CASB Report for May, 2000

Highlights for Division Meeting

· Completed scanning half of all the expected UKST-IR plates.

· Processed over 240 POSSI-E plate scans for GSC-II this month; total now is over 40% of the survey.

Community Affairs and Communications

· McLean is member of SOC for upcoming NVO workshop (June 13-16)

Project Management and Technical Infrastructure

· The new COMPASS server configuration with 2TB disk was completed.

MISSION SUPPORT

HST Operations

· Reviewed 22 Guide Star Problem Reports (GSPRs). Of these, 16 were due to fine-lock failures caused by close doubles, and 6 were search failures. Five of the search failures were galaxies misclassified in GSC-I, but the sixth object was simply fainter than expected. Bad GS Alerts were filed for all 22 GSPRs. Several more GSPRs were received at the end of the period.

· Investigated FGS search failures for several GSs derived from a Black Birch astrometric plate that includes the bright star Canopus. Processed this plate using the GSC-II pipeline processing. The results of the new astrometric and photometric calibrations indicate that the quality of the original GSC-I positions and magnitudes were adequate for FGS acquisition. The investigation continues.

· Continuing to work on the requirements of the BOP GUI, to be made available to STIS, ACS and COS for cycle 10. Ron Downes, who is a STIS contact scientist and also a member of the APT/VTT team, has provided clearer requirements for not only the automated version of the Bright Object Checker, but also for the version that hopefully will be part of the VTT. (APT is the Astronomer's Proposal Tool and VTT is the Visual Target Tuner.) These requirements will be used to create an implementation timeline with an ultimate completion of January 2001.

· Plan to have the automated version of the Bright Object Checker ready for release by next week.

NGST Guiding Studies

· Spagna is preparing report for submission to NGST division.

DIGITIZED SKY SURVEY

Plate Scanning

· Scanned POSSI-O plates until the latest ROE shipment arrived and then started scanning the UKST-IR plates in that shipment.

· As of this report, completed scanning half of all the expected UKST-IR plates. The current shipment contains both IR and moderately-short exposure R plates (henceforth referred to as RS plates) from the SERC I/SR Atlas of the Milky Way.

GAMMA Microdensitometer Systems

· GAMMA 2 is back in full production scanning. Scan diagnostics show excellent performance of the mechanical and optical system.

· GAMMA 1 is currently down for mechanical work. The scanning program was crashing due to S/W time-outs. It was thought that the time-outs were due to mechanical problems. So far, mechanical problems were found and fixed, but the time-out problem still exists. More tests are planned.

· Affected a workaround in software for the broken terminal server that controls the console on one of the scanning machines. A change to a task parameter file redirected the X, Y, and density information from the physical displays on the console to a DECTerm window. The display, while not aesthetic, is sufficient to support scanning. Additional software changes required replacing the original displays in a more user friendly fashion will be undertaken as resources permit.

Data Compression/CD Production

· Awaiting the receipt, scanning, and compressing of the final plate from Palomar before the last four J plates can be archived. Meanwhile, compression of POSS-II N AAO-R and PPARC-ER plates continues.

Archive Maintenance

· Performing quality assurance of the MOD archive, starting with the AAO-SR survey since that collection has known errors. Fixing errors when possible; some discs will be rewritten when the QA is complete.

Software Development

· No progress reported in investigation of a GETIMAGE failure with the DSS-I, first reported by the NAOJ Astronomical Data Analysis Center.

GSC-II

Plate Processing

· Processing of the POSS-I red plates continues at a very efficient pace. Over 40% of this survey has now been processed. Several small problems encountered regarding astrometric solutions are currently under investigation.

· Created a simple IDL tool to interactively flag bad or corrupted objects from the oop records. This tool was created in response to a number of POSS-I red archival data containing small regions of corrupted data that probably dates back to problems with an older media (video grade 8mm tape).

COMPASS Database

· Loaded the production GSC-II COMPASS database onto the new COMPASS 2 Terabyte server. This will provide a fully online access to all second generation processed and calibrated plates required for GSC-II catalog construction.

· Production object-matching is now in progress on the new server.

· Preparation for the next delivery of the GSC-II catalog is underway. The near term plan is to complete recalibration and object-matching on all the plates currently loaded in COMPASS. The export is scheduled to complete in the coming month.

· Benchmark testing with the new system confirmed a marked improvement in I/O performance with an increase to 50MB/sec as compared to the old system with 3-5 MB/sec. The GSC-II catalog production pipeline tasks are taking approximately one-half the time of the old system pipeline.

