Welcome to iraf.net Thursday, March 28 2024 @ 10:19 PM GMT
gruntle |
02/23/2006 08:48AM (Read 4847 times)
|
|
|
Status: offline
Registered: 02/21/2006
Posts: 2
|
Hi,
I'm using addstar in the daophot package to simulate stars (based
on a psf model computed with the psf task).The task runs ok, I get no error messages and the resulting
simulated images look nice. I'm adding ~1000 stars to a
~2000*2000 pixel image and the psf model is computed with a
varorder of 2 (i.e. the psf can vary over the frame).When I run addstar I add all of the stars using the "simple file"
option as input (i.e a list with xcoord, ycoord, mag, id) and with all
stars set to the same magnitude. The problem I have appears when
I do photometry of the simulated stars. I measure the magnitudes
of the stars using aperture photometry (with a radius of ~5FWHM)
and then compare the mean estimated errors to the scatter of the
magnitude distribution. Two things then stand out as being a bit
strange:
1. The mean/median magnitude of the sample is not the magnitude
I specified in the input file (the measurements give magnitudes
about ~0.3 mags fainter than the input, this increases slightly when
going towards the noise limit).
2. The scatter of the magnitude distribution is about two times the
mean photometric errors. I would expect the scatter to be the same
as the mean error more or less.Have anyone encountered these problems before? I need to be
able to control the brightness of my simulated sources for the work
I'm doing, so point 2 is really the most problematic thing (1. can be
solved by just remeasuring in the simulated image).Would be grateful for any help,
cheers Jens
|
|
|
|
phil |
02/23/2006 08:48AM
|
|
|
Status: offline
Registered: 11/03/2005
Posts: 13
|
Hi Jens,I'm really not sure how to answer your question. But, perusing the
source and the docs brings up some questions which you may
consider. I apologize if the following seems obvious/elementary.Using varorder=2 may be the cause of some of your results, e.g.,
while phot'ing, the larger spread would give you less flux. Hence the
median/mean magnitude of the sample would end up being fainter,
particularly due to those that are affected by the psf variability
order.Have you tried with varorder=-1 or 0 to see if you get results that
are closer to what you expect?It seems that varorder=2 would also increase the scatter in the
magnitude distribution as a result of decreasing flux for a given
aperture size. On the other hand, increasing the aperture will give
you more flux, but may end up including nearby stars.Hope that helps.Phillip Warner
NOAO
|
|
|
|
| |
|
Content generated in: 0.06 seconds |
|