Home

AKARI IRC Data User Manual

1. lll ii 31 33 34 35 35 35 43 43 44 44 44 AT 47 iv IRC Data User Manual 6 Spectroscopy pipeline cookbook 83 6 1 General overview of the pipeline processing ln 83 6 01 Dark siibtraction uates A a ERROR EE LER 83 6 1 2 Linearity correction 2 2 ee 83 6 1 3 monochromatic Flat helding uo x ox ex ee a es 83 6 1 4 Background subtraction from individual sub frames 84 6 1 5 lmapgesscreenmnge sarase gyae a A Exp ACER dq RCECRAS 84 6 1 6 Image registration among sub frames llle 85 617 Imaging stacking ia dup Dese censui pie be puo y Ron EUR dO P 85 6 1 8 Target detection position measurement 85 6 1 9 Background subtraction from stacked image len 85 6 1 10 Extracting 2D spectra rh 85 6 1 11 Wavelength calibration 22e 86 6 1 12 Flat color term correction 2 eh 87 6 1 13 local Background subtraction from extracted 2D spectra 88 6 1 14 Spectral tilt correction ooo ee 88 6 1 15 Spectral response calibration ee 88 6 1 16 Notes on slit spectroscopy s 88 6 2 How to install and to set up the IRC spectroscopy pipeline 89 6 24 1 Data preparation u lll re 89 6 3 Calibration data 2 9 etur ee qoe RA e TE ete y Se iei te RA nS 89 6 3 4 Calibration fles a ss ers ace seo Rok p Ro o9 odo E m EU RR E 90 6 4 Running the pipeline 2 2 les 92 6 4 4 Data reduction order 22e
2. Telescope axis Figure 2 1 4 Side view of the MIR S channel The beam splitter that splits the light into NIR and MIR S channels acts as a folding mirror for the MIR S 2 1 3 Long wavelength Mid InfraRed Camera MIR L The MIR L channel consists of 5 lenses of CsI and KRS 5 Fig 2 1 5 It also has a small slit for spectroscopy of diffuse emission Fig 2 0 2 Unfocused ghost images are known to exist for very bright sources whose location depends on the position of the true source in the field of view The origin of the ghost is not yet clearly identified but it is most likely to come from the scattering by the optical elements of the MIR L channel internal scattering Version 1 3 September 13 2007 9 KRS 5 Filter wheel Detector module Lj Cylindrical mirror Telescope axis Figure 2 1 5 Side view of the MIR L The folding mirror is cylindrical to correct the astigmatism Thus the pixel scale of the MIR L channel is not the same for the X and Y directions 2 1 4 Arrays operation The focal plane arrays are operated in a synchronized manner The MIR S and MIR L arrays are operated in the same clock pattern The array operation is made in a unit frame which consists of short and long exposures The short exposure is intended to increase the dynamic range by about 8 to 10 times The clock pattern of one frame is the same for all AOTs except for IRC0
3. LMC FIELD218 256 S11 80 53135 68 17282 IRCO2 1 2210218 1 F004160558_S002 LMC FIELD218 256 S11 80 53135 68 17282 IRC02 2 2210218 1 F004160558_S003 LMC FIELD218 256 S11 80 53135 68 17282 IRCO2 3 2210218 1 F004160558_S004 LMC FIELD218 256 S11 80 53135 68 17282 IRCO2 4 2210218 1 F004160559 NO01 LMC FIELD218 412 N3 80 53135 68 17282 IRCO2 1 2210218 1 F004160559_N002 LMC FIELD218 412 N3 80 53135 68 17282 IRCO2 2 2210218 1 F004160560_L001 LMC FIELD218 256 UNDEF 80 53135 68 17282 IRCO2 1 2210218 1 F004160560_L002 LMC FIELD218 256 UNDEF 80 53135 68 17282 IRCO2 2 2210218 1 F004160560_L003 LMC FIELD218 256 UNDEF 80 53135 68 17282 IRCO2 3 2210218 1 7 F004160560 L004 LMC FIELD218 256 UNDEF 80 53135 68 17282 IRCO2 4 2210218 1 F004160560_S001 LMC FIELD218 256 UNDEF 80 53135 68 17282 IRCO2 1 2210218 1 F004160560_S002 LMC FIELD218 256 UNDEF 80 53135 68 17282 IRCO2 2 2210218 1 F004160560_S003 LMC FIELD218 256 UNDEF 80 53135 68 17282 IRCO2 3 2210218 1 F004160560_S004 LMC FIELD218 256 UNDEF 80 53135 68 17282 IRCO2 4 2210218 1 F004160561_N001 LMC FIELD218 412 UNDEF 80 53135 68 17282 IRCO2 1 2210218 1 F004160561_N002 LMC FIELD218 412 UNDEF 80 53135 68 17282 IRCO2 2 2210218 1 F004160562_L001 LMC FIELD218 256 DARK 359 7475 59 3225 IRCO2 1 2210218 1 F004160562_L002 LMC FIELD218 256 DARK 359 7475 59 3225 IRCO2 2 2210218 1 F004160562_L00
4. N3 distortion database dat NA distortion database dat S11 distortion database dat ST distortion database dat S9W distortion database dat Table 4 9 14 Accuracy of the distortion correction in units of pixels In summary except for L24 the accuracy is about 0 1 pixel N4 may be slightly worse The bad value of L24 comes mostly from scarcity of good bright sources and the matter of statistics With the accumulation of the data it could be improved distortion vectors x20 distortion vectors x20 distortion vectors x20 mi MXAUIERREAAEECLLARRLRRKRYYSAAARARA OR GA Eaa N Ar gt Pehhbbbeb vr V VYAAdAa aap PPP POPP PPPs dd eat eet Pea bbhbe bee VYALaA a RR 400 Be P LLLA RTO d 200 amp A A amp Eb bkkYX3X2cx 2 2 2 20 P Pbi La drca at hb ALLLLkYYAA x xE be PIPPIN PRORA 444A LLRBRYYMAMRA RR D gt DOOD 500 Fe Piel LA dread oh 444 LLLYA ARR xxxrE 48 BR IL GF 44444 4 YN xpx yvrxP Y amp 1222212222222 B a5aazzsz44 Y LxLrx p a DPPP PIR ANER gt a hk LL LENG p gt gt HO Or 7 d ZERS53 UL a a Se ee a a a a 1222227344534 KK KKK KEE ETO TTT AE 6 in gt e ie KELLEKS lier le ie et oy wo i09 7221233 SOK KKK RRA TTT TIT 7 7 eek he Ss gt gt gt 522 28355 3 K amp HAAAM4HTTTTTTYTY Sov P Pr eden IK MAATHmTTTmTYTTYY tuat at na ia a a gt gt E N E Aas AE AO AE AR 50 K E i KALA 4 50 100 200 300 400 100 1 200 250 X pixel X pixel Figure 4
5. The observed wavelength calibrated 1D spectrum in ADU is just divided by the spectral response table Since the wavelength for a given pixel in the observed spectrum is slightly different from that in the response table simple interpolation will be performed to match the wavelength of the observed wavelength calibrated spectrum with that in the response table 6 1 16 Notes on slit spectroscopy The same pipeline can be used for both slit less and slit spectroscopy data reduction Although the reduction of the slit spectroscopy data can be made in a similar way as for the slit less spectroscopy data there are some differences in some processing steps between the two Here we summarize these differences e The slit spectroscopy flat fielding consists of a normalized feature less flat image Therefore no color term correction is necessary e Background subtraction is made on extracted 2D spectra in the case of point sources slit spectroscopy Np after stacking sub frames If dealing with slit spectroscopy for diffuse sources the background subtraction is not performed since entire slit is expected to be filled with the object that is diffuse and there is no pure background area in the image Version 1 3 September 13 2007 89 e Shift and add feature of sub frames will be disabled e In the wavelength calibration measurement of the reference image positions will not be made and pre defined slit positions will be used as a wavelength zero r
6. At present the toolkit concept consists of 3 boxes currently in a state of development e pre pipeline processor Red Boz e pipeline processor Green Boz e post pipeline processor Blue Boz e Pre pipeline Processor Red Box Produces Basic Data 50 Version 1 3 September 13 2007 51 redbox ircslice header formatting slice IRC 3D images into usual 2D ones redbox mkirclog making the observing log file irclog e Pipeline Green Box Produces Basic Science Calibrated Data greenbox anomalous_pix mask Bad Dead pixels greenbox wraparound wrap around correction greenbox dark subtract dark current greenbox ircnorm Normalize sampling and compression bit shift greenbox scatt_light Subtract scattered light pattern greenbox cosmic_ray remove cosmic rays greenbox linearity correct linearity of the detector response greenbox saturation mask saturated pixels greenbox slit mask mask slit area on detector greenbox flat flatten the pixels greenbox aspect ratio Distortion Correction Aspect ratio resampling e Post Pipeline Processor Blue Box This is a separated package but called from pipeline processor by default bluebox coadd to form a co added image source_extract Perform source extraction calcshift Calculate shift amp rotation between images adjust sky adjust sky level before stacking x irc stack stack individual frames to form a co added image The pipeline can be run automati
7. e greenbox aspect ratio Distortion Correction Aspect ratio resampling Alternatively the individual Green Box modules can be run interactively by 3 possible meth ods 1 By setting the interac parameter in the parameter list 2 By running the pipeline with the command pipeline interactive yes 3 By running the pipeline with the command pipeline interactive When running the pipeline interactively individual steps may be entered as Greenbox anomalous pix Greenbox dark etc or alternatively by typing Greenbox then running from inside the Green Box as anomalous pix dark etc When running interactively there are options to perform skip stop each process The present configuration of the Green Box pipeline modules are shown in Fig 5 9 10 Starting from the original input FITS file e g F23342 S004 fits At each step of the Green Box pipeline processing a qualifier is added as a prefix to the original filename note the original FITS files are preserved These prefixes build up over each processing step The prefixes are summarized in Table 5 9 2 1 Mask bad or dead pixels greenbox anomalous pix Bad or dead pixels were identified by using pre flight laboratory data Mask files are stored in ircroot lib anomalous pix They are shown in Fig 5 9 11 After masking bad hot pix els an a will be added as a prefix to the original filename such that F23342 8004 fits becomes aF23342 8004 fits 2 Wraparound Correction gree
8. 5 3 5 4 5 5 5 6 5 7 5 8 5 9 5 10 5 11 5 12 5 13 5 14 5 15 4 6 1 Observed standards and data processing 0208 4 6 2 Estimation of the in band flux 4 6 5 Absolute calibration ses 4 6 4 Overall accuracy of the flux calibration rn Flux calibration for extended sources 2l Color COPFecblOD ioo uoo ehem AAS web ue b blau um pru Wes Dror on Ssa 2 aoe ea Se Roos beue d wee ee e NOS ee a Ge we eps Memory effects caused by bright source observations 00 4 ASUFOInebbyw x6isk RW eds ER ee hea Ean he Sie ae RA SE eae Arrays anomalies zem ke ge a le Ge ese aaa Foe Ree ae E E AND NIR array 050 oa at A to BOL they Wg See gels eg 4 12 2 MIR Arrays d ouo X et da an a a a A BUE AUS E AN General concerns on slit less spectroscopy data Imaging toolkit cookbook ntroduciomnm yg Ao ee xs Re eg 8 he DOL qus ees General overview of the toolkit processing 2 220 4 Expected Data Processing Rate minimum expectation How to install the IRC imaging Toolkit 00 54L R quimements use eh Ee ei ae De Pel te Bea 5vb2 dasta MIRAE error ua Stet ee oe EE Meee ode Radon tecti seo oe Bees 5 4 3 Download IRC imaging data reduction software package JAA Unpackdredga fossa san foe eek ae Mh ek Rags MOIS ROS PR 5 4 5 Make irc binaries sia acai e a g e a a ia e a E nS 5 4 6 Run se
9. Ki gt o a 10 4 15 L J 20 1 1 1 1 1 1 0 2000 4000 6000 8000 10000 12000 RAW ADU MIRS linearity T x c Sg 5 gt o a 80 Ll 1 fi L fi 1 0 5000 10000 15000 20000 25000 30000 35000 RAW ADU MIRL linearity 20 oL 4 20 H 4 L c 2 5 gt o a aor 4 60 4 80 1 fi fi 1 1 fi 0 5000 10000 15000 20000 25000 30000 35000 RAW ADU Figure 4 3 4 Raw signal versus deviation from linear curve A negative deviation means the raw signals in ADU is lower than that expected if the detector were linear Red marks show before applying linearity correction and the green marks represents after applying the correction Version 1 3 September 13 2007 2T 4 5 RSRFE Figures 4 5 5 to 4 5 10 show the Relative Spectral Response functions for the different filters and dispersion elements 28 IRC Data User Manual o Oo o Oo o A Relative Response o N O 1 2 3 4 5 6 Wavelength um Figure 4 5 5 The Relative Spectral Response Function of the IRC NIR Camera for F e co T NP NG Relative Response O O O NOB o0 O T 1 2 3 4 5 6 Wavelength um Figure 4 5 6 The Relative Spectral Response Function of the IRC NIR dispersion elements per photon Version 1 3 September 13 2007 29 o co T S7 o Oo T S11 O A T Relative Response o N T o p 4 8 12 16 20 Wavelength um Figure 4 5 7 The Relative Spectral Response Function for
10. exposure frame are picked out and the corresponding pixels in the long exposure frames will be masked out to be 0 The scaled values in the exposure frames are calculated by Exposure Time short frame 5 9 3 Sshort exposure in ADU S physical saturation in ADU E Ti xposure Imejong frame This is done because the IRC uses Correlated Double Sampling and with the current IRC operating clock we cannot tell which pixels are saturated by long exposure frames alone After masking saturated pixels a s will be added such that 1CcnDwaF23342 8004 fits becomes slCcnDwaF23342_S004 fits 68 9 10 11 Therefore at the end of the Green Box correctly processed frame files should have aefmslCcnDwa IRC Data User Manual Table 5 9 2 Prefixes added during Green Box Pipeline Processing Prefix module action a greenbox anomalous pix Mask Bad or Dead Pixels w greenbox wraparound Wraparound Correction D greenbox dark Dark Subtraction n greenbox norm Normalization c greenbox scat light Subtract Scattered Light Pattern C greenbox cosmic ray Detect and Replace Cosmic Rays l greenbox linearity Linearity Correction S greenbox saturation Mask Saturated Pixels m greenbox slit mask Slit Mask f greenbox flat Flat Fielding e greenbox aspect ratio Aspect Ratio Resampling Slit Mask greenbox slit mask The slit mask processing masks the slit area in the IRC Field of View After the slit mask processing a m will be added such t
11. space_shift parameter e Other generic plot options useful in the plotting tool The plotting tool accepts any kind of IDL generic plot options for spectral plotting See plot manual in the IDL documents for full information Following are some frequently used options 102 IRC Data User Manual xrange plot range along the X wavelength axis By default plots are shown within the wavelength range set in the wavelength calibration database file along the X axis If set explicitly in the plotting tool command line the default settings will be overridden yrange plot range along the Y flux axis By default plots are shown in auto scale mode along Y One can limit fix the plot range by setting yrange xlog ylog plotting in log scale along x or y 6 5 5 Working on saved data All the processed data are saved as an IDL save file as well as FITS output files The IDL save file is actually a dump record of the IDL memory image at the end of the data process ing Therefore one can recall the pipeline results by simply issuing the following command restore lt savefile gt Here the savefile is a string of the save file name To recall other save file image issue reset session first to clear the current IDL memory contents and then issue another restore command with another save file name See IDL manual to know more on the IDL save file restore and reset session commands After restoring the IDL memory one can
12. the filter wheel at so called CAL position The spectroscopy frames are taken with dispersers inserted along the optical path by rotating the filter wheel The direct imaging frame is taken with the same ways as for the normal imaging observations with AOTOO 02 and 03 The image taken during the spectroscopy mode AOTO04 is called as the reference image The N3 S9W and L18W wide band filters will be used as the reference images of NIR MIR S and MIR L spectroscopy respectively Other combinations of the broad band filters and dispersers are not available The satellite starts to maneuver back to the survey mode according to the timer irrespective of the IRC operation The last exposure should be discarded since part of the exposure could be made during the maneuver The present FITS header does not record the status of the satellite attitude control In the imaging data pipeline the removal of the last exposure is made automatically by checking the source intensity Such an automatic procedure is not included in the current spectroscopy data reduction software users need to do it by themselves see Chapter 6 Note however that the automatic removal is not always perfect and users may have to do it manually in some cases even for the imaging data 3 3 Quick look data Together with the rawdata quick look products are distributed in the archive for both imaging and spectroscopy IRC observations These files are not intended for sc
13. while considering the source masks The image obtained after background subtraction has the following expression obs obj A x Fo z y A x R OJ UA x y x spectral feature z y 6 1 3 6 1 5 Image screening Some sub frames may be severely damaged by cosmic ray shower a satellite passing in front of the telescope etc If this is the case one can interactively set the flag by using ds9 for any sub frames to be discarded in the following processes Version 1 3 September 13 2007 85 6 1 6 Image registration among sub frames Relative image shift due to satellite attitude drift among spectroscopy sub frames is measured by means of cross correlation technique This can be made only with NP and NG due to the number of spectra present in NIR images For SG1 SG2 and LG2 pixel offsets measured with NP NG are used for matching rather than measuring their own shifts since all cameras take images simultaneously Similar shift and add technique is also needed for the reference sub frames except for N3 where there is only a single reference sub frame To find the shift values for L18W we first detect point sources in S9W measure their positions and calculate the shift by using the list of target coordinates and shift both S9W and L18W images 6 1 7 Imaging stacking After registering all the sub frames all the selected screened sub frames are combined to make higher S N stacked images For spectroscopy images a three sigm
14. 180 60000 0 980 1 200 1 204 0 925 0 994 1 180 Values are calculated for A4 7 0um As 9 0um Ag 11 0um Ay 15 0um Ag 18 0um Ag 24 0um Version 1 3 September 13 2007 41 Table 4 8 12 Color Correction factors for MIR channel Gray Body a 1 Intrinsic MIR 5 57 MIR S S9W MIR S SI1 MIR L L15 MIR L L20W MIR L L24 Temperature K K 7 0um K 9 0um K 11 0um K 15 0um K 18 0um K 24 0um 40 126 883 11 699 11 085 0 913 50 175 833 15 078 4 943 4 371 0 870 60 19 630 5 044 2 961 2 528 0 869 70 7 299 2 712 2 135 1 798 0 883 80 239 889 4 200 1 855 1 715 1 443 0 902 90 44 874 2 906 1 452 1 472 1 247 0 922 100 13 402 2 228 1 235 1 320 1 130 0 942 110 5 953 1 826 1 107 1 218 1 056 0 960 120 3 563 1 568 1 029 1 147 1 008 0 977 130 2 580 1 393 0 979 1 096 0 976 0 993 140 2 088 1 270 0 947 1 057 0 954 1 007 150 1 801 1 181 0 928 1 028 0 940 1 021 160 1 615 1 115 0 916 1 006 0 930 1 033 170 1 485 1 065 0 910 0 988 0 924 1 044 180 1 389 1 027 0 908 0 974 0 920 1 054 190 1 316 0 999 0 909 0 963 0 918 1 063 200 1 258 0 977 0 912 0 954 0 917 1 072 210 1 212 0 960 0 916 0 946 0 917 1 080 220 1 174 0 948 0 922 0 940 0 918 1 088 230 1 143 0 939 0 928 0 935 0 919 1 094 240 1 116 0 933 0 935 0 931 0 921 1 101 250 1 094 0 928 0 942 0 927 0 923 1 107 300 1 023 0 927 0 979 0 917 0 934 1 131 350 0 988 0 944 1 014 0 913 0 946 1 150 400 0 969 0 967 1 045 0 911 0 956 1 164 450 0 960 0 992 1 072 0 911 0 965 1 176 500 0 95
15. 2 B Figure 5 6 1 Example directory structure for IRC toolkit 5 6 2 Launch IRAF Launch IRAF in your home directory and check that you have an entry for the irc package in the list of IRAF packages see Fig 5 6 2 NOAO PC IRAF Revision 2 12 2 EXPORT Sun Jan 25 16 09 03 MST 2004 This is the EXPORT version of PC IRAF VZ 12 supporting most FC systems Welcome to IRAF To list the available commands type or 77 To get detailed information about a command type help command To run a command or load a package type its name Type bye to exit a package or logout to pet out of the CL Type news to find out what is new in the version of the system you are using The following commands or packages are currently defined color images noaoo softools utilities ctio irc obsolete southafrica dataio language plot stsdas doms lists proto system finder nmisc sirius tables Here Figure 5 6 2 Start up screen for IRAF showing the IRC package visible in the lsit of IRAF packages Then you can move to your working directory before starting to run the different tools by typing e cd to_your_working_directory where to_your_working_directory should follow the rules explained in Section 5 6 1 5 6 3 Load the IRC package The irc package can be loaded by typing irc You should then see the IRC pipelin splash screen welcome message and a list of the available IRC packages
16. 210 546 949 4 356 3 571 220 280 566 3 657 3 172 230 153 521 3 154 2 856 240 89 003 2 779 2 601 250 54 374 2 492 2 393 300 8 881 1 715 1 757 390 3 276 1 390 1 449 400 1 952 1 223 1 276 450 1 477 1 127 1 168 500 1 257 1 069 1 097 600 1 064 1 007 1 011 700 0 989 0 981 0 963 800 0 957 0 971 0 934 900 0 945 0 968 0 915 1000 0 943 0 970 0 902 1500 0 982 0 992 0 875 2000 1 027 1 012 0 868 2500 1 061 1 027 0 866 3000 1 087 1 037 0 865 3500 1 106 1 045 0 865 4000 1 121 1 051 0 865 4500 1 132 1 056 0 865 5000 1 142 1 059 0 865 6000 1 156 1 065 0 866 7000 1 166 1 069 0 866 8000 1 173 1 072 0 866 9000 1 179 1 074 0 866 10000 1 183 1 076 0 867 20000 1 203 1 084 0 868 30000 1 210 1 087 0 868 40000 1 213 1 089 0 868 50000 1 215 1 089 0 868 60000 1 216 1 090 0 868 Values are calculated for Ay 2 4um Ao 3 2um Aa 4 1m Version 1 3 September 13 2007 Table 4 8 10 Color Correction factors for NIR channel Gray Body a 2 Intrinsic NIR N2 NIR N3 NIR N4 Temperature K K 2 4um K 3 2um K 4 1um 40 50 60 70 80 119 020 90 236 320 100 99 825 110 E 50 445 120 339 050 29 088 130 120 115 18 531 140 51 864 12 748 150 i 26 407 9 316 160 15 397 7 143 170 B 10 012 5 694 180 i 1 092 4 686 190 5 367 3 959 200 549 483 4 275 3 418 210 264 183 3 041 3 006 220 136 893 3 025 2 685 230 75 855 2 647 2 430 240 44 673 2 361 2 224 250 27 825 2 140 2 055 300 5 291 1 528 1
17. 297 140 29 139 140 121 814 19 760 150 58 297 14 242 160 31 767 10 776 170 S 19 291 8 482 180 12 811 6 895 190 9 149 5 756 200 S 6 928 4 914 210 5 496 4 275 220 582 603 4 526 3 778 230 316 052 3 839 3 385 240 181 324 3 336 3 069 250 109 374 2 955 2 810 300 16 057 1 952 2 024 350 5 065 1 543 1 643 400 2 631 1 333 1 428 450 1 818 1 212 1 294 500 1 463 1 136 1 204 600 1 167 1 052 1 095 700 1 051 1 012 1 033 800 0 997 0 992 0 994 900 0 970 0 982 0 969 1000 0 957 0 977 0 951 1500 0 963 0 981 0 910 2000 0 990 0 992 0 896 2500 1 013 1 001 0 890 3000 1 031 1 008 0 886 3500 1 045 1 013 0 884 4000 1 056 1 017 0 883 4500 1 064 1 020 0 882 5000 1 071 1 022 0 881 6000 1 081 1 026 0 880 7000 1 088 1 029 0 880 8000 1 094 1 031 0 879 9000 1 098 1 032 0 879 10000 1 101 1 034 0 879 20000 1 116 1 039 0 878 30000 1 121 1 041 0 877 40000 1 123 1 042 0 877 50000 1 125 1 043 0 877 60000 1 126 1 043 0 877 Values are calculated for Ay 2 4um Ao 3 2um Aa 4 1 um 38 IRC Data User Manual Table 4 8 9 Color Correction factors for NIR channel Gray Body a 1 Intrinsic NIR N2 NIR N3 NIR N4 Temperature K K 2 4um K 3 2um K 4 1um 40 50 60 70 80 938 233 90 305 576 100 127 958 110 E 64 122 120 541 920 36 679 130 z 187 774 23 187 140 78 796 15 834 150 i 38 794 11 488 160 5 21 822 8 749 170 S 13 696 6 929 180 9 390 5 665 190 E 6 906 4 757 200 xa 5 367 4 083
18. 541 300 2 329 1 268 1 293 400 1 566 1 135 1 154 450 1 270 1 061 1 068 500 1 125 1 017 1 012 600 0 996 0 976 0 946 700 0 949 0 963 0 910 800 0 935 0 962 0 889 900 0 936 0 968 0 876 1000 0 945 0 975 0 868 1500 1 019 1 016 0 855 2000 1 084 1 047 0 855 2500 1 132 1 068 0 857 3000 1 166 1 084 0 860 3500 1 192 1 095 0 862 4000 1 212 1 103 0 864 4500 1 227 1 110 0 866 5000 1 239 1 115 0 867 6000 1 258 1 123 0 869 7000 1 271 1 129 0 871 8000 1 281 1 133 0 873 9000 1 288 1 137 0 874 10000 1 294 1 139 0 874 20000 1 320 1 151 0 879 30000 1 329 1 155 0 880 40000 1 333 1 157 0 881 50000 1 335 1 158 0 881 60000 1 337 1 159 0 881 Values are calculated for Ay 2 4um Ao 3 2m Aa 4 1m 40 IRC Data User Manual Table 4 8 11 Color Correction factors for MIR channel Black Body a 0 Intrinsic MIR 5 57 MIR S S9W MIR S SI1 MIR L L15 MIR L L20W MIR L L24 Temperature K K 7 0um K 9 0um K 11 0um K 15 0um K 18 0um K 24 0um 40 lt 204 598 14 514 14 825 0 942 50 373 308 21 615 5 989 5 661 0 882 60 31 533 6 632 3 516 3 178 0 870 70 9 722 3 371 2 492 2 198 0 875 80 619 872 5 209 2 217 1 971 1 720 0 887 90 110 313 3 493 1 685 1 670 1 454 0 902 100 29 730 2 629 1 399 1 480 1 292 0 917 110 11 350 2 124 1 230 1 352 1 187 0 931 120 5 796 1 803 1 123 1 262 1 116 0 945 130 3 691 1 585 1 053 1 196 1 066 0 958 140 2 731 1 430 1 006 1 147 1 031 0 969 150 2 221 1 318 0 973 1 109 1 005 0 980 160 1 917 1 233 0 952 1 079 0 986
19. DATE REF number of tracked stars in STT B at DATE REF STT A Mode status at DATE REF STT B Mode status at DAT E REF Any strings Any strings BITPIX is the number of bits per data pixel and equal to 16 in the case of IRC NAXIS is the number of axes equal to 3 for IRC NAXIS1 and NAXIS2 are the image dimensions NAXIS3 is the number of image frames EXTEND refers to the presence of extensions in the FITS file e Data type creation and processing information FMTTYPE is the type of File Format in FITS file It can be ASTRO F IMAGE IRC or ASTRO F SCAN IRC FTYPEVER is the version of the file format described in FMTTYPE DATE is the file Creation Date CREATOR is the data generator program name CNTTYPE is the type of data content It can be IRC_NIR or IRC_MIR CRIRVER is the version of CREATOR Contents TBD Version 1 3 September 13 2007 17 PIPELINE is the Data Processing Pipeline name and version DATASTAT is the Data status It describes data status mainly from complete ness of telemetry data This does not tell detailed scientific quality of the data All appropriate error status are listed otherwise GOOD is given GOOD No problem INCOMPLETE Scientific data incomplete due to telemetry loss etc NOHK HK Status not available NOADS Attitude information not available STTINI STT did not work properly More status may be added as analysis progresses Data other th
