Advection

BoundaryCondition
Abstract type for boundary conditions in MPDATA advection scheme.

mpdatasettings Struct for 2D MPDATA advection scheme Settings - ncorr::Int: Number of corrective steps in MPDATA - grid::Tuple{Int, Int}: Grid dimensions (Nx, Nz) - horizontalboundarycondition::Union{Periodic, NoFlux, Extrapolated}: Horizontal boundary condition type - verticalboundarycondition::Union{Periodic, NoFlux, Extrapolated}: Vertical boundary condition type - nonoscillatory::Bool: Flag for non-oscillatory option - infinite_gauge::Bool: Flag for infinite gauge option

mpdatasettings1d{TVar<:ThermoVariable} Struct for 1D MPDATA advection scheme Settings - ncorr::Int: Number of corrective steps in MPDATA - grid::Int: Grid dimension (Nz) - verticalboundarycondition::Union{Periodic, NoFlux, Extrapolated}: Vertical boundary condition type - nonoscillatory::Bool: Flag for non-oscillatory option - infinitegauge::Bool: Flag for infinite gauge option - thermovariable::TVar: Type of thermodynamic variable (θ, qv) - topcellboundary_condition::Union{Periodic, NoFlux, Reservoir}: Top cell boundary condition type (commented out)

mpdatatmp mpdata1d Struct for MPData Allocations - ϕ::Matrix{Float64}: Field variable - GCxstep::Matrix{Float64}: Step size in x-direction - GCystep::Matrix{Float64}: Step size in y-direction - GCxtmp::Matrix{Float64}: Temporary storage for x-direction - GCytmp::Matrix{Float64}: Temporary storage for y-direction - minmax::NamedTuple: Named tuple containing local minimum and maximum values

Rescale the 2D antidiffusive velocities in tmp (FCT limiter) so the corrected donor-cell fluxes stay nonoscillatory.

Rescale the 1D (z) antidiffusive velocity in tmp (FCT limiter) so the corrected donor-cell flux stays nonoscillatory.

Compute the 2D nonoscillatory flux limiters (β_in, β_out) at cell (i,j) that bound ϕ by its local extrema.

Compute the 1D nonoscillatory flux limiters (β_in, β_out) at cell k that bound ϕ by its local extrema.

Compute and store the 2D antidiffusive velocities in tmp for the MPDATA corrective pass.

Compute and store the 1D (z) antidiffusive velocity in tmp for the MPDATA corrective pass.

Compute the 1D (z) MPDATA antidiffusive velocity correction into tmp.GCz_tmp.

Advance 1D ϕ in place by one donor-cell flux pass along z, optionally dividing by g_factor and zeroing the top-cell flux when topcellreservoir is set.

Advance 1D ϕ in place by one donor-cell flux pass along z under extrapolated (zero-gradient) boundaries.

Advance 2D ϕ in place by one donor-cell flux pass in both x and y.

Store the local max/min of 2D ϕ over each cell's neighbors into tmp.minmax.

Store the local max/min of 1D ϕ over each cell's z-neighbors into tmp.minmax.

Upwind donor-cell flux of ψL/ψR across a face with Courant number GC.

Accumulate one donor-cell flux pass along the x-direction into ϕ_tmp, using Courant numbers GCx.

Accumulate one donor-cell flux pass along the y-direction into ϕ_tmp, using Courant numbers GCy.

Map index i into the valid range [1,N] per boundary condition bc (wrap, reflect, or clamp).

Advection is off, so grid is left unchanged.

MPDATA-advect only water vapor mixing ratio in grid (advection on, thermo feedback off).

MPDATA-advect grid's thermodynamic and dynamic fields (θ/θ_l, qv/qt, e, u, v, ρ) and reconstruct dependent state (advection on, thermo feedback on).

Advance 2D ϕ_stage in place by one full MPDATA step (donor-cell pass plus n_corr-1 antidiffusive corrections) given Courant numbers GCx, GCy.

Advance 1D ϕ_stage in place by one full MPDATA step along z (donor-cell pass plus n_corr-1 antidiffusive corrections) given Courant numbers GCz.

Recompute θ/qv (or θ_l/qt, whichever wasn't the advected pair) in grid.states after MPDATA advection.

Compute the x-component MPDATA antidiffusive velocity correction into tmp.GCx_tmp.

Compute the y-component MPDATA antidiffusive velocity correction into tmp.GCy_tmp.

Create a stream function for a non-divergent flow field based on the given parameters.

Compute the non-divergent vector field in 2D based on the given stream function and grid parameters. Returns the u and v velocity components, as well as the CFL numbers in x and y directions.