Urban Spatio-Temporal Foundation Models for Climate-Resilient Housing: Scaling Diffusion Transformers for Disaster Risk Prediction
This work addresses the critical problem of predicting climate-related disaster risk for urban housing and infrastructure, specifically for emergency response and intelligent vehicle routing systems, offering a tool for urban planners and disaster management.
This paper introduces Skjold-DiT, a diffusion-transformer framework that forecasts building-level climate-risk indicators by integrating diverse urban spatio-temporal data and transportation network information. It generates calibrated, uncertainty-aware accessibility layers, such as reachability and travel-time inflation, which can be used by intelligent-vehicle routing systems. The framework is evaluated on the new Baltic-Caspian Urban Resilience (BCUR) dataset, comprising 847,392 building-level observations across six cities.
Climate hazards increasingly disrupt urban transportation and emergency-response operations by damaging housing stock, degrading infrastructure, and reducing network accessibility. This paper presents Skjold-DiT, a diffusion-transformer framework that integrates heterogeneous spatio-temporal urban data to forecast building-level climate-risk indicators while explicitly incorporating transportation-network structure and accessibility signals relevant to intelligent vehicles (e.g., emergency reachability and evacuation-route constraints). Concretely, Skjold-DiT enables hazard-conditioned routing constraints by producing calibrated, uncertainty-aware accessibility layers (reachability, travel-time inflation, and route redundancy) that can be consumed by intelligent-vehicle routing and emergency dispatch systems. Skjold-DiT combines: (1) Fjell-Prompt, a prompt-based conditioning interface designed to support cross-city transfer; (2) Norrland-Fusion, a cross-modal attention mechanism unifying hazard maps/imagery, building attributes, demographics, and transportation infrastructure into a shared latent representation; and (3) Valkyrie-Forecast, a counterfactual simulator for generating probabilistic risk trajectories under intervention prompts. We introduce the Baltic-Caspian Urban Resilience (BCUR) dataset with 847,392 building-level observations across six cities, including multi-hazard annotations (e.g., flood and heat indicators) and transportation accessibility features. Experiments evaluate prediction quality, cross-city generalization, calibration, and downstream transportation-relevant outcomes, including reachability and hazard-conditioned travel times under counterfactual interventions.