How to get stream flow in MOSART

[MAX]

Dear all:
I'm trying to get the stream flow in MOSART; howerver I don't find useful guidance about MOSART now, such as how to set the user_nl_mosart, etc. I can run CLM5 and it works well. Now I want to output stream flow, and I want to validate the result with the dataset "Dai and Trenberth Global River Flow and Continental Discharge Dataset".
I am running based on the I2000Clm50SpGs compset, its long name is 2000_DATM%GSWP3v1_CLM50%SP_SICE_SOCN_MOSART_SGLC_SWAV_SIAC_SESP. But I can only get the file: runoff_test_v1.clm2.h0.1995-01-01-00000.nc, and this nc file only contains runoff results(unit: mm/s). I think I need to get another file: *mosart*.nc, and maybe it will include the stream flow(unit: m3/s). Any help would be greatly appreciated!

 

[oleson]

You should be getting mosart history files with that compset if you have run at least one month, e.g.,

runoff_test_v1.mosart.h0.0001-01.nc

That file should contain variables of river discharge in m3/s:

RIVER_DISCHARGE_OVER_LAND_ICE
RIVER_DISCHARGE_OVER_LAND_LIQ

Check to make sure that "MOSART_MODE" is set to "ACTIVE" in env_build.xml.

There is some documentation on MOSART here:

River Runoff Component | Community Earth System Model

www.cesm.ucar.edu

 

[MAX]

You should be getting mosart history files with that compset if you have run at least one month, e.g.,

runoff_test_v1.mosart.h0.0001-01.nc

That file should contain variables of river discharge in m3/s:

RIVER_DISCHARGE_OVER_LAND_ICE
RIVER_DISCHARGE_OVER_LAND_LIQ

Check to make sure that "MOSART_MODE" is set to "ACTIVE" in env_build.xml.

There is some documentation on MOSART here:

River Runoff Component | Community Earth System Model

www.cesm.ucar.edu

Thank you for your reply!
I have run the compset for just 10 days, now I know that's why I don't get the mosart history files.

I set MOSART_MODE=ACTIVE, and then I get the following error on trying to build:
ERROR: ROF_GRID is null MOSART_MODE not NULL

I check the env_build.xml, in this file the ROF_GRID is null, the ATM_GRID and LND_GRID are set to CLM_USRDAT, and the ATM_NX and ATM_NY are set to 1. Now I have two questions:
(1) do I need to set ROF_GRID and ROF_NX, ROF_NY to the same value as ATM_GRID and the ATM_NX and ATM_NY?
(2) some posts in the forums(Inquiry about MOSART 8th/100 degree resolution input data) mentioned that the MOSAR can have different spatial resolution with CLM, and it needs a MOSART regional input file, can you give me more details about coupling CLM and MOSART?
Any help would be greatly appreciated!

 

[oleson]

What type of compset are you trying to setup now? The information you've provided indicate that you are setting up a single point case? In which case, MOSART_MODE should be NULL.

 

[MAX]

What type of compset are you trying to setup now? The information you've provided indicate that you are setting up a single point case? In which case, MOSART_MODE should be NULL.

I am running with my own regional 0.25*0.25 degree resolution, so the ATM_GRID and LND_GRID are set to CLM_USRDAT. I created the domain file and the fsurface file of this case, and the output runoff looks good in this area. But I don't know how to couple the MOSART. Thanks!

 

[oleson]

Ok, I thought maybe it was single-point because ATM_NX and ATM_NY are set to 1.
We don't have any instructions to set up MOSART for a regional grid.
There are new capabilities coming on board to subset data for regional simulations but I don't know if they would include subsetting for MOSART.
@erik @swensosc @negins

 

[erik]

We don't have the ability to run MOSART over regional areas out of the box as Keith points out above. I think it's true that if you subset the MOSART direction file to cover just over your region you should be able to get this to work. I know @swensosc has run regional grids with MOSART. We don't currently have the ability to do that with our subset_data tool, but it's something that I think we should be able to add.