Radiative Transfer Models

Every function on this site belongs to one of a small number of named physical models. This page is what each one is – scale, what it simulates, and how it differs from its siblings – before 07. Building RTM Workflows shows how they chain together and Examples runs them.

At a glance

Model	Scale	Simulates	Typical use	Python entry point
PROSPECT-D / PROSPECT-PRO	Leaf	Reflectance/transmittance (R/T)	Pigments, water, dry matter (PRO: + protein, CBC)	`prospect_d()` / `prospect_pro()`
Fluspect-B / Fluspect-Cx	Leaf	R/T + fluorescence	Solar-induced fluorescence (SIF)	`fluspect_b()` / `fluspect_cx()`
LIBERTY	Leaf (needle)	R/T	Conifer canopies, not broadleaf	`liberty()`
fourSAIL	Canopy	TOC reflectance (BRDF)	Crops, grassland, single-layer canopies	`foursail()`
fourSAIL2	Canopy	TOC reflectance (BRDF)	Mixed green + senescent/brown canopies	`toolsrtm.canopy.foursail2`
INFORM	Canopy (forest stand)	TOC reflectance (BRDF)	Forests – explicit crown/gap geometry	`inform()`
MARMIT	Soil	Reflectance	Soil moisture, from a real dry reference spectrum	`get_marmit_rsoil()`
BSM	Soil	Reflectance	Soil moisture, from empirical brightness/shape parameters	`get_bsm()`
SPART	Soil -> canopy -> atmosphere	TOC + TOA reflectance	A vegetated scene as a real sensor would measure it	`spart_toa()`
SCOPE	Ecosystem (leaf -> canopy -> atmosphere-facing)	Reflectance + SIF + energy balance	Physiology: temperature, photosynthesis, carbon/water flux	`get_scope()`

This table is the “which model do I want” lookup; the sections below are the “what does this model actually do” reference for each row.

Leaf optical models

All leaf models take pigment/water/dry-matter traits and return a reflectance/transmittance spectrum – see the 02. Parameters & Traits for exactly what each trait (N, Cab, EWT, …) means.

Model	What it represents	How it differs from the others
`prospect_d()` (PROSPECT-D)	The reference leaf model: a stack of `N` absorbing/scattering plates, pigments (`Cab`/`Car`/`Anth`) + water (`EWT`) + one lumped dry-matter term (`LMA`).	The default – broadleaf, no fluorescence, dry matter as a single term.
`prospect_pro()` (PROSPECT-PRO)	Same physics as PROSPECT-D, but splits `LMA` into `Prot` (protein) + `CBC` (cellulose+lignin).	Only differs from PROSPECT-D in how dry matter is parameterized – useful when protein/nitrogen matters on its own.
`fluspect_b()` (Fluspect-B)	PROSPECT-D’s optics plus chlorophyll-fluorescence excitation- emission matrices (`MbI`/`MbII`) – needed wherever SIF is simulated.	Adds fluorescence on top of PROSPECT-D; reflectance/transmittance themselves are near-identical to PROSPECT-D.
`fluspect_cx()` (Fluspect-B-Cx)	Fluspect-B plus a `Cx` (xanthophyll de-epoxidation / NPQ) term, letting fluorescence yield respond to photoprotection state, not just pigment content.	The only leaf model with a photoprotection (`Cx`) term; what SCOPE’s own leaf-optics step uses internally.
`liberty()` (LIBERTY)	A structurally different model built for conifer needles (Dawson et al. 1998) – explicit cell diameter/intercellular air space instead of PROSPECT’s Fresnel-refraction layer.	Not a PROSPECT variant at all; needle-specific anatomy, a genuinely different internal structure, not just different defaults.

Canopy models

All three take a leaf model’s reflectance/transmittance plus a soil background and canopy structure, and return top-of-canopy (TOC) BRDF.

