========== DIIV9P_ENC ========== _____________________________________________________________________________ _____________________________________________________________________________ Reviewers: A. Cochran, S. Joy dataset.cat - Check DOY/exposure ID table. Should match actual data DOYs and exposure IDs. Has some typos and missing DOYs. - Verify that all processed data sets only include Tempel 1 data; the mission did not calibrate and archive calibration data/frames. - Remove one of the occurrences of "after" in the data set overview: "This data set contains... the Deep Impact....after after the Impacator..." - Correct typo under Processing: "calibration. For impact images, geometric parameter suc such as the ... - Correct typo under Processing: "The pipeline also creates a FITS image extension a signal-to-noise ratio map". - Correct misspelling: pnuemonic - Correct typo (is -> are): "Calibration steps and files applied to each reduced FITS image is provided by PROCESSING_HISTORY_TEXT..." - Correct typo (A -> The): "A 2-dimensional, reduced CCD images in this data set are..." - Correct typo (move comma afgter "eight"): "The first extension uses eight, bit flags to describe the quality of each pixel in the primary image..." /calib/abscalvs - Convert files from from one-dimensional FITS to ASCII or to FITS tables. Modify labels and calinfo.txt accordingly. /calib/adclut, /decomprs, /psf, and /xtalk - Convert FITS "images" to FITS or ASCII tables. Modify labels and calinfo.txt accordingly. /calib/gain - Check contents the files; all values are one? Are these files simply place holders to make the calibration pipeline run? The gain FITS headers indicate these may actually be gain corrections. Clarify with the SDC and update documentation and labels accordingly. /calib/drkmodel - Check contents the master dark files. Why are all of the values are the same? Are these simply place holder files to make the calibration pipeline run? Need to research with data provider (O. Groussin). Per T. Farnham, each quad of ITS has a different coefficient, but HRIV and MRI have the same coefficient for each quadrant. Therefore, these maps have the same value for all quadrants. - Consider adding the dark integration time to at least the PDS labels and ideally to the FITS headers. /calib/badpix - Why are there horizontal lines of bad pixels at the top and bottom of (at least) the mode 6 map? /calib/psf - Are these files simply dummy files to keep the calibration pipeline running smoothly? Tar files of /data DOY directories - Does 185 tar file have labels? General ======= /browse ------- - Click on thumbnail and get a JPEG version of the actual FITS file. Consider linking to the actual FITS file. - Add a note that states the JPEG images can be small. During encounter, many VIS images were taken in the 64x64 pixel subframe mode. Consider adding image sizes on the browser pages. - What is the reduced resolution on the JPEG images? If it is available, document it in at least browinfo.txt. browse.lbl - Fictional browse object is used; state this object is not currently defined in the PDS. Recommend writing a separate label for each JPEG file. - Change PDS_VERION_ID from "PDS4" to "PDS3". /catalog -------- General - Update mission, instrument host, instrument files with information through encounter. - Change references to the relative impact speed from 10.2 to 10.3 km/s; this speed changed to 10.3 because the launch was delayed to Jan 12th (10.2 km/s was for a Jan 1st launch). mission.cat - Check DOYs and start/stop dates provided for each mission phase. - Assuming all calibration targets are renamed to "CALIBRATION", modify the list of mission targets to be only 9P/TEMPEL 1 and CALIBRATION. personnel.cat - Correct AHEARNETAL2005 and AHEARNETAL2005A reference IDs. dataset.cat - Remove contact names. - Verify list of observation dates and exposure IDs. Some min/max exposure IDs do not seem to match. Check date ranges, too. - Add section for "essential reading" documents (for documents included in the archive). - Add a section for related data sets. For example, for a raw data set, provide a list of associated reduced data sets. - Add note that horizontal stripping indicates missing data. The image quality map identifies these affected pixels. - Add note to reduced data sets that calibration frames were not calibrated and are not included in these data sets; Calibration frames are only available in the raw data sets. - Clarify description of correlation between exposure ID and image number. For example, there can be multiple image numbers for one exposure ID. One HRII scan is represented by one exposure ID but many image numbers, starting at 001. - Add a note that the first 100-bytes of a raw image are the spacecraft header bytes. Beware when displaying the images. The mission will not overwrite these bytes in the raw and processed data