Update plate scale

[mgalloy]

Use IMAGESCL in Steve's results to compute an updated plate scale.

• [x] plot IMAGESCL over the mission
• [x] identify epochs where the plate scale changes
• [x] find a plate scale value for each wave region in each epoch

[mgalloy]

There was not enough data for 670, 761, 802, and 991 to establish any pattern.

I extracted CDELT1 from Steve's level 1 files along with the chi-squared of the occulter finding fit for both cameras. The below plot shows these "image scale" values, with the top 20% of the chi squared values removed.

ucomp.platescale.pdf

There is a clear change in plate scale on 2022-02-02 in 637, 656, 789, 1074, and 1079 which is consistent with the 691, 706, and 530.

[mgalloy]

Here are some preliminary plate scales:

ucomp.530.image_scale.txt: 1607 files
fit 0: 80.0% percentile chisq: 213907.00
fit 1: 80.0% percentile chisq: 209118.00
epoch starting 2021-05-26 [279 files], plate scale: 2.899 arcsec/pixel
epoch starting 2022-02-02 [819 files], plate scale: 3.067 arcsec/pixel

ucomp.637.image_scale.txt: 4236 files
fit 0: 80.0% percentile chisq: 102.78
fit 1: 80.0% percentile chisq: 124.41
epoch starting 2021-05-26 [1223 files], plate scale: 2.946 arcsec/pixel
epoch starting 2022-02-02 [2071 files], plate scale: 3.060 arcsec/pixel

ucomp.656.image_scale.txt: 2453 files
fit 0: 80.0% percentile chisq: 68047.60
fit 1: 80.0% percentile chisq: 73323.30
epoch starting 2021-05-26 [1793 files], plate scale: 2.932 arcsec/pixel
epoch starting 2022-02-02 [152 files], plate scale: 3.038 arcsec/pixel

ucomp.691.image_scale.txt: 919 files
fit 0: 80.0% percentile chisq: 337.39
fit 1: 80.0% percentile chisq: 278.49
epoch starting 2021-05-26 [644 files], plate scale: 2.936 arcsec/pixel
epoch starting 2022-02-02 [41 files], plate scale: 3.038 arcsec/pixel

ucomp.706.image_scale.txt: 3193 files
fit 0: 80.0% percentile chisq: 185.74
fit 1: 80.0% percentile chisq: 211.72
epoch starting 2021-05-26 [455 files], plate scale: 2.930 arcsec/pixel
epoch starting 2022-02-02 [1997 files], plate scale: 3.035 arcsec/pixel

ucomp.789.image_scale.txt: 3980 files
fit 0: 80.0% percentile chisq: 121.92
fit 1: 80.0% percentile chisq: 106.45
epoch starting 2021-05-26 [1095 files], plate scale: 2.901 arcsec/pixel
epoch starting 2022-02-02 [1893 files], plate scale: 3.012 arcsec/pixel

ucomp.1074.image_scale.txt: 33213 files
fit 0: 80.0% percentile chisq: 4.40
fit 1: 80.0% percentile chisq: 3.08
epoch starting 2021-05-26 [8562 files], plate scale: 2.832 arcsec/pixel
epoch starting 2022-02-02 [16430 files], plate scale: 2.943 arcsec/pixel

ucomp.1079.image_scale.txt: 3789 files
fit 0: 80.0% percentile chisq: 22.01
fit 1: 80.0% percentile chisq: 21.00
epoch starting 2021-05-26 [867 files], plate scale: 2.826 arcsec/pixel
epoch starting 2022-02-02 [2114 files], plate scale: 2.942 arcsec/pixel

ucomp.1083.image_scale.txt: 1148 files
fit 0: 80.0% percentile chisq: 43.62
fit 1: 80.0% percentile chisq: 39.75
epoch starting 2021-05-26 [879 files], plate scale: 2.836 arcsec/pixel

[mgalloy]

Currently, I am using the 50% chi squared threshold values for the plate scales and only the 2022-02-02 date as a epoch change date.