20. all the pair files for all filters pointing at one position on the sky There may be files for both the long long_exp input and short Short exp inputexposure frames e Skypair Files These log file can be found inside the separate directory logs This log file is produced during the adjustment of the sky level between frames bluebox adjust sky with a name skypair0002 N2 1ist long or similar This file lists the frames looking at a given area on the sky with the corresponding mean median and mode sky brightness and 1c standard deviation e calcshift log This log file can be found inside the separate directory logs T he calcshift log file is the log file for the calculation of shift and rotation in the bluebox calshift process Relative to the first frame in the pair file described above the log file gives the number of stars matched between 2 coo files and the corresponding x amp y offset for these frames The calcshift 1log file will also record any matching failures e darklist before 5 11 The IRC TOOL irc_tool It has not yet been decided whether an additional post pipeline processing module will be in cluded in the toolkit and released to the users Some structure for the post pipeline processing does exist for co adding source extraction photometry and catalog creation and may be devel oped at some later time The present configuration of the IRC TOOL is shown in Fig 5 11 15 e irc_tool coaddLusingS An alternative procedure wh
21. also automatically discarded so you need not examine each frame and edit irclog e darklist before In a pointing observation a dark frame is taken at the beginning and the end of the operation for both the NIR and MIR channels This file contains the names of the dark frames taken at the beginning of the operation Dark frames taken at the end of the operation may be affected by latent so they are excluded from the list Those who want to use this pre dark images insted of super dark images should turn the selfdark parameter on described later in pipeline section then the pipeline prpcessor willl read the darklist before file and will instead make a dark image by averaging over 3 images of MIR S and MIR L long exposure dark frames However super dark will be used for all short frames and NIR long frames even if selfdark is on Table 5 8 1 Sample of the irclog file ZFRAME OBJECT NAXISI FILTER RA SET DEC SET AOT EXPID TIDNUM SUBID _ F004160544 L00I LMC FIELD218 X 250 DARK 80 546088 68 17107 IRCO 1 2210218 1 7 F004160544 L002 LMC FIELD218 256 DARK 80 54688 68 17107 IRCO2 2 2210218 1 7 F004160544 L003 LMC FIELD218 256 DARK 80 54688 68 17107 IRCO2 3 2210218 1 F004160544_L004 LMC FIELD218 256 DARK 80 54688 68 17107 IRCO2 4 2210218 1 F004160544_S001 LMC FIELD218 256 DARK 80 54688 68 17107 IRCO2 1 2210218 1 F004160544_S002 LMC FIELD218 256 DARK 80 54688 68 17107 IRCO2 2 22
22. and will be mainly used for simultaneous observations of diffuse light with the MIR S camera This slit position is labeled as Ns for IRCO4 AOT observation parameter Both the NP low resolution prism and NG high resolution grism will be used with this slit The middle 1 x 1 square part referred to as Np is for spectroscopy of point sources The aperture is large compared to the absolute pointing accuracy of the satellite designed to be better than 30 arcsec to ensure that the target can be accurately guided into the area Note that for observations of faint sources confusion due to galaxies may be a serious problem The NG grism is assumed to be Version 1 3 September 13 2007 5 5Omm folding mirror unit NIR ET i MIR S By Nob detector unit MIR S P Figure 2 0 1 Bird s eye view of the IRC camera used with this aperture The rightmost outer part Nh has a 3 arcsec width and is used for the highest resolution spectroscopy of diffuse radiation with the NG grism e The MIR S has a slit of 5 arcsec width for diffuse light As this slit overlaps with the innermost slit of the NIR camera it is also referred to as Ns We assume that the point source density in the mid infrared range is low enough to avoid serious confusion such that spectroscopy of poi
23. catalog coordinates The present configuration of the Blux Box co add wrapper module is shown in Fig 5 10 13 The starting point for the Blue Box Co Add wrapper are the files created from the final step in the Green Box pipeline Processor efmsiCcnDwaF23342 9004 fits The Co Add module then performs the following tasks 1 Extract bright reference sources bluebox source extract This module extracts individual bright sources from individual frames Before extraction original 412x512 256x256 NIR MIR images are paseted to larger 1024x1024 512x512 images to prepare for the xy shift and rotation During source extraction a S will be added such that fdmslnDwaF23342 8004 fits becomes SfdmslnDwaF23342 8004 fits For each MIR image a median box car filtered image will be made which is then subtracted before source extraction for efficient source extraction in the case of variable background within an image Furthermore a simple cosmic ray detection and rejection process will be applied to the MIR images Then bright sources stars are extracted from each frame as source lists 70 IRC Data User Manual BLUE BOX Co Add Wrapper bluebox source_extract extract reference sources bluebox calshift Calculate shift amp rotation between image frames bluebox adjust_sky Adjusts sky level between individual frames bluebox irc stack Stacks the IRC images bluebox putwcs Add WCS information to an
24. command prompt The present configuration of the pre pipeline modules is shown in Fig 5 7 5 Figure 5 7 5 Present configuration of the Red Box pre pipeline modules 1 redbox ircslice The number of FITS files produced for any given single pointing will depend on the AOT Each AOT comprises of a combination of exposure cycles EC Filter Wheel changes W 58 IRC Data User Manual and Dither Maneuvers M see Fig 2 2 6 The IRC FITS data is not a usual 2D one A raw data FITS file is created for each Exposure Cycle during a pointing for the NIR and combined MIR S MIR L channels i e 1 NIR FITS file and 1 MIR FIT file 2 FITS files per Exposure Cycle The filename format is given as F _N fits or F M fits where is a distinct incremental reference number A NIR raw fits data file is a data cube containing 2 frames within it corresponding to one short and one long exposure Each MIR raw FITS file contains 4 frames within it one short and three long exposures for both the MIR S and MIR L channels respectively making a total of 8 frames per MIR FITS file per Exposure Cycle see Fig 5 7 6 In addition to the Exposure Cycles a Dark frame is taken at the beginning and the end of the operation for both the NIR and MIR channels i e an extra 2 FITS files for both NIR and MIR Therefore as shown in Fifure 2 2 6 for AOT IRCO2 for a single pointing the maximum number of raw data files will comprise of 1 NIR 1 MIR FITS EC
25. defined at the centre of the field of view A Basically yes But the Spectral response at the periphery of the field has been checked for consistency 5 Q Does the user have to use a flux calibration aperture A The toolkit can take account of the aperture correction for the slitless spectroscopy as explained in Section 6 of the IDUM For the slit spectroscopy the brightness should be preserved For the case of small extended sources the user must take special care 6 Q Is there any wavelength dependence of the PSF A No significant dependence is seen The PSF in the spectroscopy mode is slightly worse than in the imaging mode although wavelength dependence of the spectroscopy PSF is not clearly seen
26. frame NIR 2 long exposure frame MIR S L2 long exposure frame MIR L S3 long exposure frame MIR S L3 long exposure frame MIR L S4 long exposure frame MIR S L4 long exposure frame MIR L Figure 5 7 6 IRC raw FITS file data structure for NIR and MIR images Figure 5 7 7 Image of IRC raw data NIR and MIR bands The NIR raw FITS file left is 2 frames deep corresponding to 1 short and 1 long exposure The MIR raw FITS file right is 2 frames wide and 4 frames deep corresponding to 1 short and 3 long exposures each for MIR S and MIR L 5 8 Before runnning the pipeline processor After runnning prepipeline you will get 2 text files namely irclog and darklist before e irclog The irclog file is used to select the data to be reduced and any unnecessary entries should 60 IRC Data User Manual be removed from the irclog file editing it by hand Alternatively entries and unnecessary lines can be commented out by adding to the head of the entries NOTE the DARK frames and any grism prism images need to be removed commented out before running the Green Box pipeline This is automatically done by prepipeline In general bad frames may cause a failure in the pipeline at a later stage downstream the best strategy may be an iterative run where if a crash occurs the individual frames can be examined and the culprits excluded from future runs of the Green Box The data frame taken during the maneuver can be
27. image Figure 5 10 13 Present configuration of the Blux Box co add wrapper module 2 Calculate Shift amp Rotation to match frames bluebox calshift The bright reference star source lists are used to calculate the relative shift and rotation values between individual image frames looking at a particular field of view on the sky us ing the first frame as a reference After this shift and rotation process the original frame file names will receive additional R prefixes such that Sfdms1nDwaF23342 8004 fits be comes RSfdmslnDwaF23342 8004 fits There is a parameter minmatch that determines the minimum number of stars required for matching between 2 frames If there are not enough stars then a warning will be given and a log file coadd failure created containing the names of the processed frames that could not be combined A log file calcshift log is also produced listing the number of stars matched between and the corresponding x amp y offset for included frames Not that especially for the long wavelength channels MIR L there may be occasions when there are not enough guide stars for matching An alterna tive procedure can then be used which utilizes the MIR S shift amp rotation for the MIR L channel See irc tool section Adjusts Sky Level bluebox adjust sky The bluebox adjust sky module collects all frames looking at a given area on the sky and calculates the median sky from each image Each calculated median valu
28. is fixed to the center of the extracted spectroscopy image Therefore only dispersion and wavelength at origin are set in the calibration database files The parameters can be applied for all the spectra within the FOV i e the parameters are constant across the FOV Effective wave length ranges there are two kinds of range definitions sensitive and reliable ranges for each disperser are also set in the table For NP pixel positions for given A is expressed in 2nd order polynomial equation and there are three parameters Oth 1st and 2nd order coefficients Note that for data analysis convenience the equation is in pixel function A form being in inverse form for the grisms Effective wavelength ranges are also set in the table Only for NP significant spectroscopy distortion exists i e reference positions on the spec troscopy images can not be represented by a constant pixel shift dX dY from reference image positions The deviation of pixel shift along wavelength axis or Y axis from the case of constant pixel shift is expressed in 3rd order polynomial equation of reference posi tion dY gistortion function X er Yrer Polynomial coefficients are tabulated in a separate ascii table and is stored under the same directory Note that spectroscopy distortion along X dXgistortion is not so significant and constant pixel shift works rather well for NP as in the same way for NG SG1 SG2 LG2 WAVEPAR IRCWAVEPAR_COMM
29. lt targetsubid gt rawdata IRC SPECRED DATADIR lt targetid gt and lt targetsubid gt will be used in the command line of the pipeline command All the reduced data and related information will be stored in a separate directory called root dir targetid targetsubid irc specred out When the directory is missing the toolkit will create it 6 3 Calibration data When de packing the toolkit the following calibration directories are created In the following contents within the parentheses something should be changed according to user s interests 90 IRC Data User Manual 6 3 1 Calibration files e IRC_SPECRED_CALIBDIR DARK contains dark images with high S N combined with pre compiled dark images which can be applied for all the observations after applying a small correction in the count offset There are several super darks for NIR MIR S MIR L as specified in the following list files DARK DARK_NIR_long 1st DARK DARK_NIR_short 1st DARK DARK_MIRS_long 1st DARK DARK_MIRS_short 1st DARK DARK_MIRL_long 1st DARK DARK MIRL short lst e IRC_SPECRED_CALIBDIR FLAT contains two types of flats super flats for spectroscopy and reference images These super flats are made by combining a large number of blank sky images For slit less spectroscopy FLAT SPEC2DFLAT NP 1st FLAT SPEC2DFLAT NG 1st FLAT SPEC2DFLAT SG1 1st FLAT SPEC2DFLAT SG2 1st FLAT SPEC2DFLAT_LG2 1st For slit spectrosco
30. new directory stacked IM The stacking process creates 3 files for any given filter position on the sky e A co added image file e g 1757132 N2 long fits e A noise map e g sigma1757132 N2 long fits e Summary File e g p11757132 N2 long fits pl Figure 5 10 14 Example of final science grade data produced by the Green Box and Blue Box Co Add Wrapper left is the stacked image map right is the corresponding noise map 5 Convert to WCS coordinates bluebox putwcs After the Co add wrapper has been completed the module bluebox putwcs adds WCS information to the FITS images by matching with 2MASS catalog coordinates using the following procedure a source extraction of stars from the IRC image b download 2MASS catalog for the observed area c convert 2MASS RA DEC to the x y image coodinates 72 IRC Data User Manual d match 2MASS x y coordinates with image xy coordinates At this point there will be a list for extracted IRC stars corresponding to stari x1 y2 rail deci x1 y2 star2 x2 y3 ra2 dec2 x2 y2 starN xN yN raN decN xN yN for the matched stars The module then calculates a transformation matrix Nth order polynomial that will convert xy to RA and DEC The module attempts to find the best fitting solution by increasing the order of plynomial from 2 to 4 Usually the best result should be 2nd order as long as the image has been corrected for distortion e if matched put wes
31. observation 3 2 2 AKARI FITS Primary HDU common information rawdata header The contents of the AKARI FITS Primary HDU of all raw data both IRC and FIS is as follows SIMPLE T Standard FITS format BITPIX 16 number of bits per data pixel NAXIS 3 Number of axes NAXIS1 412 Image dimension NAXIS2 512 Image dimension NAXIS3 2 Image frame EXTEND T Extension may be present FMTTYPE ASTRO F IMAGE IRC Type of File Format in FITS file FTYPEVER 4 Version of FMTTYPE CNTTYPE IRC_NIR Type of data content DATE 2006 09 25T09 45 24 File Creation Date CREATOR TBD Data generator program name CRTRVER 1 0 Version of CREATOR PIPELINE 5 ircpl ver 1 0 Data Processing Pipeline name DATASTAT GOOD Data status ORIGIN ISAS JAXA Organization creating FITS file TELESCOP AKARI AKARI mission INSTRUME IRC Identifier of the instrument DETECTOR NIR Detector name OBSERVER PI Name PI Name Observer s ID PROPOSAL PRPID Proposal ID OBS CAT OT Observation Category PNTNG ID 1234567 Pointing ID TARGETID 1234567 Target ID SUBID 1 SubID OBJECT target Object name OBJ RA 320 5533 degree Target position OBJ DEC 23 3325 degree Target position AOT IRCO3 Observation AOT AOTPARAM 8 0 5 70 AOT Parameter INSTMODE TBD Instrument operation mode TIMESYS UTC Explicit
32. of iterations in statistical process obslog string default irclog NEVER CHANGE THIS Output filename of redbox mkirclog cosmicr boolean yes or no default no Detect and replace cosmic rays in MIR images Cosmic ray events in each image are detected and replaced by the average of the four nearest neighbors Type help cosmicrays inside IRAF s shell for details wcs boolean yes or no default no Try to match the 2MASS sources to calculate wcs To use this function you should have internet connection to automatically download 2mass catalog from the internet This process will not be executed as default Note although the toolkit was checked on Solaris machines also Solaris does not have curl in the original applications You need to install it to run WCS selfdar boolean yes or no default no Use an alternative dark to the super dark provided coaddsh boolean yes or no default no Coadd short exposure frames as well as long ones interac boolean yes or no default no option to run pipeline interactively With this option selected the pipeline can be run step by step e g as Greenbox anomalous pix Greenbox dark etc or alternatively by typing Greenbox then running from inside the Green Box as anomalous pix dark etc deltemp boolean yes or no default yes delete any temporary created files rejecti string none minmax ccdclip crreject sigclip avsigclip pclip default sigclip T
33. running the the irc tool coaddLusingS function will overwrite any previous coadd failure file in the working di rectory i e if you have an coadd failure from MIR S NIR image processing then this log may be lost Therefore any previous coadd failure log should be renamed prior to running the coaddLusingS function e coaddition of multiple pointing observations In principle the IRC toolkit will attempt to process and coadd all the frames from any number of pointings in the working directory It should be noted however that no such functionality is available within the coaddLusingS function and only frames corresponding to a specific pointed observation may be coadded at any one time This facility may be included in later versions of the toolkit 80 IRC Data User Manual Figure 5 14 18 Difference in flats for MIR S S11 band for epochs prior to left and after right 2007 01 07 Version 1 3 September 13 2007 81 5 15 Frequently Asked Questions and Troubleshooting 1 Q During the pre pipeline stage sometimes I find error messages such as 1s rawdata No such file or directoryorcat slice tmpO0 No such file or directory etc Although the pre pipeline seems to run correctly A Try typing unlearn all to clear these error messages 2 Q On running the pipeline processor Green Box I get the following message on the Distortion processing and the pipeline crashes 4 DISTORTION 4 454 Making the input file list C
34. the DATE OBS in PIM TI 36 bits DHUTI format OxXXXXX PIMTIEND is the DATE END in PIM TI 36 bits DHUTI same format than above 18 IRC Data User Manual PIMTIREF is the DATE REF in PIM TI 36 bits DHUTI same format than above 7 OBS and END are identical with REF for convenience REF is Reference time from TI sampled during each exposure cycle This is due to the fact that IRC only sample time information once per exposure cycle Note PIMTI is the primary information directly from telemetry AFTI and DATE are from the timing correction based on PIMTI e attitude information 3 2 3 EQUINOX is the Epoch of Coordinate RA is the Target position at DATE REF in degrees DEC is the Target position at DATE REF in degrees ROLL is the Roll Angle at DATE OBS AA SOL is the Solar avoidance Angle at DATE REF in degrees AA EAR is the Earth avoidance Angle at DATE REF in degrees AA LUN is the Lunar avoidance Angle at DATE REF in degrees TM SAA is the duration in seconds since last SAA at DATE REF Definition of SAA is different for different detectors SAA region is defined by the glitch rate map observed by the Star Tracker with arbitrary threshold level IRC follows this threshold Shifts of 30 and 60 seconds are applied to FIS SW and LW respectively SAT POSX is the Satellite position at DATE REF in km SAT POSY is the Satellite position at DATE REF in km SAT POSZ i
35. the IRC electronics onboard the satellite NIR images NP NG and N3 will be rotated by 90 deg counterclockwise at the very first stage of the data reduction for convenience in the spectroscopy pipeline This is the original orientation in raw images without rotation for NIR e NP longer wavelength comes at right side toward positive X e NG longer wavelength comes at left side toward negative X SG1 2 longer wavelength comes at higher toward positive Y e LG2 longer wavelength comes at lower toward positive Y The IRC04 data reduction pipeline makes the rotation of the NIR images at the first step Af ter the NIR rotation the dispersion directions are the same for all the dispersers in spectroscopy images Note however that NP SG1 and SG2 show positive dispersions longer wavelength comes at higher Y and NG and LG2 show negative dispersions longer wavelength comes at lower Y At the same stage the orientation of all NIR MIR S and MIR L images is also set right i e the image is neither flipped nor mirrored T hus one can match the IRC images with other WCS correct images such as 2MASS images only by shifting and or rotating the images Since the satellite is designed to scan the sky along the ecliptic latitude on the sky and the X axis of the IRC array is aligned perpendicular to the scan direction the Y axis is aligned closely with the ecliptic latitude while the X axis is aligned with the ecliptic longitude Cha
36. toolkit De activate if you want to use your own flat images instead of using the toolkit default flats In that case you have to prepare flat images for nine filters N2 N3 N4 S7 S9W S11 L15 L18W and L24 and put them in where you installed irc lib flat user Their file names should be named as those of the flat images in where you installed irc lib flat soramame ari e verbose boolean yes or no default no activate if you want to print verbose progress messages Figure 5 9 9 Area utilized for the Co added images depends upon the parameter com area The 2 options are common area red region and whole area within green dashed region 5 9 2 Running Green Box pipeline module The Green Box pipeline processor can be run by entering pipeline at the IRAF command prompt The Green Box pipeline produces the basic calibrated data and currently consists of the following steps 64 IRC Data User Manual e greenbox anomalous_pix Mask Bad Dead pixels e greenbox wraparound Corrects for wraparound e greenbox dark Subtract dark current e greenbox ircnorm Normalization for sampling and data compression e greenbox scatt light Subtract scattered light pattern e greenbox cosmic ray Detect and replace cosmic rays in MIR images e greenbox linearity Correct linearity of the detector response e greenbox saturation Mask saturated pixels e greenbox slit mask Masks the slit area of IRC field of view e greenbox flat Flatten the pixels
37. with less accurate position measurements Source identification numbers are also indicated by the region marks For clear view of the image under complicated region marks check out the region gt show regions check box in the ds9 pull down menu To display these images manually issue show_aperture_on_ds9 lt image gt imag where image could be refimage ff 1 or specimage_bg etc on the IDL command line When imag option is set the image is shown on right side of the ds9 with region marks for reference image The default no option is for spectroscopy image See appendix for more on array name conventions 6 5 2 Displaying the extracted images on ATV ATV is a general purpose interactive array image displaying tool Array should be in 2D See ATV web page http www physics uci edu barth atv for more information on the program 3 Command line syntax atv array block Here are some ATV tips e To display whole images use atv lt array gt 1 1 means long exposure frame and 0 for short Here array should be something like refimage_bg or specimage bg e To display extracted images of your desired source_id use atv lt array gt source_id Here lt array gt should be something like specimage_n_bg To check mask area use atv lt array gt source_id mask source_id e When color table looks abnormal type set_color at the IDL prompt before launching the atv 6 5 3 Ch
38. x 7 EC 4 Dark 18 FITS files for one pointing Consequently for AOT IRC03 for a single pointing the raw data files will comprise of 1 NIR 1 MIR FITS EC x 8 EC 4 Dark 20 FITS files for one pointing In addition to the raw FITS files a text file is also included giving details of the original target list for the observation The redbox ircslice module slices each raw FITS file into the individual frames separates all individual IRC frames for one Exposure Cycle Thus for every NIR raw FITS file including the Dark redbox ircslice will create 2 FITS files corresponding to the short and long exposure For every MIR raw FITS file including the Dark redbox ircslice will create 8 FITS files corresponding to single short and three long exposures for both the MIR S and MIR L channels Therefore after running redbox ircslice on a single pointing for AOT IRCOS3 you can expect as many as 100 individual FITS files The filename format takes the original format with an extension defining the channel N S L and frame number 001 004 For example an original raw FITS file for the MIR channel F23340_M fits is sliced into 8 separate fits files F23340 L001 fits F23340 L002 fits F23340_L003 fits F23340 L004 fits F23340_S001 fits F23340_S002 fits F23340_S003 fits F23340_S004 fits Fig 5 7 7 shows the images of IRC raw data for the NIR and MIR bands The orientation of the images are such that the NIR is rotated by 90 degrees rel
