Subtraction of sky from object frames

Subtraction of sky from object frames

NewToIFU (guest) 21 Apr 2010 12:18

Hmmm - first post in the forum ever. Not an encouraging sign but here goes anyway.

I am wanting to subtract sky frames from my object frames for my IFU data. I was going to do this by taking my sky frames and averaging the sky spectra over each frame, for all of the frames that I have. Then I was going to subtract this from my object frames to get the sky subtracted datacubes.

My question is - is the Starlink Datacube package capable of doing this? I've been reading the documentation, and it does seem to have the functionality for averaging groupings of spectra and plotting the results, but it doesn't seem to say anything about whether one can average the spectra and then go on to use the results to add/combine/subtract from other datacubes. Is it possible? Is there any other program out there that would be better for this function?

Re: Subtraction of sky from object frames

katrina (guest) 22 Apr 2010 11:48

I've not actually used Slink Datacube before, but since I work on data where the sky is either taken from the astro FoV (eg VIMOS) or is taken from a few isolated spaxels that where placed around the target's FoV (eg GMOS), I always simply make the best sky spectrum (singular) I can (using eg iraf or qfitsview or IDL), then I scale that spectrum to each and every astro spaxel (usually on the 5577 sky line flux, and I do this if I suspect that there are variations in the sky flux across the FoV, tho actually that should have already been corrected for), and subtract it (again using IDL or iraf).

Whether you do this by subtracting a single spectrum (scaled) from a cube, or make a cube of the same size as the astro cube but containing all the sky spectra (scaled), does not really matter - whichever is easier.

So basically, you've got the right idea but everyone I know always does it "manually". I'm using python now to do that. Once I have a script, it works for all subsequent datasets. But I alway check afterwards, as even small changes to the spectral PSF with spaxel will result in sharp residuals in the sky-subtracted spectra, tho this is usually only a problem when I am working in the Ha or red [OI] region.

Re: Subtraction of sky from object frames

NewToIFU (guest) 23 Apr 2010 03:28

OK, thanks for that Katrina. I'll have a bit more of a think about my options. If I have any more questions, I'll be back!