· Performed an assessment of the various HTM 1^st levels of storage in the database to improve understanding of the different disk capacities. This was necessary because the disk configuration maps directly to the sky partitioning of the database. Of the 8 regions on the sky, N0 was shown to contain 2x the number of calibrated sources.

· Rebuilt the testFDB on the old system under the new server host, CHART. This will provide a separate staging area to the system lock-server and make calibration development and QA tasks perform more efficiently.

· Successfully loaded GSPC-II as an external reference catalog into the testFDB.

· Grant funds were used to purchase an Objectivity license for the CASB Web server to allow external access to GSC-II.

CALIBRATIONS: Astrometry

· Nearly completed testing the equations describing the astrometric Schmidt plate distortions caused by the bending of the plates in the telescope plate holder. Essentially, the residuals determined from the mask process are used as data for testing. The standard coordinates of the mask points (and their delta xi and delta eta) are converted into polar coordinates because the distortion pattern is easier to represent in these coordinates than in rectangular coordinates.

· A general least squares routine is used to solve for the coefficients of the delta theta and delta r. The least squares routine is written to solve for one equation (one for delta theta and one for delta r) for the entire plate, or for eight equations independently (one for each symmetric region), or for eight equations while enforcing overlapping objects as a restraint on the equations.

· Various different models have been tried. The actual astrometric mask is compared to a mask created with the coefficients determined from the above least squares fit. A refinement of the selection of overlapping objects is currently underway. At this stage it does not appear that the equations will be able to completely replace the astrometric mask, but possibly using the equations before the mask is created can improve the accuracy of the mask. At least this method should help to find abnormal plates that do not fit the astrometric mask.

CALIBRATIONS: Classification

· Experimentally modified the decision tree program to use weights during the pruning process. This permits a more sophisticated assessment of tree performance in terms of contamination and completeness of various classes, rather than a simple overall right/wrong error rate. The implementation of the weighting needs improvement, but preliminary results are very encouraging. On an intermediate latitude test plate, 4.5% of the objects changed classification. Nearly all of these objects down to 15^th magnitude were visually inspected: the results were better for 68% of them, and worse for only 17%. One notable improvement was “recapturing” many galaxies that had been previously classified as defects. Most of the objects with worse results were objects that had been through the deblender. Experiments to improve the results on these objects by including more deblended objects in the training set are underway.

· The classify task was modified in develop to produce “leaf” information for each object classified, and IDL routines were written to read and display the leaf information. This provides surprising insights into the behavior of various populations of objects through the decision trees – for example, some leaves contain objects of all magnitudes, some contain objects of only a very small magnitude range, and some seem to contain bimodal populations, that is, bright objects, faint objects, but no intermediate objects.

CALIBRATIONS: Photometry (GSPC-II and DSS-I and -II)

· Photometric masks for POSS-II J and F and for southern J and F surveys are available for testing. Preliminary results show that the RMS in magnitude dispersion has an improvement of 60% for the J bandpass and 70% for the F bandpass on the basis of a single equatorial plate pair for each color.

· The GSPC2 data archiving to DLT is still in progress.

· Prepared an export fits GSPC-II catalog, which is ready to be released to the community. It contains unique names (for multiple observations), and averaged values of ra, dec, v, verr, r, and rerr, plus b and berr when b photometry is available.

CALIBRATIONS: Apply_Calibs:

·

GSC-II General Software Development

· A small bug was found and corrected in the PLATE_SOLUTION task.

· Defined a new bandpass code for the 098-04 + RG630 emulsion/filter combination used for RS survey plates recently received.

· A PUTOOP routine was developed for the XSPAM library.

GSC-II Science

· Fabrizia Guglielmetti is processing 11 plates to correlate and find the proper motion of the cluster NGC2264. This study is more difficult than expected. She started working on five plates containing the cluster: three from the POSS-II survey and two from POSS-I. The POSS-I plates contain part of the cluster only in their southeast corners; so plate-matching results are very poor. Moreover, processing of one POSS-II plate produced wrong classifications. During the coming month Guglielmetti hopes to find a consistent solution for the cluster members.

