Change cadence of KCor subtractions to 30 seconds (currently 5 minutes)

[jburkepile]

Create 30 second subtraction images in EOD pipeline. Changes to be made:

• [x] change avginterval to 0.185185e-3 This is Julian units in fraction of a day = 16 seconds. The code currently sets this to 2 minutes. Note: avginterval is the time over which to average images.

• [x] change number of images averaged to 2 (currently set to 4). This is hardwired in the kcor_create_differences.procode in comments and in for loop:

  • LINE 89 ; read in up to 4 images, get time, and average if images <= 2 min apart CHANGE TO: if images <= 16 sec apart
    • LINE 90: for i = 0, 3 do begin CHANGE TO: for i = 0, 1 do begin
    • Remove line 145 comment: ; ** NOTE: 2 minutes in julian date units = 1.38889e-03
    • LINE 147 comment ; If images are <= 2 minutes apart we average them together. CHANGE to < = 16 sec apart
    • LINE 148 comment: If images are > 2 minutes apart we stop averaging, save avg. image and CHANGE to > 16 sec apart
    • LINE 155 comment: goodheader = header ; save header in case next image is > 2 min. in time CHANGE to > 16 sec apart

• [x] change time_between_subs to 0.34722e-03 , this is Julian units for 30 sec (code is currently set to 5 minutes)

  • LINE 212 comment: Has it been 5 minutes since the previous subtraction? CHANGE TO: 30 seconds since previous subtraction

• [x] Change GIF intensity scaling min and max:

  • LINE 318: display_min = -2.0e-8 CHANGE to -5.0e-9
  • LINE 319: display_max = 2.0e-8 CHANGE TO: 9.0e-9

• [ ] test the code on these 2 days: 20210507, 20220830

NOTE: I ESTIMATE THIS WILL ADD an EXTRA ~400 GB PER YEAR (FITS + GIFs) to amount of Level 2 K-Cor data.

NOTE: I will make a separate ticket for adding the times of the 2 images being averaged to the FITS header, #335.

[mgalloy]

I re-ran 20210507 with the updated code for this issue, but no difference images were produced.

[mgalloy]

Joan has code outside the repo that should address this issue.