Above-ground Biomass Estimation with Geospatial Foundation Models

2026-08-05

AI Summary

Accurately estimating global aboveground biomass (AGB) remains a challenging remote sensing regression task, and the potential of existing geospatial foundation models (GFMs) for such quantitative applications has not been fully explored. This study presents the first systematic evaluation of GFMs for AGB estimation, distinguishing between two usage paradigms—frozen encoders and precomputed embeddings—and benchmarking 11 GFMs within the PANGAEA framework. Results demonstrate that an MLP trained on AlphaEarth Foundations (AEF) embeddings outperforms current supervised state-of-the-art (SOTA) models. Moreover, integrating AEF embeddings further enhances SOTA performance, significantly improving spatiotemporal generalization and consistency with the ESA CCI biomass product. These findings challenge the prevailing notion that GFMs are suited only for classification or segmentation tasks, highlighting their untapped potential in quantitative Earth observation.

Abstract

Accurate estimation of Above-Ground Biomass (AGB) from satellite imagery is essential for the large-scale monitoring of carbon stocks, yet it remains a challenging regression task at global scale. Geospatial Foundation Models (GFMs) have recently emerged as a promising machine learning paradigm to derive general-purpose representations from Earth observation data, but their utility for quantitative regression tasks like biomass estimation remains largely unexplored, as most benchmarks emphasize classification and segmentation. Here, we present a comprehensive benchmark of GFMs for global-scale AGB estimation using the AGBD dataset, a machine learning-ready benchmark spanning diverse biomes and geographies. We distinguish two ways in which GFMs reach practitioners: (i) models distributed as weights to be run by the user, which we evaluate as frozen encoders within the PANGAEA benchmarking framework; and (ii) models distributed as ready-to-use, pre-computed embedding products, for which we evaluate AlphaEarth Foundations (AEF) and TESSERA. We compare 11 GFMs available on PANGAEA and both embedding products against a fully supervised state-of-the-art (SOTA) model, assess their geographical and temporal generalization abilities, as well as agreement with the ESA CCI biomass product on independent reference data. Our results show that GFMs run as frozen encoders substantially underperform with respect to the supervised SOTA model, whereas pre-computed embedding products prove highly effective. An MLP trained on AEF embeddings outperforms the supervised SOTA model trained on AGBD features, and the same SOTA model trained on AEF embeddings (optionally augmented with selected raw features) achieves the best overall result, while also generalizing better across space and time.

Problem

• Above-Ground Biomass
• Geospatial Foundation Models
• satellite imagery
• regression task
• carbon stocks

Innovation

• Geospatial Foundation Models
• Above-Ground Biomass estimation
• pre-computed embeddings
• spatiotemporal generalization
• foundation model benchmarking