iraf.net

Hello,I am trying to use starlist and mkobjects (located in noao>artdata) and I have
a little trouble because I could not find more info except the help pages which
did not help my questions.I am trying to derive the completeness limit for a set of data in B,V,R,I filters
and I want to create an artificial star list. In the keywords there is a minimum
and maximum magnitude. Can this be magnitude in any filter or is it only the
V filter apparent magnitude? Also, I am not sure I understand what the mzero is.
Is it the distance modulus that enables us to convert from apparent magnitude
or is it something else?
Also, after I run the starlist and the mkobjects do I need to run "phot" or
it is not necessary and I can just include the outcome of starlist to my list of stars?Any help you be appreciated, especially if you have used the artdata package
to derive completeness limits.Thank you,
Andrei

Hello Andrei,To estimate completeness you make an artificial image(s), or add stars to a real image, then process the image(s) exactly as you would the real data. So you do have to run phot on the images. By adding a range of magnitudes that extends fainter than can be detected you then analyze how many stars of each magnitude were detected, measured, and the accuracy (how well the measured magnitude matches the known input magnitude).As far as the magnitudes the artificial data have to be matched to your expected magnitude zero point. This is the factor that converts between how many ADU are in a star aperture and what reference magnitude that corresponds for a particular filter and instrument. For example, if the instrument you are using or want to use with the expected exposure time gives 100 ADU in a 2 arcsecond aperture for a star of 24th magnitude then you want to set up mkobjects to produce 100 ADU in that aperture when starlist has a magnitude of 24. This is all done with the magnitude zero point parameter. Once you have the right value then the counts you would get at any magnitude, even 30th magnitude, will scale appropriately and you can measure completeness and report something like 90% of stars at 25.1 magnitude are detected (in uncrowded fields) at a 5 sigma level.For a completeness study you probably don't need to simulate an actual distribution of star magnitudes. In fact you probably want to have equal numbers of stars at each magnitude. The purpose of the luminousity functions in starlist is to make simulated images that have a distributions of brightness corresponding to an astrophysical situation or model. The mzero parameter is what characterizes the luminosity function in the particular magnitude system being simulated.In a study such as you are doing, you probably don't need to do each filter. the only difference in the filters (as far as the instrumental ADU) is the magnitude zero points. You can use the same simulation and analysis that you calibrated to your expected real data by determining the magnitude zero points. The only reason to do multiple filters is to simulate your reduction procedure which might depend on the way you find stars. For instance if you plan to detect the stars in V and then use the same apertures for the other bands then you would probably make different images with different noise values.Speaking of noise, be use to use the seed parameter to get different noise realizations.So, this is can be a complex analysis and you need to think through what it is you are doing and want to characterize in the analysis. It depends on your expected data, your reduction process, the science you want to do, and how you want to report it.I hope this helps.Yours,
Frank Valdes

Dear Frank,thank you for the detailed response. It seems that this is a little more complicated than I thought. Your response has been a great help.Thank you again,
Andrei