Agenda And Minutes for 12/09 10am CST¶
1. Dial Toll-Free Number: 866-740-1260 (U.S. & Canada)
2. International participants dial:
Toll Number: 303-248-0285
Or International Toll-Free Number: http://www.readytalk.com/intl
3. Enter your 7-digit access code, followed by “#”
Your access code: 7691719
- The S-G separation Challenge
- What needs to be done: SVA1 Catalog Punch List
- Mangle plans
- Possible issues with star/galaxy separation in non-homogenized coadds
Attending: Eli, Diego, Bob A., Gary, Nacho, Ross, Huan, Steve F., Steve K, Josh F., Angelo, Alex D-W, Keith B, Aurelien, others...
1. Nacho discussed The S-G separation Challenge. He will have standard training/validation data sets released very soon. The first data set is COSMOS, without SLR correction (though the resulting color shifts would be on the order of a percent). Eli asked about performing additional metrics on classifiers that release probabilistic assessments...this has been pushed back to further rounds.
We also discussed amassing other HST fields for S/G validation, including the cluster fields (Bullet, RXJ2248), and SN C3.
It was decided that this effort will continue in parallel with the general SVA1 catalog effort. For the SVA1 catalog, we will use the "modest" star/galaxy separation.
2. We discussed what needs to be done to release an SVA1 catalog in the SVA1 Catalog Punch List. Nacho pointed out that LSS has a "Copper" catalog that incorporates many of the ideas that we have discussed here. The goal is to consolidate these efforts (hopefully with a minimum of duplication!) so we can use the same basis catalog.
The issue of the incomplete mangle coverage was raised. We will probably have a compromise plan where we have a full sample which includes COSMOS + all SN fields, and a sub-sample which has mangle maps. Science needs which require mangle maps can use that sub-area.
Diego discussed strategies for combining the mangle molygons from the database with Huan's estimation of good and bad tiles to choose a footprint. However, Eli pointed out that this only gives an area coverage within each tile, and does not give the boundaries of the covered regions within that tile.
The question of area vs depth was raised, with no real comments. Eli will map out the regions and make a well-informed but arbitrary starting point.
One hold-up on a final catalog is the SLR calibration. Bob pointed out that we have been running by calibrating against a synthetic locus until now, and it would be good to use an empirical locus. Doug will assist in selecting good observations from S82 standard star observations to create the locus. This should take <2 weeks from now to a fully recalibrated survey.
We will also confirm that the locus is consistent across our chosen footprint. We know the locus is different in the LMC region, so we will have to cut out this region from our analysis (where extragalactic astronomy is also very challenging behind the high stellar density). SLR comparison tests will determine whether our declination cut should be -60 or -61 in the SPT-E region.