Published September 11, 2026 | Version v0.11.0

Breeze.jl

  • 1. Aeolus Labs, Massachusetts Institute of Technology
  • 2. University of Melbourne
  • 3. University College London
  • 4. Aeolus Labs
  • 5. University of Tasmania

Description

Breeze v0.11.0

Diff since v0.10.0

Breaking changes

  • Energy and water BCs / forcings are keyed by formulation-agnostic interface names, ρE and ρqᵗ. Other keys are available only when they are genuinely prognostic: ρs (with formulation = :StaticEnergy), ρe (using a prognostic TKE closure), and ρqᵉ/ρqᵛ (for equilibrium/non-equilibrium microphysics); otherwise an ArgumentError is raised at construction (https://github.com/NumericalEarth/Breeze.jl/pull/974)
  • Parameters/Properties suffixes are dropped from types and from the objects instantiated from them, everywhere outside microphysics (https://github.com/NumericalEarth/Breeze.jl/pull/957)
  • A scalar base_pressure is the new ReferenceState datum at z = 0; surface_pressure now names the 2D pressure field at each column's bottom face (https://github.com/NumericalEarth/Breeze.jl/pull/893)

Merged pull requests:

  • Add TKEBasedTurbulenceClosure, a CATKE-style prognostic-TKE closure with constant stability functions (#875) (@ewquon)
  • Support Lagrangian particles in AtmosphereModel (#876) (@glwagner)
  • Resolve the three readings of surface_pressure (#893) (@kaiyuan-cheng)
  • Mass-weight the vertically implicit diffusion solve, fill closure-field halos, route the implicit solve by name (#925) (@glwagner)
  • Accept any AbstractDateTime in the solar geometry (#929) (@Pangoraw)
  • [docs] Use DocumenterLandingPage for the landing page (#940) (@giordano)
  • [docs] Allow Documenter v1.19 (#946) (@giordano)
  • Bump julia-actions/cache from 3.1.0 to 3.3.0 (#949) (@dependabot[bot])
  • Bump dorny/paths-filter from 4.0.2 to 4.0.3 (#950) (@dependabot[bot])
  • Bump JuliaRegistries/TagBot from 1.25.10 to 1.25.11 (#951) (@dependabot[bot])
  • Density-weight the vertically implicit diffusion solve for ρw (#952) (@ewquon)
  • Expose the hard-coded P3 empirical parameters (#954) (@glwagner)
  • Update README.md (#955) (@awiteck)
  • Drop Parameters/Properties suffixes outside Microphysics (#957) (@glwagner)
  • Bump CloudMicrophysics in / (#961) (@dependabot[bot])
  • Add examples to the workspace (#962) (@giordano)
  • Various minor improvements to the docs (#963) (@giordano)
  • Fold the thermodynamic AIVA remainder into the acoustic substep loop (#964) (@glwagner)
  • Bulk drag, heat, and vapor fluxes on all six walls (#967) (@glwagner)
  • Add single-column / column-ensemble mode for AtmosphereModel (#968) (@glwagner)
  • [CI] Set JULIA_CPU_TARGET for GitHub-hosted x86-64 jobs (#970) (@giordano)
  • [docs] Cap compat with AtmosphericProfilesLibrary (#972) (@giordano)
  • Restore "duplicate" compat bounds in subprojects of the workspace (#973) (@giordano)
  • Add ρE and ρqᵗ interface keys, and validate boundary condition and forcing names (#974) (@glwagner)
  • Revert "[docs] Cap compat with AtmosphericProfilesLibrary" (#977) (@giordano)
  • Implement checkpointing for AtmosphereModel (#978) (@ewquon)
  • Evaluate the SSP RK3 third-stage tendency at tⁿ + Δt/2 (#980) (@kaiyuan-cheng)
  • (0.11.0) Bump version (#987) (@ewquon)
  • [CI] Move non-GPU jobs to Julia v1.13.0 (#988) (@giordano)

Closed issues:

  • Add SingleColumnModel mode (#179)
  • Checkpointing for AtmosphereModel (#443)
  • Duplicate compat entries for deps? (#563)
  • Closure-field halos are never filled, halving the vertical viscous stress in the first column (#870)
  • Vertically-implicit diffusion is not mass-flux weighted, giving a spurious c ∝ 1/ρ steady state (#871)
  • surface_pressure is read as three different quantities; flat and terrain reference states disagree on the datum (#887)
  • Adaptive implicit vertical advection NaNs immediately on a terrain-following nested domain (#897)
  • Bounds-preserving WENO + AIVA double-counts the vertical flux (transports 1 + (1−s) times) (#913)
  • AIVA implicit half omits terminal velocity: precipitating species lose their withheld sedimentation flux (#914)
  • scalar_substep! passes an off-by-one closure field index: every scalar gets its predecessor's diffusivity (#915)
  • Explicitly-discretized closure + AIVA applies vertical diffusion twice (#930)
  • Vertical momentum ρw is not mass-flux weighted in the vertically-implicit solve (#944)
  • AtmosphereModel silently drops unknown boundary_conditions keys (bit hard across the e→s rename) (#956)

Notes

If you use Breeze.jl in your research, teaching, or for fun, please cite it using the metadata below.

Files

NumericalEarth/Breeze.jl-v0.11.0.zip

Files (1.3 MB)

Name Size Download all
md5:34b41e2c9189b9560b3c38f2d1dbb959
1.3 MB Preview Download

Additional details

Related works