Support ISMIP6 Forcing Generation for Greenland

[andrewdnolan]

This PR aims to enable ISMIP6 forcing file generation for the Greenland Ice Sheet.

Checklist

• User's Guide has been updated
• Developer's Guide has been updated
• API documentation in the Developer's Guide (api.rst) has any new or modified class, method and/or functions listed
• Documentation has been built locally and changes look as expected
• The E3SM-Project submodule has been updated with relevant E3SM changes
• The MALI-Dev submodule has been updated with relevant MALI changes
• Document (in a comment titled Testing in this PR) any testing that was used to verify the changes
• New tests have been added to a test suite

[andrewdnolan]

Based on offline discussion, I don't actually think this needs a MALI PR. The elevation dependent components of the the forcing (i.e. dSMBdz and dSTdz) are processed and available to be used, but do not have to be used for the forcing to work.

[andrewdnolan]

@trhille @matthewhoffman Now I'm remembering what stalled this PR originally.

What should we do about atm/ocn data during a historical run? Should we do a different historical run for each GCM? I think I recall ISMIP6 cautioning against that.

https://theghub.org/groups/ismip6/wiki/ISMIP6-Projections-Greenland

The unforced control run should be easy to support.

[trhille]

@andrewdnolan, I'm sorry I missed this comment from long ago. We will need to see what the constraints are in the ISMIP7 protocol, but below are some relevant bits from Appendix A in Goelzer et al. (2020) . It mostly seems like a mix of (a) SMB-only forcing, (b) forcing by averaging ESMs, and (c) forcing with a single ESM. Another option would be to use time-varying RACMO SMB plus ocean forcing via observed terminus positions.

• AWI-ISSM: "The historical run employs SMB from RACMO2.3p2 (Noël et al., 2018) and no oceanic forcing."
• BGC-BISICLES: "The historical run uses atmosphere and ocean forcing averaged over nine GCMs: ACCESS1.3-rcp85, CNRM-CM6-ssp126, CNRM-CM6-ssp585, CSIRO-Mk3.6-rcp85, HadGEM2-ES-rcp85, IPSL-CM5-MR-rcp85, MIROC5-rcp26, MIROC5-rcp85 and NorESM1-rcp85"
• GSFC-ISSM: Not specified
• ILTSPIK-SICOPOLIS: "The historical run from 1990 until 2015 employs the MIROC5-RCP8.5 atmospheric forcing and no oceanic forcing."
• IMAU-IMAUICE: "The historical run (1960–2014) is forced with SMB anomalies from a downscaled RACMO product (Noël et al., 2016) and a historical extension of the retreat mask forcing."
• JPL-ISSM: "For the historical experiment, MAR 3.9 yearly anomalies in SMB (Fettweis et al., 2017) from the 1979–1988 mean are added to the spin-up SMB (i.e. the Box, 2013, 1979–1988 climatology)."
• JPL-ISSMPALEO: "From 1840 to 1979, the model is then forced with the surface mass balance history derived in Box (2013), and from 1979 to 2014, the RACMO2.3 (Noël et al., 2015) surface mass balance is used." (doesn't mention historical ocean forcing)
• LSCE-GRISLI: "Atmospheric forcing, namely near-surface air temperature and surface mass balance, is taken from the 1995–2014 climatological annual mean computed by the MAR version 3.9 regional atmospheric model. " (also doesn't mention historical ocean forcing)
• MUN-GSM: "Surface mass balance depends on both positive degree days (PDDs) and (orbitally updated) monthly mean insolation. Ocean temperatures (for submarine ice melt) are derived from scaling the results of the TRACE deglacial simulation with CSSM3 (Liu et al., 2009). "
• NCAR-CISM: " For the historical period (1990–2014), the model was run forward with SMB and surface temperature anomalies, including lapse-rate corrections, from the MAR simulation that provided the background climatology."
• UAF-PISM: Not specified
• UCIJPL-ISSM: "The historical run was performed with SMB anomalies of MIROC5 provided by ISMIP6"
• VUB-GISM: "The historical experiment is run from January 1990 to December 2014 using the yearly SMB from MAR forced by MIROC5."
• VUW-PISM: Not specified.