sets, but will in higher-level products. Target catalog files - RA and Dec should appear in the files. However, these should disappear in the final archive. Calibration targets will be changed to "Calibration". - In general, why do some parameters in the planetary catalog files have so many decimal places? - Clean up the calibration target catalog files. Recommend changing all calibration-related targets to "CALIBRATION" and adding a new keyword, e.g. ACTUAL_TARGET_NAME, to hold the original value in TARGET_NAME. This affects the product labels and index files for all raw data sets, as well as the target list in the mission.cat, voldesc.cat, and index.lbl files. /document --------- - Subdirectories are by DOY; can YY-MM-DD be included? Or, consider creating a new table that includes DOY, calendar date, exposure ID range for *all* instruments. - Provide a report that discusses the time discrepancy between the spacecraft clocks and how the problem was "resolved" and implemented in the SPICE kernels and the archive. Include this document in the SPICE archive, too. - Write a more detailed html page to present the documents. Need to help users decide which documents are "required reading". - Include a document that discusses the known deficiencies with the calibration of the IR spectral data. K. Klaasen (JPL) will write this document. - Consider extracting a copy of the flow chart calibration pipline, along with the corresponding text, and including it as a "calibration summary" document. Reviewers did not want to read the entire calibration document. - Calibration paper: Add at least a table of contents to the final version. - /document/calibration_pipeline: IDL routines and imbedded comments are adequate for this archive. /calib (for reduced data only) ------ - Convert /abscalvs files from from one-dimensional FITS to ASCII or to FITS tables. Modify labels and calinfo.txt accordingly. - Convert /abscalir, /adclut, /decomprs, /psf, and /xtalk from FITS "images" to FITS or ASCII tables. Modify labels and calinfo.txt accordingly. - Check contents of gain files for all VIS data sets; all values are one? Are these simply place holder files to make the calibration pipeline run? The gain FITS headers indicate these may actually be gain corrections. Clarify with the SDC and update documentation accordingly. - Check contents of dark files for all VIS data sets. Why are all of the values are the same? Are these simply place holder files to make the calibration pipeline run? Need to research with data provider (O. Groussin). Per T. Farnham, each quad of ITS has a different coefficient, but HRIV and MRI have the same coefficient for each quadrant. Therefore, these maps have the same value for all quadrants. Change the description for these files since these are not the "true" gain! /index ------ index.lbl - Add missing column descriptions. - Correct typo in INDEX_TABLE description: "This index table contain parameters..." - Correct typo in PRISM_TEMP description: "...parameter provides provides the temperature..." - Check for other typos and correct. index.tab - Median values match maximum values in all data sets. Why? - Table is hard to use and print because it has over 100 columns. Consider removing columns that are typically not necessary for science or break it into multiple parameter tables. indxinfo.txt - The INDEX.TAB description has two occurrences of "provide". Remove one. /data ----- - Check values of min, max, median keywords in FITS, labels, and index files. IDL and DS9 disagree with values in headers and labels. This may be due to the 100-byte spacecraft header being excluded from the range of values in the FITS headers. May need to explain this in all dataset.cat files. _____________________________________________________________________________ _____________________________________________________________________________ Liens by SBN/S.McLaughlin for the DI Data Sets delivered Dec 2005 Affected Raw Data Sets -------------------------------- DIF-CAL-HRII-2-9P-CRUISE-V1.0 DIF-CAL-HRIV-2-9P-CRUISE-V1.0 DIF-CAL-MRI-2-9P-CRUISE-V1.0 DIF-C-HRII-2-9P-ENCOUNTER-V1.0 DIF-C-HRIV-2-9P-ENCOUNTER-V1.0 DIF-C-MRI-2-9P-ENCOUNTER-V1.0 DII-C-ITS-2-9P-ENCOUNTER-V1.0 Affected Calibrated Data Sets -------------------------------- DIF-C-HRII-3/4-9P-ENCOUNTER-V1.0 DIF-C-HRIV-3/4-9P-ENCOUNTER-V1.0 DIF-C-MRI-3/4-9P-ENCOUNTER-V1.0 DII-C-ITS-3/4-9P-ENCOUNTER-V1.0 ======= General ======= Documentation Set ----------------- o Add quaternion.asc and quaternion.lbl to /document. It was omitted in the delivery. /data/*fit and *lbl ------------------- The following changes require the DI Science Data Center to correct the FITS headers for all raw and processed flight data. PDS labels will need to be regenerated to capture these changes. The revised data sets must be submitted as version 1.0 and will replace the data sets delivered in Dec 