis script 18 ISDC SPI Analysis User Manual Issue 9 0 cd obs spi_analysis spi_science_analysis After a few seconds the GUI will appear provided that the DISPLAY environment variable is set properly otherwise you will be prompted for the individual parameters on the command line In general but especially for learning SPI data analysis we strongly recommend using the GUI Figure 10 spi science analysis main GUI spi science analysis og spifits RADEC The top section holds the most basic parameters of the pipeline 1 the list of detectors which are to be used in the analysis 2 the coordinate system Do not change the default values for now The middle section concerns the catalogue extrac tion specification part For most bright Gamma ray sources the available catalogue source posi ISDC SPI Analysis User Manual Issue 9 0 19 tions are more accurate than the positions which can be derived from the SPI data due to the limited spatial resolution The best results are therefore obtained by fixing the known source positions in the image reconstruction process However as some sources are highly variable and the catalogue fluxes not always reliable it is sometimes difficult to predict which sources
132. t catalogue name source cat fits in the SPIROS input catalogue field It is not necessary to run the pointing step again However the number of considered energy bins is typically much larger in spectral extractions than in imaging reconstructions Select the binning background and spiros steps In the energy binning step you can now select a binning suitable for a spectrum of a bright source with a broader energy range and a relatively high number of bins For our example set the number of regions to 71 a 20 1000 keV boundary and 100 as the number of bins Using a negative bin number implies a logarithmic binning with 100 bins from 20keV up to IMeV in our case Open the spiros GUI and set spiros mode to spectra instead of imaging Click Ok and Run the pipeline After spiros has finished successfully the pipeline will automatically generate the appropriate response matrix and attached it to your spectra Now launch XSPEC to analyze your spectra In the following example we fit a single power law from 30 to 1000keV Here is a verbatim of the XSPEC session Xspec 11 3 1 17 41 21 26 May 2005 For documentation notes and fixes see http xspec gsfc nasa gov Plot device not set use cpd to set it Type help or for further information XSPEC gt data spectrum_Crab fits POISSERR keyword not found assuming FALSE Net count rate cts s for file 1 0 3903 3 3501E 03 using response RMF file sp
133. ta points but entering AUTO in the pointing subset You can also use the AUTO filter in addition to specifying the pointings to use For example AUTO 5 25 30 50 will use pointings 5 25 and 30 50 and apply the AUTO filter to these pointings All other pointings will not be used The bottom section allows you to choose the Background method the Optimization statistic and the Bins for src location For the background method you must use method 3 if you selected the flat field model If you are using the GEDSAT model use method 2 you can also use background method 5 also known as MCM but be aware that this method may produce wrong results if you have bright variable sources in the field of view With spiros background method 3 one background scaling coefficient is derived for each time interval specified as a number of pointings in the background model step assuming that detector to detector background variations follow those from the flat field model ISDC SPI Analysis User Manual Issue 9 0 25 26 With spiros background method 2 one background scaling coefficient is derived for each detector in the image reconstruction process In this way we assume that the background variations in all detectors follow that of the background model see Sect 7 8 or the SPIROS User Manual for more explanations For the optimization statistic you can choose between maximum likelihood or chi square The lat
134. ted all parameters can be tested for relevance and parameters which do not show an effect on the fit function can be excluded Using a maximum likelihood fit which is initialised from a y fit the model is then fit to the data and the statistical errors are determined from the covariance matrix For more details see the spimodfit specific handbook The entire analysis can be carried out independently in each energy bin such that boundary conditions such as variability timescales and the input point source catalog can in principle be chosen differently for each energy bin This also allows for parallel computation However spectra should then be collected by the user with tools which should be developed ad hoc If the user should run again spimodfit outside the spimodfit_analysis chain since its products are not attached to the observation group he she should manually delete the previous results e g with the command rm spectr back_model spimodfit res fitquality fits spimodfit_rmfgen csh lightcurves fits catalogue spimodfit fits 8 7 5 Known issues spimodfit handles time variability through the use of splines The spline order can be 0 to 5 0 corresponding to a piecewise constant function with one scaling parameter per interval and 3 corresponding to cubic splines In many cases when using n order splines n gt 1 the fitting algorithm fails to find the optimal parameters This is thought to be due to over parametrized time variability b
