Ecological mapping with geospatial foundation models

TL;DR

This paper systematically evaluates Earth observation foundation models for ecological applications, demonstrating their superior performance over traditional models across multiple ecological tasks and highlighting key factors influencing their effectiveness.

Contribution

It is among the first to benchmark and analyze the performance, limitations, and practical considerations of foundation models in ecological mapping tasks.

Findings

Foundation models outperform ResNet baseline in ecological tasks.

TerraMind shows advantages with multimodal inputs.

Performance depends on dataset alignment and input resolution.

Abstract

The value of Earth observation foundation models for high-impact ecological applications remains insufficiently characterized. This study is one of the first to systematically evaluate the performance, limitations and practical considerations across three common ecological use cases: forest functional trait estimation, land use and land cover mapping and peatland detection. We fine-tune two pretrained models (Prithvi-EO-2.0 and TerraMind) and benchmark them against a ResNet-101 baseline using datasets collected from open sources. Across all tasks, Prithvi-EO-2.0 and TerraMind consistently outperform the ResNet baseline, demonstrating improved generalization and transfer across ecological domains. TerraMind marginally exceeds Prithvi-EO-2.0 in unimodal settings and shows substantial gains when additional modalities are incorporated. However, performance is sensitive to divergence between downstream inputs and pretraining modalities, underscoring the need for careful dataset alignment. Results also indicate that higher-resolution inputs and more accurate pixel-level labels remain critical for capturing fine-scale ecological dynamics.