arXiv Computer Vision
Earth embeddings are decomposable records of the land surface
Read the original on arXiv Computer Vision →The Flow has not summarised this story yet — read it at arXiv Computer Vision.
Related stories
arXiv Computer Vision
Sep 23
Annual Earth-observation embeddings encode wildfire disturbance and support simplified burned area mapping
Annual Earth‑observation embeddings, specifically Tessera, can encode wildfire disturbance signals well enough to map burned areas without needing curated fire‑specific imagery or dense time‑series analysis. In tests, linear models using a single Tessera embedding matched or outperformed paired pre‑ and post‑fire HLS imagery and post‑fire imagery alone, achieving high F1 scores for burn‑scar delineation and regional mapping. The approach successfully mapped all same‑year fires in benchmark scenes, recovered 97% of California burned area without California training data, transferred to European fires with high accuracy, and even estimated ignition timing within a 13‑day error margin. "whyItMatters":"The study demonstrates that pre‑trained annual embeddings can simplify and scale burned‑area mapping, reducing reliance on dense time‑series data and enabling more efficient wildfire monitoring."
arXiv Machine Learning
Aug 14
Evaluating AlphaEarth Foundations Embeddings for Wildfire Susceptibility Mapping
arXiv:2608. 12663v1 Announce Type: cross Abstract: Wildfire susceptibility mapping typically relies on physical variables assembled from multiple remote-sensing, climate, and geospatial products.
arXiv Machine Learning
Aug 6
Above-ground Biomass Estimation with Geospatial Foundation Models
arXiv:2608. 04792v1 Announce Type: new Abstract: Accurate estimation of Above-Ground Biomass (AGB) from satellite imagery is essential for the large-scale monitoring of carbon stocks, yet it remains a challenging regression task at global scale.
arXiv Computer Vision
1d ago
National-scale mapping of tree species composition in Switzerland from high-resolution aerial imagery and multimodal data
arXiv:2611.08095v1 Announce Type: new Abstract: Tree-species composition shapes forest productivity and how forests respond to climate disturbances such as drought, windthrow and insect outbreaks. Ye...
Hugging Face Trending Papers
Jun 28
Miti360: A Comprehensive Dataset for Improved Reforestation Monitoring
Over the past decade, interest in applying machine learning (ML) to automate forest monitoring has grown significantly. However, existing training datasets are predominantly drawn from North America, Europe, Asia, and Australia, leaving a critical gap in African forestry data.
arXiv Machine Learning
Sep 29
Below-ground Fungal Biodiversity Can be Monitored Using Self-Supervised Learning Satellite Features
The study demonstrates that self‑supervised learning (SSL) applied to satellite imagery can predict below‑ground ectomycorrhizal fungal richness across diverse environments, explaining over half the variance in species richness from ~12,000 field samples in Europe and Asia. SSL‑derived features outperform traditional climate, soil, and land‑cover predictors and provide a 10,000‑fold increase in spatial resolution, moving from 1 km averages to 10 m habitat‑scale observations. The approach enables temporal monitoring of underground biodiversity and was applied to two UK National Park woodlands, highlighting ancient forests with high but declining ectomycorrhizal diversity for further field verification.