135. ter is faster and more robust but can be problematic if you have lots of energy bins with zero counts For imaging in broad energy bands chi square is definitely better while you might want to compare results obtained with chi square and likelihood options when extracting spectra The bins for source location determines in which energy bins spiros locates the sources The typical setting will be FIRST where spiros uses just the first energy bin But you may also use ALL where spiros will search for sources in each bin independently or you may give a range of bins to be used For our example we use background method 3 chi square as optimization statistic FIRST for the source location bins The options specific for imaging are available by clicking on the button imaging in the spiros GUI Figure 17 SPIROS imaging mode GUI imaging General imaging options No of sources 39 Sigma threshold 3 4 Image Options Projection CAR FOY POINTING FCFOV Orient STANDARD y Pole longitude 0 Pole latitude 30 gt Other options Iteration output No Location max error 0 1 The top section allows you to select the maximum number of sources you want spiros to look for in addition to those that might be provided through a catalogue see Sect 6 4 and a ISDC SPI Analysis User Manual Issue 9 0 lower sigma threshold SPIROS stops its iterative source search process either when it has fou
136. thly timescale variation to be fitted In case the user would like to use the external background the parameter background_input_file should be set to back model idx fits 1 Either the flatfield template or model options can be used For the flatfield option the option spi flatfield single should be set to yes as done in the default settings Also a combination of several background models can be used by setting the 70 ISDC SPI Analysis User Manual Issue 9 0 pimodfit_analysis 900 X Energy definition Genaral Parametere Parameters energy binning Number of energy regions List of pseudo detectors 0 16 ees Coorelnate system RADEC gt Regions energy ooundanes z5 5000 Numbers of bins in each region 20 Help hidden eos TSE OPTIONAL first task check output before proceeding with further tasks p Select type of Background Estimation Dk Helo CATA catalog extraction I Catalog optione Use fiatfielde to model background 7 SPIMDDEIT Input Catalog e tia Use tracers to model background T Background Model options Select analysis tacks Poin pointing definon F Pointing options Flat field model dotoctor variation model hunter or poings vin const veck 54 _ Hep Bin event binning 7 REAPER Histogram options nte single background extension 7 EKG background modeling FO Bac er spimo
137. tion of the ISDC homepage 10 7 12 Converging on a good solution checking residuals At the end of any SPI data analysis the SPIROS solution residuals provided in the log file must be checked DO NOT TRUST any solution with residuals larger than 3 standard deviations The only possible exception is when considering one of the few cases with relatively high signal to noise and signal to background ratio such as the Crab or Vela In these cases large residuals may come from non poissonian errors which are not taken into account by SPIROS The first analysis of any dataset will in most cases have too large residuals The first thing to check is the table provided in the SPIROS log file containing the solution residual for each pointing Although bad pointing are filtered in the first pointing steps there might still be a few pointings with large residuals The next step is to activate the AUTO filter of spiros by entering AUTO in the pointing subset entry field Re run the pipeline and check the before mentioned table in the log file It is important that no high residuals remain Inappropriate background modeling will also lead to large residuals Try to improve the background modeling changing time scales and or methods Check the list of detected sources Large residual can be induced by the presence of bright source located just outside the field of view Another case is when SPIROS is limited to a number of sources through t