39. 0 990 170 1 717 1 168 0 937 1 055 0 972 0 999 180 1 577 1 118 0 927 1 035 0 961 1 007 190 1 474 1 078 0 921 1 019 0 953 1 015 200 1 395 1 047 0 918 1 006 0 947 1 022 210 1 332 1 022 0 917 0 995 0 942 1 029 220 1 282 1 002 0 918 0 986 0 939 1 035 230 1 240 0 986 0 919 0 978 0 937 1 040 240 1 206 0 974 0 922 0 972 0 935 1 046 250 1 176 0 964 0 925 0 966 0 934 1 050 300 1 081 0 939 0 947 0 947 0 933 1 070 350 1 032 0 938 0 970 0 937 0 936 1 085 400 1 004 0 946 0 992 0 931 0 940 1 096 450 0 987 0 958 1 011 0 928 0 944 1 104 500 0 977 0 972 1 028 0 925 0 947 1 111 600 0 967 0 998 1 056 0 923 0 954 1 122 700 0 963 1 021 1 076 0 922 0 959 1 130 800 0 961 1 040 1 092 0 922 0 963 1 135 900 0 961 1 057 1 105 0 921 0 966 1 140 1000 0 962 1 070 1 115 0 921 0 969 1 144 1500 0 966 1 113 1 146 0 922 0 977 1 155 2000 0 969 1 136 1 162 0 922 0 981 1 161 2500 0 971 1 149 1 171 0 923 0 984 1 165 3000 0 972 1 158 1 177 0 923 0 985 1 167 3000 0 973 1 165 1 181 0 923 0 987 1 169 4000 0 974 1 169 1 184 0 923 0 987 1 170 4500 0 975 1 173 1 186 0 924 0 988 1 172 5000 0 975 1 176 1 188 0 924 0 989 1 172 6000 0 976 1 180 1 191 0 924 0 990 1 174 7000 0 977 1 184 1 193 0 924 0 990 1 175 8000 0 977 1 186 1 195 0 924 0 991 1 175 9000 0 977 1 188 1 196 0 924 0 991 1 176 10000 0 978 1 189 1 197 0 924 0 991 1 176 20000 0 979 1 196 1 201 0 924 0 993 1 178 30000 0 979 1 198 1 202 0 924 0 993 1 179 40000 0 979 1 199 1 203 0 924 0 993 1 179 50000 0 979 1 199 1 204 0 925 0 994 1
40. 01 N3 list long shift O The pair0001 N3 1ist file lists all long and short frames that point at the same area of sky The contents of the file are the original frame names e g F23343_N001 IRC channel and the coordinates The files pair0001 N3 list short amp pair0001 N3 list long are extracted and segregated from F23343 N001 on the basis of the exposure time long or short exposure frames The file pair0001 N3 1ist long shift lists the result of the coordinate matching before coadding Each line entry in the file consists of a filename x shift pix y shift pix rotation deg and the number of stars used for the shift amp rotation angle calculation For example Version 1 3 September 13 2007 73 fdmslnDwaF01001_N003 fits coo 1 0 0 0 0 0 0 0 fdmslnDwaF01007_N003 fits coo 1 19 25942609 0 00474931 359 99288567 126 The values are calcualted relative to the first frame listed in the file such that the first entry is always filename 0 0 0 0 0 0 O There may also be an additional pair file pair0001_N3 list_long shift 0 which is almost the same as the pair0001 N3 1ist long shift but it also contains detector information as filename detectorID x shift pix y shift pix rotation deg and the number of stars used for the shift amp rotation angle calculation i e fdmslnDwaF01001 NO03 fits coo 1 1 0 0 0 0 0 0 0 fdmslnDwaFO1007 NO03 fits coo 1 1 19 25942609 0 00474931 359 99288567 126 e exp input files These files contain a list of
41. 1 164 1 593 6000 1 030 1 541 1 473 0 966 1 167 1 608 7000 1 031 1 548 1 476 0 967 1 170 1 620 8000 1 032 1 552 1 479 0 967 1 172 1 628 9000 1 033 1 556 1 481 0 968 1 173 1 635 10000 1 034 1 559 1 483 0 968 1 174 1 641 20000 1 037 1 572 1 490 0 970 1 180 1 667 30000 1 038 1 576 1 493 0 971 1 182 1 676 40000 1 038 1 578 1 494 0 971 1 183 1 681 50000 1 039 1 580 1 495 0 971 1 183 1 683 60000 1 039 1 581 1 495 0 972 1 183 1 685 Values are calculated for A4 7 0um As 9 0um Ag 11 0um Ay 15 0um Ag 18 0um Ag 24 0um Version 1 3 September 13 2007 43 4 9 Distortion Because of geometric distortion in the IRC images the original detector pixel solid angle varies over the field of view We observed globular clusters and Galactic center with the IRC where many stars are expected to be detected Then we matched their 2MASS corrdinates with the detector xy coordinates by fitting polynomials In the fitting we fixed the pixel field of view to be 1 46 2 38 and 2 40 arcsec pix for the NIR MIR S and MIR L respectively The deviation from an ideal grid square are up to 2 0 6 and 16 pixels at the edge for NIR MIR S and MIR L respectively There is little difference with filters Fig 4 9 11 shows the distortion vector for N2 S7 and L24 respectively Table 4 9 14 shows the accuracy of the distortion correction L15_distortion_database dat L18W_distortion_database dat L24 distortion database dat N2 distortion database dat
42. 10218 1 3 F004160544 8003 LMC FIELD218 256 DARK 80 54688 68 17107 IRCO2 3 2210218 1 F004160544_S004 LMC FIELD218 256 DARK 80 54688 68 17107 IRCO2 4 2210218 1 F004160545_N001 LMC FIELD218 412 DARK 80 54688 68 17107 IRCO2 1 2210218 1 F004160545_N002 LMC FIELD218 412 DARK 80 54688 68 17107 IRCO2 2 2210218 1 F004160546_L001 LMC FIELD218 256 L15 80 54688 68 17107 IRCO2 1 2210218 1 F004160546_L002 LMC FIELD218 256 L15 80 54688 68 17107 IRCO2 2 2210218 1 F004160546_L003 LMC FIELD218 256 L15 80 54688 68 17107 IRCO2 3 2210218 1 F004160546_L004 LMC FIELD218 256 L15 80 54688 68 17107 IRCO2 4 2210218 1 F004160546_S001 LMC FIELD218 256 S7 80 54688 68 17107 IRC02 1 2210218 1 F004160546_S002 LMC FIELD218 256 S7 80 54688 68 17107 IRC02 2 2210218 1 F004160546_S003 LMC FIELD218 256 S7 80 54688 68 17107 IRCO2 3 2210218 1 F004160546_S004 LMC FIELD218 256 S7 80 54688 68 17107 IRC02 4 2210218 1 F004160547_NO001 LMC FIELD218 412 NP 80 54688 68 17107 IRCO2 1 2210218 1 7 F004160547 N002 LMC FIELD218 412 NP 80 54688 68 17107 IRCO2 2 2210218 1 F004160558 L001 LMC FIELD218 256 L24 80 53135 68 17282 IRCO2 1 2210218 1 F004160558_L002 LMC FIELD218 256 L24 80 53135 68 17282 IRCO2 2 2210218 1 F004160558_L003 LMC FIELD218 256 L24 80 53135 68 17282 IRCO2 3 2210218 1 F004160558_L004 LMC FIELD218 256 L24 80 53135 68 17282 IRCO2 4 2210218 1 F004160558_S001
43. 3 LMC FIELD218 256 DARK 359 7475 59 3225 IRCO2 3 2210218 1 7 F004160562 L004 LMC FIELD218 256 DARK 359 7475 59 3225 IRCO2 4 2210218 1 F004160562_S001 LMC FIELD218 256 DARK 359 7475 59 3225 IRCO2 1 2210218 1 F004160562_S002 LMC FIELD218 256 DARK 359 7475 59 3225 IRCO2 2 2210218 1 3 F004160562 8003 LMC FIELD218 256 DARK 359 7475 59 3225 IRCO2 3 2210218 1 F004160562_S004 LMC FIELD218 256 DARK 359 7475 59 3225 IRCO2 4 2210218 1 F004160563 N001 LMC FIELD218 412 DARK 359 7475 59 3225 IRC02 1 2210218 1 F004160563_N002 LMC FIELD218 412 DARK 359 7475 59 3225 IRCO2 2 2210218 1 Version 1 3 September 13 2007 61 5 9 The pipeline processor Green Box 5 9 1 Configuration Before running the pipeline you can configure the parameters for the pipeline by typing epar pipeline O000 X XGterm IRAF v 2 12 1 NFS Darwin IRAF Image Reduction and Analysis Facility PACKAGE irc TASK pipeline irccons J constants database IRC constants database file name com_mod median combine mode average median com_are 2 stack area 1tcommon area 2twhole area sky_are 2 sky matching areat 1 common area 2 whole area det sig 4 Threshold in sigma for source detection sig rej 3 Rejection unit in sigmas max itr 10 Maximum number of iterations obslog irclog output of the mkirclog selfdar no Use selfdark instead of Super Dark coaddsh no Coadd short exposure frames int
44. 31 2 228 1 847 1 500 0 897 80 95 149 3 460 1 582 1 508 1 238 0 923 90 19 552 2 451 1 275 1 313 1 097 0 949 100 6 900 1 909 1 110 1 192 1 015 0 974 110 3 696 1 586 1 016 1 112 0 966 0 996 120 2 559 1 379 0 961 1 057 0 936 1 017 130 2 034 1 240 0 928 1 017 0 919 1 037 140 1 740 1 142 0 910 0 989 0 909 1 054 150 1 553 1 073 0 901 0 968 0 904 1 070 160 1 424 1 023 0 899 0 952 0 903 1 085 170 1 330 0 987 0 901 0 939 0 903 1 099 180 1 258 0 961 0 907 0 930 0 906 1 111 190 1 202 0 942 0 914 0 923 0 909 1 123 200 1 158 0 929 0 923 0 917 0 913 1 133 210 1 122 0 921 0 933 0 913 0 918 1 143 220 1 093 0 915 0 943 0 910 0 923 1 152 230 1 068 0 913 0 954 0 907 0 928 1 161 240 1 048 0 913 0 965 0 905 0 933 1 169 250 1 032 0 914 0 976 0 904 0 938 1 177 300 0 980 0 939 1 031 0 902 0 962 1 208 350 0 957 0 976 1 079 0 904 0 983 1 233 400 0 948 1 016 1 121 0 907 1 000 1 253 450 0 946 1 055 1 157 0 910 1 015 1 269 500 0 947 1 092 1 187 0 914 1 028 1 284 600 0 953 1 156 1 235 0 920 1 048 1 309 700 0 960 1 209 1 272 0 926 1 063 1 329 800 0 967 1 251 1 300 0 930 1 076 1 348 900 0 974 1 286 1 322 0 934 1 086 1 364 1000 0 980 1 315 1 340 0 937 1 094 1 379 1500 0 999 1 404 1 394 0 948 1 120 1 439 2000 1 009 1 450 1 421 0 953 1 135 1 483 2500 1 015 1 478 1 436 0 957 1 144 1 514 3000 1 019 1 496 1 447 0 959 1 151 1 538 3500 1 022 1 509 1 454 0 961 1 155 1 557 4000 1 025 1 519 1 460 0 963 1 159 1 571 4500 1 026 1 526 1 464 0 964 1 162 1 583 5000 1 028 1 532 1 468 0 964
45. 5 whose frame is twice the period of other AOTs see next section Except for IRC05 one frame is about 63 sec in which NIR has one short and one long exposures and MIR S and MIR L have one short and three long exposures The short exposure is made with the Fowler 1 sampling scheme and the long exposure is made with the Fowler 4 sampling For the IRC05 the clock of the MIR S and MIR L is the same as for other AOTs but repeated twice in a frame The short exposure of the NIR in IRCO5 is taken once and the exposure scheme is the same for other AOTs The long exposure of the NIR in IRCO05 is made with the Fowler 16 sampling scheme and is taken once in a frame 2 2 Instrument AOTs In a pointed observation the filter and dithering combinations for IRC have been fixed to a few patterns which are called Astronomical Observation Template AOT The duration of a pointed observation and the frame time are well determined Each AOT consists of a combination of frames of the IRC operation with dithering and filter wheel rotation operations which is well fixed in a pointed observation Figure 2 2 6 shows an illustration of the pre fixed sequences in each AOT In all the AOTs except for AOT11 a dark frame is carried out before and after the observation pre dark and post dark observations In a pointed staring observation the IRC observation is started once it receives the notifi cation of the stabilization of the attitude from the attitude and orbital
46. 5 1 016 1 095 0 912 0 973 1 185 600 0 953 1 059 1 132 0 913 0 985 1 200 700 0 955 1 094 1 160 0 915 0 995 1 211 800 0 958 1 124 1 181 0 917 1 002 1 220 900 0 961 1 148 1 198 0 919 1 008 1 227 1000 0 964 1 168 1 212 0 920 1 013 1 233 1500 0 975 1 232 1 253 0 924 1 027 1 254 2000 0 982 1 264 1 273 0 927 1 035 1 267 2500 0 986 1 284 1 285 0 929 1 040 1 275 3000 0 988 1 297 1 293 0 930 1 043 1 281 3500 0 990 1 306 1 298 0 931 1 046 1 285 4000 0 992 1 313 1 303 0 931 1 047 1 288 4500 0 993 1 318 1 306 0 932 1 049 1 291 5000 0 994 1 322 1 308 0 932 1 050 1 293 6000 0 995 1 328 1 312 0 933 1 051 1 297 7000 0 996 1 333 1 315 0 933 1 053 1 299 8000 0 997 1 336 1 317 0 933 1 053 1 301 9000 0 997 1 339 1 318 0 934 1 054 1 302 10000 0 998 1 341 1 320 0 934 1 055 1 303 20000 1 000 1 350 1 325 0 935 1 057 1 309 30000 1 001 1 353 1 327 0 935 1 058 1 311 40000 1 001 1 355 1 328 0 935 1 058 1 312 50000 1 001 1 356 1 329 0 935 1 059 1 312 60000 1 001 1 356 1 329 0 935 1 059 1 313 Values are calculated for A4 7 0um As 9 0um Ag 11 0um Ay 15 0um Ag 18 04m Ag 24 0um 42 IRC Data User Manual Table 4 8 13 Color Correction factors for MIR channel Gray Body a 2 Intrinsic MIR 5 57 MIR S S9W MIR S SI1 MIR L L15 MIR L L20W MIR L L24 Temperature K K 7 0um K 9 0um K 11 0um K 15 0um K 18 0um K 24 0um 40 80 578 9 482 8 362 0 890 50 88 714 10 821 4 108 3 417 0 864 60 13 417 3 932 2 513 2 044 0 875 70 866 347 5 7
47. 5 4 4 Unpack irc tgz e mv ircYYMMDD tgz where you want to install e cd where you want to install e tar xvzf ircYYMMDD tgz 5 4 5 Make irc binaries e cd where you want to install irc sre e make This will create binary files in where you want to install irc bin 5 4 6 Run setpath pl e cd where you want to install irc lib e perl setpath pl This will rewrite the setpath dat Please check the following line in setpath dat is the following format set irchome where you want to install irc This line should indicate the location where you installed the irc package Version 1 3 September 13 2007 53 5 4 7 Perl path Typing which perl in the unix environment indicates where the perl is installed in your system The toolkit assumes that the perl is installed in usr local bin perl Therefore those who have the perl binary elsewhere should do as follows e If you know the root password please create a symbolic link in usr local bin by typing cd usr local bin In s which perl perl It will create a symbolic link perl in usr local bin e If you don t know the root password please modify the first line of every perl script file in where you want to install irc perl For example if you have perl in usr bin perl please modify it as usr local bin perl gt usr bin perl As long as you have perl in usr bin perl you can also achieve the same result by runn
48. 6602227 AOV 5 751 2006 04 22 02 03 59 NPM1p65_0451 16533704 6538175 K2III 6 524 2006 12 20 01 36 45 HD158485 172604844 5839069 A3V 6 145 2006 08 22 17 47 59 Bp66_1060 17560018 6655430 K2III 6 720 2006 11 07 16 16 54 NPM1p67_0536 17585466 6747368 K2III 6 409 2006 04 24 14 43 03 HD165459 18023073 5837381 A1V 6 584 2007 03 22 01 44 58 Bp66_1073 18030959 6628119 KIIII 7 544 2006 06 28 19 41 30 KFOIT4 18040314 6654459 K1 5III 8 067 2006 06 24 23 13 10 HD166780 18083882 5758468 KAII 3 963 2006 09 29 19 43 33 IRC Data User Manual Table 4 6 3 SAGE standard stars Star 2MASS ID Sp K Obs Date for N3 S7 S11 HD34461 05121801 6705415 KIIII 6 927 2006 06 08 18 21 09 HD34555 05125331 6744362 A3V 9 288 2006 05 31 05 38 19 HD34943 05153759 6804070 MOIII 3 970 2006 05 26 06 46 47 HD35094 05164382 6811142 A3V 8 554 2006 05 23 07 47 52 HD35094 2 2006 05 25 06 01 07 HD35183 05172303 6828190 A3V 8 754 2006 05 20 23 40 20 HD35323 05183268 6732320 MOIII 5 524 2006 11 29 13 25 55 HD35461 05194979 6626353 KIIII 5 683 2006 12 20 04 04 52 HD35461 2 2006 12 18 23 57 17 HD35665 05205607 6759034 KI1 5III 5 887 2006 11 24 15 59 18 HD35905 05223623 6721285 KIIII 6 800 2006 12 01 05 07 47 HD36207 05244398 6753507 KIII 6 451 2006 11 23 05 15 04 HD37122 05300077 6958319 K2III 5 128 2007 04 25 09 40 22 HD269704 05315890 6909392 K2III 6 760 2006 11 03 18 51 36 HD269704 2 2006 11 03 22 10 04 HD269757 05335125 6946468 KIII 8 187 2007 04 22 07 07 5
49. 9 HD37722 05344387 6928187 A4V 8 640 2007 04 25 03 03 00 HD37762 05344709 7010197 KOI 5 565 2007 04 18 07 03 34 HD269788 05345367 6846395 KAII 6 331 2007 05 02 05 40 40 HD269820 05355068 6929178 KIIII 7 185 2006 10 29 19 47 54 HD269820 2 2006 10 30 18 56 54 HD38861 05423256 7022555 KOIII 6 780 2007 04 08 06 53 36 HD38993 05431866 7027254 KI1 5III 5 478 2007 04 06 05 12 23 HD39980 05495921 6941060 K2III 5 409 2006 10 09 18 39 20 HD270186 05501123 6934296 KOI 7 774 2006 10 09 18 39 20 for L15 L24 HD34461 05121801 6705415 KIIII 6 927 2006 06 03 06 16 38 HD34461 2 2006 06 04 05 23 26 HD34489 05122388 6756520 K2III 6 169 2006 05 24 06 54 23 HD34943 05153759 6804070 MOIII 3 970 2006 05 21 06 16 33 HD35323 05183268 6732320 MOI 5 524 2006 12 04 17 29 51 HD35323 2 2006 12 07 11 40 07 HD269352 05193283 6752441 KO5III 6 014 2006 11 29 13 25 55 HD35461 05194979 6626353 KIII 5 683 2006 12 31 13 05 43 HD35665 05205607 6759034 KI1 5III 5 887 2006 11 28 15 55 38 HD35905 05223623 6721285 KIIII 6 800 2006 12 06 15 48 56 HD36207 05244398 6753507 KIII 6 451 2006 11 27 18 25 19 HD37122 05300077 6958319 K2III 5 128 2007 04 21 07 56 30 HD269704 05315890 6909392 K2III 6 760 2006 11 07 20 25 50 HD37722 05344387 6928187 A4V 8 640 2006 11 02 19 42 32 HD37722 2 2006 11 03 05 37 54 HD37762 05344709 7010197 KOI 5 565 2007 04 12 06 57 26 HD269788 05345367 6846395 KAII 6 331 2006 11 08 22 53 20 HD269820 05355068 6929178 KIII 7 185 2006 11 02 19 42 32 V
50. 9 11 Distortion from ideal grid square for N2 S7 and L24 from left to right respec tively For clearity the length of distortion vectors are multiplied by 20 4 10 Memory effects caused by bright source observations MIR S and MIR L channels show an anomalous sign after they observe bright sources An exam ple is shown in Figure 4 10 12 It seems to be a decrease of the sensitivity Although the amount is less than 196 it could severely affect detection of faint sources because of high background in the MIR We recognize that the effect last for hours The threshold level to produce this is not yet clearly known but IR AS 12um sources definitely produce it At this stage no recipe for the 44 IRC Data User Manual correction is available If your data show this pattern it should come from previous observa tions A list of potentially affected observations is available If you would like to check whether or not your observations are possibly affected please contact us at iris_help ir isas jaxa jp or http akari esac esa int esupport Figure 4 10 12 Transients effects seen in MIR S after a bright source observation top left The following images correspond to observations taken 2 4 and 5 hours after the first one A deep can be seen where the bright source was located in the first observation 4 11 Astrometry A corrected and final astrometry can be achieved at the end of the pipeline cross correlating with the 2MASS catalogu
51. 92 6 4 2 Running the pipeline 22e 93 6 4 8 Summary of interactive operations within the pipeline 96 6 4 4 Warning messages ofthe pipeline 2n 97 6 5 Working on the pipeline output 2e 98 6 5 1 Displaying the whole images onds9 98 6 5 2 Displaying the extracted images on ATV 99 6 5 3 Checking for wavelength zero reference point with the zero th order light IIHAge 2 he eon e dh seg a tu Sol ee Ye oe tm aso Y Eee amp 99 6 5 4 Spectral plotting tool 2e 100 6 5 5 Working on saved data 22 222 es 102 6 6 Append x x UE ee RR Ada er eR he RR ame deed 102 6 6 1 Variable name conventions es 102 6 7 Frequently Asked Questions and Troubleshooting 103 Chapter 1 Introduction 1 1 Purpose of this document This document is intended to provide a comprehensive guide to data from the Infrared Camera IRC onboard AKARI for observers to get started quickly It includes a brief summary of the instrumentation a summary of the data products and the pipeline software overview together with the calibration of the instrument Part of the data and information given here will be updated as the data reduction and calibration will get improved The latest information may be posted on the web till the revised version will be prepared A comprehensive overview of the instrumentation and operation is described in the ASTRO F Observers Manual and will b
52. AKARI IRC Data User Manual Version 1 3 Rosario Lorente Takashi Onaka Yoshifusa Ita Youichi Ohyama Toshihiko Tanab and Chris Pearson with contributions from Martin Cohen Daisuke Ishihara Hideo Matsuhara Itsuki Sakon Takehiko Wada Issei Yamamura European Space Astronomy Centre ESAC ESA Tokyo University Japan 3Institute of Space and Astronautical Science ISAS JAXA University of California Berkeley National Astronomical Observatory Japan NAOJ September 13 2007 Version 1 3 September 13 2007 07 March 2007 Release of version 1 0 21 March 2007 Updated Table 4 6 7 photometric conversion factors 18 June 2007 Updated Sections 4 6 3 4 10 4 13 5 13 2 5 14 6 1 06 September 2007 Version 1 3 Updated Chapter 4 5 and 6 Included ghosts in Np observations section 4 13 Date Revision Comments Contents 1 Introduction 1 1 1 2 Purpose of this document 2 22s Relevant information 4 lle es 2 Instrument overview 2d Focal plane arrays 4n X e teinana Rex X WE XGE Bue y oer 2 1 4 Near InfraRed Camera NIR a 2 1 2 Short wavelength Mid InfraRed Camera MIR S 2 1 8 Long wavelength Mid InfraRed Camera MIR L 2 1 4 Arrays operation oaoa rss 2 2 Instrument AQXES 3 nuce RR uec A RUE Ce eet de MEQUE RO REESE ELS 2 24 MRO OOM ress necem rst rete oper ets M A paier dic eye sia es 2 22 IRO02 24x fe ee RR ek ae BOX
53. Check the source detection parameters for better source detections This message is not so serious for most cases Warning sigma of specbox Y shift measurement seems too large This warn ing appears when the toolkit fails to measure the relative Y shift of NP spectra with respect to N3 reference image Check the reference image on ds9 to know if source de tection could be made successfully Sometimes the program might detect only cosmic ray hits not stars if the parameters are not optimized Warning sky level is larger than 1 5 times sky sigma After extracting spec troscopy images remaining sky will be subtracted off locally around the source If every thing works fine the average of the remaining sky should be very small and is typically less than 1 5 times sky rms fluctuation This warning message will appear if sky level is larger than the typical value When this warning appears check the ds9 image to see if there remains large scale sky level variation Presence of such sky often indicates failure in monochromatic flat fielding and or global sky subtraction or any big unexpected debris are found Note that this warning will not appear for slit spectroscopy including NG at Np Warning Available sky area is too small Due to clouded source masking re maining sky area is too small for measuring sky Check the source detection parameters for fewer source detection This message is not so serious for many cases Warning Cros
54. FITS format observation data and a Readme file describing the contents etc During the prioritized data use period data packages are encrypted and compressed by gpg GnuPG program The naming convention for the IRC package is AKARLIRC TargetID SubID tar gpg where TargetID is a 7 digits number and SubID is a 3 digits number given by the observation database Combination of TargetID and SubID give a unique identification of the observation When extracting an IRC data package a directory named AKARLIRC TargetID SubID is created Two subdirectories rawdata and irc_ql contain the raw FITS data files and the processed result files respectively 3 2 Raw data description 3 2 1 Raw data naming convention The naming convention for the IRC raw data files is common for all the IRC AOT ans it is the following FVVxxxxxx N M fits where e F is a fixed character e VVVxxxxxx Extended frame counter decimal degits This is a unique identifier of the exposure 14 Version 1 3 September 13 2007 15 e xxxxxx frame conter in the telemetry file e VVV maintained by the FITS creation program Incremented when xxxxxx is reset to 0 N M NIR MIR Scan mode data may have extra characters to this NIR data is in a separate file while MIR S and MIR L are stored in the same FITS file The README file describes e file list in the data package e observation summary extracted from the FITS file e comments specific for the
55. Fy of the IRC MIR S Camera S61 SG2 Relative Response Oo Oo Oo o no D Oo on T T T T O T 12 B Wavelength um Js co Figure 4 5 8 The Relative Spectral Response Function of the IRC MIR S dispersion elements per photon 30 IRC Data User Manual Relative Response O O D gt oO co o N O 10 15 20 25 30 Wavelength um Figure 4 5 9 The Relative Spectral Response Function of the IRC MIR L Camera for F Relative Response O O D ox co Oo oV O 20 24 28 Wavelength um On Figure 4 5 10 The Relative Spectral Response Function of the IRC MIR L dispersion elements per photon LG1 will not be used for astronomical observations due to degradation Data shown here is 2nd order light subtracted Version 1 3 September 13 2007 31 4 6 Flux calibration for point sources 4 6 1 Observed standards and data processing The standard stars for the absolute flux calibration were selected in the North and South Ecliptic Pole NEP and SEP regions which were established by M Cohen originally for the calibration of the IRAC onboard Spitzer We list the observed NEP and SEP standards in Table 4 6 2 The observations of these standards were carried out with IRC AOT03 which performs either imagings of all the NIR and MIR S bands or all the MIR L bands In addition to these standards we have included standard stars in the Large Magellanic Cloud A fairly large area i
56. ON dat WAVEPAR IRCWAVEPAR_NP dat WAVEPAR IRCWAVEPAR_NG dat WAVEPAR IRCWAVEPAR SG1 dat WAVEPAR IRCWAVEPAR SG2 dat WAVEPAR IRCWAVEPAR LG2 dat IRCSPECBOXDYPAR NP dat e IRC_SPECRED_CALIBDIR RESPONSE contains ascii files tabulating A vs ADU long expo sure time in seconds mJy for each disperser There are five such tables NP NG SG1 8G2 LG2 The spectral response is measured by observing flux standard stars with known flux energy distribution This directory also contains a NP template spectrum wavelength calibrated spectrum before calibrating spectral response that will be used to find relative image shift along dispersion direction or Y between spectroscopy and reference images RESPONSE RESPONSE_NP 1st RESPONSE RESPONSE_NG 1st RESPONSE RESPONSE SG1 1st 92 IRC Data User Manual RESPONSE RESPONSE _SG2 1st RESPONSE RESPONSE_LG2 1st e IRC_SPECRED_CALIBDIR DISTPAR contains ascii tables tabulating spectral tilt informa tion of the grism insertion angle DISTPAR IRCDISTPAR_NP dat DISTPAR IRCDISTPAR_NG dat DISTPAR IRCDISTPAR SG1 dat DISTPAR IRCDISTPAR SG2 dat DISTPAR IRCDISTPAR_LG2 dat e IRC_SPECRED_CALIBDIR APCOR contains ascii tables tabulating spectral aperture cor rection table APCOR APCOR_NP dat APCOR APCOR_NG dat APCOR APCOR_SG1 dat APCOR APCOR_SG2 dat APCOR APCOR_LG2 dat 6 4 Running the pipeline 6 4 1 Data reduction order Th
