Julia Sloan
Julia Sloan
partitioned fluxes (`PartitionedComponentModelGrid`) are 60x slower, partly bc of the way it has to interface with current atmos (e.g. performing flux calc twice). See this build: https://buildkite.com/clima/climacoupler-ci/builds/1418#018993df-ed13-4b19-b1ac-3f8bb0d31576. We need to...
## Purpose We want to implement distributed matrix multiplication to enable parallel online regridding in the coupler. ## Cost/Benefits/Risks * Costs: * Developer time * Risks: * Incorrect regridding results...
It would be nice to be able to track conservation in our AMIP runs, especially given the discrepancies in conservation we've seen between CPU and GPU runs for slabplanet. We...
# The Climate Modeling Alliance ## Software Design Issue 📜 ### Purpose We want to compare our coupler to external couplers and models, including: - CESM/CMEPS --> paper? - MCM...
ClimaLand recently changed their turbulent fluxes to separately store SHF and LHF. ClimaAtmos currently just provides one combined turbulent fluxes field. We should choose one of these conventions to follow...
Our windows GHA are failing with `ERROR: LoadError: Unable to open libLLVM!`. For now, the windows GHA is no longer required for merging a PR into main. We need to...
We currently have a test for Eisenman sea ice ([eisenman_seaice_tests.jl](https://github.com/CliMA/ClimaCoupler.jl/blob/main/test/component_model_tests/eisenman_seaice_tests.jl)) which constructs and steps a simulation. This requires functionality from SciMLBase, so that package has to be added to the...
**Is your feature request related to a problem? Please describe.** In https://github.com/CliMA/ClimaLand.jl/pull/651, we convert the `SoilCanopy` model from being entirely explicitly-stepped to using explicit and implicit timestepping for different variables....
**Is your feature request related to a problem? Please describe.** In https://github.com/CliMA/ClimaLand.jl/pull/651, we hardcode which variables are to be stepped explicitly and which will be stepped implicitly in our Jacobian...
**Is your feature request related to a problem? Please describe.** There is a significant amount of duplicated code between the `update_jacobian` functions for `RichardsModel` and for `EnergyHydrology`. It would be...