IPAC 2MASS Working Group Meeting #77 Minutes

IPAC 2MASS Working Group Meeting #77 Minutes, 10/31/95

Attendees: T. Chester, R. Cutri, T. Evans, J. Fowler, G. Kopan, B. Light, H. McCallon

AGENDA

Friday SDS Delivery
D. Monet SDS Responses

NOTE: The Working Group met as usual on October 17 and 24, but the issues discussed all involved topics that are intended to be documented elsewhere (e.g., SDS and SIS topics); since no minutes needed to be issued, these meetings will not be included in the numbering.

DISCUSSION

Friday SDS Delivery -- R. Cutri reported that he will need the latest SDS version for each subsystem on Friday, November 3, since the team members going to Amherst for the upcoming review will be leaving Monday morning.
D. Monet SDS Responses -- A substantial response to the POSMAN and PIXPHOT/FREXAS SDS's was received from D. Monet, and the remainder of the meeting was spent discussing some of the points made in this communication. The issues raised by Monet can be placed in four categories:
- A.) items requiring better communication, after which they are probably of no real concern;
- B.) items that have already been corrected;
- C.) items that are debatable;
- D.) items of obvious merit that should be incorporated immediately.
Categories (A.) and (B.) were not discussed beyond identifying which items should be classified under those headings. The existence of items in category (C.) requires that SDS's include a section on liens, so that we may go forward on the general concepts while details are resolved. Items in category (D.) were accepted for adoption in the design and will appear in revisions of the SDS's affected.
An example of an item in category (D.) is the need to reestimate Read1 point source positions in FREXAS according to the same procedure as that applied to Read2-Read1 point sources. Everyone agreed with Monet's statement that everything possible must be done to minimize differences between the way Read1 and Read2-Read1 point sources are handled, since the majority of high quality astrometric point sources will probably be processed as Read1 objects.
This same principle runs into category (C.) however, when it comes to processing Read1 point sources for PSF fitting in PROPHOT in the same way that Read2-Read1 point sources are processed. PROPHOT computes improved positions for the point sources it processes, and this could potentially put the Read1 and Read2-Read1 sources on a different astrometric scale as the design stands. On the other hand, the Read1 source positions are much more vulnerable to seeing disturbances, since their much shorter integration time results in greatly reduced averaging over the seeing jitter. The additional computational burden of passing Read1 sources through the full PROPHOT processing, furthermore, is a potential backbreaker. It needs to be established if the potential improvement in R1 positions is worth the potentially large hit in resources and development time. It is anticipated, therefore, that this item may not be resolved in the near future, and the task of settling it to the project's satisfaction may have to be carried as a lien in the meantime.
Another example of an item in category (C.) is inclusion of scale factor and twist angle reestimation on a frame-by-frame approach in POSFRM. Monet has pointed out a number of phenomena that could cause these parameters to vary too rapidly to be tracked by the longer-term estimation algorithms described in the POSMAN SDS. These algorithms were designed with nominal hardware performance in mind, and it was the understanding at IPAC that (a.) the project policy was to design that way, and (b.) the hardware requirements limit such effects to an extent that eliminates this particular concern. In this case, it is felt that a ruling by the science team is needed before making any major design changes in POSFRM, and the item should be carried as a lien in the meantime.
Other items were discussed, but documentation of them will be deferred to the appropriate channels (primarily SDS's).