57. a clipping averaging method is used For reference images a simple median averaging is used due to the small number of sub frames 6 1 8 Target detection position measurement Target positions can be provided by the user in a form of the source table see below or automatically computed within the toolkit using the daofind method for target detection Even if the target positions are set by users the toolkit performs re centering of the source positions by means of Gaussian peak search This functionality can be disabled with the command line option This process is coupled with the following step background subtraction 6 1 9 Background subtraction from stacked image Although background subtraction has been already applied before stacking images we here remove any remaining background The background subtraction and target detection position measurement explained in the previous section are made iteratively in the following way First the target is tentatively de tected by the automatic source detection program not based on the user supplied source table and source masks will be created for all detectable sources Then the background is measured while considering the source masks and is subtracted off from the original stacked image The background subtracted image is used for better source detection position measurement and better source masks will be created for better background subtraction After sources are detected sour
58. an GOOD may not be in the archive at the first stage e instrument information ORIGIN is the organization creating FITS file TELESCOP is the AKARI mission Satellite Name INSTRUME is the Identifier of the instrument DETECTOR is the detector name either NIR or MIR e observation details OBSERVER is the PI Name Observer s ID PROPOSAL is the Proposal ID OBS CAT is Observation Category either LS MP OT DT CAL or ENG PNTNG ID is the Pointing ID Usually it is identical with the Target ID but is different for parallel mode observations TARGETID is the Target ID SUBID is the Target Sub ID OBJECT is the Object name OBJ RA is the RA Target position in degrees recorded in the database double pre cision OBJ DEC is the DEC Target position in degrees recorded in the database double precision AOT is the Observation AOT AOTPARAM is the AOT Parameters set INSTMODE is the Instrument operation mode Contents TBD TIMESYS is the Time system used in this file DATE OBS is the Observation start date time with format YYYY MM DDTHH MM SS DATE END is the Observation end date time Same format than above DATE REF is the Reference time in the Observation Same format than above AFTM OBS is the DATE OBS in ASTRO F Time AFTM END is the DATE END in ASTRO F Time AFTM REF is the DATE REF in ASTRO F Time PIMTIOBS is
59. ative to MIR S L Please note the dark areas in each images These areas are reserved for slit spectroscopy and they are supposed not to receive any light Therefore these regions are useful to monitor the dark level Hereafter we refer to this this dark area as the slit area and the other as the imaging area redbox mkirclog The redbox mkirclog creates the observing log file irclog This is a text file with content shown in Table 5 8 1 The contents of the irclog file summarize the nature of the processed files in the working directory The table entries correspond e FRAME The filename corresponding to the sliced frame e g following the format such as F23340_L001 fits as described above e OBJECT Target name taken from the original target list e NAXIS Number of pixels in cross scan direction 256 for MIR and 412 for NIR e FILTER Filter name i e N2 N3 N4 S7 L24 or DARK e RA SET Right ascension coordinates e DEC SET Declination coordinates Version 1 3 September 13 2007 59 e AOT AOT type e g IRC02 IRC03 etc item EXPID Sequential frame number during an exposure cycle e g 1 or 2 for NIR and 1 2 3 or 4 for MIR S and MIR L images e IDNUM Pointing ID e SUBID sub Pointing ID greater than 1 for multi pointing observations 1 NIR fits image 1 MIR fits image NI SI LI S4 LA N1 short exposure frame NIR S1 short exposure frame MIR S L1 short exposure frame MIR L N2 long exposure
60. cally or interactively step by step 5 3 Expected Data Processing Rate minimum expectation e NIR 412 512 pixels 16 bit pixel 8 bit byte 412 Kbyte frame 1 NIR fits 412 Kbyte frame 2 frame fits 824 Kbyte fits file e MIR 256 512 pixels 16 bit pixel 8 bit byte 256 Kbyte frame 1 MIR fits 256 Kbyte frame 4 frame fits 1024 Kbyte fits file e AOTO2 6 exposure cycles pointing 1 NIR fits 1 MIR fits exposure cycle 10 8 Mbyte pointing e AOTO3 6 exposure cycles pointing 1 NIR fits 1 MIR fits exposure cycle 10 8 Mbyte pointing e AOTO05 9 exposure cycles pointing 1 NIR fits 1 MIR fits exposure cycle 16 2 Mbyte pointing The toolkit requires 3 or 4 times of disk space compared to the data 52 IRC Data User Manual 5 4 How to install the IRC imaging Toolkit 5 4 1 Requirements The toolkit is developed in a Linux PC environment and has also been successfully run in the OSX Unix environment The toolkit requires the following environment for its full function e unix Solaris MacOS X Linux BSD e IRAF 2 12 2 or later gcc 3 2 or later e perl e curl for WCS calculation on Solaris machines 5 4 2 Install IRAF e http iraf noao edu 5 4 3 Download IRC imaging data reduction software package The latest version of the toolkit with installation and operating instructions can be obtained from the AKARI Observer s web site see section 1 2
61. ce image If ext source table is set to for automatic source detection null string then the pipeline activates its sub program There are some important tips in preparing the source table Coordinates counts from 1 not 0 i e coordinates of the lower left corner of the image is 1 1 not 0 0 For NIR the toolkit interprets by default the pixel coordinates as measured in the raw image before the image rotation When one measures the source positions in the processed images after the image rotation set the rotated NIR source table option at the pipeline command line For MIR L one can use Y range of either 257 512 for images before detaching MIR S L or 1 256 for images after detaching MIR S L to set Y position of the targets The pipeline subtracts 256 from the Y input if Y 256 94 IRC Data User Manual e refimage_list An ascii file listing FITS file names of reference image Example N3 1st cat N3 1st F000001 N fits A default list is provided with the data distribution and is found in rawdata directory specimage list an ascii file listing FITS file names of spectroscopy images Example NP 1st cat NP 1st F000002 N fits F000004 N fits F000006 N fits A default list is provided with the data distribution and is found in rawdata directory Even if you find some images being damaged severely and you do not want them to be included in the pipeline processing you must li
62. ce masks are created for better background subtraction from individual sub frames section 6 1 4 in the second pass of the pipeline processing 6 1 10 Extracting 2D spectra By using the reference image positions and pre defined coordinates offsets in the calibration database rectangle areas around the source spectra are extracted on the spectroscopy images For NP spectral distortion see below is taken into account along the dispersion direction to find best Y offset when extracting 2D spectra X offset adjustment in spectroscopy image extraction In real data one need to further adjust the offset of the source extraction boxes One may some times find 2D spectra slightly away from the expected position at the center of the extraction 86 IRC Data User Manual box along X axis This kind of shift can not be corrected in the previous image registration processes and the correction can be made at this stage The center position of the 2D spectra is measured for each extracted 2D spectra and calculate the mean X offset from the center of the extraction box If the pipeline successfully finds the shift value the shift will be applied in extracting the 2D spectra However for some cases where only very faint objects were detected calculating this additional shift may fail and no further shift will be applied Measuring the X offset and making good source masks as explained in the previous section are closely related to each other durin
63. ch as Jy We will provide a table of the conversion of ADU s to Jy for point sources in addition to the integration time for each frame The in orbit calibration is made based on aperture photometry with an aperture radius of 10 pixels for the NIR and 7 5 pixels for the MIR S L For other aperture sizes observers have to make an aperture correction which is currently not known owing to the unknown PSF not stable and not well defined Similar to the Spitzer IRAC arrays reflection of the light within the arrays exists also for the IRC arrays The calibration for the diffuse light is currently not available The toolkit including the entire source code will be released and be progressively updated reflecting the user s feedback The user will receive the raw data plus automatic toolkit to make basic science data The user may customize the toolkit at his her own risk However we will provide the super dark and super flat They will be updated periodically and will be given the version number e g Super flat Y YMMDD fits Super darkY YMMDD fits The goal for the toolkit is to achieve an absolute flux accuracy of better than 3096 25 and 2596 at the 5 sigma detection limit for NIR MIR S and MIR L channels respectively However these goals are at present the optimal values and we do not guarantee the numbers 5 2 General overview of the toolkit processing The toolkit runs on IRAF and it consists of two main steps plus an optional step
64. clock is running just around pointing observations Therefore the temperature conditions of the array could be a function of both time and pixel positions since only a part of the entire array is used while in the survey mode The effect can be clearly seen in the pre dark image where series of columns of higher dark current are found around the center of the Y axis in MIR S L and the excess of the dark current decreases with time and is essentially invisible in post dark images taken after 20 min from the start of the imaging mode clock operation In the case of spectroscopy observations the post dark image is taken just after finishing imaging exposures and the image shows a memory pattern of the bright background within the slit less area Therefore it can not be used as the dark image to subtract from other images 4 2 Flatfield 4 2 1 Flatfileds for imaging data The IRC super flats for the MIR bands were derived by observing the high surface brightness zodiacal background For NIR we used hundreds of pointing data from the North Ecliptic Pole survey because the observing chances near ecliptic plane are limitted and we could not erase the stellar contribution from the flat images with only a few pointings of data Unfortunately stray 21 22 IRC Data User Manual Figure 4 1 1 Dark images taken in April 2006 above and in August 2006 below of NIR MIR S and MIR L from left to right respectively Cleary the number of hot
65. controlling system The IRC observation continues till the angle between the telescope axis and the earth rim becomes 10 IRC Data User Manual less than a certain value Thus the last image in a pointed observation may be taken during the maneuver and cannot be used for astronomical observations This will be correctly treated in the pipeline software Manuever z Figure 2 2 6 Observation sequences of the AOT IRCOO 02 03 04 and 05 Yellow boxes labeled as Exposure cycle indicate exposure frames Orange boxes with M are Micro Scan operations including stabilization and light blue boxes with W are Filter Wheel rotations Dead time for a Filter Wheel change depends on the relative position of the elements The Green area on the right side is the extra observation time which is not guaranteed 2 2 1 IRCO0O0 The IRC00 mode was designed for deep imaging observations After performance investigation in orbit it has been replaced by IRCO05 and is not in use 2 2 2 IRCO02 The IRC02 mode was designed for general purpose imaging observations that take images with two fixed filters in a pointed observation It provides at least three images for a filter with dithering operations 2 2 3 IRCO3 The IRC03 mode was designed for general purpose imaging obse
66. ctive wavelength for broad band filters The equation 6 1 3 leads to obs obj X x R X x Fa z y A spectral_feature x y x Fi x y A 6 1 5 and the color term is expressed as Fia y A Euren uo peeks spectral feature z y x Fi z y spectral_feature x y x Fi x y idi Therefore the color term correction is calculated by two broad band super flats and one spec troscopy super flat Note that the product PFj x y x spectral feature r y always appears together i e we do not have to separate spectral_feature term from the super flat After the color term correction the image is as follows obs obj X x R X 6 1 7 For the NG spectra with the point source aperture Np flat fielding will be made in a similar way to the slit spectroscopy of diffuse sources since the aperture size is much smaller than the 88 IRC Data User Manual size of dispersed spectroscopy images For spectroscopy with NG at Np and other slit spectroscopy data calibration the super flat is normalized to be unity at each wavelength or Y pixel and there are no spectral features in the flats On the other hand NG spectra at Ne will be reduced in the same way as for other slit less spectroscopy 6 1 13 local Background subtraction from extracted 2D spectra Although the background has been subtracted and it should be close to zero at this stage we here locally subtract any remaining background on the extracted 2D spectroscopy images H
67. d Ai 4 6 2 i Xie A i i R Ad A i is where f A is the flux density of a standard star Cohen template and R A is the spectral response the transmission of the optics and the response of the detector unit electron 34 IRC Data User Manual photon of the band i Here f v x v7 or f A x AW is assumed The adopted effective wavelengths of each band Aj are listed in Table 4 6 5 along with the range of the integration Ais Aie Table 4 6 5 The effective wavelength A and the range of integration A and Aj band i Ais i N2 2 40 1 60 5 770 N3 3 20 1 60 5 770 N4 4 10 1 60 5 770 S7 7 00 2 50 23 860 SOW 9 00 2 50 23 510 S11 11 00 2 50 24 000 L15 15 00 2 50 23 760 L18W 18 00 2 50 28 720 L24 24 00 2 50 28 720 4 6 3 Absolute calibration e The observed ADUs are converted to ADU t7 amit according to the unit number Table 4 6 4 We assumed that the error of the estimated flux density of the standard star is 5 76 Observational errors ranging from 5 96 to 100 96 were assigned according to their ADU values A straight line was fitted to the estimated flux density vs the normalized ADU ADU Ed The slope of the fitted lines provides the conversion factors fo ADU b Jy which are tabulated in Table 4 6 6 Using these factors we calculated conversion factors f ADU to Jy for short exposure data and f ADU to Jy for long exposure data which are tabulated in Table 4 6 7 Table 4 6 6 Con
68. d for SG1 SG2 and LG2 Satellite attitude stability instability Position shifts among sub frames are frequently observed due to the satellite attitude drift in the pointing attitude control mode The drift is as large as several pixels in the worst cases To correct the drift one needs to Version 1 3 September 13 2007 49 register images among spectroscopy sub frames and between spectroscopy and reference images The second correction is very important to determine the wavelength reference point Note that it takes about 30 sec to switch the spectroscopy mode to from the imaging mode for the filter wheel rotation Thus the time interval between the last spectroscopy sub frame and the first imaging sub frame is longer by this period than the interval between taking spectroscopy sub frames To measure the image drift among spectroscopy images we use cross correlation image matching technique To measure the image drift between spectroscopy and reference images we could use the same cross correlation technique In this cross correlation we use a template spectrum that was extracted from the data with negligible drifts Then by comparing the template with spectra extracted from the spectroscopy data in question we estimate the relative drift Since all the channels operate simultaneously any image drifts along the X and Y directions found in NIR or MIR S should be seen in MIR L with the different pixel scale For spectra of diffuse sources at
69. e This option is turned off by default This is because the software to retrieve the 2MASS catalog may be platform dependent To enable this option curl has to be installed in the system The instructions for it are beyond this manual Ask your computer administrator for the installation The current pipeline has a capability to first include the pointing information from the attitude and orbital control system AOCS directly in the WCS format If matching with the 2MASS data is successful then the pointing information will be replaced by the matching data The parameter WCSROOT in the FITS header indicates which information AOCS or 2MASS is used to determine the coordinates 4 12 Arrays anomalies 4 12 1 NIR array NIR anomalies are shown in Figures 4 12 13 and 4 12 14 Most of them also affects the Spitzer IRAC instrument e Muxbleed Version 1 3 September 13 2007 e Muxstripes e Column pulldown e Banding Figure 4 12 13 Banding in NIR array 45 IRC Data User Manual 46 Figure 4 12 14 NIR array anomalies Muxbleed Muxstripes and column pulldown Version 1 3 September 13 2007 47 4 12 2 MIR arrays e vertical stripes as shown in Figure 4 12 15 e short exposure frame data There is a problem in the short exposure frame data taken with IRCO00 IRC04 and IRC05 The phenomena is only seen in the short frames immediately following the frames without any operations such as filter wheel operatio
70. e given in the AKARI first result volume 2 IRC Data User Manual 1 2 Relevant information AKARI Observer s Web The ISAS Web page contains the most up to date information URL Attp wwuw ir isas jaxa jp A K A RI Observation The ESAC page also includes up to date information URL Attp akari esac esa int observers Helpdesk Any questions and comments on AKARI observations and user support are addressed to the AKARI Helpdesks iris help ir isas jaxa jp http akari esac esa int esupport Version 1 3 September 13 2007 Chapter 2 Instrument overview The Infrared Camera IRC onboard AKARI was originally designed to make wide field deep imaging and low resolution spectroscopic observations in the pointing mode of the AKARI satel lite Its unique wide field coverage of 10 x 10 arcmin is ideal for survey type observations or multi object spectroscopic programs The capability for the use in the survey mode has later been explored and now is also being used to carry out mid infrared all sky survey observations Each channel has a filter wheel on which medium band filters dispersive elements and a blank window as a shutter are installed Table 2 0 1 summarizes the parameters of the IRC filters and dispersion elements The NIR camera covers three independent wavelength bands that very roughly correspond to the well known K L and M bands Each of the two MIR cameras have two narrow filters which cover the shorter and longer ha
71. e Fowler 4 sampling 2 2 6 IRCI11 The IRC11 was designed for wide area observations or slow scan observations with the IRC Only the MIR S and MIR L channels can be used in the IRC11 The arrays are operated in the same manner as in the all sky survey mode making binning of 4 pixels in the cross scan direction On the orbit the data downlink capacity was found to be sufficient to transmit the full resolution data in the IRC slow scan The unbinning mode is now designated as IRC51 and all the IRC slow scan observations from 2007 January are executed with the IRC51 mode 2 2 7 IRCS51 This is the same slow scan mode as the IRC11 except that the IRC51 provides the full spatial resolution without binning in the cross scan direction All the IRC slow scan observations after 2007 January are executed with IRC51 2 3 In orbit sensitivity Optical thougputs of all IRC channels are confirmed to be as expected by observations of stan dard stars The sensitivity values for the imaging AOTs are collected in Table 2 3 3 The IRC05 numbers are preliminary In the case of IRC04 Figures 2 3 7 and 2 3 8 show the sensitivity lo noise with 2 x 2 binning and the trhoughput of the system respectively as a function of A 12 1 o Noise Equivalent Flux mJy 0 1 0 01 IRC Data User Manual IRC SPECTROSCOPY LIMITTING FLUX LEVEL 3 4 5 6 7 8 9 10 12 14 16 18 20 22 2426 Wavelength um Figure 2 3 7 IRC spectroscopy sensitivity a
72. e is then subtracted from each individual corresponding frame From these median values an average sky brightness is calculated that is then re added to all the frames A log file is produced during this process with a name skypair0002 N2 1ist long or similar This file lists the frames looking at a given area on the sky with the corresponding mean median and mode sky brightness and lo standard deviation Those who want to subtract median box car filtered image instead of adjusting sky level should configure the following parametes by typing epar adjust sky before running pipeline e submeds boolean yes or no default no Subtract median filtered sky e rmmedsk boolean yes or no default yes Version 1 3 September 13 2007 71 Remove median filtered image Those who want to check box car median filterd image set this parameter no e x_box integer min 1 max 100 default 21 x box car size e y_box integer min 1 max 100 default 21 y box car size After adjusting the sky level the R prefix is removed and an A is added such that RSfdmslnDwaF23342 8004 fits becomes ASfdmslnDwaF23342 8004 fits 4 Image Stacking Co adding bluebox irc stack Once every frame pointing at the same area of sky in the irclog list has been correctly matched and the sky brightness adjusted every frame is stacked to produce the final co added image The FITS images and files created from the stacking process can be found within a
73. e pipeline has to be run more than once for NP AOT0a or NG AOTOb first then SG1 and SG2 and finally LG2 We need to start processing on shorter wavelength cameras NIR and MIR S where larger number of brighter sources are expected to be observed within the FOV as position reference sources The information derived there will be used for registering longer wavelength cameras MIR S and MIR L e to measure relative X and Y shift among spectroscopy sub frames of NP NG for registering spectroscopy sub frames of NP NG SG1 SG2 LG2 e to measure relative X and Y shift among reference sub frames of S9W with larger number of brighter sources for registering imaging sub frames of S9W and L18W e to measure relative X and Y shift of the spectroscopy image with respect to the reference image of NP or NG for registering reference and spectroscopy images of NP NG SG1 SG2 LG2 Therefore the data reduction order should be as follows e First run run the pipeline for NP AOT04a or NG AOT04b without a target table for enabling automatic target detection sub program For NG data with point source aperture Np one needs to run the pipeline first without Np spec option For other types of slit spectroscopy one may skip this run because shift and add feature is disabled for this observing mode Version 1 3 September 13 2007 93 e Second run run the pipeline for your desired targets Make target tables for N3 S9W and L18W on the raw image
74. ecking for wavelength zero reference point with the zero th order light image The toolkit calculates the wavelength zero reference point based primary on positions of the objects on the reference image It also takes into account a wavelength zero point drift due to satellite attitude drift as well as coordinate rounding effect when extracting 2D spectra In most cases these methods work well to estimate the wavelength zero reference point with accuracy of 0 5 pixel or less Since positions of the zero th order light image are also calculated in the same way as for the wavelength zero reference point one may check the wavelength zero reference point accuracy by comparing positions of expected zero th order light image and actually observed ones Unfortunately this check can be made only when bright sources with measurable zero th order light images are observed at some part of the FOV by chance When the shift is found use the change wave offset command with the estimated wave length shift in pixel The shift is calculated from the current position This command changes the internal variable that records the wavelength offset updates an offset database on disk and Items in parentheses are optional 3In some systems block option may be required for interactive operation on the ATV The conditions for which block requirement seems to depend on local Linux or X11 system settings The author does not have any good ideas on the correct usage o
