CESM Models

Known Problems in CESM1.2.0

The CESM1.2 release has numerous new key features among which are the addition of CLM4.5, new science changes to CAM5 running with the CAM-SE dynamical core, and new scripting infrastructure for the generation of component sets, grids and model testing.

Show/Hide All

Component Sets (COMPSETS)

The B_1850_BGC_BDRD, B_1850_BGC_BPRP, B_1850-2000_BGC-BDRD and B_1850-2000_BGC-BPRP compsets must be run from cold start conditions in CESM 1.2.0. If you need to start from a spun-up initial condition, please use CESM 1.0.5.


(Bugzilla 1750) The ne240_ne240 grid as defined in the cesm 1.2.0 release is incomplete.


(Bugzilla 1872) RRTMG gives biased results on big-endian architectures
(Bugzilla 1760) F_1850_CAM5_PM has hard coded yellowstone paths for prescribed_aero_datapath and aerodep_flx_datapath.
(Bugzilla 1727) CLUBB cannot be run with OMP_NUM_THREADS greater than 1.


Known Issue - namelist description of "do_circulation_diags" in CAM's namelist_definition.xml should include:
(Bugzilla 1768) The correct reference case for the nuclear winter compset is:
(Bugzilla 1766) The SE version of gravity_waves_sources in CESM 1.2 has a bug and workaround.
(Bugzilla 1744) Potential problem with WACCM-SE and threading
(Bugzilla 1612) There is a known problem with the following WACCM compsets when the ESMF time manager is used.


(Bugzilla 1761) The variable orb_velp should be orb_mvelp in the namelist_definition_drv.xml file.
(Bugzilla 1740) Limitations with different coupling frequencies for different models.
(Bugzilla 1733) Driver needs a different list hardcoded in for hist_a2x3hr_flds (important when trying to save data to run CLM later, i.e. CLM spinup).
(Bugzilla 1339) Problem running with "too many files" for datm for over 83 years.


(Bugzilla 1867) irrigated soybean not treated same as rainfed soybean
(Bugzilla 1838) The population density is set to 1850 for 2000_control namelist use-cases.
(Bugzilla 1815) For both CLM4.0 and CLM4.5, crop GDD is incorrect for the first ~20 years.
(Bugzilla 1805) There is a bug in clm4.5 fire model connected to fpftdyn/=' '
(Bugzilla 1776) Issue with CLM45CNDV trees over Amazon.
(Bugzilla 1759) For CLM4.0, the RCP 8.5 file pftdyn for f09 was NOT correctly updated.
(Bugzilla 1634) clm4_0_60 interpinic fails to properly map from 1.9x2.5 to the 360x720 grid for CRUNCEP.
(Bugzilla 1753) Problem with an initial file used in CROP simulations.
(Bugzilla 1734) Problem with branch startup for CLM single-point simulations.
(Bugzilla 1732) Value of the VOLR diagnostic variable not correct when passed from RTM to CLM. Also effects TWS.
(Bugzilla 1719) Minor -- false overestimation of fire C emissions for natural vegetation PFTs without transient land-use change.
(Bugzilla 1717) Minor -- lake hydrology bug when sudden intense snowstorm occurs preceeded by warm weather.
Please refer to the Known Bugs for CLM document for a detailed list of known issues specifically addressing custom configurations for CLM.


(Bugzilla 1871) Issue re-running a multi-instance case that has produced rpointer file.
(Bugzilla 1870) domain_size.F90 and POP_DomainSizeMod.F90 are not handled correctly in SourceMods/
(Bugzilla 1866) riverine Fe flux overwrites atm Fe flux, instead of being added to it
(Bugzilla 1861) Issue using RESUBMIT option in env_run.xml when setting certain variables in user_nl_pop2
(Bugzilla 1802) global Si imbalance in ecosys
(Bugzilla 1803) scalelength for depth dependent remin length scale not applied consistently.
(Bugzilla 1622) CFC initialization incorrect for 20C branch & hybrid runs


The CICE history files have some incorrect history fields (albpnd, albice, albsno). These should not be used.


(Bugzilla 1611) Restarts can only happen on year boundaries.


(Bugzilla 1777) There is a problem with setting up a branch simulation when using RTM. The problem occurs when ./cesm_setup is invoked.

Machines / Scripts

(Bugzilla 1882) Bug in config_grid for a couple of tri-grid setups
(Bugzilla 1793) The pgi compiler is currently failing in calls to mpi_init on yellowstone.
Information regarding port of CESM 1.2.0 to edison.
(Bugzilla 1748) CESM on stampede is still in development.
(Bugzilla 1594) Problem and fix when running create_clone
(Bugzilla 1743) Coordination of variables EXEROOT and RUNDIR in env_build.xml and env_run.xml
(Bugzilla 1737) Decomposition problem on Mira using compset BG1850CN at resolution f09_g16.
(Bugzilla 1736) Intrepid: There is an xml compiler problem with threads, that prevents ne30 F compsets from being able to restart correctly.
(Bugzilla 1735) Janus is not supported for this release.