Sort out issues with 15 minute accumulated precip records

[MatthewPyle-NOAA]

May post the same issue in the UPP repository.

 wgrib2 rrfs.t12z.prslev.f003.conus_3km.grib2 -match APCP
820:737929943:d=2024040212:APCP:surface:2-3 hour acc fcst: (max 29.375)
821:738016911:d=2024040212:APCP:surface:0-3 hour acc fcst: (max 41.9375)

wgrib2 rrfs.t12z.prslev.f003-15-00.pr.grib2 -match APCP
820:59191477:d=2024040212:APCP:surface:10815--2147472656 min acc fcst: (max 626.188)
821:59199007:d=2024040212:APCP:surface:15-18 min acc fcst: (max 38.6875)

wgrib2 rrfs.t12z.prslev.f003-30-00.pr.grib2 -match APCP
820:59587233:d=2024040212:APCP:surface:10830--2147472641 min acc fcst: (max 842.375)
821:59596225:d=2024040212:APCP:surface:30-33 min acc fcst: (max 44.5)

The run time total (second APCP record) mostly seems to be okay (though the run total drops between f03 and f03:15), but is mislabeled in the 15 minute output. Maybe when we trim down the 15 minute product list, can eliminate this partial bucket record if remains problematic.

[MatthewPyle-NOAA]

More updates...sub hourly product (running total) all zero within first hour. Running total drops from 1:15 to 1:30 into forecast. Problematic time labels for all of the subhourly

wgrib2 -V rrfs.t06z.prslev.f000-15-00.conus.grib2 -match APCP
19:18445403:vt=2024041106:surface:15-15 min acc fcst:APCP Total Precipitation [kg/m^2]:
    ndata=1905141:undef=0:**mean=0:min=0:max=0**

wgrib2 -V rrfs.t06z.prslev.f000-30-00.conus.grib2 -match APCP
19:18474763:vt=2024041106:surface:30-30 min acc fcst:APCP Total Precipitation [kg/m^2]:
    ndata=1905141:undef=0:**mean=0:min=0:max=0**

wgrib2 -V rrfs.t06z.prslev.f000-45-00.conus.grib2 -match APCP
19:18543802:vt=2024041106:surface:45-45 min acc fcst:APCP Total Precipitation [kg/m^2]:
    ndata=1905141:undef=0:**mean=0:min=0:max=0**

wgrib2 -V rrfs.t06z.prslev.f001.conus.grib2 -match APCP
758:643509165:vt=2024041107:surface:0-1 hour acc fcst:APCP Total Precipitation [kg/m^2]:
    ndata=1905141:undef=0:mean=0.132625:min=0:max=98.5625

wgrib2 -V rrfs.t06z.prslev.f001-15-00.conus.grib2 -match APCP
19:18736740:vt=2024041107:surface:15-16 min acc fcst:APCP Total Precipitation [kg/m^2]:
    ndata=1905141:undef=0:mean=0.134905:min=0:max=101.5

wgrib2 -V rrfs.t06z.prslev.f001-30-00.conus.grib2 -match APCP
19:18796085:vt=2024041107:surface:30-31 min acc fcst:APCP Total Precipitation [kg/m^2]:
    ndata=1905141:undef=0:mean=0.136489:min=0:max=91.5

[EricJames-NOAA]

@MatthewPyle-NOAA thanks for the info. I see that we currently have these four 15-min acc variables in RRFS-A:

ASNOW APCP FRZR TSNOWP

Can you confirm we need all of these? Or do we just need APCP?

[MatthewPyle-NOAA]

@EricJames-NOAA I'd say APCP is the only critical one - the others were on the "nice to have" pile, but I've been wondering if they should be trimmed. HRRR has these accumulation fields...probably why I included some winter weather fields:

26:25857641:d=2024050114:APCP:surface:1020-1035 min acc fcst: 27:26245238:d=2024050114:WEASD:surface:1020-1035 min acc fcst: 28:26313753:d=2024050114:FROZR:surface:1020-1035 min acc fcst:

[EricJames-NOAA]

@MatthewPyle-NOAA thanks. I'll include them for now, and we can decide to remove them later if we want.

[EricJames-NOAA]

@MatthewPyle-NOAA I have the UPP changes ready which are required to clean up the labeling. Should I open a PR, or wait until we have the final bucket values to finalize the 15-min fields? I guess there will be some minor UPP changes to pass through the correct bucket fields

[MatthewPyle-NOAA]

@EricJames-NOAA Definitely open a UPP PR for this fix. That will motivate me to the finish the follow on work on a separate bucket code.