vSmartMOM

Appearance

Quick Start: vSmartMOM Radiative Transfer ​

vSmartMOM is a polarized radiative transfer solver using the adding-doubling method. This 5-minute tutorial runs a small forward RT simulation on CPU.

1) Load the package ​

julia
using vSmartMOM

2) Load parameters from YAML ​

Parameters define the atmosphere, surface, geometry, and spectral bands. See the IO Schema documentation for the full YAML format.

julia
yaml_path = joinpath(pkgdir(vSmartMOM), "config", "quickstart.yaml")
params = read_parameters(yaml_path)
params.architecture = vSmartMOM.Architectures.CPU()

3) Build the model and run forward RT ​

julia
model = model_from_parameters(params)
R, T = rt_run(model)

4) Interpret results ​

  • R = upwelling radiance at top-of-atmosphere (TOA) [nVZA × nStokes × nSpec]

  • T = downwelling radiance at bottom-of-atmosphere (BOA)

Dimensions: view zenith angles × Stokes components (I,Q,U,V) × spectral points.

julia
println("R shape: ", size(R))
println("T shape: ", size(T))
println("R(VZA 60°, I, ν₁) = ", R[1, 1, 1])

The documentation build renders the same two arrays as an interactive Plotly view. Hovering the bars shows the numeric Stokes-I values returned above.

5) Next steps ​

The shipped quickstart scene intentionally avoids absorption and aerosol Mie setup. For richer spectra and plots, use the absorption, scattering, surface, and Jacobian tutorials. For analytic derivatives, start with the Compute Jacobians page.

6) GPU support ​

For NVIDIA GPUs with CUDA.jl, set params.architecture = vSmartMOM.Architectures.GPU() before building the model. The same model_from_parameters and rt_run calls apply. See the Run on GPU task page for more.

This page was generated using Literate.jl.