Initializing MPAS with ERA5 data: ungrib cannot find date at specified time

[aroseman]

Trying to use ungrib for surface LSM data, and got this error below. Usually the ungrib.exe file is used for GFS and works fine, is there maybe another version to use for ERA5?
*** Starting program ungrib.exe ***
Start_date = 1979-01-01_00:00:00 , End_date = 1979-01-01_00:00:00
output format is WPS
Path to intermediate files is ./

###############################################################################

Inventory for date =
PRES GEOPT HGT TT UU VV RH DEWPT LANDSEA SOILGEO SOILHGT PSFC PMSL SKINTEMP SEAICE SST SNOW_DEN SNOW_EC SNOW SNOWH ST000007 ST007028 ST028100 ST100289 SM000007 SM007028 SM028100 SM100289
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------


Subroutine DATINT: Interpolating 3-d files to fill in any missing data...
Looking for data at time 1979-01-01_00
ERROR: Data not found: 1979-01-01_00:00:00.0000


HERE IS MY namelist.wps FILE:


&share
wrf_core = 'ARW',
max_dom = 1,
start_date = '1979-01-01_00:00:00',
end_date = '1979-01-01_00:00:00',
interval_seconds = 21600
io_form_geogrid = 2,
/


&ungrib
out_format = 'WPS',
prefix = 'LSM',
/

 

[Ming Chen]

Please download ERA5 data from NCAR RDA: NSF NCAR GDEX Dataset d633000

Then get the python script from: GitHub - NCAR/era5_to_int: A simple Python script for converting ERA5 model-level netCDF files to the WPS intermediate format

This script is created specifically for ERA5 data. You can follow the instruction to run this script and produce intermediate files for MPAS run.

Please try and let me know if you have any questions.

 

[aroseman]

Hi Ming, thanks for the response!

I have downloaded the dataset (directly from Derecho directory), though upon running era5_to_int I do still get some errors:

(1) python era5_to_int.py --path ../ERA5_Data/ -v LANDSEA 1979-01-01_00 1979-01-01_00 ==> works fine!
[
datetime = 1979-01-01 00:00:00
until_datetime = 1979-01-01 00:00:00
interval_hours = 6:00:00
Processing time record 1979-01-01_00
../ERA5_Data/e5.oper.invariant.128_172_lsm.ll025sc.1979010100_1979010100.nc
]

(2) python era5_to_int.py --path ../ERA5_Data/ -v SOILGEO 1979-01-01_00 1979-01-01_00 ==> fails!

[
datetime = 1979-01-01 00:00:00
until_datetime = 1979-01-01 00:00:00
interval_hours = 6:00:00
Processing time record 1979-01-01_00
Error: Could not find file e5.oper.invariant.128_129_z.regn320sc.2016010100_2016010100.nc needed for ERA5 variable Z
]

Notes:
(a) I don't have any files with the name regn320sc (they are labeled ll025sc on Derecho).
If I change the name of my file ("e5.oper.invariant.128_129_z.ll025sc.1979010100_1979010100.nc"), to that required by era5_to_int, I get the following error:
Converting SOILGEO at 1979-01-01_00:00:00
SOILGEO is NOT being written to the intermediate file at level 1.0

 

[Ming Chen]

Would you please clarify why you run the script to process ERA5 data one by one? I suppose it is better that you run the script to process all the input files.

I will drop by your office to discuss the issue.

 

[aroseman]

"./era5_to_int.py -i 2022-06-30_00 2022-07-02_23 1" works very well. Very convenient too.

Thanks for the help!