Main menu


error in running deep time simulation

2 posts / 0 new
Last post
error in running deep time simulation
Hi,  I have been trying to do a deep time experiment in which both North American and South American continents are shifted eastward. I prepared the atmosphere boundary topography data, land surface data, ocean domain file, land domain file and mapping data following the instruction of the paleosource documentation, but the model crashed at time step 0. Could anyone give any help?   Here is how I did the experiment: 
case description: Component set Description: CAM: CLM: SROF: SICE: DOCN: SGLC: SWAV: pre-industrial: cam5 physics: clm4.0 physics: clm4.0 Satellite phenology: docn slab ocean mode:

Grid: a%1.9x2.5_l%1.9x2.5_oi%gx1v6_r%r05_m%gx1v6_g%null_w%null (1.9x2.5_gx1v6)



  • Starting from topography data, shift the continents using my own python script

  • run definesurf with Ross Ice Shelf turned OFF (-r option) by calling

  • Add SGH30 by calling add_SGH30_paleo.ncl



  • starting from the mksrf files, shift the continents eastward using my own python script

  • mkmapgrid: Create a SCRIP grids by running ncl mkscripgrid_paleo.ncl

  • mkmapdata: by running bsub <

  • mksurfdata: by running ./mksurfdata_map <

  • mksurfdata_map.namelist  (I modified 

  • pftdyn_hist_simyr1850-2005.txt to included only year 1850 before generating the surfdata)

land/ocean domain files and mapping files:

  • changing the kmt file topography_20090204.ieeei4 with my own python script

  • Create SCRIP grid file using mk_SCRIPTgrid.csh

  • generate the mapping files using

  • generate the domain files using gen_domain

Error: NetCDF: Numeric conversion not representable

   1: pio_support::pio_die:: myrank=          -1 : ERROR:

   1: pionfwrite_mod::write_nfdarray_double:         246 :

Debugging I have tried
  1. checked the latitude matching between land domain, atmosphere initial file and land surface data. The latitude is from -90 to 90 for all. 
  2. ran scrip_test (by the way, the PaleoTookit package in GitHub missing the scrip_test_in file). Mapping file has an error in the order of 1e-3. 
  3. wrote instantaneous history files for cam and clm. Dclm stopped after timestep 1; cam stopped after timestep 0. 
  4. set info_debug = 3 in env_run.xml; didn't help. 
  5. turned on debugging; didn't help. 
  6. changed the source code to write extra output; didn't help. 

Update: I checked the bnd_topo data and surfdata using np.sum(isnan(var)). There is no missing value in the files. Any thought where the error might be from?

Log in or register to post comments

Who's new

  • shona.mackie@...
  • hgadhavi@...
  • q7peng@...
  • jingyu.yao@...