NCAR / ucomp-pipeline

Data processing pipeline for UCoMP
Other
6 stars 3 forks source link

Reprocess test dates #210

Closed mgalloy closed 1 year ago

mgalloy commented 1 year ago

Things to check:

Dates:

Test 2:

Test 3:

Rest wavelength computation done for 1074 by improved POLY_FIT linear model.

Test 4:

Rest wavelength computation done for 1074 by improved POLY_FIT quadratic model. 1074 only.

Test 5:

Rest wavelength computation done for 1074 by individual data file method. 1074 only.

Test 6:

Tweaks to constants and bug fix.

Test 7:

Same as 6 except only 1074 nm and using image data for rest wavelength calculation instead of the model.

Test 8:

Over mask occulter for rest wavelength calculation by 8 pixels. Use image data for rest wavelength calculation for 1074 nm. 789 nm enhanced intensity image display fix.

Test 9:

Test display 706 nm IQUV all images and new mechanism to add user guide to tarballs.

mgalloy commented 1 year ago

In #175, @detoma suggests using the following days for testing the rest wavelength computation, so they would most likely be good for this issue as well:

mgalloy commented 1 year ago

Results are in process.commissioning-test.

detoma commented 1 year ago

The new color table for the FWHM does not work. It is basically all the same color at the low and high end. We need a color table with more variation.

Velocity .png do not look good. Are we sure we apply the correct rest-wavelength? If so, the median rest wavelength from file should be very close to the number we applied from the fit.

Another thing I noticed about the velocity is that we extend too far in the fov. We are including bad pixels that should be masked. That is easy to fix.

mgalloy commented 1 year ago

Sorry, wrong issue.