- There is some turbulence at higher frequencies as well
- You can emulate a faster chopping by software by only using the end of the 'T' frames and the beginning of the 'S' frames -- but of course you lose Signal/Noise!
- The automatic background subtraction performed by
oirChopPhotoImages might not be good enough. oirChopPhotoImages just averages the counts in the skybands atop and beneath the expected source location and subtracts this value from all pixels. If you are not satisfied with this result you can do the background subtraction by hand and fit a higher-order function (best: second order) to the background profile.
Fancy background subtraction
First get some chopped (but not masked) photometry data using chop = oirgetdata(/path/to/choppedphotometry.fits), i.e. the output of oirChopPhotoImages. Select only the best estimate of target - sky from the 7 element array: chop=chop[0].
Define a weighting function m=fltarr(41) and find from inspection of the intensity profile on the chip, i.e. plot,total(chop.data1,1), which parts you consider background. Since we will only be fitting a second order function (higher order functions tend to easily go wrong), try to find those parts of the intensity profile across the detector chip that look like they can be fitted by a second order function. Say you consider the pixels 5 to 9 and 18 to 24 to be background. Then do m(5:9)=1. and m(18:24)=1. and copy the chopped image since you will be editing it: dchop=chop. Then you can run the fit routine
for i=0, 170 do begin q = polyfitw(findgen(41), chop.data1(i,*), m, 2, y) & dchop.data1(i,*) = chop.data1(i,*) - y & endfor
The polyfit function will thus fit a 2nd order polynomial function to the points of chop.data1(i,*) specified by the weighting function m. The values of that fit function are put into y. We then do the additional background subtraction with the second statement in the above line of commands.