75. edge This can be explained by the fabrication of the aperture mask The correction for the scattered light is being investigated and will be included in the nex versions of the pipelines 3 MIR L The flat pattern of the MIR L in orbit is different from that in the laboratory It shows a large scale gradient over the array Part of it may come from the scattering of light in the MIR L camera but we have not been able to identify the cause yet The investigation is ongoing For the time being observers should take care of diffuse structures in the MIR L data as they could be spurious In the worst cases overlapping MIR L observations can show 10 difference corresponding to the flatfield inaccuracy in the corners 24 IRC Data User Manual 4 2 2 Flats for spectroscopy images The spectroscopy flatfields are images made by a large number of blank sky spectroscopy images combined and normalized so that any faint object spectra are removed by clipping averaging techniques There are five super flats corresponding to NP NG SG1 SG2 and LG2 As will be described at the end of this chapter these flats show spectral features which are not due to the sensitivity variation Therefore object spectra should not follow the flat pattern However this super flat with spectral features is used in the data reduction because the color term correction actually cancel out the spectral feature correction For flat fielding slit spectra including NG w
76. eference point 6 2 How to install and to set up the IRC spectroscopy pipeline So far only Linux environment has been tested for running the package However the package seems to run on the MAC OS X environment though it is not officially supported Ask you local computer administrator for the IDL installation There are no special re quests in the IDL installation Ask you local computer administrator for the ds9 xpa installation There are no special requests in the ds9 xpa installation Get the irc specred package from the AKARI Observer s web site see section 1 2 Extract and store it under your favorite directory Set the following environment variables in your command line shell Below is an example for csh setenv IRC SPECRED ROOT somewhere setenv IRC SPECRED LIB IRC_SPECRED_ROOT LIB setenv IRC SPECRED HOME IRC_SPECRED_ROOT ASTRO F setenv IRC SPECRED CALIBDIR IRC_SPECRED_HOME IRC_SPECRED CALIBDIR setenv IRC SPECRED DATADIR lt anotherwhere gt setenv IDL PATH lt IDL system path gt IRC_SPECRED_HOME IRC_SPECRED_LIB IRC SPECRED ROOT IRC SPECRED DATADIR and IDL_PATH have to be modified by substi tuting somewhere lt anotherwhere gt and lt IDL system path according to your local system settings 6 2 1 Data preparation As explained in Chapter 3 when de packing the data from the archive under IRC SPECRED DATADIR the raw data will be stored in IRC_SPECRED_DATADIR lt targetid gt
77. elected from the toolkit parameter list to ensure valid coaddition Coadd failure of the MIR L images The coaddLusingS function Often the coadd images procedure fails for MIR L images due to the lack of a sufficient number of bright stars in the FOV of a MIR L image In response within the irc_tool package there is additional independent software the coaddLusingS function that performs the calcula tion of the shift and rotation of each frame using the information from the MIR S channel to coadd the MIR L images In the working directory typing irc tool at the IRAF prompt will enter the irc_tool environment and the coaddLusingS function should be visible in the iraf terminal To use this software type coaddLusingS at the IRAF prompt and it will ask for the name of a reference MIR S list name e g pair0002 S7 list long from an already successfully coadded MIR S image to perform the coadding Other additional parameters may be requested but the default will be sufficient for most users After the parameters have been set coaddLusingS will coadd the corresponding MIR L images automatically and move them to the stacked IM directory Note One needs to make sure that the correct corresponding MIR S file is used with the desired MIR L file As an example Version 1 3 September 13 2007 79 In the irclog the appropriate corresponding filter pair can be selected using the the same file number F40813 As raw data all these fi
78. en data set As a check the files and their counterpart corresponding MIR L images listed in the pair file e g pair0001_S7 list_long should be examined to see if they have the appropriate FILTER keywords in their header Chapter 6 Spectroscopy pipeline cookbook The spectroscopy pipeline is being developed and maintained by the IRC spectroscopy data reduction team It is mostly written in IDL and uses IDL ASTRO library maintained at the GSFC among others Although it is developed under Linux environment it may be portable to other platform after some modifications though we have no plan to increase the exportable platforms by ourselves The toolkit also uses the ds9 FITS viewer for reviewing the images and the xpa package for communication between the ds9 and the IDL main program The spectroscopy data reduction requires calibration database FITS images and ascii tables distributed and updated also by the IRC spectroscopy data reduction team The database is based on observations of calibration objects and calibration frames taken during PV DT phase observations as well as pre launch calibration experiments in our laboratory Therefore observers do not have to make their own calibration observations 6 1 General overview of the pipeline processing The main pipeline processing consists of several well defined steps which are explained in the following sections 6 1 1 Dark subtraction Scaled super dark images are subtracted f
79. erac no Run pipeline interactivelu deltemp yes Delete intermediate files rejecti sigclip rej in coaddiminmaxlccdcliplerrejectlsigcliplavs verbose no Print verbose progress messages mode al list Figure 5 9 8 Screen for parameter configuration for pipeline The pipeline has several parameters e irccons string default constants database NEVER CHANGE THIS This is the irc constant file name e com mod string average median default median Method of combination of frame images average median e com area int 1 or 2 default 2 Coadded image area of stacked frames used for extraction see Fig 5 9 9 NOTE although strictly speaking a Blue Box process but is run from the Green Box Pipeline script THIS PARAMETER IS NOT SUPPORTED YET AND YOU WILL GET THE WHOLE AREA IMAGE EVEN IF YOU CHOOSE 1 TO GET THE COMMON AREA 1 Only the common area red region is extracted 2 The whole area areas within green dashed line is extracted e sky are int 1 or 2 default 2 Area of sky to be used for statistics to adjust sky level before coadding frames 1 Only the common area is used 62 IRC Data User Manual 2 The whole area is used det sig real min 1 0 max 1000 0 default 4 0 Source detection threshold in sigmas sig_rej real min 0 0 max 100 0 default 3 0 Rejection unit in sigmas max itr int min 1 max 10 default 10 Maximum number
80. ere we consider source masks for better determination of the local background Background is measured and subtracted off by assuming constant background value across the images for all dispersers 6 1 14 Spectral tilt correction Although we do not reform image shape along wavelength direction during the course of wave length calibration because of simple linear grism dispersion we do reform along X or space direction to correct spectral tilt The tilt occurs due to misalignment of grism insertion direction with respect to chip orientation The tilt is notably seen in NG image in which as large as AX 2 pixel shift is observed over the longer dispersed image length AY 250 pixel Similar tilt is also observed for other prism grisms NP SG1 SG2 and LG2 although the tilt looks very small due to short dispersed image length However tilt correction will be made for spectra of all types of dispersers 6 1 15 Spectral response calibration The spectral response calibration R X depends exclusively on the wavelength Therefore the response calibration table is a 1D vector lambda vs ADU s mJy This means that response variations of individual pixels have been removed beforehand by flat fielding processes Being different from spatial flat correction spectral response varies significantly along wavelength due to change in quantum efficiency of chip and change in optical transmission along the camera optical trains including the disperser
81. ersion 1 3 September 13 2007 33 HD38861 05423256 7022555 KOI 6 780 2007 04 02 06 48 19 HD38993 05431866 7027254 K1 5III 5 478 2007 04 02 05 08 54 HD39980 05495921 6941060 K2III 5 409 2006 10 16 22 35 50 HD270186 05501123 6934296 KOIII 7 774 2006 10 18 19 14 15 The raw data were reduced with the IRC imaging pipeline The pipeline produces one coadded image for each band exposure configuration using median as the combine mode Each configuration image corresponds to the exposure unit listed in Table 4 6 4 The actual exposure time is the unit number x the unit exposure time tunit which is approximately 0 5844 s Table 4 6 4 Exposure time of each band exposure configuration Band Exposure unit number NIR short 8 NIR long 76 112 IRC05 MIR short 1 MIR long 28 One unit time corresponds to about 0 5844 s An aperture photometry IRAF phot was performed for each standard star The radius of the aperture adopted is 10 pixels for NIR band and 7 5 pixels for MIR S and MIR L band respectively We determined the sky value in an annulus outside the aperture with a width of 5 pixels A simple mean of the measured ADU values is used if the star was observed more than once 4 6 2 Estimation of the in band flux e Estimated flux density The in band flux density of each band i was calculated by the following equation e Riv f v dv quoted __ hv vi molar B RO 4 6 1 Vis V hv Or Aie Ry A f A d quote
82. ersions of the toolkit will be available to users throughout the course of the mission On receiving a new version of toolkit package e g ircY YMMDD tgz the following commands are required to be input e mv ircYYMMDD tgz where you installed previous version e cd where you installed previous version e tar xvzf ircY YMMDD tgz These commands will OVERWRITE any old files directories Then follow the original proce dure described in 1 4 5 and 1 4 7 Finally please type following commands when you use a new version of the toolkit for the first time e launch IRAF e load the ircpackage by typing irc e type unlearn_all e load the irc_tool package by typing irc_tool e type unlearn coaddLusingS 5 6 Setting up your toolkit environment and running the pipeline The steps to reduce the raw data are outlined below 5 6 1 Creating the directory structure The toolkit assumes the following directory structure see Fig 5 6 1 anyname rawdata must be this name anyname your working directory e g data rawdata data working The toolkit should be run in the working directory Thus you may create sets of these directory structures for different sets of IRC data Version 1 3 September 13 2007 55 polairetyita pud fhome yita polairetyita tar xzf sample_data tgz polairetyita cd sample data polairetyita Z mkdir yita polairezyita ls rawdata yita polairetyita
83. f this option 100 IRC Data User Manual re draws the images on ds9 with updated zero th order light marks Repeat this command until you get satisfactory result Then one can use the spectral plot tool for reviewing spectra with updated wavelength and spectral response calibration See below for the plotting tools Command line syntax change_wave_offset lt wave_offset gt specimage n_sff source_ table specimagen_wc wave array specimage bg where lt wave_offset gt is a pixel number e g 1 0 0 5 0 5 to be shifted 6 5 4 Spectral plotting tool The toolkit s spectral plotting tool can handle many IRC spectroscopy specific features and we recommend to use it for creating spectral plots These are some of the functionalities of the tool e source masks can be considered see also no mask option e error bars statistical plus systematical ones can be plotted e wavelength offset along wavelength or Y axis due to rounding effect of coordinates when extracting the spectroscopy images is considered e simple image filters for removing isolated spike pixels can be applied e aperture stacking along spatial direction and smoothing along wavelength direction can be applied for higher S N spectra e shifting aperture positions along spatial direction X axis can be made Command line syntax a command in a single line plot spec with image wave array specimage n wc mask specimage n source table source id o
84. fective bandwidth Dispersion power of NP depends on wavelength The quoted value corresponds to 3 5m Renamed from L20W No change of the wavelength profile itself SOO Nara NEA 6 7 8 Table 2 0 2 General characteristics of the IRC focal plane arrays j Dark current e s 0 2 26 same as MIR S a Estimated by Fowler 1 sampling b Estimated by Fowler 4 sampling c Pre launch performance the beam splitter The brightness of the ghost is about 0 8 of the true source for S7 and S9W and 3 8 for S11 band IRC Data User Manual Folding mirror Detector module Si Ge Si Telescope axis 50mm Figure 2 1 3 Side view of the NIR channel Detector module Filter wheel Beam splitter Ge 7 1 N 3 ES i f L f E Jd 3 rr 50mm
85. for both short and long exposure frames even with this flag set This flag could be useful for measuring source positions of bright and saturated objects in the long exposure frame savefile set this option to a named variable that will contain the file name of the IDL save file section 6 5 5 The simplest way to restore the IDL save file is to run irc specred with the savefile savefile option and issue restore savefile command after irc_specred finished and before e g plotting the spectra Version 1 3 September 13 2007 95 Outputs e Whole image products Processed and stacked reference image and corresponding mask and residual refer ence image x mask images Dark subtraction flat fielding and background sub traction are made lt targetid gt lt targetsubid gt lt filter_spec gt refimage_bg fits lt targetid gt lt targetsubid gt lt filter_spec gt refimage_mask fits lt targetid gt lt targetsubid gt lt filter_spec gt residual_refimage_bg fits Processed and stacked spectroscopy image and corresponding mask and residual spectroscopy x mask images Dark subtraction flat fielding and background sub traction are made lt targetid gt lt targetsubid gt lt filter_spec gt specimage_bg fits lt targetid gt lt targetsubid gt lt filter_spec gt specimage mask fits lt targetid gt lt targetsubid gt lt filter_spec gt residual_specimage_bg fits These images are in 3D and third Z dimension is
86. for short Z 0 and long Z 1 exposure frames NaN Not a Number represents masked pixel area e Extracted image products Extracted reference images for individual targets and corresponding mask images lt targetid gt lt targetsubid gt lt filter_spec gt refimage_indiv fits lt targetid gt lt targetsubid gt lt filter_spec gt refimage_mask_indiv fits Extracted spectroscopy images for individual targets and corresponding mask im ages There are two kinds of images One is wavelength calibrated WC image for which flat color term correction and wavelength calibration were applied The other are flux calibrated _FC images for which flux calibration was applied as well as flat color term correction and wavelength calibration lt targetid gt lt targetsubid gt lt filter_spec gt specimage_wc_indiv fits lt targetid gt lt targetsubid gt lt filter_spec gt specimage_mask_indiv fits lt targetid gt lt targetsubid gt lt filter_spec gt specimage_fc_indiv fits Note that our spectral plotting tool see below uses _WC image as an input not the FC image since the tool does on the fly flux calibration after considering various plotting conditions options Since the default FC images does not care these we strongly recommend to use our plotting tool to review the spectra These images are in 3D and third Z dimension is for source id NaN Not a Number represents masked pixel area e Others Target
87. g the pipeline processing In the first pass of the source detection automatic source detection sub program works even if the target table is supplied in the pipeline command line The X offset is measured with the tentatively detected sources at this stage In the second pass measured X offset as well as Y offset found in adjusting the wavelength zero reference point see the next section between spectroscopy and reference images is taken into account in extracting the 2D spectra Only the sources specified in the source table or sources detected with the sub program will be processed extracted in the second pass This two stage process ensures good source masking for better sky subtraction and good offset measurement with brighter and larger number of sources even for the cases in which observer is just interested in small number of faint sources Measurement of the center position of the 2D spectra can be made easily for NP NG and SG1 spectra for which larger number of brighter sources can be detected even at blank sky region Therefore the toolkit measures the X offset at NP and NG for correcting both NP and NG and at SG1 for correcting SG1 SG2 and LG2 Therefore user must start pipeline processing for NP NG first then SG1 and finally SG2 and LG2 See section 6 4 1 for more explanations 6 1 11 Wavelength calibration In the case of grism spectra the dispersion equation is almost linear theoretically and a linear equation for expres
88. greenbox pipeline redbox Figure 5 6 3 Start up screen for the IRC package showing the available modules and tools The start up splash screen gives the current version of the toolkit and current versions of flat dark linearity distortion files etc Version 1 3 September 13 2007 57 5 7 The pre pipeline processor Red Box Note that prepipeline processor is a one time only process Do not run prepipeline twice 5 7 1 Configuration Before running prepipeline you can configure the parameter for prepipeline by typing epar prepipeline This displays the parameter screen shown in Figure 5 7 4 Image Reduction and Analysis Facility PACKAGE ire TASK prepipeline verbose B no Print verbose progress messages mode al Het E HEE for HELP Figure 5 7 4 parameter screen for the Red Box showing the available parameters Please do not change the parameter irccons Prepipeline has only one changeable param eter verbose If you want to see verbose progress messages change it to yes The default is no 5 7 2 Running the prepipeline processor The prepipeline processor consists of two functions e redbox ircslice slice IRC 3D images into usual 2D ones e redbox mkirclog making the observing log file irclog The prepipeline processor is run by entering prepipeline at the IRAF command prompt Alternatively the individual Red Box modules can be run by entering redbox at the IRAF
89. hat s1CcnDwaF23342 8004 fits becomes ms1CcnDwaF23342_S004 fits Flat Fielding greenbox flat The flat fielding is made using the imaging super flats described in section 4 2 Note that there was a noticeable pattern in the lower right corner of MIR S flat images A similar pattern also appeared in NIR flat images However these patterns suddenly dissapeared on the 8th January 2007 Thus the pipeline now incorporates two sets of super flats one for the data taken before 8th January 2007 and the other for the data taken after the 8th January 2007 By default the toolkit automatically selects flat images appro priate for the observation data by reading the observation dates in the FITS header However it was noticed that the pattern also slightly changes with time and in some cases the pipeline default flats can not wipe out the pattern completely In this case the data may require flat images made by using the AKARI NEP survey data taken close to the observing date In the near future we plan to attach the most suitable flat im ages to every distributed data set These flats can then be selected by putting them in where you installed irc lib flat user and deactivating the default_flat parameter After the flat fielding process a f will be added such that ms1CcnDwaF23342 8004 fits becomes fms1lCcnDwaF23342_S004 fits Aspect ratio resampling greenbox aspect_ratio This is a distortion correction The module the corrects as
90. he super dark is used to get a high S N Due to various reasons 1st frame effect and so on the dark level may vary during a pointing Therefore we monitor the slit area as a reference of the dark level and we shift the super dark and or self dark by adding subtraction a constant to them and then subtracting this from each image to account for the changes in the dark level After the dark subtraction a D will be added such that waF23342_S004 fits becomes DwaF23342 8004 fits Normalization greenbox ircnorm To reduce the readout noise of the detectors the IRC data is read with the Fowler sampling method non destructively reading and summing the array multiple times determined by the Fowler number then dividing this resulting sum by the Fowler number However when the data is transmitted to the ground for the purpose of data compression the least significant bit is dropped The normalization process accounts for this bit loss by correcting the data value by Auncorrected gbitehift Fowler number 5 9 2 Acorrected Where the bitshift represents the number of bit shifts emloyed Version 1 3 September 13 2007 67 After the normalization process a n will be added such that DwaF23342 8004 fits becomes nDwaF23342_S004 fits 5 Scattered light pattern rejection greenbox scatt light Due to the scattering of light from the edges of the detectors lattice shaped sky patterns are present in the imaging area The pattern
91. ich utilizes the MIR S shift amp rotation for the MIR L channel This function is especially useful when bluebox calcshift fails to calculate the xy shift and rotation angle for the MIR L channel because there are few reference stars detected The coaddLusingS has several parameters but they are identical to those in the pipeline except the s_list s_list string default pair0005_S9W list_long The name of MIR S pair list for which the calcshit has successfully matched frames 74 IRC Data User Manual Figure 5 11 15 Present configuration of the Blux Box post Pipeline Processor Version 1 3 September 13 2007 75 5 12 Toolkit structure ro IRC imaging data reduction pipeline i er dark Paes ate aot05 Hong fits short fits distortion distortion database dat flat fits slit mask slit mask fits constants database setenv dat setpath dat setpath pl welcomeorg perl pkg redbox greenbox bluebox irc tool 5 13 Working on the output Users will receive both the original raw data and processed data up to but not including the co add stage for the toolkit The data is is basic processed with the default parameters of the toolki
92. ience and the users are strongly encouraged to run the correponding pipelines in order to obtain the best out of the raw data 3 3 1 IRC imaging IRCO5 IRCO02 and IRCO03 As explained in the README file distributed with the data in the archive the irc ql subdirectory contains together with a set of log files a main quick look fits image per detector with the following naming convention Its name is fdmslnDwaFV V Vxxxxxx NSL yyy fits The firts prefixes refer to the performed steps during the processing DARK correction D linearity In distortion correction d and flatfiedling f NSL refer to each detector The irclog file contains a brief description of each file from the data reduction pipeline 3 3 2 IRC spectroscopy IRC04 The quick look products distributed in the case of spectroscopy observations IRC04 are the following e TARGETID SUBID FILTER_DISPERSER refimage_bg fits dark linearity and flatfiled corrected reference image e TARGETID SUBID FILTER DISPERSER specimage bg fits dark linearity and flatfield corrected spectroscopy image This quick look image is usable only for slit less spec troscopy For slit spectroscopy further processing with the spectroscopy pipeline is needed Auxiliary files are described in the README file distributed in the archive 20 IRC Data User Manual 3 4 IRCOA image orientation and dispersion direction Since raw NIR images are rotated for technical reasons of data handling in
93. ific purpose They may be used to check the saturation in long exposure frame data Specifically the IRC MIR pointing is composed of several exposure cycles filter changes and dithers Each exposure cycle is made up of 1 short and 3 long frames For AOT IRCO02 IRC03 exposures are always separated by dithers etc and the problem does not arise however for IRC05 IRC00 IRC04 they are not necessarily sepa rated The problem is occurring on contiguous exposure cycles i e nothing in between For example short1 long1 long1 long1 short2 long2 long2 long2 Version 1 3 September 13 2007 77 The bad frames are the short2 frames i e those following a previous contiguous exposure cycle This effect extends to the short frame of the final dark even This problem may be critical for spectroscopy of bright sources with IRC04 however for IRC00 amp IRCO5 users the effect may not prove critical since users may disregard the short frame since these AOTs are intended for deep imaging the processing of the short frames may be disabled from the Green Box pipeline parameter list a NEP 2006 5 17 G Y d x Time In one pointed observation min Figure 5 14 16 Change in the background level in a pointing observation 5 14 2 Toolkit Limitations and Caveats e MIR S flat changed As informed via the the AKARI user support web page the flat of MIR S changed early this year 2007 In response new flat data for MIR S ha
