Postproc multiyear optics

[plazzari]

Output frequency for optics

• Kd daily
• Rrs(0+) daily
• Light at the bottom multispectral daily
• Raw optical files (Ed,Es,Eu) 3D daily lossy compression when archived

New features

• Script for rrs calculation to be upgraded using lat as correction.
• Script to compute multispectral light at the bottom for all components Ed, Es, Eu (at n+1 layer)
• Script to save Ed,Es,Eu with higher compression (5 significant digit)

[gbolzon]

With 5 significant digits compression for 3D files

3.2M  ave.19991101-12:00:00.Ed_0400.nc
 13M  ave.19991101-12:00:00.Es_0400.nc
 11M  ave.19991101-12:00:00.Eu_0400.nc

The resulting archive space is 1TB / year, the total about 20TB.

With 4 significant digits we have 0.7TB/ year, the total about 14TB. With 3 significant digits we have 0.5TB/ year, the total about 10TB.

(Original file 202 MB, with no-loss compression 30MB)

Here are these case of compression

/g100_scratch/userexternal/gbolzon0/NECCTON/testCA/run_link/wrkdir/POSTPROC/bin/archive/2digits/ave.19990501-12:00:00.Eu_0400.nc
/g100_scratch/userexternal/gbolzon0/NECCTON/testCA/run_link/wrkdir/POSTPROC/bin/archive/3digits/ave.19990501-12:00:00.Eu_0400.nc
/g100_scratch/userexternal/gbolzon0/NECCTON/testCA/run_link/wrkdir/POSTPROC/bin/archive/4digits/ave.19990501-12:00:00.Eu_0400.nc
/g100_scratch/userexternal/gbolzon0/NECCTON/testCA/run_link/wrkdir/POSTPROC/bin/archive/5digits/ave.19990501-12:00:00.Eu_0400.nc

Can be useful to compare them with the original one? /g100_scratch/userexternal/camadio0/Neccton_hindcast1999_2022/wrkdir/MODEL/AVE_FREQ_3/ave.19991101-12:00:00.Eu_0400.nc