The study presents a method for estimating building heights in a large Brazilian city using freely available satellite data, including TerraSAR-X StripMap, PlanetScope, and Sentinel-1. A geographically weighted random forest model achieved an RMSE of 5.34 m and an R² of 0.756 against LiDAR reference data, with local feature importance varying by building type and context. The results highlight that no single sensor dominates across all scenarios, offering guidance for selecting satellite-derived products in different urban settings.
By Guilherme Iablonovski, Pierre-Louis Frison, Tatiana Silva da Silva
arXiv:2606. 11534v1 Announce Type: cross Abstract: Urban heat is amplified by impermeable surfaces and heterogeneous built environments, yet street-level variability remains difficult to quantify because multi-sensor observations are rarely available in consistent, analysis-ready form at the necessary spatiotemporal scales.
By Jonathan Starfeldt, Maria J. Molina, Alexander Kerr, Adam Yang, Thomas R. H. Holmes, Christopher R. Hain
arXiv:2607. 24180v1 Announce Type: cross Abstract: Monitoring post-disaster recovery is essential for understanding how urban systems rebuild and progressively return to functionality.
By Luigi Russo, Deodato Tapete, Silvia Liberata Ullo, Paolo Gamba
The paper presents a method for estimating building heights in a large Brazilian city using freely available satellite data, including TerraSAR‑X StripMap, PlanetScope, and Sentinel‑1. By integrating these sources in a geographically weighted random forest, the authors achieve an RMSE of 5.34 m and an R² of 0.756 against a LiDAR reference. The study also reveals that different predictors dominate in different urban contexts, offering guidance on sensor selection for building‑height mapping.
DeepC4 is a deep learning-based spatial disaggregation method that uses local census statistics as cluster-level constraints and incorporates multiple conditional label relationships in a multitask learning framework. Applied to Rwandan urban morphology, it achieves macro‑F1 scores of 0.63, 0.78, and 0.45 for roof, wall, and height prediction, respectively, and estimates national dwelling and occupant counts within about 1.1% error compared to census records. The approach outperforms existing GEM and METEOR methods and covers 32‑49% more 500‑meter grid pixels across provinces.
By Joshua Dimasaka, Christian Gei{\ss}, Emily So
The study evaluates how well wavelet-diffusion models (WDM) for precipitation downscaling generalize across different U.S. regions and precipitation regimes. A WDM trained only on Oklahoma data performs competitively in other regions, while a model trained on all six regions achieves the best overall image-domain and detection performance, though improvements vary with precipitation intensity. The analysis shows a strong correlation between spatial autocorrelation (Moran's I) and detection success, indicating that model performance is linked to the spatial organization of precipitation fields.
By Weikang Qian, Yixin Wen, Chugang Yi, Zhi Li, Lingcheng Li, Haizhao Yang
arXiv:2601.07268v2 Announce Type: replace
Abstract: Data-driven landslide susceptibility mapping (LSM) typically relies on landslide conditioning factors (LCFs), whose availability, heterogeneity, an...
By Yusen Cheng, Qinfeng Zhu, Lei Fan
The paper presents a lightweight method to adapt general‑purpose vision‑language models (VLMs) for multispectral and synthetic aperture radar (SAR) image understanding. By rendering each observation as five optical views and one SAR view, naming them in the prompt, and applying LoRA to the language network and selected visual transformer blocks, the authors enable VLMs to process band composites, spectral indices, and radar backscatter without retraining a new foundation model. On a balanced six‑class land‑cover benchmark from BigEarthNet‑v2, the adapted Qwen3‑VL achieves a micro F1 of 0.8275, and the same protocol improves four other VLMs and transfers to flood verification and captioning tasks.
"whyItMatters":"The study shows that existing VLMs can be repurposed for multispectral and SAR tasks through simple input rendering and compact LoRA adaptation, avoiding the need for dedicated encoders and domain pretraining."
By Shanji Liu, Kelu Yao, Junxiao Xue, Chenghui Lv, Xiangyang Miao, Yekai Huang, Yaying Chen, Chao Li
arXiv:2606. 20523v1 Announce Type: cross Abstract: Multimodal foundation models have advanced rapidly thanks to large optical benchmarks, but comparable resources for synthetic aperture radar (SAR) remain limited.
By Sol\`ene Debuys\`ere, Nicolas Trouv\'e, Nathan Letheule, Elise Colin, Georgia Channing
The paper demonstrates a lightweight method to adapt general-purpose vision‑language models (VLMs) for multispectral and synthetic aperture radar (SAR) image understanding. By rendering each observation as five optical views and one SAR view, naming them in the prompt, and applying LoRA to the language network and selected visual transformer blocks, the authors enable VLMs to process band composites, spectral indices, and radar backscatter without retraining a new foundation model. On a balanced six‑class land‑cover benchmark from BigEarthNet‑v2, the adapted Qwen3‑VL achieves a micro F1 score of 0.8275, and the same protocol improves four other VLMs and transfers to flood verification and captioning tasks.
arXiv:2607. 16080v1 Announce Type: new Abstract: Precipitation nowcasting over the immediate 10-90 min period is important for flood management and real-time decision-making in urban regions.
By Akshay Sunil, Muhammed Rashid, Raja Sekhar Sivaraju, Sushma Nair, Subimal Ghosh
Effective flood monitoring is critical for minimizing the impacts of flood disasters on populations and infrastructure. Yet reliable remote sensing across extensive and environmentally diverse regions remains challenging, as most segmentation algorithms lack the generalisation capacity required for large-scale application, while annotated flood data are scarce and unevenly distributed.