94. ig a script e cd where you want to install irc perl e perlpath sh which will create a directory temp containing perl files whose first lines are modified as above Then you can overwrite these new perl files by typing e mv temp pl Those who do not have perl in usr bin perl please modify perlpath sh and indicate the location of your perl 5 4 8 Add IRC entry to IRAF e If you know root password and you installed IRAF package into for example iraf cd iraf iraf unix hlib edit the file extern pkg Please add the following 2 lines to A iraf iraf unix hlib extern pkg reset irc where you want to install irc task irc pkg irc irc cl DO NOT FORGET THE BEFORE THE IRC PKG DO NOT FORGET THE LAST SLASH e If you don t know root password then you should launch iraf from your home directory everytime you want to use irc package cd change directory to your home directory mkiraf only when using IRAF for the very first time 54 IRC Data User Manual This will create the login cl file in your home directory Then edit login cl and add 2 lines reset irc where you want to install irc task irc pkg irc irc cl DO NOT FORGET THE BEFORE THE IRC PKG DO NOT FORGET THE LAST SLASH the toolkit is now installed and is ready to be run 5 5 How to UPGRADE the version of IRC imaging toolkit Upgrades to the IRC imaging toolkit is a progressive and ongoing process and new v
95. ight nor a series of narrow band filters covering the passband of the disperser Because of these difficulties we will make a flat calibration in two steps starting with the monochromatic correction To correct for the pixel to pixel variation of the monochromatic response response or flat that does not depend on wavelength the whole image is divided by the super flat The super flats are made by combining a larger number of background spectroscopy images Note that the background sky image is not flat along Y or the wavelength axis i e there are several spectral features on the super flats due to e g contributions of zero th or higher order light in cases of grism spectroscopy images Although the features should be removed to correct for the pixel to pixel variation of the response we first divide the dark subtracted images by the super flat with the spectral features As shown below in equation 6 1 1 this process makes the background flat facilitating the sky subtraction Assume that the dark subtracted images are represented by the following equation obs sky x Fi z y x spectral f eature x y obj A x Folz y A x R X 6 1 1 where Fj x y is the flat in which the background spectral features are not present Fi z y x spectral_feature x y is the super flat F x y A is the wavelength sensitive flat see below for color term correction and R A is the spectral response Since the super flat is only a functi
96. in the header Note that it is possible to shorten the time to download the relevant 2MASS catalog by changing the download mirror in the toolkit file irc perl get2mass pl Find the line that determines the data server where the 2MASS catalog will be downloaded The default is Tokyo Japan To choose the nearest server un comment the relevant URL and comment out TokyoJapan Note for multiple pointings and deep imaging data many pointings of data may have to be co added In such cases all the images must be rotated and the toolkit will have to take care of this before co adding The final result is the basic science grade images see Figure 5 10 14 5 10 2 Log files produced from the toolkit In addition there are various log files produced by the toolkit which are described below e coo Files The coo files e g SfdmslnDwaF23343_N002 fits coo 1 are the results for robust source extraction that extracts x and y coordinates amp robust photometry One coo file is created for each individual exposure frame These files can then be used to calculate the shift and rotation of frames that are being co added by matching bright stars in the extracted source lists Pair Files The pair files collect together and list all long and short frames that point the same area of sky Taking the N3 band as an example the following files may be seen pair0001N3 list pair0001 N3 list short pair0001 N3 list long pair0001 N3 list long shift pair00
97. information table a table of target information such as positions after source position tuning flux and size Gaussian FWHM 96 IRC Data User Manual lt targetid gt lt targetsubid gt lt filter_spec gt target_table tbl Its format is the following ID image_x image_y image_mask_dx image_mask_dy spec_x spec_y spec_mask_dx spec_mask_dy flux image_ FWHM spec bgnoise ADU spec_x_pos spec_x_FWHM bad sourcepos flag Target table If the source detection sub program is used within the pipeline a target table will be written The file format is similar to the input target table When source position tuning option is on this file contains the updated source coordinates lt targetid gt lt targetsubid gt lt filter_spec gt source_table tbl IDL save file The output of the pipeline will be also be stored in the IDL save file which is basically a dump file of the IDL memory image at the end of the pipeline processing See also 6 5 5 to see how to work on the save file lt targetid gt lt targetsubid gt lt filter_spec gt IRC_SPECRED_OUT sav Log of the toolkit processing a copy of the irc_specred logger is saved as an ascii file lt targetid gt lt targetsubid gt lt filter_spec gt log DS9 region files DS9 region files that have been used to locate targets on reference and spectroscopy images on ds9 lt targetid gt lt targetsubid gt lt filter_spec gt refimage reg lt targetid gt lt targetsubid gt
98. ing observation a dark frame is taken at the beginning pre dark and the end of the operation for both the NIR and MIR channels Those who want to use these pre dark images in preference to the super dark images should turn on the self dark parameter and the pipeline processor will make dark images by averaging over the 3 images of MIR S and MIR L long exposure dark frames Hereafter we refer to this dark images as the selfdark image However the super dark will still be used for all short exposure frames and also the NIR long exposure frames even if the selfdark parameter is on Although the S N of the self dark images may be lower than that of the super dark image since the MIR images are usually sky noise limited so the S N of the dark should not significantly change the results unless dealing with very faint objects the self dark may be able to remove hot pixels more efficiently than using the super dark especially for the MIR S and MIR L because the number of hotpixels is increasing since the launch as mentioned in section 4 1 and shown in Figure 4 1 1 There are at least two known flaws in the current super dark a It does not correct for the increased dark signal due to the IRC all sky survey obser vations in MIR S L This effect corresponds to the belt like pattern in the horizontal direction b It does not accommodate the increasing number of hot pixels It may be better to use the pre and or post dark frames T
99. ion we shift both the wavelength array and spectral response curve both of which should show rather smooth change along Y or wavelength axis and not perform sub pixel shifting of the images As a result since object positions change slightly among different pointing observations wavelength at the same Y pixel of the extracted 2D spectra or the wavelength array also changes with different pointing observations 6 1 12 Flat color term correction The presence of significant color variation in the flat images can be found in the ratio images of the broad band flats e g S7 flat 15 flat Therefore although monochromatic flat fielding can flat the background the object spectrum is affected by the color term of the sensitivity The correction for this could be done after extracting the 2D spectra for each target and applying the wavelength calibration However since we have only two broad band filters for MIR S MIR L S7 and 15 for MIR S and L15 and L24 for MIR L and three for NIR N2 N3 and N4 we can derive only global trends of wavelength dependence of the flats Note that the wide band filters S9W and L18W are not suitable for deriving the color dependence of the flat within the spectral coverage of the channel Two broad band flat images are interpolated to estimate the flats for a given wavelength in the following way F z y A F z y 2 A F x y 1 22 A1 x A z A1 F x y M 6 1 4 where Ag and A are effe
100. ith the point source aperture Np the super flats are typical for conventional slit spectroscopy These flats are also made by combining a large number of blank sky spectra At present there are only two slit flats for NG at Np and Np at Ns The others NG at Nh SG1 2 at Ns and LG2 at Ls are in preparation 4 3 Instrument linearity Detector linearity was measured in the laboratory before launch and later in flight Measure ments wre made with a calibration lamp that illuminates the detector and an increasing range of integration times At first it was assumed that the detecter behaves linearly below 5000A DU and was fitted with a linear curve by a least squares method Fig 4 3 3 shows the raw signal in ADU versus the fit in ADU signal expected if the detectors were to behave linearly The deviation of the raw signal from the linear expectation was then calculated and correction equa tions by fitting polynomials up to 7th order were applied The red lines in Fig 4 3 3 represent the calculated correction equations Fig 4 3 4 shows how the calculated correction equations work After the correction the error from the ideal linear curve is better than 5 at 12000 20000 and 20000 ADU after ircnorm for NIR MIR S and MIR L respectively Note that from Fig 4 3 3 the physical detector saturation occurs around 12500 33000 and 33000 ADU after ircnorm for NIR MIR S and MIR L respectively No reports have been given against its reliabili
101. l these images contaminate other spectra aligned along with the columns of pixels where the objects exit These contaminations can be ignored in most cases and cautions are needed actually only around very bright sources For the NG spectroscopy with point source aperture Np there is little chance for this kind of contamination to occur since the aperture size is much smaller than the size of the dispersed spectroscopy image along the Y or the dispersion direction Spectral smearing for extended sources In the slit less spectroscopy mode the spectral resolution is determined by the size of the sources or the PSF structure plus satellite attitude drift Thus the spectral resolution is lower for extended objects If an extended object shows resolved structures the interpretation of the 2D spectra becomes very difficult because of the convolution of the spectrum over the object spatial structure Background spectral features Although the background sky is essentially flat in the imaging mode it is not true for spectroscopy images Note that these features would not be observed with the point source aperture Np and other narrow slits since the aperture size is much smaller than the size of the dispersed spectroscopy image along the Y or the dispersion direction Faint background areas near the Y 0 or Y Ymax edge of FOV For the slit area the background is dominated by the zodiacal light The spectral response curve shows a simple patter
102. l be shown on a ds9 for reviewing To remove some sub frames use the Frame gt show hide frames pull down menu of ds9 to hide the frame s you need to remove The toolkit checks the hidden sub frames and remove them from the internal sub frame combination list e Retrying automatic source detection with modified detection parameters If you use au tomatic source detection sub program within the pipeline you can retry source detection after tweaking the detection parameters You will see the reference image on the ds9 with detected sources marked and you are asked to answer if the detection is satisfactory or not If you answer no then another dialog with several detection parameter entries pops up You can edit the parameters and press OK to re find the sources You can repeat this detection process as many times as you like until you get satisfactory source list The sub program uses daofind algorithm daofind parameters in more detail threshold Threshold intensity for a point source should generally be 3 or 4 sigma above background fwhm FWHM to be used in the convolve filter sharplim 2 element vector giving low and high cutoff for the sharpness statistic Default 0 2 1 0 Change this default only if the stars have significantly larger or smaller concentration than a Gaussian roundlim 2 element vector giving low and high cutoff for the roundness statistic Default 1 0 1 0 Change this default
103. lename ps of the plot in the working output directory under irc_specred_out png lt filename gt creates a png image filename png of the plot in the working output directory ascii lt filename gt creates an ascii file A vs flux and flux errors filename spc of the plot in the working output directory User may use gnuplot program to review the ascii output with the command of plot filename spc with yerrorbar e with_image tvbottom lt tvbottom gt tvtop lt tvtop gt The 2D image will be shown on TV an IDL graphic window below the spectrum plot Extraction box for plotting 1D spectrum is overlaid The default is with_image 0 i e no TV display The top and bottom ADU counts for displaying image can be specified with tvtop and tvbottom These options are active only when with_image is set e diffuse For slit spectroscopy data spectra will be shown in units of MJy str rather than mJy with this option As of 2007 summer diffuse flux calibration is not yet fully established e no_aperture_correction Aperture correction for point like sources will be applied au tomatically within the plotting tool The aperture size set by nsum is taken into account to find the correction factor When the no_aperture_correction option is set the cor rection will be disabled To perform good aperture correction the aperture is centered on the source and nsum should be 3 or more Note that the aperture can be shifted by
104. les are packed into a single fits file Pre pipeline slices the MIR fits file into the 8 4MIRS 4MIRL files F40813_L001 NGC104 256 L18W 5 293348 72 44307 IRCO3 1 5020012 1 F40813_L002 NGC104 256 L18W 5 293348 72 44307 IRCO3 2 5020012 1 F40813_L003 NGC104 256 L18W 5 293348 72 44307 IRCO3 3 5020012 1 F40813_L004 NGC104 256 L18W 5 293348 72 44307 IRCO3 4 5020012 1 F40813_S001 NGC104 256 S7 5 293348 72 44307 IRCO3 1 5020012 1 F40813 8002 NGC104 256 87 5 293348 72 44307 IRCO3 2 5020012 1 F40813 8003 NGC104 256 87 5 293348 72 44307 IRCO3 3 5020012 1 F40813 8004 NGC104 256 S7 5 293348 72 44307 IRCO3 4 5020012 1 In this case using pair0002 S7 list long for the reference the coaddLusingS function will make the L18W_long image e coaddLusingS function parameter list In a similar manner to the general pipeline the parameters for the coaddLusingS function may be set via the epar iraf command Please note that in some instances some of the parameters may not be set as desired On executing the function you will be asked to confirm the parameters and users should check at run time that indeed the parameters have the required values and edit them if neces sary In addition note that the option to subtract median filtered sky is not included in the parameter list i e cannot be set using epar but appears during the parameter confirmation at run time e coaddLusingS function coadd failure log Please note that
105. lf of the wavelength range of the cameras and a wide filter that overlaps the two narrow filters The S9W and L18W bands are used for the all sky survey observations The pixel scale and the imaging area of the focal plane arrays are summarized in Table 2 0 2 The imaging area is the rectangular area excluding the slit region The dispersion elements of the IRC are set into the filter wheel so that all the light in the FoV is dispersed A spectrum is obtained in the direction parallel to the scan path in scan direction Slits are provided for each camera Figure 2 0 2 in order to avoid contamination by nearby sources diffuse radiation The two dispersion elements of the NIR camera provide different spectral resolutions over a similar wavelength range In the MIR S and MIR L each dispersion element covers about half of the camera s wavelength range However unfortunately one of the elements of MIR L LG1 was degradated during the ground tests Thus only LG2 will be used for observations resulting in a gap in the wavelength range corresponding to the LG1 element The slits are primarily designed for extended sources and it should not be assumed that they can be used to guide a point source into the slit except for the NIR camera which has an aperture for point sources The current design is the following e The slit for the NIR camera consists of three parts of different widths The left most closest to the imaging area has a 5 arcsec width
106. ll sky survey data and the MSX data 10jum The 204m bands calibration will be more difficult because of no good reference data at these wavelengths The absolute calibration is still being in progress with new observations and will be updated 4 8 Color correction s mentioned in previous section the quoted values of the IRC flux calibration assume fy A For cases of other incident spetra the color correction is required The quoted value at a reference frequency is given by fal oed n urs where R is the response in units of electron per photon The correction factor K is calculated as flv y LY ay et E fal actual _ m fi actual K A i actual Tuy Riv 8 Tee dme 5 d 36 IRC Data User Manual Thus K is given by We calculate the correction factors for the following incident spetrum cases f A To ae 4 8 3 with a 0 1 2 The results are shown in the following tables The reference wavelengths are fixed as A1 2 4um Ag 3 2um Aa 4 lum A4 7 0um Ag 9 0um Ag 11 0um Az 15 0um Ag 18 0um Ag 24 0um Version 1 3 September 13 2007 Table 4 8 8 Color Correction factors for NIR channel Black Body a 0 Intrinsic NIR N2 NIR N3 NIR N4 Temperature K K 2 4um K 3 2um K 4 1um 40 50 60 70 80 90 396 131 100 164 506 110 z 81 786 120 873 058 46 429 130
107. lt filter_spec gt specimage reg The region files and saved FITS images of whole image products can be used to review the targets on ds9 manually e Pipeline work files The following files will be created by the pipeline The files will be overwritten without notice User are asked not to delete any of these files NP SHIFT XY dat for registering NP SG1 SG2 LG2 sub frames in AOT04a NG SHIFT XY dat for registering NP SG1 SG2 LG2 sub frames in AOT04b S9W SHIFT XY dat for registering S9W and L18W sub frames NP SPECBOX SHIFT X dat for shifting SG2 LG2 along X in AOTO4a NP SPECBOX SHIFT Y dat for shifting the wavelength zero reference point for NP SG1 8G2 LG2 in AOT04a NG SPECBOX SHIFT X dat for shifting SG2 LG2 along X in AOT04b NG SPECBOX SHIFT Y dat for shifting the wavelength zero reference points of NG SG1 SG2 LG2 in AOTOAb 6 4 3 Summary of interactive operations within the pipeline Although the pipeline program works as a pipeline i e without interactive operation by uses some operations in the irc specred do require interactive operation Version 1 3 September 13 2007 97 e Removing one or more sub frames before combining the sub frames In some sub frames one may find severe damage by cosmic ray shower nearby satellite passing in front of the telescope etc To check the sub frames and remove from the sub frame combination list all the dark subtracted flat fielded sky subtracted sub frames wil
108. n and dithering Therefore the very first short frame in all modes and the short frame after the image frame in IRC04 should be fine since they follow the filter operation These short exposure frames can be used for checking the saturation but should not be used for scientific purpose Figure 4 12 15 MIR array vertical stripes 4 13 General concerns on slit less spectroscopy data This section is dealing mainly with problems related to the slit less spectroscopy data Issues for the slit spectroscopy matters are described separately e Wavelength reference point In the slit less spectroscopy mode the wavelength reference point depends on the location of objects within the FOV Therefore the determination of the source posistions on the reference image is very important Errors in the source positions leads to errors in the asigned wavelengths and hence the flux calibration 48 IRC Data User Manual e Contamination by nearby sources Spectra of more than two objects aligned along with the Y axis could overlap on the same pixels with different wavelengths It is impossible to separate the spectral overlap on the observed image without knowing the spectrum of each object Contamination by zero th 2nd order light For the grisms images of the zero th and the 2nd order light can be seen as well as the 1st order image our prime target for data reduction although the efficiency for the zero th and the 2nd order light is very low Al
109. n in the hot pixel mask database you will find outliers hot or cold pixels even after sub frame combination Version 1 3 September 13 2007 101 e space shift shift of the plot extracting box along X spatial axis in pixel Default is zero Although the pipeline adjusts the X coordinates of spectrum extraction box by measuring positions of spectra one may find remaining X offset in some cases Change this offset to find peak position of the flux If one changes nsum and space shift issue the display command to see the modified spectrum extraction boxes graphically on the spectroscopy images over the ds9 show aperture on ds9 specimage bg source table space_shift lt space_shift gt nsum lt nsum gt a command in a single line e no_mask by default plots are shown after applying spectral overlapping masks by nearby sources The no mask option disables this masking functionality and plot spectra regard less of the possible source overlapping Since masks are created without examining the source brightness and or spectral shape e g line emitters continuum emitters with break etc one may often find the situation where no significant change in plots is found when comparing plots with and without the masking If this is the case you can just disable the masking by setting no mask option e ps png ascii Plots will be recorded on the files not on the IDL plot window ps lt filename gt creates a postscript file fi
110. n the LMC have been surveyed by one of the AKARI s Large Area Survey programs LSLMC The standard stars in the LMC the so called SAGE standards were also established by M Cohen for the Spizter SAGE program We list observed SAGE standards in Table 4 6 3 This survey was done with IRC AOT02 and only N3 7 S11 L15 and L24 images are available Table 4 6 2 Observed NEP and SEP standard stars Star 2MASS ID Sp K Obs Date for NIR MIR S HD42525 06060937 6602227 A0V 5 751 2006 04 22 04 08 28 NPM1p60_0581 17245227 6025508 A1V 9 645 2007 02 13 23 48 41 1757132 17571324 6703409 A3V 11 155 2006 04 26 22 49 47 KFO3T1 17574394 6626553 KOLI 9 923 2006 08 03 19 30 00 KF03T2 17575147 6631034 K1 5III 8 963 2006 08 03 19 30 00 KFO6T1 17575849 6652293 K1 5III 10 872 2006 04 26 19 31 48 KFO6T2 17583798 6646522 KI1 5III 11 149 2006 04 26 19 31 48 KFO06T3 17585021 6649406 KIII 10 348 2006 04 26 19 31 48 KFO6T3 2 2006 07 06 19 15 10 KF03T3 175901144 6633262 K1 5III 10 925 2006 08 11 19 08 13 KFO3T4 17590395 6630593 KIIII 10 091 2006 08 11 19 08 13 KFO9T1 17592304 6602561 KOIII 8 114 2007 04 15 02 03 34 KFO9T1 2 2007 06 02 01 28 46 KFO6T4 17592606 6654581 KOIII 11 240 2006 07 06 19 15 10 KFOIT4 18040314 6654459 K1 5III 8 067 2006 06 24 19 55 00 KFO1T4 2 2007 04 15 00 24 01 KFOIT4 3 2007 06 01 00 37 25 KFO1T5 18040388 6655437 KIII 11 072 2006 06 24 19 55 00 KFO1T5 2 2007 04 15 00 24 01 KFOIT5 3 2007 06 01 00 37 25 for MIR L HD42525 06060937
111. n with the decreasing sensitivity at the highest and lowest wavelength ends of the disperser s spectral coverage Thus one will see a background pattern which becomes faint at the top and bottom of the image in untis of ADU For the slit less area both the wavelength and the spatial axis Y go along with the same direction Y and thus the observed background image is a spatially convolved background spectrum Since the length of the spectrum along Y 50 70 pix is much smaller than the aperture size 256 or 412 pix the resultant background spectrum is almost constant across Y after being convolved spatially by the large aperture The regions around Y 0 and Y Yyax are exceptions where the edge of the aperture prevents full convolution of the background spectrum along the Y direction and hence the background signal becomes faint near Y 0 and Y Y max is seen Note that this kind of pattern at the very Y edge of the FOV does occur only for the background light but not for object spectra Jump of the background ridge near the center of FOV Another background feature is a ridge seen near the Y center stretching along the X axis seen in grism spectroscopy images This is caused by the zero th order light of the grisms Since the zero th order light image forms at an offset position from the reference image only about a half of the FOV suffers from its contamination This jump of the background flux level in ADU is about 2 396 of the total backgroun
