unrealistic LWDN and TV, TG at first timestep

[a24ao09002-coder]

I have downloaded the era5 land data at 3 hourly interval separately for two monthes, but when i am trying to run the hrldas , it is blowing off giving the avove issue . I am attaching the namelist files . I have tried changing the NOAHLSM_TIMESTEP , but the error continued.

namelistinput_era5.txt

namelist.hrldas.txt

[cenlinhe]

are you running the global domain (i.e., entire ERA5-land domain) or a small area? Are there coastal lines and small islands in your simulation domain? If so, you may want to increase the EXPAND_LOOP value in your namelistinput_era5.txt file. This error often occurs when you have unrealistic forcing data over unresolved land points (particularly over coasts and small islands). i.e., your forcing file includes a few water points with unrealistic values that should have been identified as land points based on your geo_em file in your study domain.

[a24ao09002-coder]

Yes sir , I am trying to simulate for a smaller domain. I am trying to download and use the forcing files at 3 hourly interval due to the vast size of the files. Can i process the files separately at a monthly scale and simulate the HRLDAS for longer periods?

[cenlinhe]

yes, you can process forcing files for each monthly and run the model for longer periods as long as the forcing files are continuous in time.

[shekhar-cloud]

Sir, I increased the 'EXPAND_LOOP' and ran the simulation for one month. I have created forcing files differently for different months, after one month of simulation the simulation stopped, with the same error as before. As the file sizes are very large, it is difficult to merge all the files. Kindly help..

[cenlinhe]

Are your forcing files continuous in time for every 3 hours. What is your model simulation error? What model timestep (NOAH_TIMESTEP) in the namelist.hrldas are you using? If you are using 3 hourly or hourly model timestep, please reduce it to half hourly.