dataset · Global

BioMassters

Hugging Face dataset ibm-nasa-geospatial/BioMassters

Claim this listing

Represent BioMassters? Send a claim for review so we can discuss listing updates. Claims do not automatically grant ownership or verification.

Data offered docs-verified: Sentinel-1 SAR imagery (12 monthly images per chip) · Sentinel-2 MSI imagery (10 selected spectral bands plus cloud probability band) · Above Ground Biomass (AGB) reference measurements (LiDAR calibrated with in-situ data) · Chip metadata: cloud_percentage, corrupt_values, red_mean, split
Coverage docs-verified: Finnish forests (Finland) · 2016-2021
Resolution llm-inferred: approx. 10 m spatial (derived from 256x256 px over 2,560 m patches); monthly temporal (1 image per month for Sentinel-1)
Formats docs-verified: CSV (chip metadata tracker) · Hugging Face dataset files (image format not stated)
Access docs-verified: Open; licence CC-BY-4.0

Use cases llm-inferred: Benchmarking Sentinel-1/Sentinel-2 time-series models for forest above-ground biomass against LiDAR reference patches · Training or validating biomass regression models on 12-month multi-sensor stacks before adapting them to UK woodland · Testing cloud and snow filtering workflows using the provided chip tracker metadata
Audience llm-inferred: remote sensing analysts · GIS analysts · ecologists · machine learning engineers · forestry and carbon consultants

Constraints docs-verified: Coverage is Finnish forests only (2016-2021), so direct UK applicability is limited · Sentinel-2 data does not have full temporal coverage across all months · Dataset viewer is unavailable for this dataset · Citation is still under review · Total size is about 201 GB

Verified 2026-10-15 via HF API: exists; 275 downloads/30d, 2 likes.

Metadata observed 14/10/2026 · research sweep: manual-2026-10-15 (Hugging Face datasets)