[gradsusr] New GFS accumulated precipitation variable

[Adams, Jennifer M. (GSFC-610.2)[ADNET SYSTEMS INC]]

Hi, Everyone –
I’m sure many of you are aware of the changes to the GFS that went operational with yesterday’s 12Z run. In the new data files, there are two records for accumulated precipitation that look identical except for the beginning-of-accumulation information. Here is what the wgrib2 output looks like for the APCP variable in the 12hr forecast and 15hr forecast files:

# wgrib2 gfs.t12z.pgrb2.0p25.f012 | grep APCP
449:248062227:d=2019061212:APCP:surface:6-12 hour acc fcst:
450:248429916:d=2019061212:APCP:surface:0-12 hour acc fcst:

# wgrib2 gfs.t12z.pgrb2.0p25.f015 | grep APCP
449:246395924:d=2019061212:APCP:surface:12-15 hour acc fcst:
450:246700803:d=2019061212:APCP:surface:0-15 hour acc fcst:

The variable in record #449 is the ‘old’ version, with different accumulation periods depending on the forecast hour. The variable in record 450 has the ‘new’ version, with accumulations from the beginning of the forecast. There is no way to distinguish between these two based on the grib2 codes in a descriptor file. Here’s what I use:

p           0,1,0            0,1,8,1    Total Accumulated Precipitation [kg/m2]

As gribmap examines the records in the grib file to look for matches in the descriptor file, it will find record 449 and index that position for the variable p. Then it will find record 450 and overwrite the index for the variable p with the new position. So, as long as the ‘new’ version of APCP appears after the ‘old’ version in the grib file, gribmap will index the ‘new’ records, and you can work with the precip data assuming the new accumulation standards. NCEP says that the ‘old’ variables will be phased out in the near future, so there’s no point in trying to match with the old style records.

In many ways, the new standard is easier to deal with, although those of us who have written a boatload of scripts to work around the awkwardness of the old standard have some revisions to do. I am working on modifications to the scripts behind the maps at wxmaps.org and wx.gmu.edu.  I hope to have things looking right by the end of the week.

Jennifer

--
Jennifer Miletta Adams
ADNET Systems, Inc.
NASA/GSFC, Code 610.2
Building 32, Room S159
(301) 614-6070

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://gradsusr.org/pipermail/gradsusr/attachments/20190613/68e20133/attachment.html>