Interpolation of GEDI Biomass Estimates with Calibrated Uncertainty Quantification

TL;DR

This paper introduces Attentive Neural Processes for calibrated biomass estimation from GEDI data, effectively capturing uncertainty and adapting to landscape heterogeneity across diverse ecosystems.

Contribution

It presents a novel probabilistic meta-learning model that explicitly models spatial covariance, improving uncertainty calibration over traditional ensemble methods.

Findings

ANPs achieve competitive accuracy across five biomes.

The method provides well-calibrated uncertainty estimates.

Few-shot adaptation effectively transfers models across regions.

Abstract

Reliable wall-to-wall biomass density estimation from NASA's GEDI mission requires interpolating sparse LIDAR observations across heterogeneous landscapes. While machine learning approaches like Random Forest and XGBoost are widely used, they treat spatial predictions of GEDI observations from multispectral or SAR remote sensing data as independent without adapting to the varying difficulty of heterogeneous landscapes. We demonstrate these approaches generally fail to produce calibrated prediction intervals. We show that this stems from conflating ensemble variance with aleatoric uncertainty and ignoring local spatial context. To resolve this, we introduce Attentive Neural Processes (ANPs), a probabilistic meta-learning architecture that explicitly conditions predictions on local observation sets and exploits geospatial foundation model embeddings. Unlike static ensembles, ANPs learn a flexible spatial covariance function, allowing estimates to be more uncertain in complex landscapes and less in homogeneous areas. We validate this approach across five distinct biomes ranging from tropical Amazonian forests to boreal, temperate, and alpine ecosystems, demonstrating that ANPs achieve competitive accuracy while maintaining near-ideal uncertainty calibration. We demonstrate the operational utility of the method through few-shot adaptation, where the model recovers most of the performance gap in cross-region transfer using minimal local data. This work provides a scalable, theoretically rigorous alternative to ensemble variance for continental scale earth observation.