Model	What it represents	How it differs from the others
`foursail()` (fourSAIL)	The classic PROSAIL turbid-medium canopy: a single, statistically homogeneous “cloud” of leaves at a given LAI and angle distribution – no explicit 3D structure.	The reference/default; single-layer, single leaf-biochemistry profile.
`toolsrtm.canopy.foursail2` (fourSAIL2)	Two-layer canopy (a green layer + a brown/senescent layer, `fraction_brown`-weighted) – e.g. a canopy with visible dead/dry material mixed in with live foliage.	Same turbid-medium idea as fourSAIL, but two vertically-stacked layers instead of one.
`inform()` (INFORM)	Forest-stand extension (Atzberger): explicit tree crowns (stem density, crown diameter, height) over an understorey + background, rather than one homogeneous canopy.	The only one of the three with real gap/shadow geometry – produces visibly lower reflectance than fourSAIL at the same LAI, matching a discontinuous forest stand’s real physics.

Soil

Model	What it represents	How it differs from the others
`get_marmit_rsoil()` (MARMIT)	Starts from a real dry reference soil spectrum and adds a physically modelled liquid-water film, so the same soil can be simulated at any moisture level.	The only soil model driven by an actual measured reference spectrum rather than empirical shape parameters.
`get_bsm()` (BSM, Brightness-Shape-Moisture)	Builds a soil spectrum from three empirical parameters (`BSMBrightness`, `BSMlat`, `BSMlon`) plus a wetting term – no reference spectrum needed.	Purely parametric, not spectrum-driven – SPART’s and SCOPE’s own soil model.

Soil-Plant-Atmosphere

Model	What it represents	How it differs from the others
`sentinel2a_msi()` + `get_smac` (SMAC)	Atmospheric radiative transfer (gas absorption + aerosol scattering) that converts top-of-canopy reflectance into what a real satellite sensor would measure above the atmosphere.	Not a soil or canopy model – the atmosphere step, only relevant when going all the way to top-of-atmosphere (TOA).
`spart_toa()` (SPART)	Not a new physical model, but the full chain: BSM soil -> fourSAIL canopy -> SMAC atmosphere -> TOA reflectance, already resampled to a real sensor’s bands in one call.	The end-to-end soil-plant-atmosphere pipeline; no separate “simulate native, then convolve” step, unlike plain fourSAIL.

Energy balance / fluorescence: SCOPE

SCOPE (scopeinpython.scope.get_scope()) is a different kind of model from everything above, not just a bigger one: instead of assuming leaf/soil temperature and computing reflectance alone, it iteratively solves leaf and soil temperature so that absorbed radiation balances sensible + latent heat + photosynthesis (the energy balance), then derives fluorescence and carbon flux from that solved state. Five distinct components, chained together:

Component	What it does	Depends on
`get_fluspect_cx_scope()` (+ `fluspect_mscope()` for multi-layer)	Leaf optics: reflectance/transmittance + fluorescence excitation- emission matrices, per canopy layer. Same PROSPECT/Fluspect physics as above – just the SCOPE-specific wrapper.	Leaf biochemistry traits (02. Parameters & Traits)
`run_rtmo()`	Optical top-of-canopy BRDF – physically the same turbid-medium idea as fourSAIL, re-implemented to plug into the layers below. Stops at reflectance; assumes no temperature yet.	Fluspect-Cx output + canopy structure + BSM soil
`ebal()`	The energy balance: iterates leaf/soil temperature until the flux budget closes, calling the biochemistry model at every candidate temperature.	RTMo output + aerodynamic resistances + meteorology
`get_biochemical()`	Leaf-level photosynthesis (Farquhar/Collatz) and fluorescence yield, given a leaf micro-environment.	Photosynthesis + NPQ traits (02. Parameters & Traits)
`rtmf()` / `rtmz()` (optional)	Canopy-level fluorescence radiance/flux, and a small zeaxanthin (photoprotection) correction to the TOC spectrum.	ebal output + Fluspect-Cx’s fluorescence matrices

What’s next

07. Building RTM Workflows – how these models chain together into the four standard simulation pipelines.
02. Parameters & Traits – what every input trait to these models means, its unit, and its realistic range.
Examples – every model above, run with real, verified code.