112. nbox wraparound Due to data volume constraints the IRC compresses the data when it is transmitted to the ground by discarding the data sign bit This will cause obvious ambiguities in the corresponding flux values Since the IRC data structure uses the 2 s complement method to represent negative numbers discarding the sign bit most significant bit will result Version 1 3 September 13 2007 65 GREEN BOX Pipeline Processor Figure 5 9 10 Present configuration of the Green Box pipeline modules Figure 5 9 11 Mask files for NIR MIR S and MIR L from left to right Outlier pixels have value of 0 white and others 1 black so any outlier pixels will be masked out by mutipling these mask files to images of concern 66 IRC Data User Manual in apparently extremely large positive numbers for negative values thus pixels suffering from this effect may appear as hot or dead in the image plane However these values are designed to be smaller than the saturation limit of the detectors such that values higher than the maximum saturation levels must in fact be wrapped negative numbers This correction is made by Acorrected Auncorrected 2 5 9 1 for any pixels that have a pixel value smaller than 11953 After the wraparound processing is completed a 4 will be added as a prefix to the processed filename such that aF23342 8004 fits becomes waF23342 8004 fits Dark Subtraction greenbox dark In a point
113. nd subtracted Version 1 3 September 13 2007 103 e n fc flux calibrated after color term corrected residual background subtracted and wavelength calibrated Finally these are the most frequently referred arrays e refimage_bg Flat fielded background subtracted whole reference image in 3D X Y short long ID e specimage_bg Flat field background subtracted whole spectroscopy image in 3D X Y short long ID e specimage n wc Wavelength calibrated extracted 2D spectra in 3D X Y source id e specimage n fc flux calibrated extracted 2D spectra in 3D X Y source id 6 7 1 Frequently Asked Questions and Troubleshooting Q Two kinds of flats exist in the spectroscopy data reduction the imaging flats and the wavelength dependence flats How should we consider these flats A The IRC slitless spectroscopy differs from more standard slit spectroscopy in that in the latter case the pixel position and the wavelength correspond on a one to one basis The IRC spectroscopy does not follow this relation and needs to be applied in three steps e Many spectroscopic flats have been combined to form a super flat which can be used to remove the pixel to pixel variation of the monochromatic response of the detector e After subtracting the background extracting the sources and performing the wave length calibration the wavelength flat can be used The wavelength dependency of the flats is due to the same mechanism
114. nt sources can be made in the imaging field e The MIR L has a 7 arcsec slit for diffuse light similar to that of the MIR S camera This position is referred to as Ls 6 IRC Data User Manual Slit MIR L 7 x0 4 Scan Direction Telescope Axis EM Slit NIR 3 x1 Slit NIR 1 x1 Slit NIR MIR S 5 x0 8 9 3 0 7 for N2 9 5 0 5 for N3 amp N4 9 1 for MIR S all bands Figure 2 0 2 Field of view location of the three channels of the IRC The vertical arrow indicates the scan direction in the survey mode The NIR and MIR S share the same field of view by means of a beam splitter See text for the usage of the slits 2 1 Focal plane arrays The IRC comprises three channels Fig 2 0 1 The near infrared NIR channel operates in 2 5 um the mid infrared short MIR S channel covers 5 12 um and the mid infrared long MIR L channel works in 12 26 wm The NIR uses 512 x 412 format InSb array and MIR S and MIR L both employ 256 x 256 format Si As impurity band conduction arrays The three channels have a field of view of about 10 x 10 and the NIR and MIR S share the same field of view by the beam splitter whereas the MIR L observes a sky about 25 away from the NIR MIR S sky Figure 2 0 2 2 1 1 Near InfraRed Camera NIR Figure 2 1 3 shows the side view of the NIR channel The light from the telescope is split by a germanium beam splitter see Figure 2 1 4 and the near infrared radia
115. on of the position in the detector this step is called monochromatic flat fielding If we divide the observation by the super flat we obtain obs sky obj A x Fo z g A x R A Fi 2 y x spectral f eature z y 6 1 2 6 1 4 Background subtraction from individual sub frames After monochromatic flat fielding background should be flat over the aperture However in real data this is not the case for most of the cases For NP NG and SG1 outlier tolerant low order polynomial surface fitting is performed for each sub frame and is subtracted off from the flat fielded images For SG2 and LG2 background suffers from so called stray light of the earth shine contamination and the flat fielded images are much more structured To make good sky subtraction median running box filter is applied to estimate the local sky and is subtracted off from the stray light covered background Such background should be removed from individual sub frames since contribution of the earth shine depends on telescope pointing angle from the earth direction and amount of the stray light changes within a single pointing maneuver or within a single AOT operation In the first pass of the processing within the pipeline source masking will not be applied in determining the background because no source position information is available at the stage In the second pass after processing through section 6 1 9 once the background is measured again
116. only if the stars are significantly elongated By default threshold is three times the measured background noise When too much or too less sources are found threshold change would give better results and one does not have to take care of sharpness roundness parameters in most cases 6 4 4 Warning messages of the pipeline Some common and frequently appearing warning messages are listed below e Warning irc specred Wrong number of arguments Check the command line ar guments and try again e Warning No lt gt was found set Stop Some files that should be provided for the pipeline seem missing Check if list files are properly set in the command line This warning will also appear if you run SG1 SG2 LG2 data reduction before performing NP or NG data processing and some mandatory database files e g describing relative image shifts are missing e Warning Offset search box is too small When telescope drifts too much the image matching sub program fails to find the best shift values among sub frames Normally the sub program resumes searching the shift with larger search box This message is not so serious for most cases 98 IRC Data User Manual e Warning No data are available for measurement Returning 0 or Warning Data seems too noisy sigma lt sigma gt Returning 0 These messages will ap pear if e g detected sources are too faint and their positions could not be measured with good enough accuracy
117. orrecting distortion This may take a while ERROR on line 56 Cannot open file c iraf irc lib distortion DARK distortion database dat distortion ircconst constants database logfile irclog prefixs mslnDwa verbose no pipeline A Remove the DARK entries from the irc log file 3 Q Running the pre pipeline processor the following error is encountered at the co add stage Hitt COADD jt Making the input file list Extracting sources Calculating XY shift Adjusting sky level Coadding images ERROR on line 148 parameter direction not found A This problem may be caused by earlier versions of RAF the toolkit requires version IRAF 2 12 2 or later If your version of IRAF is earlier then please upgrade The problem is with georytran task You can type eparimages immatch georytran in the I RAF shell to see if the parameter exists Even after installing the newer version of IRAF you may have to type unlearn geoxytran before the toolkit runs correctly 4 Q During the pre pipeline run I find some apparently missing perl scripts and the irclog file is not produced or is empty HHH MKIRCLOG 44454 Making the file list Reading header This may take a while tcsh iraf irc perl formatlog pl Command not found tcsh iraf irc perl checkname pl Command not found 4 MKIRCLOG finished A Primarily please check you are running the toolkit from your working directory and not for example in iraf If
118. pect ratios of pixels to be square i e Aspect ratio is corrected to 1 to 1 by resampling the image Thus the toolkit corrects the linear distortion only Non linear distortion is not corrected for at present since it is assumed to be negligible After the resampling a e will be added such that fmslCcnDwaF23342 8004 fits be comes efmslCcnDwaF23342 5004 fits prefix before the original filename i e F23342_S004 fits gt gt gt efmslCcnDwaF 23342 S004 fits The resulting data is the basic processed data with individual frames not pointings corrected for detector and instrumental effects Version 1 3 September 13 2007 69 5 10 The pipeline processor Blue Box bluebox coadd wrap per Module 5 10 1 The Blue Box Co Add Wrapper It has not yet been decided whether an additional post pipeline processing module will be released to the users The pre requisite for the IRC data reduction team is at present solely to deliver basic science grade processed data to the users It is not the responsibility of the team to provide modules to produce mosaiced images or to provide tools for photometry etc The output of the green box produces the basic processed data for individual frames Ad ditional processing is required to produce the basic science grade processed data There exists a jitter between frames in the IRC images causing frames to become unaligned with each other note this is not an intentional dither which is a separate procedu
119. pixels are incleasing Version 1 3 September 13 2007 23 Figure 4 2 2 Representative Flatfields for each of the channels N3 S7 and L15 light is present in all detectors and an unexpected high background is seen This is probably due to external Earth light which is uniform over the Field of View Therefore observations of the diffuse background may be severely affected The S N of the superflat is estimated to be 5 N2 N3 10 N4 to 100 MIR S L and will limit the signal to noise ratio of the brightest stars There are at least the following problems 1 NIR We notice that there is stray light scattered off the telescope baffle This owes to the short baffle design and was not unexpected The amount depends on the direction and season of the observation worst at the ecliptic poles at the solstices and the wavelength A preliminary report is available in Japanese It cannot be ruled out that the current super flat may include a contribution from the stray light particularly for NIR 2 MIR S There is a noticeable pattern in the lower right quarter A similar pattern may appear even after flat fielding in some cases mostly for non dithering observations which are obviously artifacts The periphery of the MIR S images is affected by scattered light and should be discarded The portion affected is about 5 pixels from the left edge and 40 pixels from the bottom and right edges No scattered light is appreciated in the top
120. pter 4 IRC Calibration and Accuracy This chapter offers an overview of the main issues related to the IRC calibration and in orbir performance The instrument calibration is addressed in an standard way dark level flat fields linearity point spread function absolute flux calibration and distortion correction Spectroscopy and imaging are addressed at the same time Specific topics for each of them are explicitely indicated Caveats and general concerns are also discussed 4 1 Dark image Superdark is made from the pre dark measurements of LMC observations It is used by default during the automatic processing both for imaging and spectroscopic data after shifting its level by checking the slit area in each image Since the number of hot pixels is increasing along the mission as shown in Figure 4 1 1 the superdark does not correct for them Coadding media filter different images IRC03 05 different observations or IRC02 different images will filter out non corrected hot pixels As an alternative dark current data taken before pointed observations can be used and can provide with better results since it corrects for hot pixels Its SNR is lower than in the superdark one especially in NIR since each observation has only one pre dark frame Another problem related to the dark current arises from the IRC array operation It is not stationary since the scan mode clock is running for most of the time during the survey mode and the imaging
121. py FLAT SPEC2DFLAT_NP_slit 1st For reference images and color term correction of the flats FLAT IMAG2DFLAT_N3 1st FLAT IMAG2DFLAT_N4 1st FLAT IMAG2DFLAT_S7 1st FLAT IMAG2DFLAT_SOW 1st FLAT IMAG2DFLAT_S11 1st FLAT IMAG2DFLAT_L15 1st FLAT IMAG2DFLAT_L18W 1st FLAT IMAG2DFLAT_L24 1st e IRC_SPECRED_CALIBDIR MASK contains images of known bad pixels hot pixels cold pixels etc Masked area are marked by NaN Not a Number in the mask images MASK SLITMASK_NIR 1st MASK SLITMASK MIRS 1st MASK SLITMASK_MIRL 1st MASK OUTLIERMASK NIR 1st MASK OUTLIERMASK MIRS 1st MASK OUTLIERMASK MIRL 1st Version 1 3 September 13 2007 91 e IRC_SPECRED_CALIBDIR COORDOFFSET contains a table of coordinate offsets in pixels dX dY used to extract spectroscopy images for each object based on target positions on the reference image for all dispersers The table also includes sizes of source extraction boxes AX AY on reference and spectroscopy images and offsets dX dY for zero th order light image position COORDOFFSET IRCCOORDOFFSETPAR dat e IRC_SPECRED_CALIBDIR WAVEPAR contains the wavelength calibration tables For grisms NG SG1 SG2 LG2 wavelength um is expressed by 1st order polynomial lin ear equation and parameters are 1 dispersion dA dY 2 wavelength at origin Ao and 3 position of the origin Yo In the toolkit position of the origin
122. r plot spec with image wave array specimage n wc short mask specimage n source table source id short with the following options e nsum Number of pixels along X axis combined for plotting Default is 3 Larger nsum wider aperture collects more photons from the object but this also collects more back ground noise Therefore the best nsum for highest S N is typically 2 or 3 for point sources corresponding typical full width of image PSF e smooth Boxcar smoothing width in pixel along wavelength direction Default is snooth 0 no smoothing For spectra with higher S N without loosing spectral resolution smooth should be 3 corresponding typical full width of image PSF Larger boxcar smoothing win dow will loose the spectral resolution Note that when even number is set in the smooth e g smooth 2 the actual smoothing box size would be smooth t1 e g smooth 3 e sigma filter This option enables sigma filter operation at 3 sigma significance level over 2D spectra to remove spatially isolated high or low count pixels Default is off If you find too narrow emitting absorbing features in your plots try this option to see if this is a real feature or not Note that if there is a cosmic ray hit at the position in one of the sub frames such count should be removed as an outlier while combining sub frames with sigma clipping averaging method However if there are temporal hot pixels or weaker hot cold pixels that are not show
123. raging the offsets measured for many stars Once the offset is measured extraction of the 2D spectra is made again after considering the Y offset as well as X offset as explained earlier Zero reference point is much difficult to be found in NG spectra due to lack of notable spectral features before applying the flux calibration Therefore although the toolkit can estimate the Version 1 3 September 13 2007 87 reference point by examining the observed spectroscopy images it is strongly recommended to check the zero th order light images for better wavelength calibration accuracy The chance of detecting zero th order light image at significant level for SG1 SG2 and LG2 is not so large Therefore the drift of the wavelength zero reference point is calculated by using the drift measured in NP or NG for MIR S L grisms after correcting the pixel scale difference Another issue related to measuring the wavelength zero reference point is the finite pixel resolution Although the source positions can be measured with an accuracy of less than one pixel size unit on the reference image extraction of the 2D spectroscopy images can only be made on integer pixel number to avoid erroneous image interpolation This means that as large as 0 5 pixel error could be introduced in the wavelength calibration process if not corrected and is not so small comparing with the full length of the dispersed spectroscopy images 50 pixel As a first order correct
124. re The attitude of these frames must be matched for any shift and rotation in position and stacked This processing is carried out by the modules called from the Blue Box pipeline Note this is not the same as actually running the Blue Box Post Pipeline processor In reality this process occurs automat ically at the end of the Green Box pipeline processor In order to do this bright sources stars are extracted from each frame as source lists These source lists are then used to calculate the shift and rotation for each frame looking at a particular field of view on the sky using the first frame as a reference The processing currently consists of the following steps almost entirely included in the blue box co add wrapper module e Co add individual frames bluebox coadd This module calls further individual modules to coadd the image frames together There is also an alternative module that may be called to coadd the image frames for MIR L using the information from the MIR S channel in the case that there are not enough bright stars for an independent calculation of the shift and rotation of each frame bluebox source_extract Extracts bright reference sources bluebox calshift Calculate shift and rotation value between image frames bluebox adjust_sky Adjusts sky level between individual frames bluebox irc_stack Stacks the IRC images e bluebox putwcs To add WCS information to an image FITS file by matching 2MASS
125. rom the raw data For the scaling dark count offset is calculated within the pipeline by comparing average counts at the slit mask covered portion of the pre dark image and the corresponding area on the super dark image Additionally the average dark counts of the mask covered portion of each observed sub frame is measured except for LG2 for which there is no good dark area on the spectroscopy images for measuring the dark level The averaged offset is added to or subtracted from the super dark to make the scaled super dark 6 1 2 Linearity correction Linearity correction is made following the method adopted for imaging data reduction 6 1 3 monochromatic Flat fielding In the case of the slit less spectroscopy the entire FOV is the aperture for the background sky but the object image itself is the aperture for the object This aperture size difference makes difficulties in flat fielding the slit less spectroscopy images since unlike a conventional long slit spectrograph a given pixel can be illuminated by background photons with a range of wavelengths within the disperser s passband and the mixing fraction of photons of different 83 84 IRC Data User Manual wavelengths depending on the spatial distributions and spectra of sources We need 3D flat information X Y along space plus Z along the wavelength for full flat calibration However it is impossible to obtain such a detailed flat since there are no good monochromatic flat l
126. rvations that take images with three filters in a pointed observation For each filter two imaging observations are made with dithering operations 2 2 4 IRC04 The IRC04 mode was designed for general purpose spectroscopic observations This mode does not have dithering operations It has an imaging observation sandwiched by spectroscopic Version 1 3 September 13 2007 11 Table 2 3 3 IRC sensitivity and image quality N2 N3 N4 S7 SOW S11 L15 L18W L24 5 0 sensitivity uJy AOTO02 16 16 16 74 76 132 279 273 584 AOT03 20 19 19 91 93 162 341 335 716 AOTO05 5 57 9 43 44 76 161 158 337 The values for N2 N3 and N4 of IRCO05 are tentative observations observations with the dispersive elements of 4 frames The imaging observation will be used to determine the wavelength reference point for slit less spectroscopy 2 2 5 IRCO5 The IRC05 mode was designed for deep imaging observations with a filter in a pointed observa tion This mode has neither dithering operation nor filter change Thus observers are requested to make a number of pointed observations for a give target for the redundancy This mode replaces IRCO00 after confirming its high performance for faint source observations for the NIR channel The exposure times for the MIR S and MIR L are the same as for other AOTs for both short and long exposures whereas that for the long exposure of NIR is longer than others and is made with the Fowler 16 sampling scheme instead of th
127. s at this stage only N3 image has been processed Run the pipeline again for NP NG SG1 SG2 LG2 with target lists Image shift database created previously with NP NG will be used for the 2nd run If your target is compact and bright enough to be detected by the source detection sub program you do not have to perform this 2nd run For the 2nd run begin processing for NP NG first then SG1 and finally LG2 e Third run run the pipeline again if you want to revise the target table This is optional Now you got processed reference images for revising the target tables Run the pipeline again with the updated target tables to see final results 6 4 2 Running the pipeline Type the following in the IDL command line a command in a single line irc specred lt targetid gt lt targetsubid gt lt ext_source_table gt refimage list specimage list lt filter_spec gt Mandatory arguments e targetid ID of the pointing observation The information will be provided with the data distribution Example 1331048 e targetsubid Sub ID of the pointing observation The information will be provided with the data distribution Example 1 e ext source table An ascii list describing source position X Y in pixel or the target table Example myobjects tbl cat myobjects tbl 100 0 100 0 200 5 200 5 150 0 150 5 This will extract spectra of sources located at 100 0 100 0 200 5 200 5 and 300 0 300 5 in pixel on the referen
128. s a function of X Version 1 3 September 13 2007 System Throughput 13 IRC SPECTROSCOPY SYSTEM THROUGHPUT 0 5 NP NG x SGI o SG2 x LG2 scite x i 01 F x 4 X f X k i i x x PO A 0 05 x i i x 4 x i x x lt i z i X E 5 y A x x T d i 1 x Tox x T 3 i x x x EE T i x i T 1 i x 4 0 01 l I l Eo i i o l L dq o1 0 14d 2 3 4 5 6 7 8 9 10 12 14 16 18 20222426 Wavelength um Figure 2 3 8 IRC spectroscopy throughput as a function of A Chapter 3 Distributed Data Products Two levels of data products are delivered to the users raw data and processed data The raw data are intended to be used to run the pipelines described in Chapters 5 and 6 to get interactively the best science out of them The quick look data are merely a rough approximation of what the user will get from his observations They are the result of running the pipeline automatically and with a default setting An interactive and fine tuned run of the pipeline is recommended in order to improve the quality of the results The quick look data header contain the version of the pipeline used Since it is continously evolving it could be different from the version made available to the users 3 1 IRC FITS file naming rule Pointed observation data are distributed to the users as a tar gz package format hearafter data package per observation A data package contains
129. s are especially prominent in the MIR S images The intensity of the pattern is proportional to the sky background level Therefore the scattered light pattern has been modeled and template images made for the three MIR S filters They are stored in where you installed irc lib scatt Figure 5 9 12 shows the modeled scattered light template images These template images are multiplied by the mode of the sky background level which is then subtracted from each image to eliminate the pattern After the scatt light process a c will be added such that nDwaF23342 8004 fits becomes cnDwaF23342_5004 fits Figure 5 9 12 The modeled scattered light pattern images from left to right for the S7 S9W and S11 bands 6 Cosmic ray rejection greenbox cosmic ray Detect and replace cosmic rays in the MIR images After the comic ray rejection a C will be added such that cnDwaF23342 8004 fits becomes CcnDwaF23342 8004 fits 7 linearity correction greenbox linearity The linearity correction is made as explained in section 4 3 After the linearity correction a will be added such that CcnDwaF23342 8004 fits becomes 1CcnDwaF23342 8004 fits 8 Saturation greenbox saturation Any pixels that are marked as saturated are masked Physical detector saturation occurs around 12500 33000 and 33000 ADU after irenorm for NIR MIR S and MIR L re spectively Therefore any pixels with values greater than the scaled values in the short
130. s been made available and should be used in preference to the old flat data for observations after 2007 01 07 02 49 00 In particular the pattern in the lower right corner of the MIR S has disappeared after this epoch see Figure 5 14 18 We are planning to include the flat data in each observation but at this stage the flats may be replaced by either 1 replacing the fits files in irc lib flat directory or 2 changing the entries of flat in constants database in irc lib The new flat data is avalable from http www ir isas jaxa jp ASTRO F Observation DataReduction IRC e Super or Self dark and coadding data The Super dark images were obtained from more than 100 pointings of data of pre dark images taken at an early stage in the mission and should provide superior single to noise However it is known that the number of hot pixels in the detector arrays are gradually increasing throughout the mission and the 78 IRC Data User Manual 170 aan fa BOD f n D amp Figure 5 14 17 Background levels Super dark image now contains incomplete information on their number and position This can have a severe effect on the coaddition of MIR L frames in certain cases In this scenario it is possible that a coaddition may indeed appear successful but in fact what has happened is that hot pixels may have been included in the coaddition process resulting in a meaningless final image In this case the self dark option should be s
