v0.5.3

NumericalEarth v0.5.3

Diff since v0.5.2

Closed issues:

• Baroclinic example definition of balanced u (#289)

v0.5.2

NumericalEarth v0.5.2

Diff since v0.4.1

Closed issues:

• Unify JRA55 with all the other datasets (#18)
• Cleaning imports (#229)
• ERA5 pressure-level ingest: per-column geopotential and surface-pressure masking (#236)
• OceanSeaIceModel() default atmosphere=nothing not supported (#268)

v0.5.1

NumericalEarth v0.5.1

Diff since v0.4.1

Closed issues:

• Unify JRA55 with all the other datasets (#18)
• Cleaning imports (#229)
• ERA5 pressure-level ingest: per-column geopotential and surface-pressure masking (#236)

v0.5.0

NumericalEarth v0.5.0

Diff since v0.4.1

Breaking changes

• Flipped arguments so syntax reads OceanSeaIceModel(ocean, sea_ice; kw...).

Closed issues:

• Unify JRA55 with all the other datasets (#18)
• Cleaning imports (#229)
• ERA5 pressure-level ingest: per-column geopotential and surface-pressure masking (#236)

v0.4.1

What's Changed

• Combine all ERA5 examples by @ewquon in #209
• Extract pure build_era5_request helper for unit testability by @ewquon in #212
• Temporarily use older versions of Reactant and GPUCompiler by @simone-silvestri in #217
• Add skin temperature to ERA5 single-levels by @ewquon in #218
• Fix SpeedyWeather output frequency set up by @maximilian-gelbrecht in #216
• Set JULIA_CUDA_USE_COMPAT=false in test / docs by @simone-silvestri in #225
• Add SSH key to TagBot configuration by @navidcy in #221
• Integrate ClimaSeaIce snow model by @simone-silvestri in #164
• Fix a typo in docs/Project.toml by @simone-silvestri in #227
• Fix ESM reset_clock! component clocks and add test by @taimoorsohail in #211
• Add bibtex refs + admonitions in ERA5_hourly_data.jl example by @navidcy in #238
• Avoid computing frazil fluxes when sea ice model does not exist by @taimoorsohail in #226
• Reinitialize download progress when dowloading a new file by @simone-silvestri in #243
• Replace XESMF with ConservativeRegridding by @simone-silvestri in #160
• Minor cleanup of test_diagnostics_2.jl by @navidcy in #240
• Add AbstractPrescribedComponent abstract type by @navidcy in #228
• CompatHelper: bump compat for Breeze in [weakdeps] to 0.5, (keep existing compat) by @github-actions[bot] in #245
• Add AGENDS.md/CLAUDE.md by @navidcy in #137
• Quality control using ExplicitImports (part 1) by @navidcy in #246
• Correct orca grid for Oceananigans 0.107.7 by @simone-silvestri in #252
• Fix global_climate_simulation example by @simone-silvestri in #251
• Update Oceananigans version in Project.toml by @navidcy in #253
• Quality control using ExplicitImports (part 2) by @navidcy in #248
• Format NumericalEarth module docstring by @navidcy in #250

New Contributors

• @github-actions[bot] made their first contribution in #245

Full Changelog: v0.4.0...v0.4.1

v0.4.0

NumericalEarth v0.4.0

Diff since v0.3.0

Breaking changes

• Moved radiation to a top-level model components (#200). This completely changes the syntax for building EarthSystemModel with radiation and specifying surface properties.

Closed issues:

• Add radiation as a top-level EarthSystemModel component (#30)
• ERA5 data wranglers only set up for single levels (#88)
• Docs not building on main (#185)
• CPU CI takes over an hour and seems to overlap with data downloading (#193)
• Inconsistent return types across download_dataset implementations (#201)

v0.3.0

NumericalEarth v0.3.0

Diff since v0.2.3

Breaking changes

• Adds a land component, and splits out JRA55PrescribedLand from JRA55PrescribedAtmosphere
• adds Column to BoundingBox for Metadata, and renames Metadata.bounding_box to Metadata.region

Closed issues:

