I1850Clm50Bgc error

[qian yang]

Hello, when I run I1850Clm50Bgc with CESM2.0, an error,
Writing to file yang.cism.initial_hist.0001-01-01-00000.nc at time 0.0000000000000000
MCT::m_SparseMatrixPlus:: FATAL--length of vector y different from row count of sMat.Length of y = 25 Number of rows in sMat = 13824
000.MCT(MPEU)::die.: from MCT::m_SparseMatrixPlus::initDistributed_()
application called MPI_Abort(MPI_COMM_WORLD, 2) - process 0
MCT::m_SparseMatrixPlus:: FATAL--length of vector y different from row count of sMat.Length of y = 25 Number of rows in sMat = 13824
001.MCT(MPEU)::die.: from MCT::m_SparseMatrixPlus::initDistributed_()
application called MPI_Abort(MPI_COMM_WORLD, 2) - process 1
MCT::m_SparseMatrixPlus:: FATAL--length of vector y different from row count of sMat.Length of y = 25 Number of rows in sMat = 13824
003.MCT(MPEU)::die.: from MCT::m_SparseMatrixPlus::initDistributed_()
application called MPI_Abort(MPI_COMM_WORLD, 2) - process 3

But I do not know where it is wrong. I don not know how to solve it . Can anyone help me ?

 

[sacks]

From what I can tell from the log files you have posted, it looks like you are running a regional case (I see reference to surfdata_16x16pt_Hulunbuir_simyr2000_c210816.nc). But to do that, there is more you need to do than just point to the new surface dataset. There is some information in 1. CTSM1 User’s Guide — ctsm CTSM master documentation . One thing needed in a regional case is to use a stub glacier model (SGLC), instead of the default CISM (I'm not sure if that is called out in the user's guide). For the compset you listed, the compset long name is 1850_DATM%GSWP3v1_CLM50%BGC_SICE_SOCN_MOSART_CISM2%NOEVOLVE_SWAV (you can find this in README.case). You can change this to use stub glc by creating a new case where you use this long name for --compset, but changing CISM2%NOEVOLVE to SGLC: 1850_DATM%GSWP3v1_CLM50%BGC_SICE_SOCN_MOSART_SGLC_SWAV. Similarly, you will need a stub runoff model (SROF) in place of MOSART, but I think this may happen automatically with the processes documented for single point and regional runs in the User's Guide. If you have further questions about running a regional case, you can post a question to the land forums (CTSM, CLM, MOSART, RTM).

If you would like further help, please look at Information to include in help requests and provide the information requested there, including all steps you took to set up your case.

Also, a few other minor points:

- In user_nl_clm you don't need "&clm_inparm"

- For most work, we recommend using the latest CESM2.1 version (CESM2.1.3) instead of CESM2.0. CESM2.1 has a number of bug fixes and is the scientifically supported release version.

 

[qian yang]

I am so grateful for your help. @sacks. I create new case with your suggestion .And It runs fine.