Accurate Above-Ground Biomass (AGB) mapping at both large scale and high spatio-temporal resolution is essential for applications ranging from climate modeling to biodiversity assessment, and sustainable supply chain monitoring.
At present, fine-grained AGB mapping relies on costly airborne laser scanning acquisition campaigns usually limited to regional scales. Initiatives such as the ESA CCI map attempt to generate global biomass products from diverse spaceborne sensors but at a coarser resolution.To enable global, high-resolution (HR) mapping, several works propose to regress AGB from HR satellite observations such as ESA Sentinel-1/2 images.
We propose a novel way to address HR AGB estimation, by leveraging both HR satellite observations and existing low-resolution (LR) biomass products. We cast this problem as Guided Super-Resolution (GSR), aiming at upsampling LR biomass maps (sources) from 100 to 10 m resolution, using auxiliary HR co-registered satellite images (guides).
We compare super-resolving AGB maps with and without guidance, against direct regression from satellite images, on the public BioMassters dataset. We observe that Multi-Scale Guidance (MSG) outperforms direct regression both for regression (-780 t/ha RMSE) and perception (+2.0 dB PSNR) metrics, and better captures high-biomass values, without significant computational overhead. Interestingly, unlike the RGB+Depth setting they were originally designed for, our best-performing AGB GSR approaches are those that most preserve the guide image texture. Our results make a strong case for adopting the GSR framework for accurate HR biomass mapping at scale.
Our code and model weights are made publicly available at github.com/kaankaramanofficial/GSR4B.
We compare several biomass estimation approaches to produce high-resolution biomass maps. Guided Super-Resolution (GSR) upsamples an input low resolution biomass map using higher resolution satellite imagery as guidance. Super-Resolution (SR) takes biomass map and upscale it to a higher resolution. Biomass Estimation (BE) directly regresses biomass from a satellite imagery as input.
Our results indicate that learning-based GSR approaches taken from the depth estimation literature perform best. Interestingly, we observe that texture copying, which is hurtful for depth estimation from natural images, is actually beneficial at the resolution at which our present biomass upsampling setting is formulated.
Learning-based GSR methods better reconstruct capture high frequency patterns with the help of the satellite guide image.
A common issue for canopy height and biomass estimation works in the literature, is that methods tend to overestimate low-biomass values and underestimate high values. However, we observe that GSR methods mitigate this behavior throughout the distribution of errors.
@article{karaman2025gsr4b,
title={GSR4B: Biomass Map Super-Resolution with Sentinel-1/2 Guidance},
author={Karaman, Kaan and Jiang, Yuchang and Robert, Damien and Sainte Fare Garnot, Vivien and Santos, Maria João and Wegner, Jan Dirk},
journal={ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences},
year={2025},
}