131. s correlation peak is weak The cross correlation function will be calculated while registering sub frames and or short and long exposure frames The peak value of the cross correlation function is a good measure of the frame registration The value is 1 for ideal image matching and is typically 0 6 or larger for real data with noise but peak of 0 3 is still good When this warning appears check the quality of the registered and stacked images So far registration of the short exposure frame can not be made with good enough accuracy and one should ignore this warning for short exposure frames This message is not so serious for many cases 6 5 Working on the pipeline output 6 5 1 Displaying the whole images on ds9 Here we explain the region marks on the ds9 that shows processed whole image products e e spectroscopy and reference images before source extraction In the spectroscopy image side left e Blue rectangle shows spectroscopy image extraction area e Green rectangle shows area reserved for the object or masked area for other objects e Red rectangle shows plotting area as defined by the spectral plotting tool e Yellow circle shows the expected positions of the zero th order light In the reference image side right marked circles correspond source positions with diameter of the source FWHM Version 1 3 September 13 2007 99 e Green circles show sources with good position measurements e Red circles show sources
132. s the Satellite position at DATE REF in km DAYNIGHT is the Day night status at DATE REF These fields are updated as pointing analysis goes on from On board AOCS to G ADS Pointing reconstruction for Survey mode STTA NUM is the number of tracked stars in STT A at DATE REF STTB NUM is the number of tracked stars in STT B at DATE REF STTA MOD is the STT A Mode status at DATE REF either TRK ACQ STB INL R or ININ STTB MOD is the STT B Mode status at DATE REF same as above COMMENT Any strings HISTORY Any strings Raw data dimensions FITS dimension of the NIR frames is 412 x x 512 y x 2 short long exposure sub frames FITS dimension of the MIR S and MIR L frames is 256 x x 256 y x 4 1 short 3 long exposure sub frames Note that for convenience of data handling in the electronics onboard the satellite MIR S and MIR L frames are attached to each other along Y axis and form a single FITS of 256x512x4 dimensions The lower half portion of the 256x512 pixel images is for MIR S and the upper half portion is for MIR L With the AOT04 one will obtain 11 or 12 sets NIR and MIR S MIR L of frames e 1 set of pre dark frame e 8 or 9 sets of spectroscopy exposure frames Version 1 3 September 13 2007 19 e 1 set of reference imaging exposure frame e 1 set of post dark frame The dark frames will be taken during pre and post satellite maneuvering period with the shutter closed
133. see Fig 5 6 3 The screen shows the version of toolkit super flat super dark linearity and distortion Please let us know these numbers if you send any email to the help desk iris_help ir isas jaxa jp to inquire any trouble with the toolkit 56 rane X XGterm IRAF v 2 12 1 NFS Darwin IRC Data User Manual cl ir NOAO PC IRAF Revision 2 12 2 EXPORT Sun Jan 25 16 09 03 MST 2004 This is the EXPORT version of PC IRAF V2 12 supporting most PC systems Welcome to IRAF To list the available commands type or To get detailed information about a command type help command To run a command or load a package type its name Type bye to exit a package or logout to get out of the CL Type neus to find out what is new in the version of the system you are using The following commands or packages are currently defined dataio irc noao proto utilities dbms language obsolete softools images lists plot system irc 4 AKARI IRC imaging data reduction pipeline Pipeline version 060801 flat version 060626 dark version 060428 linearity version 051130 distortion version 060529 Copyright C 2005 Institute of Space and fistronautical Science Japan Aerospace Exploration Agency For help send email to yita ir isas jaxa jp 4 bluebox irc tool prepipeline unlearn all
134. sing the wavelength is assumed Therefore once the 2D spectra extraction is made at sufficient accuracy it is straightforward to make a wavelength calibration For prism NP spectra since the dispersion equation is highly deviated from the linear one especially at shorter wavelength end it is better not to transform images to avoid introduc ing extra uncertainties in the image interpolating extrapolating processes Rather a separate array whose length is equal to the Y length of the extracted 2D spectra will be created to store the wavelength values for each Y pixel One wavelength array is applicable for all the extracted spectra within the FOV This kind of wavelength array is also used for grism spectra NG SG1 SG2 LG2 Finding wavelength zero reference point Some problems arise for accurate wavelength determination One is the satellite attitude stability problem If the satellite pointing moves between reference and spectroscopy images reference image does not provide good wavelength zero reference point any more NP spectrum before flux calibration shows a notable peak around 2 4m and there are lots of fairly bright stars with almost identical spectra regardless of the type of stars within a single FOV Therefore it is possible to find the best offset of the wavelength zero reference point by measuring the NP peak positions with respect to the spectral template that was taken when satellite attitude stability was good enough by ave
135. st all the images in the input list Then you should specify the sub frames to be removed on ds9 within the pipeline This is because the file names and their order in the list are used to relate FITS files to the exposure timing along the AOT operation sequence filter spec a string specifying a disperser for the processing Set one of the following N3 NP N3 NG S9W SG1 S9W SG2 L18W_LG2 Options e root dir a string specifying a directory in which a set of data is located If set this overrides the setting found in the environment variable IRC SPECRED DATADIR Example root dir DATA ASTRO F IRC SPEC Nh spec Ns spec Ls spec flags for slit spectroscopy data reduction Np spec a flag for Np spectroscopy data reduction See section of notes on slit spectroscopy for more information of slit spectroscopy data reduction no tune sourcepos a flag for disabling source position tuning subprogram within the pipeline By default irc specred tries to measure accurate source positions by searching Gaussian peak around the coordinates set in the target table This flag disables this functionality use short refimage use a short exposure image to measure source positions By de fault irc specred uses the long exposure frame for measuring source positions in the reference image With this flag set the toolkit uses the short exposure frame for measur ing the source positions Note that the data reduction will be made
136. t and users are advised to re process using the latest version of the IRC toolkit Note the co add products are not initially distributed since the co add often fails especially for the MIR L images Users may therefore use the additional coaddLusingS toolkit feature or other 76 IRC Data User Manual independent methods 5 14 Limitations of the functionalities in the current version of the imaging toolkit 5 14 1 Instrumental characteristics and artefacts in the data First frame effect The first frame of the dark current prior to the pointing observation shows larger values than others particularly for MIR S and MIR L detectors This effect could be related to the detector temperature Increase of hot pixels As shown in Figure 4 1 1 the number of hot pixels is increasing along the mission They should be corrected when coadding individual images or correcting the image by its own dark image Memory effects No noticeable effects have been reported Dark Level The dark level is not stable during the pointing observations The NIR dark level shows no correlation with temperature The MIR detector dark level show a weak correlation with temperature The Dark level is also very high after passing SAA Effects of high energy particle hits The toolkit does not perform any deglitching Glitches should disappear when coadding individual images median filter Even so their effect on the responsivity are not yet well investigated Fu
137. the narrow slits Nh Ns and Ls we do not have to take account of the satellite attitude stability if the size of the drift is much smaller than that of the objects e Ghosts in Np observations Ghosts relating the Np slit have also been recognized Left panel of Fig 4 13 16 shows an example of the imaging data There are sources seen on the slit mask region which must be ghosts from the sources in the Np slit Right panel of Fig 4 13 16 shows the corresponding spectral data in which ghost spectra overlap with the source spectrum A similar ghost is also reported for SG2 in which the effect of the ghost should not be significant since the ghost spectrum appears at much shorther wavelengths lower part of the image and does not overlap the source spectrum No definite origin for these ghosts have been elucidated and thus no clear recipe has been prepared to correct for them Figure 4 13 16 Ghosts generated in Np observations Left panel Imaging data Right panel spectroscopic data Chapter 5 Imaging toolkit cookbook 5 1 Introduction The IRC imaging data reduction toolkit is developed to address and correct any IRC Instrument features such as linearity flat fielding distortion etc and converts the raw ADU signal to physically meaningful units ie flux However in the current version the data number of resultant image is ADU per frame and they should be multiplied by a certain constant to convert it to a physical unit su
138. the problem persists then it may be because your perl is in the wrong place Perl should be in usr local bin per1 therefore please set up a symbolic link from your perl library to usr 1ocal bin perl type whichperl to find out where perl is currently hiding 5 Q I have installed a new version of the toolkit and now it is crashing A Always as a first fix type unlearn_all from within IRC 82 IRC Data User Manual 6 Q My new version of the IRC toolkit crashes when I try and run the original Sample Data through it Coadding images This may take a while ERROR on line 272 Attempt to access undefined local variable angle A Note that due to changes in the FITS Headers etc the original Sample Data Files are no longer compatible with the IRC toolkit The sample data is no longer available for download Q The IRC toolkit does not run on the new Intel Mac machines A The IRC toolkit fails in the Macbook pro environment due to differences in the IRAF build in the intel binaries Replacing the intel binaries with the original binaries causes iraf to run under emulation but solves the problem RAF 2 13 has been successfully run on the intel machines Q During the coaddition process or using the coaddLusingS function the toolkit crashes with the error ERROR on line 128 Attempt to access undefined local variable filter A This error will occur if the pre pipeline is attempted to run twice on a giv
139. time scale specification 16 DATE OBS DATE END DATE REF AFTM OBS AFTM END AFTM REF PIMTIOBS PIMTIEND PIMTIREF EQUINOX RA DEC ROLL AA SOL AA EAR AA LUN TM SAA SAT POSX SAT POSY SAT POSZ DAYNIGHT STTA NUM STTB NUM STTA MOD STTB MOD COMMENT HISTORY END YYYY MM DDTHH MM SS YYYY MM DDTHH MM SS YYYY MM DDTHH MM SS double double double UxXXXXX UxXXXXX UxXXXXX 2000 0000 320 5533 23 3325 30 553 90 0021 180 2083 210 6821 1829 2903 5528 1704 3092 1968 4286 DAY 4 4 TREK TREK The header keywords are sorted as follows e FITS basic information data size information SIMPLE refers to the standard FITS format TOC IRC Data User Manual Observation start date time Observation end date time Reference time in the Observation DATE OBS in ASTRO F Time DATE END in ASTRO F Time DATE REF in ASTRO F Time DATE OBS in PIM TI 36 bits DHUTI DATE END in PIM TI 36 bits DHUTI DATE REF in PIM TI 36 bits DHUTI Epoch of Coordinate degree Target position at DATE REF degree Target position at DATE REF degree Roll Angle at DATE REF degree Solar avoidance Angle at DATE REF degree Earth avoidance Angle at DATE REF degree Lunar avoidance Angle at DATE REF sec Duration since last SAA at DATE REF km Satellite position at DATE REF km Satellite position at DATE REF km Satellite position at DATE REF Day night status at DATE REF number of tracked stars in STT A at
140. tion is introduced to the NIR channel The NIR consists of silicon and germanium lenses There are color aberrations among N2 N3 and N4 and the telescope focus is adjusted in between them Faint ghosts of bright sources are also present which come from internal reflections in the beam splitter The brightness is about 0 7 of the true source and the position is well determined and in a good agreement with the ray tracing simulation There are also ghosts that seem to come from the internal scattering in the NIR optics Details are under investigation 2 1 2 Short wavelength Mid InfraRed Camera MIR S The light reflected by the beam splitter is lead to the MIR S It consists of two aspherical germanium lenses Figure 2 1 4 shows the side view of the MIR S channel Part of the narrow slit used for the NIR channel Fig 2 0 2 will also be shared by MIR S for spectroscopy for diffuse emission The MIR S also has ghost images of bright sources due to the internal reflections in Version 1 3 September 13 2007 Y 1 Channel NIR Table 2 0 1 IRC Filters and Dispersion Elements 3 4 Name Aret Wavelength Dispersion um um um pix 2 4 3 2 2 N2 N3 N4 4 1 NP 0 06 3 5 pm NG 0 0097 S7 SOW MIR S S11 SG1 SG2 12 6 19 4 13 9 25 6 MIR L 20 3 26 5 17 5 25 7 4 Reference wavelength 5 Defined as where the responsivity for a given energy is larger than 1 e of the peak Isophotal wavelength of the filter band Ef
141. to that of imaging mode and is corrected by interpolating extrapolating the imaging superflat taken with two different filters for every pixel and the wavelength colour correction e Finally the spectral response specific to the spectroscopic mode observation is cor rected by the spectral response function obtained from the observations of standard stars More detailed explanations can be found in the IDUM Q When extracting the spectrum can the usage of the masks can be avoided A The no_mask option is especially relevant in spectral images where the source of interest is much brighter than the others Results can be compared with and without the no_mask option Q What can be done if the spacecraft attitude shifts after the reference frames have been taken A The effect will be apparent in the wavelength calibration For NP only the signal at the peak of the RSF is used to correct the wavelength reference by default The result will be inserted into the log file which will be further utilized to correct the drift in SG1 2 and LG2 A shift of more than 1 2 pixels can happen The actual correction is made by shifting the wavelength array rather than the image array The SG1 2 and LG2 use the NP NG as a reference thus one should process the NIR data first Similarly for NG in the point source aperture the NG slitless spectroscopy field should be processed first 104 IRC Data User Manual 4 Q Was the spectral response
142. tpath Pl e sit ma siian a e i a Eea ee ee SAT Perl paths zv e 2 uos Lm Ale te a RS tee gt Sus At A 5 4 8 Add IRC entry to IRAF 2e How to UPGRADE the version of IRC imaging toolkit Setting up your toolkit environment and running the pipeline 5 6 1 Creating the directory structure les 60 2 3Eaunch DEA s cu uem eem xp f eU nien Tu iq 5 6 3 Load the IRC package es The pre pipeline processor Red Box lens 5 51 Configuration 204 243 mn oh Paha Xo uomo X RR XU SE 5 7 2 Running the prepipeline processor 2e Before runnning the pipeline processor 2e The pipeline processor Green Box 2l 5 9 L Configurations z 2o weed e bee Gee e EARUM EUM GS 5 9 2 Running Green Box pipeline module 2 The pipeline processor Blue Box bluebox coadd wrapper Module 5 10 1 The Blue Box Co Add Wrapper es 5 10 2 Log files produced from the toolkit 22e The IRC TOOL GBesbonl uos uem a avem ematis dus de BAM Mee Ur cae Toolkit StruCt Ee n eue et ete un RUDI a ol reves Re med a recae Worl ng on thevotitputs siae eges rose det IgE ae RR P RE haere D ROT eas Limitations of the functionalities in the current version of the imaging toolkit 5 14 1 Instrumental characteristics and artefacts in the data 5 14 2 Toolkit Limitations and Caveats c n Frequently Asked Questions and Troubleshooting
143. ture versions of the toolkit will involve a more careful treatment of cosmic ray hits Ghosts Ghosts appear in all three bands NIR and MIR S ghosts originate from the beam splitter while those in the MIR L arise from the lenses The positions which depend on the real source positions are well determined for all detectors and also depends on the filter The intensity of the ghosts is well determined for NIR a few and MIR S 4 of the parent source for S11 1 for the rest of the filters In the cases where ghosts are detected in MIR L the parent source is saturated therefore there is no quantification of the relative intensity of the ghost in this case Background Level The background level changes during the pointing observations in all bands It is due to the Earth shine reflection therefore it depends on the angle between satellite and Earth and the epoch of the observation The effect from the Earthshine is worse at northern ecliptic latitudes during June July amp August The Background level is different at the begining and at the end of the observation see Fig 5 14 16 at different pointings see Fig 5 14 17 and in differernt seasons strongest in solstices Problem with short exposure frames with IRC A problem exists with the con tiguous short exposure frames of the MIR S and MIR L data taken in IRC00 IRC04 and IRCO5 At present the cause is uncertain however it is advised not to use short exposure frame data for scient
144. ty Observations of very bright standard stars seem to be compatible with those of medium brightness 4 4 Instrument Point Spread Function Table 4 4 1 shows the FWHM of the in flight PSF in the imaging mode based on observations of standard stars performed in May and September 2006 They are partly affected by th attitude control stability and indicate the worst cases Table 4 4 1 In orbit PSF in pixels The PSF are not spherical Therefore the users are recommended to average individual images in the coadding process if photometric accuracy is concerned In some cases mainly in deep survey the median filtered image will lose some signal if the images are rotated with respect to each other To avoid this problem users should use average instead of median filter when combining individual images The PSF in the spectroscopic mode is worse by one or two pixels compared to imaging data Version 1 3 September 13 2007 25 Figure 4 3 3 Raw signal versus fit signal expected if detectors were to behave linearly The white line shows Raw signals equals to fit and the red line shows the calculated correction equations Note that the physical detector saturation occurs around 12500 33000 and 33000 ADU after irenorm for NIR MIR S and MIR L respectively 26 IRC Data User Manual NIR linearity i i before correction 5 after correction x or 4 E s c 2
145. use for example the following irc specred com mands to review the results e show_aperture_on_ds9 specimage bg e show aperture on ds9 refimage bg imag e plot spec with image wave array specimage n wc mask specimage n source_table lt source_id gt 6 6 Appendix 6 6 1 Variable name conventions As a general rule the extension of the IDL variables _ff bg wc _fc indicates the finally processed calibration status Examples e ff means that the data is processed all the way to flat fielding e _bg means that the data is processed all the way to background subtraction i e flat fielded and background subtracted e _wc means that the data is processed all the way to wavelength calibration i e flat fielded background subtracted and wavelength calibrated e fc means that the data is processed all the way to flux calibration i e flat fielded background subtracted color term corrected wavelength calibrated and flux calibrated Whole image products ff and bg are in 3D format X Y short long ID Short exposure frame is in Z 0 and long exposure frame is in Z 1 Extracted image products n images are in 3D The third dimension is for indicating the source id They are indicated by n extension of the variable names The following extensions refer to different processing stages e mn bg background subtracted on extracted images e n wc wavelength calibrated after color term corrected residual backgrou
146. version factor ADU tz unit to Jy Band fo error 76 Nstar N2 3 226 x 107 2 94 14 N3 2 506 x 10 5 2 33 17 N4 1 676 x 10 5 3 44 16 S7 2 601 x 10 5 2 33 28 S9W 1 561 x 10 5 97 11 S11 2 223 x10 2 36 23 L15 4 145 x 10 5 2 82 33 LI8W 2 736 x 10 4 56 13 L24 1 461 x 10 7 4 71 21 Number of standard stars used Version 1 3 September 13 2007 35 Table 4 6 7 Conversion factor ADU exposure to Jy calculated from fo Band short exposure long exposure long exposure IRCO5 error fs fi fi N2 4 033 x 10 9 4 245 x 107 2 881 x 107 2 94 N3 3432x109 3 297 x 107 2 237 x 107 2 33 N4 2 095 x 10 9 2 205 x 107 1 496 x 107 3 44 S7 2 601 x 10 5 9 290 x 107 2 33 S9W 1 561 x 10 5 575 x 107 5 97 S11 2 223 x 10 7 938 x 107 2 36 L15 4 145 x 105 1 480 x 10 9 2 82 L18W 2 736 x 10 5 9 770 x 107 4 56 L24 1 461 x 1077 5 217 x 10 9 4 71 Long exposure for IRC05 4 6 4 Overall accuracy of the flux calibration The absolute accuracy for point sources based on observations of standard stars is less than 5 for all the bands The stability of the instruments are being monitored by observations of the same stars Observations show no indications of any change in the reponsivity within 5 96 for all the three channels over more than a year 4 7 Flux calibration for extended sources A correction factor should be applied to convert the point sources flux calibration into extended sources flux This calibration will be done using the a
147. x XUL EG 2 2 97 ARCS se ny hU qu UR Phe eek eS Ns eue Eus ue Bae EbNIau 2 220 ARCO Llc ESAMI DR Rut EE eut hux d p Ee S 2 2 5 RECOS a nies terio uet rra ted on tm eer vri ph dte d 2 2 6 IRGIEL e 42 ACERBA Oe eb US Gee eae ER ROO 220 MID S esu soos RIA do SERLO EUR atado ue he dob Ue 2 89 In orbit sensitivity eee Sere 3 Distributed Data Products 3 1 IRC FITS file naming rule ss 3 2 Raw data description 2 22s 3 21 Raw data naming convention 4 2s 3 2 2 AKARI FITS Primary HDU common information rawdata header 3 2 3 Raw data dimensions i6 ae PoP eB Sie Soke Ye OR de bee CORO 3 9 Quick look data ous es wi le uem ae oe eke Be ee ee pee A Elide 3 3 1 IRC imaging IRC05 IRCO2 and IRCOS 3 3 2 IRC spectroscopy IRC04 aoaaa 3 4 IRC04 image orientation and dispersion direction 4 IRC Calibration and Accuracy 4l Darkimage sorora og moane a xc a eha BS He Ere Rom a ee Rob Eg ALD sBlatheld rs mx 6 he gs BOA ed hind SO eo Esa qd dns d 4 2 91 Flatfileds for imaging data les 4 2 0 Flats for spectroscopy images ln 4 3 Instrument linearity aoaaa 4 4 Instrument Point Spread Function ls ASD RSRE eue ode Ye Mire ee mec de ob eee pon d cu e b b Su A rise dms 4 6 Flux calibration for point sources 2e N m 000000148 10 10 10 11 11 11 11 Version 1 3 September 13 2007 4 7 4 8 4 9 4 10 4 11 4 12 4 13 5 1 5 2
148. ype of rejection operation performed in coadding none No rejection minmax Reject the nlow and nhigh pixels ccdclip Reject pixels using CCD noise parameters crreject Reject only positive pixels using CCD noise parameters sigclip Reject pixels using a sigma clipping algorithm avsigclip Reject pixels using an averaged sigma clipping algorithm Version 1 3 September 13 2007 63 pclip Reject pixels using sigma based on percentiles e rej sky boolean yes or no default yes activate if you want to reject any images which have outlier sky level from the coaddition process Some frames may be taken during maneuver and the pipeline will detect such frames by watchig their sky level e skip_L boolean yes or no default no activate if you want to skip the process of coadding MIR L images It will save you great time and you can use a tool to coadd the L image after the pipeline has finished e submeds boolean yes or no default no activate if you want to make a median filtered sky image and subtract it from each image before coadding This option is useful for deep imaging observation where the sky has no structure e coadd boolean yes or no default yes de activate if you do not want to try to coadd images This option is useful for users who have their own coaddition strategy and software e default boolean yes or no default yes This parameter defines which flat is used in the

AKARI IRC Data User Manual

Contents

Download Pdf Manuals

Related Search

Related Contents