ReflectGAN: Modeling Vegetation Effects for Soil Carbon Estimation from Satellite Imagery

TL;DR

This paper introduces ReflectGAN, a novel GAN-based framework that reconstructs bare soil reflectance from vegetated satellite images, significantly improving soil organic carbon estimation accuracy in vegetated regions.

Contribution

ReflectGAN is the first paired GAN model designed to transform vegetated soil reflectance into bare soil reflectance, enhancing SOC estimation from satellite imagery.

Findings

ReflectGAN outperforms existing vegetation correction methods.

Best model achieved R^2 of 0.54, RMSE of 3.95, RPD of 2.07.

Model performance improved by over 35% in R^2 compared to previous methods.

Abstract

Soil organic carbon (SOC) is a critical indicator of soil health, but its accurate estimation from satellite imagery is hindered in vegetated regions due to spectral contamination from plant cover, which obscures soil reflectance and reduces model reliability. This study proposes the Reflectance Transformation Generative Adversarial Network (ReflectGAN), a novel paired GAN-based framework designed to reconstruct accurate bare soil reflectance from vegetated soil satellite observations. By learning the spectral transformation between vegetated and bare soil reflectance, ReflectGAN facilitates more precise SOC estimation under mixed land cover conditions. Using the LUCAS 2018 dataset and corresponding Landsat 8 imagery, we trained multiple learning-based models on both original and ReflectGAN-reconstructed reflectance inputs. Models trained on ReflectGAN outputs consistently outperformed those using existing vegetation correction methods. For example, the best-performing model (RF) achieved an $R^2$ of 0.54, RMSE of 3.95, and RPD of 2.07 when applied to the ReflectGAN-generated signals, representing a 35\% increase in $R^2$, a 43\% reduction in RMSE, and a 43\% improvement in RPD compared to the best existing method (PMM-SU). The performance of the models with ReflectGAN is also better compared to their counterparts when applied to another dataset, i.e., Sentinel-2 imagery. These findings demonstrate the potential of ReflectGAN to improve SOC estimation accuracy in vegetated landscapes, supporting more reliable soil monitoring.