A 1 km resolution dataset of Northern Hemisphere permafrost active layer thickness (2000–2024)

This dataset provides continuous, 1 km resolution annual maps of Active Layer Thickness (ALT) across the Northern Hemisphere permafrost regions from 2000 to 2024. The dataset was generated using an ensemble spatiotemporal machine learning framework trained on 2,435 historical in situ observations. To ensure high reliability and spatiotemporal continuity, a spatiotemporal feature-matching strategy was employed to fuse multi-source dynamic climate drivers and static land surface properties. Furthermore, the dataset includes spatially explicit, pixel-wise uncertainty maps derived from a repeated random resampling framework, providing a transparent quality benchmark for downstream users.