[Submitted on 19 Mar 2025 (v1), last revised 29 Oct 2026 (this version, v3)]
Abstract:Understanding natural Earth surface processes requires integrating topography, imagery, hydrology, and human infrastructure while remaining robust to geographic variation, yet few benchmarks support both challenges. We introduce EarthScape, a multimodal benchmark for geologic mapping and surface-aware learning under geographic domain shift. EarthScape combines aerial imagery, LiDAR-derived elevation, multi-scale terrain derivatives, and contextual vector data into 37 co-registered input channels per sample, together with pixel-level geologic masks. Its geographically separated regions share a common label space, enabling controlled evaluation of in-domain and cross-region generalization under realistic class imbalance and spatial complexity. Benchmark experiments show that engineered terrain representations transfer more robustly than raw elevation, while spatial support and multimodal terrain inputs strongly influence performance. Fusion effectiveness depends on backbone and evaluation domain, with more complex cross-attention providing no consistent advantage over simpler strategies. These representation trends extend across geospatial foundation models and semantic segmentation, while class-level results show that no single terrain feature or scale is optimal across geologic classes. Together, EarthScape provides a benchmark for studying representation design, multimodal fusion, and geographic robustness in surface-aware geospatial learning. Our source code is available at this https URL and the dataset at this https URL.
Submission history
From: Matthew Massey [view email]
[v1] Wed, 19 Mar 2025 18:23:48 UTC (37,743 KB)
[v2] Thu, 5 Mar 2026 20:44:58 UTC (27,072 KB)
[v3] Thu, 29 Oct 2026 17:07:53 UTC (39,099 KB)
