# Neural Processes Maintain Calibrated Biomass Estimates Across Spatiotemporal Gaps and Disturbance

Source: [Neural Processes Maintain Calibrated Biomass Estimates Across Spatiotemporal Gaps and Disturbance](<https://www.sota2.com/research/paper/69d48f747d0172140ec74230>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-04-15T05:00:04.591Z

Authors: Robin Young, Srinivasan Keshav

Published: 2026-04-04T21:41:38.000Z

## Abstract

Monitoring deforestation-driven carbon emissions requires both spatially explicit and temporally continuous estimates of aboveground biomass density (AGBD) with calibrated uncertainty. NASA's Global Ecosystem Dynamics Investigation (GEDI) provides reliable LIDAR-derived AGBD, but its orbital sampling causes irregular spatiotemporal coverage, and occasional operational interruptions, including a 13-month hibernation from March 2023 to April 2024, leave extended gaps in the observational record. Prior work has used machine learning approaches to fill GEDI's spatial gaps using satellite-derived features, but temporal interpolation of biomass through unobserved periods, particularly across active disturbance events, remains largely unaddressed. Moreover, standard ensemble methods for biomass mapping have been shown to produce systematically miscalibrated prediction intervals. To address these gaps, we extend the Attentive Neural Process (ANP) framework, previously applied to spatial biomass interpolation, to jointly sparse spatiotemporal settings using geospatial foundation model embeddings. We treat space and time symmetrically, empirically validating a form of space-for-time substitution in which observations from nearby locations at other times inform predictions at held-out periods. Our results demonstrate that the ANP produces well-calibrated uncertainty estimates across disturbance regimes, supporting its use in Measurement, Reporting, and Verification (MRV) applications that require reliable uncertainty quantification for forest carbon accounting.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Spatiotemporal Biomass Density Interpolation on Guaviare, Colombia 2021 (test)](<https://www.sota2.com/research/sota/spatiotemporal-biomass-density-interpolation-on-guaviare-colombia-2021-test.md>): Spatiotemporal Biomass Density Interpolation; Guaviare, Colombia 2021 (test); Log R2: 0.75.

- [Spatiotemporal Biomass Density Interpolation on Queensland, Australia 2021 (test)](<https://www.sota2.com/research/sota/spatiotemporal-biomass-density-interpolation-on-queensland-australia-2021-test.md>): Spatiotemporal Biomass Density Interpolation; Queensland, Australia 2021 (test); Log R2: 0.5.

- [Spatiotemporal Biomass Density Interpolation on Ucayali, Peru 2021 (test)](<https://www.sota2.com/research/sota/spatiotemporal-biomass-density-interpolation-on-ucayali-peru-2021-test.md>): Spatiotemporal Biomass Density Interpolation; Ucayali, Peru 2021 (test); Log R2: 0.42.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2604.03874>)

[PDF](<https://arxiv.org/pdf/2604.03874.pdf>)
