Fine-tuning of Geospatial Foundation Models for Aboveground Biomass Estimation

TL;DR

This study demonstrates that fine-tuning a geospatial foundation model with a Swin-B transformer for aboveground biomass estimation achieves comparable accuracy to a U-Net while using significantly fewer parameters, highlighting efficiency and transfer-learning potential.

Contribution

It introduces a novel fine-tuning approach for geospatial foundation models in biomass estimation, showing efficiency gains and transfer-learning capabilities across eco-regions.

Findings

Fine-tuned model matches U-Net performance with 13x fewer parameters.

Frozen encoder during fine-tuning reduces computational resources.

Transfer learning enables adaptation across different eco-regions.

Abstract

Global vegetation structure mapping is critical for understanding the global carbon cycle and maximizing the efficacy of nature-based carbon sequestration initiatives. Moreover, vegetation structure mapping can help reduce the impacts of climate change by, for example, guiding actions to improve water security, increase biodiversity and reduce flood risk. Global satellite measurements provide an important set of observations for monitoring and managing deforestation and degradation of existing forests, natural forest regeneration, reforestation, biodiversity restoration, and the implementation of sustainable agricultural practices. In this paper, we explore the effectiveness of fine-tuning of a geospatial foundation model to estimate above-ground biomass (AGB) using space-borne data collected across different eco-regions in Brazil. The fine-tuned model architecture consisted of a Swin-B transformer as the encoder (i.e., backbone) and a single convolutional layer for the decoder head. All results were compared to a U-Net which was trained as the baseline model Experimental results of this sparse-label prediction task demonstrate that the fine-tuned geospatial foundation model with a frozen encoder has comparable performance to a U-Net trained from scratch. This is despite the fine-tuned model having 13 times less parameters requiring optimization, which saves both time and compute resources. Further, we explore the transfer-learning capabilities of the geospatial foundation models by fine-tuning on satellite imagery with sparse labels from different eco-regions in Brazil.