Skip to content

Below-ground Fungal Biodiversity Can be Monitored Using Self-Supervised Learning Satellite Features

Apr 2026 · arXiv.org · Vol abs/2604.09818 · 0 citations
Computer Science

TL;DR

It is shown that self-supervised learning applied to satellite imagery can predict below-ground ectomycorrhizal fungal richness across diverse environments, and SSL-derived features are the most informative tested predictor group, and outperform each of the established climate, soil, and land cover baselines.

Abstract

Mycorrhizal fungi are vital to terrestrial ecosystem functioning. Yet monitoring their biodiversity at landscape scales is often unfeasible due to time and cost constraints. Current predictions suggest that 90% of mycorrhizal diversity hotspots remain unprotected, opening questions of how to broadly and effectively map underground fungal communities. We show that self-supervised learning (SSL) applied to satellite imagery can predict below-ground ectomycorrhizal fungal richness across diverse environments. Our models explain over half the variance in species richness across ~12,000 field samples spanning Europe and Asia. SSL-derived features are the most informative tested predictor group, and outperform each of the established climate, soil, and land cover baselines. We achieve a 10,000-fold increase in spatial resolution over existing techniques, moving from 1km landscape averages to 10m habitat-scale observations. As satellite observations are dynamic rather than static, this enables temporal monitoring of below-ground biodiversity at landscape scales for the first time. We apply this approach across two UK National Park woodlands, where ancient forests are predicted to have high but declining ectomycorrhizal diversity, marking key areas for further field verification. These results establish SSL satellite features as a scalable tool for extending sparse field observations to continuous, high-resolution biodiversity maps for monitoring the invisible half of terrestrial ecosystems.

View source

Similar papers

Conference Open access Sep 2026

Scalable Mapping of Tree Traits to Study the Dynamics of Protected Forest Areas in India

Monitoring forest functional diversity is essential to understand ecosystem resilience in the face of rapid environmental change. While existing remote sensing approaches primarily track structural attributes such as canopy density and tree height, functional traits like leaf phenology (evergreen vs. deciduous) and lea...

Dhruvi Goyal, Aaditeshwar Seth · 0 citations
Open access Sep 2026

Satellite-Based Aboveground Biomass Estimation in Mountain Pastures of Armenia

Accurate estimation of aboveground biomass (AGB) is essential for sustainable pasture management, but remains challenging in heterogeneous mountain environments. This study evaluated a remote-sensing framework for estimating AGB across five landscape zones of the Aragats Volcanic Massif, Armenia, using PlanetScope imag...

G. Ayvazyan, A. Medvedev, V. Muradyan et al. · 0 citations
Review Open access Oct 2026

Species distribution models of ESA-listed corals highlight niche differences among taxa that may guide recovery planning

Understanding the biogeography of living marine resources is essential for informed management, particularly in data-limited environments. Here, we use two corals listed under the U.S. Endangered Species Act (ESA) around Tutuila, American Samoa, to evaluate a transferable workflow that combines presence-only occurr...

Kisei R. Tanaka, Thomas A. Oliver, C. Couch et al. · 0 citations
Open access Sep 2026

Predicting forest biodiversity from spaceborne GEDI LiDAR and machine learning: structural drivers and regional transferability across Pacific Northwest forests

Abstract Remote sensing offers important opportunities for large-scale biodiversity monitoring, yet the capacity of spaceborne laser scanning to predict forest biodiversity remains insufficiently understood. This study evaluates relationships between forest structural complexity and tree species diversity using data fr...

Satyam Shah · 0 citations
Open access Sep 2026

Occurrence-based models reveal greater spatial details in tropical species richness

Species richness maps are essential for biodiversity assessment and conservation planning, but their accuracy is limited by uneven sampling effort and large data gaps, especially in tropical regions. Here, we combined the Uniform Sampling from Sampling Effort (USSE) framework with deep neural networks to generate occur...

Ubirajara Oliveira, B. Soares-Filho · 0 citations
Open access Oct 2026

Two decades-long NDVI anomalies to understand Mediterranean forests ecosystem shifts

Mediterranean forest ecosystems are increasingly exposed to climate change, land-use transformations and altered disturbance regimes. Understanding vegetation shifts across these complex landscapes requires scalable, high-resolution tools. In this study, we analyzed a 24-year NDVI time series (2000–2023) from Landsat i...

Onofrio Cappelluti, Vincenzo Giannico, Giovanni Sanesi et al. · 0 citations

Related blog posts

MIT News · Artificial Intelligence Nov 17, 2026

Learning the language of machine learning

By studying the computational foundations of language learning, Associate Professor Jacob Andreas builds intelligent systems that can communicate effectively with humans.

Microsoft Research Blog Nov 16, 2026

TokenEconomizer: Learning from Experience to Make AI Coding Agents More Efficient

AI coding agents are increasingly help developers fix bugs and build features largely on their own, but every interaction with their models consumes tokens. In each session, an agent sends instructions, tool descriptions, and session history to its model. Most efforts to reduce this cost focus on individual tasks. Yet developers return to the same […] The post TokenEconomizer: Learning from Experience to Make AI Coding Agents More Efficient appeared first on Microsoft Research.

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.