The Design Document page provides a description of the algorithms, implementation and planned testing including unit, verification, validation and performance testing. Please read Step 1.3 Performance Expectations that explains feature documentation requirements from the performance group point of view.
...
ELM: modify the land shortwave code to include a subroutine that adopts the new radiative transfer algorithm.
MPAS-seaice: modify the sea-ice shortwave radiative transfer code to include a new subroutine that
- provides new input files for snow and ice optics adopted by ELM;
- expands the shortwave radiative calculation from three to five subdivided solar bands adopted by ELM;
- removes empirical adjustment for radiative-effective snow grain size;
- adopts a new algorithm to correct for the bias of near-infrared albedo and absorption when solar zenith angles are larger than 75 degrees.
Date last modified: Jun 1st, 2019
Contributors: Cheng Dang (Unlicensed) Charlie Zender
Algorithmic Formulations
Design solution: described in the following article
Dang, C., Zender, C. S., and Flanner, M. G.: Inter-comparison and improvement of 2-stream shortwave radiative transfer models for a unified treatment of cryospheric surfaces in ESMs, The Cryosphere Discuss., https://doi.org/10.5194/tc-2019-22, in review, 2019.
Date last modified: July 1st, 2019
Contributors: Cheng Dang (Unlicensed) Charlie Zender Mark Flanner
...
Design and Implementation
Implementation:
In the sea-ice and land shortwave code (ice_shortwace.F90, SnowSnicarMod.F90), we implemented SNICAR_AD that can be turned on/off by configuration toggles in the model namelists.
MPAS-source code: https://github.com/chengdang/MPAS-Model/tree/seaice/cdang_SnicarAD
ELM-source code: https://github.com/chengdang/E3SM/tree/chengdang/lnd/SNICAR_AD
Verification and Unit Testing:
- The offline radiative transfer model has been tested against a multi-stream benchmark model. (Dang, C., Zender, C. S., and Flanner, M. G.: Inter-comparison and improvement of 2-stream shortwave radiative transfer models for a unified treatment of cryospheric surfaces in ESMs, The Cryosphere Discuss., https://doi.org/10.5194/tc-2019-22, in review, 2019.)Results of the
Date last modified: July 1st, 2019
Contributors: Cheng Dang (Unlicensed)
Planned Validation Testing
Validation Testing:
- The modified E3SM model are tested against E3SM_v1 released pre-industrial simulations:is tested with Pi-Control configuration, against the released Deckv1b simulation (mpas-analysis, e3sm diagnostics, and land comparison carried by the developer)
- 30-year MPAS-analysis: https://portal.nersc.gov/project/m2833/cdang/30yr.mpas_analysis.m2m.SNICAR_AD_landseaice_vs_DECKv1b.piControl.ne30_oEC.edison/
- 30-year E3SM diagnostics: https://portal.nersc.gov/project/m2833/cdang/30yr.e3sm_diags.m2m.SNICAR_AD_landseaice.SNICAR_AD_seaice.ne30_oEC.edison/viewer/
- Results of the modified E3SM model will be tested against E3SM_v1 released historical simulations (Fully-coupled historical simulations are in progress).
Date last modified: July 1st, 2019
Contributors: Cheng Dang (Unlicensed)
Planned Validation Testing
Validation Testing:
- Results of the modified E3SM model are tested against observation provided by mpas-analysis, e3sm diagnostics.
- Results of the modified E3SM model will be tested against observations/reanalysis/model simulations of snow depth, sea-ice concentration/fraction, dust and black carbon concentration (Fully-coupled historical simulations are in progress).
- BC/Dust in snow: in-situ observations (Doherty et al., 2010, Wang et al., 2013, Doherty et al., 2014, Dang et al., 2017)
- Sea-ice concentration/extent: PIOMAS
- Snow depth: to be added.
Date last modified: July 1st, 2019
Contributors: Cheng Dang (Unlicensed)
Planned Performance Testing
Performance Testing:
The new implementation is not expected to have a significant impact on the computational cost/performance of E3SM.
Date last modified: July 1st, 2019
Contributors: Cheng Dang (Unlicensed)
...