Add Ion Pump diagnostic measurements to FITS headers

[awhoward]

I just uploaded a Jupyter Notebook that creates plots of the ion pump current and posted an Issue for the plots to be added to the QLP. I will submit a separate Issue for the task of taking the measurements (electrons per hour, etc.) and putting them into the 2D FITS headers.

The purpose of this Issue is to use the same code in the Jupyter Notebook put measurements of the 2D images into their FITS headers. This module will take a 2D file as input and have the same 2D file as output, but with twenty new FITS headers added. I'm not sure if this means that the data model needs to be modified.

I suggest that this module be run only on dark images (Texp != 0 and no source) or bias images (Texp=0 and any source since the shutters won't open with Texp=0) to avoid confusion. This could be accomplished in the recipe not the module though.

Do we have other modules that only make measurements and add information to the FITS headers, while leaving the input file unchanged? We will have many more modules of this type.

The Jupyter Notebook is part of a pull request that I expect will be approved soon. I'm attaching a copy of it here. Note that this version is slightly updated from the version that I submitted last week.

KPF_Data_Tutorial_2D_and_Ion_Pumps.ipynb.zip

Here are the measurements and their associated variables in the Notebook code. There are ten regions in each CCD where an average flux per pixel is measured. For dark images, the units are electrons per hour. For bias images, the units are electrons.

FITS Keyword	FITS Comment	Variable name
FLXCOLLG	Flux (e- [bias] or e-/hr [dark]) in Ion Pump reg., coll. side [3700:4000,700:1000], Green CCD	reg['ref1'][''med_elec'']
FLXECHG	Flux (e- [bias] or e-/hr [dark]) in Ion Pump reg., coll. side [3700:4000,3080:3380], Green CCD	reg['ref1'][''med_elec'']
FLXREG1G	Flux (e- [bias] or e-/hr [dark]) in Region 1 [1690:1990,1690:1990], Green CCD	reg['ref1'][''med_elec'']
FLXREG2G	Flux (e- [bias] or e-/hr [dark]) in Region 2 [1690:1990,2090:2390], Green CCD	reg['ref2'][''med_elec'']
FLXREG3G	Flux (e- [bias] or e-/hr [dark]) in Region 3 [2090:2390,1690:1990], Green CCD	reg['ref3'][''med_elec'']
FLXREG4G	Flux (e- [bias] or e-/hr [dark]) in Region 4 [2090:2390,2090:2390], Green CCD	reg['ref4'][''med_elec'']
FLXREG5G	Flux (e- [bias] or e-/hr [dark]) in Region 5 [80:380,700:1000], Green CCD	reg['ref5'][''med_elec'']
FLXREG6G	Flux (e- [bias] or e-/hr [dark]) in Region 6 [80:380,3080:3380], Green CCD	reg['ref6'][''med_elec'']
FLXAMP1G	Flux (e- [bias] or e-/hr [dark]) in Amp Region 1 [3700:4000,700:1000], Green CCD	reg['amp1'][''med_elec'']
FLXAMP2G	Flux (e- [bias] or e-/hr [dark]) in Amp Region 2 [3700:4000,3080:3380], Green CCD	reg['amp2'][''med_elec'']
FLXCOLLR	Flux (e- [bias] or e-/hr [dark]) in Ion Pump reg., coll. side [3700:4000,700:1000], Red CCD	reg['ref1'][''med_elec'']
FLXECHR	Flux (e- [bias] or e-/hr [dark]) in Ion Pump reg., coll. side [3700:4000,3080:3380], Red CCD	reg['ref1'][''med_elec'']
FLXREG1R	Flux (e- [bias] or e-/hr [dark]) in Region 1 [1690:1990,1690:1990], Red CCD	reg['ref1'][''med_elec'']
FLXREG2R	Flux (e- [bias] or e-/hr [dark]) in Region 2 [1690:1990,2090:2390], Red CCD	reg['ref2'][''med_elec'']
FLXREG3R	Flux (e- [bias] or e-/hr [dark]) in Region 3 [2090:2390,1690:1990], Red CCD	reg['ref3'][''med_elec'']
FLXREG4R	Flux (e- [bias] or e-/hr [dark]) in Region 4 [2090:2390,2090:2390], Red CCD	reg['ref4'][''med_elec'']
FLXREG5R	Flux (e- [bias] or e-/hr [dark]) in Region 5 [80:380,700:1000], Red CCD	reg['ref5'][''med_elec'']
FLXREG6R	Flux (e- [bias] or e-/hr [dark]) in Region 6 [80:380,3080:3380], Red CCD	reg['ref6'][''med_elec'']
FLXAMP1R	Flux (e- [bias] or e-/hr [dark]) in Amp Region 1 [3700:4000,700:1000], Red CCD	reg['amp1'][''med_elec'']
FLXAMP2R	Flux (e- [bias] or e-/hr [dark]) in Amp Region 2 [3700:4000,3080:3380], Red CCD	reg['amp2'][''med_elec'']

I'm not sure if the FITS Comments exceed some standard length specification.

Tagging @bjfultn for coordination.

[awhoward]

The code for this is in the measure_2D_dark_current() method in this part of the DRP: https://github.com/Keck-DataReductionPipelines/KPF-Pipeline/blob/develop/modules/Utils/analyze_2d.py

The next step is to run this analysis on 2D files as they are created and put the measured dark currents into the headers. It will be very helpful to analyze the time series of these from the 20-minute dark frames that we take every day.