University of Tasmania

File(s) under permanent embargo

Downscaling of solar-induced chlorophyll fluorescence from canopy level to photosystem level using a random forest model

journal contribution
posted on 2023-05-19, 20:02 authored by Liu, X, Guanter, L, Liu, L, Damm, A, Zbynek MalenovskyZbynek Malenovsky, Rascher, U, Peng, D, Du, S, Gastellu-Etchegorry, J-P
Solar-induced chlorophyll fluorescence (SIF), an electromagnetic signal that can potentially indicate vegetation photosynthetic activity, can be retrieved from ground-based, airborne and satellite measurements. However, due to the scattering and re-absorption effects inside the leaves and canopy, SIF measured at the canopy level is only a small part of the total SIF emission at the photosystem level. Therefore, a downscaling mechanism of SIF from the canopy level to the photosystem level is important for better understanding the relationship between SIF and the vegetation gross primary production (GPP). In this study, firstly, we analyzed the canopy scattering effects using a simple parameterization model based on the spectral invariant theory. The probability for SIF photons to escape from the canopy was found to be related to the anisotropic spectral reflectance, canopy interception of the upward solar radiation, and leaf absorption. An empirical approach based on a Random Forest (RF) regression algorithm was applied to downscale SIF constrained by the red, red-edge and far-red anisotropic reflectance. The RF was trained using simulations conducted with the Soil Canopy Observation, Photochemistry and Energy fluxes (SCOPE) model. The performance of the SIF downscaling method was evaluated with SCOPE and Discrete Anisotropic Radiative Transfer (DART) model simulations, ground measurements and airborne data. Results show that estimated SIF at the photosystem level matches well with simulated reference data, and the relationship between SIF and photosynthetically active radiation absorbed by chlorophyll is improved by SIF downscaling. This finding in combination with other evaluation criteria suggests the downscaling of canopy SIF as an efficient strategy to normalize species dependent effects of canopy structure and varying solar-view geometries. Based on our results for the SIF-APAR relationship, we expect that such normalization approaches can be helpful to improve estimates of photosynthesis using remote sensing measurements of SIF.


Australian Research Council


Publication title

Remote Sensing of Environment



Article number









School of Geography, Planning and Spatial Sciences


Elsevier Science Inc

Place of publication

360 Park Ave South, New York, USA, Ny, 10010-1710

Rights statement

Copyright 2018 Elsevier Inc.

Repository Status

  • Restricted

Socio-economic Objectives

Assessment and management of Antarctic and Southern Ocean ecosystems; Expanding knowledge in the environmental sciences

Usage metrics

    University Of Tasmania


    Ref. manager