Issues are ordered from most recent to least recent. 

NOTE: The SE Advection error and CLM time-averaging errors are fixed in recent versions of CESM2.2.

CLUBB interface and water vapor

Date of report: June 2024
Title: CLUBB conversion from wet to dry mixing ratios
Issue: CLUBB (Cloud Layers Unified By Binormals) is a sub-gridscale parameterization of clouds and turbulence. It includes a conversion from wet to dry mixing ratios, that was incorrectly set to "double dry", and wet tracers were not converted to dry.
Model tags affected: The issue was committed in tag cam6_3_094 and remained later versions.
Model response: Impacts on stratospheric water vapor and potential influence on the "tape recorder".
Fix: Correct code in clubb_intr.F90.
More information: Issue described here: ESCOMP GitHub issue 1053
Code fix here: AMWG GitHub Issue 545

FHIST and F2000climo compsets not defined properly

Date of report: December 2023
Title: Failure to build FHIST and F2000climo compsets
Issue: For CESM2.2 on Derecho, the default FHIST and F2000climo compsets are not defined properly. 
Fix: Try using: 
--compset HIST_CAM60_CLM50%SP_CICE%PRES_DOCN%DOM_MOSART_SGLC_SWAV_SIAC_SESP

Regridding error

Date of report: November 2023
Title: IPT NCL regridding code does not recognize ncl
Issue: Casper has problems loading modules within a script (should be fixed by Dec 1), such as https://github.com/NCAR/IPT/blob/master/Meterological_Reanalysis_Data/Spectral_element_dycore/Gen_Data_ne0CONUSne30x8/Gen_MERRA2_ne0CONUSne30x8_mt12.csh
Model tags affected: Pre-simulation input data processing.
Fix: within a PBS script use:

source /etc/profile.d/z00_modules.csh 
module load ncl

More information: 

Sometimes the script crashes because it cannot find the log directory. Add the following lines to fix:

if( ! -d $MYLOGDIR ) then
     mkdir $MYLOGDIR
else
     echo "Directory exists"
endif

Topography file error

Date of report: February 2023
Title: Topography file error 2022
Issue: Regional refined topography was in the incorrect location/orientation
Model tags affected: NCAR_Topo_2_0_1 files for regional refinement. An erroneous topography file was specified for use when creating variable resolution grids between April 2022 and Jan 2023. 
Model response: Gravity waves impacted, incorrect transport, chemistry fields displaced
Fix: Reprocess topography and re-run model simulations. If you created your own grid during that period please checkout the updated code at:  https://GitHub.com/NCAR/Topo and recreate your topography file for your grid and re-run any simulations. To check if your topography file has this error, plot ANGLL vs ANGLX.  In a "good" topo-file these variables should have large differences. In a "bad" file they will be very close everywhere.

Also be sure to recreate any regridded met (GEOS or MERRA2) files that used the incorrect topo file.
More information: AMWG Github issue 213

Advection error in SE

Date of report: 2022
Title: Advection error in Spectral Element dycore
Issue: Spurious vertical transport was found over steep topography.  The issue is described in this AMWG GitHub issue #151. 
Model tags affected: ???
Model response: ???
Fix: For CESM2.2.0 source code the correction is, in src/dynamics/se/dycore/prim_advance_mod.F90 (at l.578), replace:

                 elem(ie)%derived%dpdiss_ave(i,j,k)=elem(ie)%derived%dpdiss_ave(i,j,k)+&
                      rhypervis_subcycle*eta_ave_w*elem(ie)%state%dp3d(i,j,k,nt)

with:

                 elem(ie)%derived%dpdiss_ave(i,j,k)=elem(ie)%derived%dpdiss_ave(i,j,k)+&  
                      rhypervis_subcycle*eta_ave_w*(elem(ie)%state%dp3d(i,j,k,nt)-dp3d_ref(i,j,k,ie))

More information: The correction for current development tags is provided in CAM github issue #633. 
Update Nov 1, 2022: Please see additional code fix described in CAM github issue #678.

CLM time-averaging bug affecting MEGAN emissions

Date of report: 8 July 2022
Title: Incorrect calculation of MEGAN emissions
Issue: A bug has been identified in CLM that probably has affected most MUSICAv0 simulations. 
Model tags affected: MEGAN emissions, especially in regional refined simulations. MUSICAv0 simulations before July, 2022.
Model response: This bug results in erroneous calculations of the time-averaged (1-day, 10-days) temperature, photon flux, and other variables, that are used in the MEGAN biogenic emissions calculation (and elsewhere in CLM).
Fix: Added a fix to accumulation code CTSM/src/main/accumulMod.F90
In general, Isoprene emissions are reduced with the bug fix.

More information: See discussion at the CTSM GitHub issue #1789. 

Chemistry mechanism bug

Date of report: 2021
Title: MPAN+M chemistry bug
Issue: In the TS1 and TSMLT1 chemistry mechanisms (compsets FCnudged, FCHIST, FWHIST, etc.) the MPAN+M reaction is not calculated because the user reaction refers to a reaction not in the mechanism (tag_MCO3_NO2 instead of usr_MCO3_NO2).  This error does not exist in the TS2 chemical mechanisms.
Model tags affected: Not in versions CESM2.1 and prior. Found in version CESM2.2, May 2021. Fixed in version CAM6_3_019.
Model response: Missing chemistry.
Fix: Update chemical mechanism from:

[usr_MCO3_NO2] MCO3 + NO2 + M -> MPAN + M

to:

[tag_MCO3_NO2] MCO3 + NO2 + M -> MPAN + M. ; 9.7e-29, 5.6, 9.3e-12, 1.5, 0.6

More information:  CAM GitHub issue #364 
Contact Louisa Emmons with any questions.



  • No labels