138. urrent directory i e REP BASE PROD as explain above The file spi lst will be used later as an argument for the og_create program In this file you find the lines scw 0102 010200200010 001 swg fits 1 scw 0102 010200210010 001 swg fits 1 scw 0102 010200220010 001 swg fits 1 scw 0102 010200230010 001 swg fits 1 scw 0102 010200240010 001 swg fits 1 scw 0102 010200250010 001 swg fits 1 scw 0102 010200260010 001 swg fits 1 scw 0102 010200270010 001 swg fits 1 scw 0102 010200280010 001 swg fits 1 scw 0102 010200290010 001 swg fits 1 which is the list of DOLs Data Object Locators of the Science Window Groups you want to analyze The DOLs includes the path and name of the FTTS file and the location of the SWG grouping table inside the file remember that a FITS file can include more than one table or image arranged in a number of extensions For example the last DOL in the above list in dicates a SWG located in the first extension of the FITS file swg fits lying in the directory scw 0102 010200290010 001 You can download the data pressing the Request data products for selected rows button In the Public Data Distribution Form provide your e mail address and press the Submit Request button You will be e mailed the required script to get your data and the instructions for the settings of the IC files and the reference catalogue 6 1 3 Setting the software environment Before you run any O
139. ut image projection type possible values CAR TAN AIT default AIT image fov string Enter extent and location of output image possible values POINTING FCFOV POINTING ZCFOV POINTING USER default POINTING FCFOV image orient string Enter output image orientation possible values STANDARD USER default STANDARD image pole long real Enter longitude of output image pole degrees 68 default 0 0 ISDC SPI Analysis User Manual Issue 9 0 image pole lat real Enter latitude of output image pole degrees default 90 0 iteration output string Enter whether output should be produced at each iteration default NO location max error real Maximum admissible input error source location default 0 1 8 7 Spectral Extraction SPMODFIT The tool spimodfit is intented to perform model fitting with SPI data By model fitting we mean the fit of a linear combination of sky emission models point sources and instrumental back ground The coded mask imaging y ray spectrometer SPI on the INTEGRAL space observatory will detect point sources and diffuse extended emission with an angular accuracy of about 1 over its energy range of 20 8000 keV The purpose of the software package described here is to repre sent the measurement in terms of spatial models of the sky fitting parameters of these models and estimating the
140. want to run the analysis yet This procedure is just a convenient way to enter parameters for the later analyses Alternative If you do not want to use the GUI you can specify the above parameters by editing the spi science analysis par file located in your PFILES directory 6 Now the grb_analysis par file located in the current directory now contains the main parameters you want to use in your burst analysis This file will be used by the script 7 Run spi grb analysis Without any command line argument you get the following help text Figure 37 Outputs of spi grb analysis when invoked without any input parameters 9900 X xterm isdesf7tgrb030501 313 gt spi_grb_analysis E a RE dd od od od od do od od o aa SG RA od do od od o od od do EE od od od od do od do EE EE EE EE spi grb analysis analysis of a single pointing only Version 1 3 7 Date 1 September 2006 Author Pierre Dubath Add oo de ee dd dd oo dd dd dd od dd dd od dd dd od dd ok oo oo oo dde Usage spi_grb_analysis lt grb_start gt lt grb_stop gt UTC IJD 4 lt sec_to_avoid_before_grb gt lt sec_to_avoid_after_grb gt the two last parameters are optional defaulted to 20 and 120 seconds respectively UTC format for lt grb_start gt lt grb stop is like 2003 05 01703 10 10 000 gt Check your inputs and make sure your observation group include only one pointing nj isdes tgrb030501 314 gt E You know the input parameters
141. x gt 10 is obtained a warning is issued and the user should consider to vary the input catalog the variability time scales and or the background settings For the user interested in spectral analysis of point sources the relevant output are the spectra named spectrum_SRC fits and the response matrix spectral_response rmf fits no ancillary effective area is necessary In our case the Crab spectrum is spectrum_Crab fits whose fit with a power law yields a spectral index 2 10 0 03 and normalization at 1 keV 11 0 0 7 Fig 26 The ISDC SPI Analysis User Manual Issue 9 0 35 00900 X spimodfit analysis m General Parameters Save As List of pseudo detectors 0 1 8 elo a Coordinate System RADEC Bun Quit Help hidden OPTIONAL first task check output before proceeding with further tasks CAT catalog extraction 7 Catalog options SPIMODFIT Input Catalog source_catiits 1 m Select analysis tasks POIN pointing definition IV Pointing options BIN event binning M Energy definition Histogram options BKG background modeling I Background options Fite spimodfit analysis IV Spimodfit options Figure 22 GUI of spimodfit_analysis with the default setting response matrix is computed by the script spimodfit_rmfgen csh which should be run manually in case the file spectral response rmf fits is not present To run the fitting step an

INTEGRAL Science Data Centre SPI Analysis User Manual

Contents

Download Pdf Manuals

Related Search

Related Contents