Scheduled Downtime
On Friday 21 April 2023 @ 5pm MT, this website will be down for maintenance and expected to return online the morning of 24 April 2023 at the latest

issues of running MPAS-A with ERA5 ("Segmentation fault - invalid memory reference")

LihuiJi

New member
I have two issues.

I have a similar problem when preparing init.nc with ERA5 data with this thread.
1. It turns out I can proceed by changing config_extrap_airtemp from "lapse-rate" to "linear". I just want to know why it works. Will this setting possibly impede my simulation?

2. However, when I run the regional simulation with initial and boundary conditions, I got "Segmentation fault - invalid memory reference." I also tried to do the same thing but with GFS data, I can run the model without problem.

I used era5_to_int.py --isobaric to generate the intermediate files.
I attached relevant files. You can also check the running directory. Thank you in advance.
/glade/derecho/scratch/lihuiji2/mpas_tutorial/regional_era5
 

Attachments

  • log.atmosphere.0000.out.txt
    20.9 KB · Views: 0
  • namelist.atmosphere.txt
    2 KB · Views: 1
  • namelist.init_atmosphere_config_7.txt
    1.5 KB · Views: 0
  • run_model.e9131699.txt
    53.3 KB · Views: 0
  • streams.atmosphere.txt
    1.8 KB · Views: 1
  • streams.init_atmosphere_config_7.txt
    912 bytes · Views: 0
Hi,
The issues you have are not model issues, ---- I repeated your case and found that the input data might not be correct.
We will revisit the python script (era5_to_int.py) and hopefully we can fix this issue soon. It may take some time and please remind me if you didn't get any update in a week. Thanks for your patience.
 
I just wonder, can I just use WPS to get the intermediate file myself? Do I need to specify "--isobaric" somewhere or WPS will do this for me?
 
WPS can only process GRIB format data. CISL RDA has removed all ERA5 data in GRIB format. This is why we provide the python package to process data in netCDF format.
 
Top