Week summary!
So! A week summary.Red sequence fit ... there are some discrepancy with models in some cases, but that needs more attention than I can give now.
Applications.
Studying for my exam. That will have to go faster, otherwise I don't finish it on time.
Carolina Nunes, student at ESO, found here at this blog that I would be able to help here with GALFIT problems :-) This is not the first time that happens, some months ago, Mathias Zetzl, a colleague from Goettingen found information about KCORRECT here. Unfortunately that time it was out of my knowledge of the code. Also some IDL hints were asked here. I guess I could help Carolina.
Good to know that some information here is useful to the people!
With Carolina's problem, I decided to but basic hints here about GALFIT.
To run it properly, aside from the good initial guesses, a stamp that covers as much as possible from the galaxy and a good decision about which objects should be simultaneously modeled and which should be masked there are a list of things you must take care.
Coming to masking. As I learned from Marco Barden (Innsbruck) if an object is inside the isophotal radius of the galaxy it should be simultaneously modeled, otherwise it should be masked.
So, you need a correct header indicating the GAIN, RON and NCOMBINE, image in COUNTS and not COUNTS/SEC or ELECTRONS and the original sky level.
That is critical if you're using HST images. Multidrizzle, by default gives you the output image in counts/sec and sky subtracted. So you must multiply it by the exposure time and put back the sky level which is written in the header under "MDRIZSKY".
In the case of Carolina, her fits were giving a CHI^2 of about 500 ... putting a dumb sky back it went down to 5 ... so with the correct sky level it will go down to the proper value.
Labels: applications, GALFIT_hints, hi-z morph
posted by Cris Da Rocha @ 11:59 PM
0 Comments:
Post a Comment
<< Home