"ERRWAT is positive" - "Water budget problem in NOAHMP LSM" #1214
-
Bug DescriptionI am facing an error " The model is gaining water (ERRWAT is positive)" while running LISF 7.4 with NoahMP3.6 and HyMAP2 in restart configuration. I am using ESMWF's ERA5 with MERRA2 in overlay at 1 hour model timestep with outputs at daily timestep. LIS works fine in "coldstart" mode but in "restart" mode, it runs for a few days and then stops abruptly. I have attached the relevant logs, config and restart files below. Please help me to resolve the issue. The mpirun error log records the following :
And the mpirun output log records :
The lislog.xxxx appears like this :
Following zip file contains restart and LIS config files. |
Beta Was this translation helpful? Give feedback.
Replies: 1 comment 4 replies
-
Hello, I also encountered the problem "The model is gaining water (ERRWAT is positive)" when running LISF 7.4 with NoahMP3.6 as you had before. I am glad to see that you have solved this problem by increasing the NoahMP3.6's water balance check threshold to 10. May I ask where you modify this threshold parameter?Thank you very much. |
Beta Was this translation helpful? Give feedback.
Hi @sudarsan3010
The LIS reader for the GPM IMERG data assumes that the files have NOT been spatially sub-setted. If you try to read in a file that was spatially sub-setted, then the code will not perform correctly, as you have seen.
You will need to download the entire region of the data, but NOT clicking on "refine region". I realize that this will take more disk space and longer to download.