Case run error: rtmini ERROR no downstream ocean cell

[Zhibo Li]

Hello, DiscussCESM

I'm a Ph.D student at Peking University, and I'm doing some research in geological deep-time paleoclimate by CESM1.2.2. I revised the land_sea mask, elevation, and kmt (in ocean) before I build the run. But, no matter how I try to make runoff downstream (resolution at 0.5*0.5 degree) to the ocean, there is always the only one error. The last few lines of my cesm.log are shown as belows:

........................................................................................................................
hist_htapes_build hist tape = 1 written with dov2xy= T
hist_htapes_build hist tape = 1 written with dov2xy= T
hist_htapes_build hist tape = 1 written with dov2xy= T
hist_htapes_build hist tape = 1 written with dov2xy= T
hist_htapes_build hist tape = 1 written with dov2xy= T
hist_htapes_build hist tape = 1 written with dov2xy= T
hist_htapes_build hist tape = 1 written with dov2xy= T
hist_htapes_build hist tape = 1 written with dov2xy= T
hist_htapes_build hist tape = 1 written with dov2xy= T
Opened existing file
/gpfs/share/home/2001110284/390manew/390manew_clm_fort.13.nc 458752
rtmini ERROR no downstream ocean cell 259200 258120 1
rtmini ERROR no downstream ocean cell 259200 258120 1
258840 258120 1 258840
(shr_sys_abort) WARNING: calling shr_mpi_abort() and stopping
rtmini ERROR no downstream ocean cell 259200 258120 1
258840 258120 1 258840
rtmini ERROR no downstream ocean cell 259200 258120 1
258840 258120 1 258840
(shr_sys_abort) WARNING: calling shr_mpi_abort() and stopping
(shr_sys_abort) WARNING: calling shr_mpi_abort() and stopping
rtmini ERROR no downstream ocean cell 259200 258120 1
258840 258120 1 258840
(shr_sys_abort) WARNING: calling shr_mpi_abort() and stopping
rtmini ERROR no downstream ocean cell 259200 258120 1
258840 258120 1 258840
(shr_sys_abort) WARNING: calling shr_mpi_abort() and stopping
rtmini ERROR no downstream ocean cell 259200 258120 1
258840 258120 1 258840
(shr_sys_abort) WARNING: calling shr_mpi_abort() and stopping
rtmini ERROR no downstream ocean cell 259200 258120 1
258840 258120 1 258840
(shr_sys_abort) WARNING: calling shr_mpi_abort() and stopping
application called MPI_Abort(MPI_COMM_WORLD, 1001) - process 17
application called MPI_Abort(MPI_COMM_WORLD, 1001) - process 28
........................................................................................................................

It seems like something wrong with the river runoff file. I had tried no less than 20 times to revise the elevation or kmt. Unfortunately, the bug is still there.
Is there anyone has ideas? Thanks a lot!!!

Zhibo

 

[shields@ucar_edu]

Hi Zhibo, My first guess is that you have an infinite loop in your RTM file. Have you used the RTM tool that creates and checks your vector directions? Here is the "recipe" for CESM1.2.2:

paleoToolkit/cesm1_2/PaleoToolkit_Recipe_2020Jan1.pdf at master · CESM-Development/paleoToolkit

paleo tools for cesm. Contribute to CESM-Development/paleoToolkit development by creating an account on GitHub.

github.com

I would double check you are using the latest tool. 0.5 degree resolution can be hard and tedious -- another option is that if you don't need that high resolution for your river runoff, to use 1 degree instead.

Good luck!

 

[Zhibo Li]

Hi Zhibo, My first guess is that you have an infinite loop in your RTM file. Have you used the RTM tool that creates and checks your vector directions? Here is the "recipe" for CESM1.2.2:

paleoToolkit/cesm1_2/PaleoToolkit_Recipe_2020Jan1.pdf at master · CESM-Development/paleoToolkit

paleo tools for cesm. Contribute to CESM-Development/paleoToolkit development by creating an account on GitHub.

github.com

I would double check you are using the latest tool. 0.5 degree resolution can be hard and tedious -- another option is that if you don't need that high resolution for your river runoff, to use 1 degree instead.

Good luck!

Hi, shields
Thanks for your good advice. The bug occurred due to the elevation problem in the polar region. But I got new bugs when the simulation output 15~20 model years data.



For the left one, I tried to revise "k.e. > 100" into "k.e. > 10000" in the file "POP.F90". For the right one, I tried to increase the dt_count from 23 to 26 in the file "user_nl_pop". Then I build these cases and submit them. But the errors are still there. I'm trying to find new solutions. Do you have any ideas?

Zhibo

 

[shields@ucar_edu]

Hi Zhibo,

It looks like you may have some major instabilities in the ocean model. We have published some recommendations in our Deep time "Recipe" found on the CESM paleo resources webpage.

Here is the direct link. Got to the end of the document and look for our suggestions in the "Gotcha's" section.

Good luck! c.

 

[Zhibo Li]

Hi Zhibo,

It looks like you may have some major instabilities in the ocean model. We have published some recommendations in our Deep time "Recipe" found on the CESM paleo resources webpage.

Here is the direct link. Got to the end of the document and look for our suggestions in the "Gotcha's" section.

Good luck! c.

Hi, shields

I found both of the problems are due to the big kinetic energy in ocean. I revised the "kmt" by putting a middle-ocean ringe in the north hemisphere, then all bugs gone. I already got 100 model years output.

Thanks for your help!

Zhibo