jswhit / build_gsinfo Goto Github PK

@frolovsa Found some discrepancies between noaa and nasa choices for black list/white list selection.
See discussion document here

per era5 and gmao goes sndrd is assimilated from goes 13. Follow nasa's notes.

Turn noaa 19 back on per era5 and ncep
Email to amal sent on why nasa doesnt use it

Hi @jswhit
On of the convinfo file in this directory (https://github.com/jswhit/build_gsinfo/tree/GMAO/convinfo/historical) doesnt have a date attached to it. Is this in error and the file can be removed? Or this is a feature and it should stay?
This is giving Adam some errors when generating a convinfo file.

Looks like we only have goesnd files going back to November 1997. Disable goes 8 and 10 before that.

For some reason all ssmi and ssmis data are on except from f16. turning it on.

HIRS2 on noaa 14 is turned off prematurely on 1995050100. Shold be off as in readme file on 20060930

looks like the best convinfo we have is for gefs12 which starts in 2000. For example we need to adjust to pre gps era.

turn on iasi per nasa in may 2020

change satinfo file to start on 2017/1/1
last file dec 1 2021

we now have nasa airs data. the dates should be
20021012 to 21001231

MSU and SSU: Turn on noaa 7, 8, 9, 10 in satinfo files following nasa's notes.
only noaa 10 was still active in 1990s.

Here are some things i noticed about the ozoneinfo

• need documentation for what is changing in the ozone info files and what are these choices are based on
• need to create ozone_info files prior to 1999. Maybe just for noaa11?
• geos choices are very different from ours. I dont kn

ow where our choices come from? Apparently this also depends on the version of the ozone retrievals.

for reference this is the nasa table for ozone choices for 21c.

SBUV/2   NOAA-14  19970101 19981231              
SBUV/2   NOAA-11  19990101 20000930              
SBUV/2   NOAA-16  20001001 20040930              
SBUV/2   NOAA-17  20020801 20040930              
OMI      Aura     20041001 21001231                
MLS      Aura     20041001 21001231              
OMPS-LP  SNPP     20130101 21001231              
OMPS-NM  SNPP     20130101 21001231              
AIRS     Aqua     20021012 21001231              
IASI     Metop-A  20080801 21001231              
IASI     Metop-B  20130731 21001231              
CrIS-FSR SNPP     20170106 21001231              
CrIS-FSR NOAA-20  20180217 21001231         

GMAO branch contains NASA choices based on table from Amal with the following differences:

• assimilate amsua ch 14 without bias correction (iuse=4) - this is also different from NCEP ops
• for amsua, assimilate chs1-3 and 15
• for n15 amsua don't stop assimilating 1-5 and 15 (since these are still being used in NCEP ops)
• for MSU use ch 1
• for HIRS use chs 2-15 (include 9,13-15)
• for ATMS use channels 1-22 (include 1-4 and 16)
• assimilate ATMS channel 15 without bias correction (iuse=4) - this is different than NCEP ops
• use different channel selection for airs_aqua
• if NASA turns off an instrument, but NOAA is still assimilating it operationally I've elected to keep it on
• use cris_npp dumps for the roughly 6 year period before cris-fsr_npp dumps are available.

looks like data is available on 1/1/2013. era5 turn on in 2015.