• Add land as a top-level EarthSystemModel component (#36)
• Wrong signature for SpeedyWeather.parameterization_tendencies! in SpeedyWeatherExt for SpeedyWeather v0.18+ (#132)
• Simpler bounding box creation (#138)
• Documentation erroring on main (#176)
• WOA Tests flakiness (#183)

v0.2.3

NumericalEarth v0.2.3

Diff since v0.2.2

This release has been identified as a backport.
Automated changelogs for backports tend to be wildly incorrect.
Therefore, the list of issues and pull requests is hidden.

v0.2.2

NumericalEarth v0.2.2

Diff since v0.2.1

Merged pull requests:

• Add free surface output to the one degree example (#118) (@simone-silvestri)
• Add filename property to Metadata (#121) (@glwagner)
• Informative error when interpolating out-of-bounds (#124) (@simone-silvestri)
• Using ParallelTestRunner to (immensely [edit by navidcy]) speed up tests (#135) (@simone-silvestri)
• Remove various unused dependencies (#136) (@giordano)
• (0.2.2) Allow using Oceananigans 0.106 (#139) (@simone-silvestri)

Closed issues:

• Should we spawn more agents for the tests? (#28)
• One-degree simulation initialised with World Ocean Atlas gives NaN immediately (1 timestep) (#96)
• Have Metadata path support in DataWrangling. Support multiple variables (names) inside Metadata. (#115)

v0.2.1

NumericalEarth v0.2.1

Diff since v0.2.0

Merged pull requests:

• Use cpu_interpolating_time_indices for ECCOPrescribedAtmosphere (#56) (@xkykai)
• Add ERA5 wave variables and switch to CDSAPI.jl (#70) (@glwagner)
• Added heat and freshwater flux diagnostics (#71) (@taimoorsohail)
• Build ERA5 winds+stokes drift demo every commit + add CDSAPI in docs/Project (#72) (@glwagner)
• Pass grid in sea ice SplitExplicitSolver (#82) (@navidcy)
• Adding the option to use the ORCA1 grid and bathymetry (#83) (@simone-silvestri)
• Adding WOA to the datasets via extension for WorldOceanAtlasTools.jl (#84) (@simone-silvestri)
• Mask out immersed cells in flux computations (#87) (@simone-silvestri)
• Use zstar in the one degree example (#91) (@simone-silvestri)
• Generalizing the ORCA grid (#92) (@simone-silvestri)
• Update author information in CITATION.cff (#98) (@navidcy)
• Inferred salt fluxes --> Freshwater volume fluxes (#99) (@taimoorsohail)
• Fix docstring rendering (#100) (@navidcy)
• Fix ecco atmosphere (#101) (@simone-silvestri)
• Correct WOA bounds (#104) (@simone-silvestri)
• Refactor BreezeExt for Breeze v0.4 moisture variable rename (#110) (@glwagner)
• Revert build status for one-degree ocean simulation to false (#122) (@navidcy)
• Move CI to self-hosted AWS runners for GPU tests (#127) (@awiteck)
• Add proper test/ environment project file (#129) (@giordano)
• Require a much more recent version of Reactant (#130) (@giordano)
• Bump version from 0.2.0 to 0.2.1 (#134) (@simone-silvestri)

Closed issues:

• ECCOPrescribedAtmosphere causes error when time indices are interpolated in OceanSeaIceModel (#48)
• ECCOPrescribedAtmosphere gives NaNs after one timestep for global simulations (#68)
• Switch ERA5 download backend from CopernicusClimateDataStore.jl to CDSAPI.jl (#69)
• Checkpointing for sea ice with distributed is not working (#74)
• Add ORCA grid option in NumericalEarth (#76)
• Add Word Ocean Atlas in DataWrangling module (#77)
• Output heat fluxes in W/m^2 and salinity fluxes in kg/(m^2*s) (#78)
• Penetrative radiation + z-star vertical coordinate (#81)
• Cook lunch for the Milan hackathon participants on Wed March 4 (#95)
• One-degree simulation initialised with ECCOPrescribedAtmosphere gives NaN immediately (1 timestep) (#97)
• Refactor Breeze Ext now that total moisture variable was dropped (#107)
• One degree ocean sea ice model example crashes (#109)
• Without XESMF the atmospshere_simulation is not loaded correctly (#131)