Message-Id: <E9BE3A06-F24C-4855-8426-A9E6FA2778B2@iap.fr>
From: Henry Joy McCracken <hjmcc@iap.fr>
To: teralist@iap.fr
Content-Type: text/plain; charset=US-ASCII; format=flowed; delsp=yes
Content-Transfer-Encoding: 7bit
Mime-Version: 1.0 (Apple Message framework v919.2)
Subject: [teralist] bilan tardif hjmcc
Date: Sun, 27 Apr 2008 23:45:09 +0200
Reply-To: teralist@iap.fr
Precedence: list
Sender: petidomo GRAND POSTIER <petidomo@iap.fr>


bonsoir emmanuel,

mea culpa et desole pour le retard-- un sortir imprevu au resto...

- This week I was working on the astrometric and photometric solution  
of the WIRCAM d3-data and also trying to resolve problems related to  
cross talk.
- Many mails were exchanged between loic, chris willot and myself over  
the origin of the black marks on the final d3 stacks. Loic thinks that  
the black marks come from incorrect crosstalk removal. He will supply  
to us images which have not been crosstalk corrected and then the  
stacks will be redone.
- If the black marks are due to to persistence effects, then they  
should be less important with a larger number of images in the sky  
background stack. What are the optimal numbers of images to include in  
the sky background? To answer this question, QF has been run on 2000  
Ks images BEFORE PROCESSING; the plots have yet to be made showing the  
background variation as a function of time. This should help us to  
quantify how rapidly the sky varies during each night. Then,  
potentially, we will re-do the sky subtraction with a larger number of  
sky frames.
- With patrick, tests were made with the new script which computes the  
zero-points chip-by-chip. There are some important differences between  
the two scripts; it turns out that Herve's script uses a very small  
matching radius and consequently finds many smaller numbers of objects  
wrt to patrick's. Another question which is not yet resolved is what  
to do for the cases where there are very small numbers of stars in  
each detector. It seems the reason the solution is unstable for the d3- 
images is precisely because of the small numbers of matching stars in  
some D3 frames
- PI data: Data from Mr. Hutchings should have been processed at the  
end of September (when the request was lodged: unfortunately this  
slipped under the carpet and was not processed; the pre-processed data  
has also been removed from CFHT, so loic will have to re-do this part  
of the processing. I think we have to be careful how the processing  
priorities are set.