Berkeley Deepsearch Meeting Notes, 2001 January 17

Agenda

General News

There is a tentative Spring Observing Schedule on the web. Stop by Jeane's office to tell her when you will NOT be available to help out with the search run (either at the telescope or at LBL).

CFHT: We proposed for 4 deep field and 20 shallow but we are looking at more like 3 deep fields and 10 shallow. As Greg sees it we can try to do 4 deep fields 3.5 hours, 1 itermediate 10 min (all of these are in the HST region), and 8 shallow. We already have I-band 6-hour ref in 1 field already and 2 others have 2-hour I-band. All other shallow fields have archived data

Greg's idea is to move the HST field a little so as to cover all the high-resistivity CCD's on CFHT. We'll lose only a small bit of coverage.

ESO has given us two half-nights on SUZI2. Rob suggests using them for final refs.

Rob notes that the groupwork and collab web pages need to be majorly overhauled, even though it is unlikely anyone will ever do this.

Alex asks should we think about getting photometric redshifts for our search fields? Seems like there are only 2 colors with poor signal to noise available. Probably not worth doing it, although someone might want to look into whats avaiable.

AAS News

A poster at the AAS claimed that light curve error bars can be improved by almost root two by fitting differences instead of subtracting a final ref. Saul's discussions with the authors indicate that the procedure is essentially what we do (when we let our zero point float) as long as you are using aperature photometry instead of PSF fitting. Preprints are on the way.

Lou Strolger's poster hinted at a new parameter in the light curve correlated with SN1a type. It may be possible to distinguish subtype by light curve shape alone with good enough late time coverage.

Rob and HST photometry

 The following is a recap of the talk Rob presented at the AAS meeting last week.

We have several SN with ~4 HST points. Alex Conley's transformations were only able to get positions to ~1 pixel, larger than we'd like. So Rob PSF fit the images using Alex's positions as an initial guess and these free parameters: PSF center(2), background height and slope, SN height

The SN used were: SET F: (3 in R&I, 1@z=0.86) -- SET G: (7 in R&I, 2@z=0.7) (note: .988 one not a 1a)

The PSF Fits seemed to work -- chi^2/v was on average about 1.2 (with v a few hundred)-- a little high, but doing okay. (Aperture fitting would be a good check of this but presently not working.)

LTCV FITS: Lightcurves were fit in R&I with SNminuy -- they looked okay visually.

one issue: the SN distibrution was too BLUE (3 reddened, 8 bluened) As a future check Rob plans to make fake images and run the procedure on them.

Another thing to think about -- what is the best way to fit the light curve for combined HST and ground photometry.

The next group meeting is two weeks from today -- Michael will fill us in on the status of the SN factory and problems occuring in the subtractions.