Characterize dark noise pattern from ion pumps

[awhoward]

Characterize the noise pattern from the ion pumps. This could involve making a map of photons/hour for the green and red CCDs using a series of long darks (filtered for cosmic rays). Most of the structure in these images is likely diffuse and could be characterized at higher SNR by doing some median smoothing. Also note that there are some sharp features due to shadowing hear the edges. There's also bright amplifier glow within a few pixels of some of the edges. This task is essentially making a very deep master-dark.

This map will be useful in needed to subract out this noise pattern (leaving significant Poisson noise) and noting the increased uncertainty in the L1 variance.

Examples of 1-hour dark exposures are shown below. green_dark_1hr.pdf red_dark_1hr.pdf

[awhoward]

My sense from looking at Quicklook plots is that the intensity of this effect is diminishing, perhaps as the pressure in the dewars goes down. It would be good to check this.

[bjfultn]

nightly master darks should be good enough to perform this analysis

[bjfultn]

Should be a DRP module that writes to a 2D file header value. Then we will ingest this into jump to track over time.

[awhoward]

I've made good progress on this. Below are some example plots from my code.

[awhoward]

I have a Jupyter Notebook added in a branch that I'll start a pull request for shortly.

A second task is to use the code on the notebook to create a pipeline module to do aperture photometry on the specified regions and put the values in the fits files.