Wednesday 29 October 2003 ========================= 12:10 Intro Present: Anne Raugh Telecon: Dan Scheeres Tony Farnham Frank Crary Mike DiSanti Ray Newburn Dan Britt Boris Semenov Mike A'Hearn Tyler Brown Chuck Acton Ed Grayzeck STARDUST GENERAL: ================= Ensure that all data sets have the most recent versions of all applicable catalog files. See notes from Raugh and Brown as well. SDDFMI_0001 (DiSanti) ===================== Data tables are both long and wide. Probably should be split. In column 24, (ON/OFF values) quotes are misplaced for the field size ("ON" rather than "ON ") FORMAT statement values for fields with multiple items are not accurate. Split COLUMNs with items for vectors into separate COLUMNs. Check if the NOT_APPLICABLE_CONSTANT and UNKNOWN_CONSTANT values actually occur in the data file, and make sure the flag values actually fit in the field. The CSV versions have one more column header than actual column (an extra angle label on the end). Chuck notes: There are cut-and-past errors in the aareadme.txt file. Verify the number of rows in each table file. START/STOP_TIME in label match .tab but not .csv. Remove "software/" subdirectory and delete references to it from aareadme.txt, voldesc.cat, etc. DISCUSSION: What level of effort is appropriate for correcting liens? This is a negative result; it is unclear what data might come from the Wild 2. RECOMMENDATION: The PI's null report after the Annefrank encounter needs to be formally captured and incorporated into the dataset documentation. 01:17 Alfredo Vinas arrives. SDNC_0001 (Britt) ================= General comment: This is potentially a very useful and interesting engineering data set. Would like to see the information about the filter wheel failure at a higher level, to explain the lack of color data in a mission that advertised color filter observations. Also needs to be mentioned repeatedly in relevant areas throughout catalog files. Annefrank index is not as complete as early cruise, and should be. Expand COLUMNs with ITEMS in the index labels. TB notes that the Annefrank index label description doesn't match columns in the table. Problem starts at column 3. DB asks why there are different numbers of columns in the Annefrank vs. early cruise data. CA notes there were different data preparers, changing WINDOW standards and modified software. Index files for both these data sets should have identical columns and formats. Remove software/ directory and references to it. TB notes there are keywords issues. Data file labels need LABEL_RECORD keywords. Note: Small bodies users have strong preference for FITS images. If resources allow, these data should be converted to FITS files. Also, Wild 2 images should be supplied in FITS format. SDNC_0002 (Britt) ================= Personel.cat file seems pointless. Try to avoid including this if possible. Find and include reference to Annefrank paper. If JGR will let us republish, include the paper itself. Similarly for any and all other relevant papers. Gather any available derived data products and include them in this archive. SBN will need to contact team leaders to ask for these. There is a mention of this in the SDMP. If there are corrected and desmeared images, they should also be gathered and included. CA notes the desmearing problem is still being worked, but DB suggests that the current state should be captured anyway. Data label style is very good. Can this be applied across all data files? CA notes the varying label styles are representative of the changing data preparation circumstances. Would like to see the same keywords used consistently throughout the various data sets from this mission from science encounters. These labels do not indicate anything about processing history. Add keywords that will clearly indicate that no processing has been done. Is it possible to separate asteroid images from other images in the data directory? This is a convenience issue. Ray N. will examine archive documentation to make sure it accurately reflects what has been provided (images, i.e.). DB asks that the calibration files be included for Wild 2 if at all possible. Ray N. notes that imaging is a low-priority task for the mission. BORRCTIO (Britt) ================ Seek out and change all "Muller" to "Mueller". Change TARGET_NAME to be "CALIBRATION FIELD" for "SA 112" targets. SDSP_1000 & DS1SP_1000 (Scheeres) ================================= Add statement to errata.txt file that states this line termination treatment is standard practice for SPICE volumes in the PDS. NOTE: Image time in the labels is based on ephemeris time, and was not calculated with reference to the NAIF SCLK kernels at all. The onboard ephemeris counter was periodically re-synchronized with ET. Images were tagged with this ET. (Other telemetry was tagged with SCLK times, which wandered.) The precise history of the ET synchronization is not known. NATIVE_TIME in these images is ET referenced to 1998 with "a little error" (per Boris Semenov). There is a log file which shows time difference larger than would be expected (30s rather than a few seconds). This needs investigating. These are the final SCLK kernels, so there is no update required for these, and confidence level in using this to produce ephemeris time is high. We (SBN) need to check the provenance of the discrepant log files. Mounting evidence indicates it is incorrect. SDSP_1000 release.cat has the wrong DISTRIBUTION_TYPE. SDSP_1000 index file does not list all labels on the disk. SDCIDA_0001 (Scheeres) ====================== Include a brief tutorial on measurement technique and scientific application, if possible. Need more and better calibration examples and explanations. Without this, the data are of questionable value. "Example" directory in the documentation folder is a bit obtuse as a name. At the least this should be renamed to more specifically indicate the content. The octave script is obscure. More information is needed to document the octave application, so that some sense can be made of the script itself. The actual algorithm would be much more useful. If it is possible to get a better explanation of the current example, do so. Determine whether the calibration PDF file should be converted to ASCII. calexmpl.pdf ------------ Fix typos in calexmpl.pdf. The procedure outlined in calexmpl.pdf is not sufficiently explicit for the entire process (the phrase "make sure that" is used without explicit instructions for how to make sure, e.g.). Mathematical steps should also be explicitly stated where they are not currently. Need details of how to find Hydrogen and Silver spectral peaks are. Section 2, step 6 is a comment, not a processing step. Section 3.1 is missing the error estimates they apparently intended to include (contains only "Error estimates!!!!" note). ... See scheeres.ppt for detailed list of reviewer liens. Consensus that the description of the calibration procedure is not adequate. If an adequate description of the calibration is not supplied, the data will be rejected. Labels should be detached from table files. These data cannot be accepted without additional review. ___________________________________________________________________________ ___________________________________________________________________________ ___________________________________________________________________________ Future Stardust Data Sets ========================= DB wants to see derived products, or at least a way to get to them via the archived data. This is particularly important for the comet encounter. Are there mission plans to collect ground-based data? The opportunity for simultaneous observations will be severely limited by solar elongation. SBN will confer with Ray N. about data sets that might already have been identified. SBN will also contact Ray N. about Wild 2 design models.