Appendix – CASB Statistical Summary

SCANNING

Survey		Goal	Previous Month	Current Month	Current Total	% done	Notes	Priority
POSS-II J	897		898	0	898	100		1
POSS-II F	897		907	0	907	100	S4	1
POSS-II N	897		787	0	787	88		1
AAO-SES	606		606	0	606	100	S4	1
PPARC-ER	288		270	0	270	94		1
AAO-SR	118		118	0	118	100	S2	-
UKST-IR	894		412	37	449	50		2
SERC-J/EJ	447		443	0	443	99	S1	2
POSS-I E	106		106	0	106	100	S3	3
POSS-I O	880		115	26	141	16		3
Rescans			18	0	18

• S1. Low priority program to rescan crowded fields (|b|<30) originally done with 25 mu scans. The total number of S scans includes the 2 for the LMC and SMC plus 4 others with |b|>30 performed for research or STIS BOP support. Another 30 plates have been added to replace those that were repaired before inclusion in DSS-I.
• S2. This survey is an on-going low level-of-effort collaboration between the STScI and the AAO. About 25-50 plates have been added in each of the past few years.
• S3. Low priority scans of the POSS-I -24 deg and -30 deg zones.
• S4. Includes recent scans of better grade plates.

COMPRESSION

Survey	Goal	Previous Month	Current Month	Current Total	% done	Notes	Priority
POSS-II J	897	893	0	893	99		2
POSS-II F	897	877	0	877	98		1
POSS-II N	897	179	32	211	24		3
AAO SES	606	572	0	572	94		1
PPARC ER	288	245	0	245	85		1
UKST-IR	894	0	0	0	0		3

MO DISK ARCHIVING

Survey	Goal	Previous Month	Current Month	Current Total	% done	Notes	Priority
POSS-II J	897	902	0	902	100		4
POSS-II F	897	878	0	878	98		3
POSS-II N	897	194	11	205	23		4
AAO-SES	606	576	11	587	97		3
PPARC-ER	288	250	0	250	87		3
AAO-SR	109	73	0	73	67		4
UKST-IR	894	0	0	0	0		5
SERC-J/EJ	894	892	0	892	100
SERC-J/REPAIR	48	48	0	48	100
SERC-J/15mu	449	449	0	449	100		5
POSS-I E	933	928	0	928	99		2
POSS-I REPAIR	198	198	0	198	100		2
POSS-QV	613	613	0	613	100		-
SERC-QV	94	94	0	94	100		-
GPO/BB	88	88	0	88	100		-

GSC-II PLATE PROCESSING

Survey	Goal	Previous Month	Current Month	Current Total	% done	Notes	Priority
POSS-I E	880	121	241	362	41		3
POSS-QV	613	2	0	2	0		3
POSS-II J	897	887	0	887	99		1
POSS-II F	897	897	0	897	100		1
POSS-II N	897	44	0	44	5		4
SERC-J/EJ	894	822	0	822	92		1
South-F	894	815	0	815	91		1
UKST-IR	894	2	0	2	0		4
South-SR	118	110	0	110	93		2
South-QV	94	94	0	94	100		2

GSC-II COMPASS DATABASE LOAD (total number entries = 1,089,000,000)

Survey	Goal	Previous Month	Current Month	Current Total	% done	Notes	Priority
POSS-I E	880	5	0	5	1		3
POSS-QV	613	2	0	2	0		3
POSS-II J	897	511	0	511	57		1
POSS-II F	897	743	0	743	83		1
POSS-II N	897	4	0	4	0		4
SERC-J	894	128	0	128	14		1
South-F	894	473	0	473	53		1
UKST-IR	894	0	0	0	0		4
South-SR	109	13	0	13	12		2
South-QV	94	57	0	57	61		2

GSC-II Dataset for V2.1.n delivery (1791 Fields, 3438 plates) (P1)

Item	Count	% done	Notes
Plates available	3417	99
Plates processed	3355	98
DB Load	1907	55
Current Calibration or Recalibration	1887	55

• P1. Defined as all 897 POSS-II fields with 1794 plates, plus the 894 southern

Fields, minus the 72 southern equatorial fields, with 1644 plates.

GSPC-II OBSERVATIONS

Survey	Goal	Previous Month	Current Month	Current Total	% done	Notes	Priority
POSS-I	584	559	0	559	96	P1	1
POSS-II	314	169	0	169	54		1
SERC	894	868	0	868	97		1

GSPC-II REDUCTIONS

Survey	Goal	Previous Month	Current Month	Current Total	% done	Notes	Priority
POSS-I	584	559	0	559	99	P1	1
POSS-II	314	169	0	169	54		1
SERC	894	868	0	868	97		1

• P1. The current totals have been revised from earlier ones.