2005. o Correct the values for the target-to-earth position and velocity keywords (TAREARRX* and TAREARV*) in the FITS headers. This affects all data sets. o Correct the comments for the TARSUNR* and TARSUNV* keywords. Recommend improving the comments for all of the best geometry keywords to avoid confusion about the direction of the vectors. o Reverse the sign of the TARSCR* and TARSCV* target-to-spacecraft position and velocity in teh FITS headers. The values will then have the same sign as the NOMinal and REConstructed FITS keywords. SBN found the flipped sign in the TARSC* keywords to be confusing. o In the labels, change TARGET_EARTH_POSITION_VECTOR to EARTH_TARGET_POSITION_VECTOR o In the labels, change TARGET_EARTH_VELOCITY_VECTOR to EARTH_TARGET_VELOCITY_VECTOR o In the labels, add SPACECRAFT_GEOCENTRIC_DISTANCE. The source is EARDIST in the FITS headers). o In the labels, correct the value of the TARGET_HELIOCENTRIC_DISTANCE keyword. Currently, it is set to the the spacecraft-to-Earth distance. The correct source is the FITS header keyword TARSUND. /catalog -------- mission.cat o Verify the estimated time-of-impact provided in the labels is noted in the data set description. Consider checking if the science team wants a particular value specified for the mission. The current value was taken from the initial results paper published in Science 2005. dataset.cat o Shorten data_set_terse_desc to 255 characters; in some cases it is too long. o Verify this note was added to the raw and processed data sets: For encounter images, geometry values relating the target were calculated using the impact site as the origin of the vector. This was the information available in the encounter SPICE SP-kernel delivered by the DI Nav team. The comet center wasn NOT used when calculating geometry for encounter images. o Verify the estimated time-of-impact provided in the labels is noted in the data set description. reference.dat o Change AHEARNETAL2005 to AHEARNETAL2005B. Make necessary changes to mission, host, instrument, and dataset.cat files. This referenence is the for SSR chapter "Deep Impact: A Large-Scale Active Experiment on a Cometary Nucleus". bethyitarget.cat o EN made a motion to change the target name from "BET HYI" to "BETA HYI". The value "BET HYI" is used the the FITS headers, PDS labels, and index files. However, since the project has been directed to replace all calibration targets with the name "CALIBRATION" in the labels, there is no longer a reason to implement this lien. Additionally, Simbad gives "BET HYI" as the accepted name for this star. index.tab and .lbl ------------------ o Change name of column TARGET_EARTH_POSITION_VECTOR to EARTH_TARGET_POSITION_VECTOR o Change name of column TARGET_EARTH_VELOCITY_VECTOR to EARTH_TARGET_VELOCITY_VECTOR o Add column for SPACECRAFT_GEOCENTRIC_DISTANCE o Column 88: Modify the source for, TARGET_HELIOCENTRIC_DISTANCE, to be the value of TARGET_HELIOCENTRIC_DISTANCE in the data labels. Currently, the value for the TARGET_HELIOCENTRIC_DISTANCE keyword in the data labels in not correct (it is the spacecraft-to-Earth distance). o Columns 64-47: Change the NAME suffixes to "COS", "SIN1", "SIN2", and "SIN3", respectively. The original values of "SIN1", "SIN2", "SIN3", a and "COS" are not the correct order. errata.txt ---------- o Update errata.txt after noted /data errors are fixed. ==================== Calibrated Data Sets ==================== /catalog -------- dataset.cat o Verify this note was added for the ENCOUNTER data sets: For encounter images, geometry values relating the target were calculated using the impact site as the origin of the vector. This was the information available in the encounter SPICE SP-kernel delivered by the DI Nav team. The comet center wasn NOT used when calculating geometry for encounter images. /calib ------ o Some calibration FITS files must be modified because the FITS format is inapproprite or incorrectly implemented as 'images'. Consider converting the affected files to ASII tables or FITS table extensions without a primary data unit. Modify the labels accordingly to correct the descriptions. Here are the suggested conversions: /calib subdir suggested conversion --------- -------------------------------------------------------------- /abscalvs Ascii text since there are only two values /abscalir Ascii text or FITS table since there are 3 types of values and >3300 entries /adclut For VIS and IR, 16384 entries so use FITS table? /decomprs FITS table? /psf It's a dummy for IR, so toss it. For VIS, ask Jianyang how to make this into an ascii table. /xtalk For VIS and IR, ask Olivier how to make this into an ASCII table? o Question for Mark Desnoyer: How are IR adclut (FITS) and decompres tables used? For IR and VIS /decomprs, got 256 samples and 2 lines. What does each line represent? For IR and VIS adclut, what are the 16384 values?