Terramesh: A Planetary Mosaic of Multimodal Earth Observation Data
Authors:
- 11 June 2025
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 2385-2393
Abstract
Large-scale foundation models in Earth Observation can learn versatile, label-efficient representations by leveraging massive amounts of unlabeled data. However, existing public datasets are often limited in scale, geographic coverage, or sensor variety. We introduce TerraMesh, a new globally diverse, multimodal dataset combining optical, synthetic aperture radar, elevation, and land-cover modalities in an Analysis-Ready Data format. TerraMesh includes over 9 million samples with eight spatiotemporal aligned modalities, enabling large-scale pre-training and fostering robust cross-modal correlation learning. The dataset spans nearly all terrestrial ecosystems and is stored with Zarr to facilitate efficient, HPC-friendly loading at scale. We provide detailed data processing steps, comprehensive statistics, and empirical evidence demonstrating improved model performance when pre-trained on TerraMesh. The dataset will be made publicly available with a permissive license.Keywords
All Related Versions
Funding Information
- European Space Agency (4000143501/23/I-DT)
This publication has 15 references indexed in Scilit:
- MMEarth: Exploring Multi-modal Pretext Tasks for Geospatial Representation Learning2024
- Bridging Remote Sensors with Multisensor Geospatial Foundation Models2024
- Detecting Marine pollutants and Sea Surface features with Deep learning in Sentinel-2 imageryISPRS Journal of Photogrammetry and Remote Sensing, 2024
- SatlasPretrain: A Large-Scale Dataset for Remote Sensing Image Understanding2023
- A high-resolution canopy height model of the EarthNature Ecology & Evolution, 2023
- Dynamic World, Near real-time global 10 m land use land cover mappingScientific Data, 2022
- Panoptic Segmentation of Satellite Image Time Series with Convolutional Temporal Attention Networks2021
- Sen1Floods11: a georeferenced dataset to train and test deep learning flood algorithms for Sentinel-12020
- Bigearthnet: A Large-Scale Benchmark Archive for Remote Sensing Image Understanding2019
- An Ecoregion-Based Approach to Protecting Half the Terrestrial RealmBioScience, 2017