1376cm-1; high-med; staremode; N6 S0 N5 D0
sky.sci.30076.fits elevation=40 deg, 16 repeats
sky.sci.30077.fits elevation=20 deg, 10 repeats
sky.sci.30078.fits elevation=25 deg, 8 repeats
sky.sci.30079.fits elevation=25 deg, 4 repeats
To complete the analysis, I first made each frame into an MCDS 7, which meant that sky.sci.30076.fits had 16 frames to examine, sky.sci.30077.fits had 10 frames to examine, etc... Below is a picture of an MCDS 7 frame for each image.
Figure 1 ( Top left: sky.sci.30076.fits; Top right: sky.sci.30077.fits; Bottom left: sky.sci.30078.fits; Bottom right: sky.sci.30079.fits)
To examine any variations with respect to time, I subtract each subsequent MCDS frame from the previous one for each image. This means that for sky.sci.30076.fits there are 15 data points, for sky.sci.30077.fits there are 9 data points, etc... The frames visually subtract very well (see Figure 2 below). When examining the median value across a 200x200 pixel section in the middle of each difference image, the values appear consistent with zero (see Figure 3 below). One interesting feature is how the frames have larger offsets at the beginning of the exposure. This may point to some sort of reset issue that takes time to become stable, but the difference in counts is still very small.
Figure 2 ( Top left: sky.sci.30076.fits; Top right: sky.sci.30077.fits; Bottom left: sky.sci.30078.fits; Bottom right: sky.sci.30079.fits)
Figure 3 (Comparison of difference in counts between subsequent frames taken at the same elevation.)
The next comparison is with respect to changing elevation. For this analysis I compared:
sky.sci.30076.fits - sky.sci.30077.fits (elevation 40 - elevation 20)
sky.sci.30076.fits - sky.sci.30078.fits (elevation 40 - elevation 25)
sky.sci.30078.fits - sky.sci.30077.fits (elevation 25 - elevation 20)
sky.sci.30078.fits - sky.sci.30079.fits (elevation 25 - elevation 25)
I only examined the first 4 MCDS images from each frame since sky.sci.30079.fits has only 4 repeats. Figure 4 shows the difference images. What can be easily seen is a small offset between the different elevation. The difference between the same elevations does not appear to have an offset. The offset is around ~5 pixels. I examined the median value in a 50x50 pixel area on the bright line that is shown near the top of each image in Figure 4. Those results are displayed in Figure 5.
Figure 4 (Top left: elevation 25 - elevation 20; Top right: elevation 25 - elevation 25; Bottom left: elevation 40 - elevation 20; Bottom right: elevation 40 - elevation 25)
Figure 5 (Comparison of difference in counts between frames taken at different elevations.)
1 comment:
The offset with elevation is something we should pursue. We should be able to do a test with line ops.
In the difference with elevation the atmospheric lines are pretty clearly not subtracting, as you might expect for different airmasses. If you raise one image to the power of the ratio of the two airmasses, do the atmospheric lines disappear? If only some of them disappear, which molecules? If none of the lines disappear, is there a different power that gets rid of more?
Post a Comment