ChronoEarth-492K: A Large Scale and Long Horizon Spatiotemporal Hyperspectral Earth Observation Dataset and Benchmark

TL;DR

ChronoEarth-492K is a comprehensive, long-term hyperspectral Earth observation dataset and benchmark designed to advance spatiotemporal modeling and self-supervised learning in remote sensing.

Contribution

It introduces the first large-scale, temporally calibrated hyperspectral dataset and a unified benchmark for long-horizon spatiotemporal analysis.

Findings

Established baseline results for state-of-the-art hyperspectral models.

Provided a standardized evaluation protocol for long-term hyperspectral analysis.

Created a platform enabling systematic research in long-horizon spatiotemporal hyperspectral learning.

Abstract

Hyperspectral imaging (HSI) provides dense spectral information for the Earth's surface, enabling material-level understanding of land cover and ecosystem dynamics. Despite recent progress in hyperspectral self-supervised learning (SSL), existing datasets remain temporally shallow, limiting the development of long-horizon spatiotemporal modeling. To address this gap, we introduce ChronoEarth-492K, the first large-scale, temporally calibrated hyperspectral SSL dataset built upon NASA's EO-1 Hyperion mission, the world's longest continuous hyperspectral archive up to date (2001-2017). ChronoEarth-492K comprises 492,354 radiometrically harmonized patches across 185,398 global locations over 17 years, with 28,786 sites containing multi-temporal sequences ($\geq 3$ observations) that enable both short- and long-horizon temporal analysis. Building on this foundation, we establish the ChronoEarth-Benchmark, a unified evaluation suite spanning static, short-horizon, and long-horizon temporal tasks, constructed from six open-source geospatial products covering land cover, crop type, forest dynamics, and soil properties. We further introduce a standardized evaluation protocol and report extensive baseline results across state-of-the-art hyperspectral foundation models. Together, ChronoEarth and benchmark provide the first large-scale, temporally grounded platform for systematic spatiotemporal hyperspectral representation learning.