Agenda And Minutes for 4/14 10am CDT¶
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
- UFIG - Chihway
- Testing Completeness with UFIG
a. Plus... [[des-lss:Systematics_tests#Detection-completeness|some old (and new) plots on detection completeness using DC6B and Gold]] - Nacho.
- Trying out new SG separation codes on SVA1-SPTE? (see attached files for discussion) - Nacho
- Other updates?
Attending: Eli, Chihway, Diego, Nacho, Angelo, Recardo, Gus, Keith, Alex D-W, Douglas, Sahar, Tommaso, + others
1. Chihway introduced the UFIG SVA1 coadd simulations. The general description is here and the specific SVA1 stuff is here . The Ultra-Fast Image Generator is currently set up to make coadd-equivalent images with systematics matched to SVA1. Chihway emphasized the caveat that whatever we do with these runs should be kept to the limitations of the simulations! Most importantly, be aware that these are constant PSF and noise, no artifacts and CRs, etc. (See list of limitations on the SVA1 UFIG page). However, these could/should still be useful for (eg) testing completeness and star/galaxy separation over a wide range of systematics. That is, the seeing, airmass, and depth parameters will be well sampled, even if the local correlations are not correct.
In order to test the local correlations, Chihway will do a small test region with SE exposures that can then be swarped and sextracted. Although these SE images will also have a constant psf, the coadds generated from them will not, and will be very useful to test our runs on the coadds with psf varying in a (relatively) realistic way.
One question that came up was whether it would be helpful to host these data on the portal. At the moment, the UFIG images themselves are available, but in the future these large images may not be distributed unless there is a need for them.
Gus asked about plans for the 5000deg^2 simulations. At the moment, the question is how to do the systematics: for SVA1 we know the depth and the systematics everywhere so these can be simulated. Should we use the target ranges for the DES survey? At the moment, Chihway is going to focus on Y1 depth.
2. We discussed completeness. In particular, the first test field on UFIG: Testing Completeness with UFIG. This looks promising, but there is some worry about the fact that SExtractor is underestimating the errors for the galaxies (very much so at the bright end). Is this a problem with the UFIG sims or with SExtractor? Can it be fixed with a simple systematic error floor? And what happens with MAG_MODEL and MAG_DETMODEL? Eli will investigate.
3. Nacho briefly brought up completeness tests on DC6B and with COSMOS. The dip in the bright end seen in the data is being investigated, but is not the same cause as the drop in the UFIG catalogs. Nacho also showed the galaxy obscuration around (fainter) non-masked stars, which is significant within ~10". Eli will investigate how these effects look in the UFIG catalogs.