Geospatial Diffusion for Land Cover Imperviousness Change Forecasting

2025-08-14

AI Summary

This study addresses the challenge of long-term impervious surface (IU) change prediction for climate risk assessment. We propose the first generative AI–based framework leveraging diffusion models to model land-use/land-cover (LULC) dynamics at decadal scales. Methodologically, the framework integrates multi-source, long-term (≥30-year) remote sensing time series with socioeconomic drivers to generate high-resolution (0.7 × 0.7 km²) IU distribution maps across 12 metropolitan regions and supports multi-scenario simulations. Its key innovation lies in the first application of diffusion models to geospatial synthesis—explicitly capturing the spatiotemporal non-stationarity and path dependence inherent in IU expansion. Quantitative evaluation demonstrates significantly lower prediction error compared to the “no-change” baseline, validating the framework’s effectiveness and scalability for regional hydrological and climate risk assessment.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Humans and AI: Human-Aware Planning and Behavior Prediction
• Machine Learning: Large Multimodal Models (LMMs)

Application Category

• Social Networks and Social Media: Generative AI / large language models and their impact on social systems
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Economics, Online Markets and Human Computation: Economic ramifications for generative AI infrastructure and applications

Abstract

Land cover, both present and future, has a significant effect on several important Earth system processes. For example, impervious surfaces heat up and speed up surface water runoff and reduce groundwater infiltration, with concomitant effects on regional hydrology and flood risk. While regional Earth System models have increasing skill at forecasting hydrologic and atmospheric processes at high resolution in future climate scenarios, our ability to forecast land-use and land-cover change (LULC), a critical input to risk and consequences assessment for these scenarios, has lagged behind. In this paper, we propose a new paradigm exploiting Generative AI (GenAI) for land cover change forecasting by framing LULC forecasting as a data synthesis problem conditioned on historical and auxiliary data-sources. We discuss desirable properties of generative models that fundament our research premise, and demonstrate the feasibility of our methodology through experiments on imperviousness forecasting using historical data covering the entire conterminous United States. Specifically, we train a diffusion model for decadal forecasting of imperviousness and compare its performance to a baseline that assumes no change at all. Evaluation across 12 metropolitan areas for a year held-out during training indicate that for average resolutions $geq 0.7 imes0.7km^2$ our model yields MAE lower than such a baseline. This finding corroborates that such a generative model can capture spatiotemporal patterns from historical data that are significant for projecting future change. Finally, we discuss future research to incorporate auxiliary information on physical properties about the Earth, as well as supporting simulation of different scenarios by means of driver variables.

Problem

• Forecasting land-use and land-cover change (LULC) for risk assessment
• Using Generative AI to model impervious surface changes
• Improving accuracy of decadal imperviousness predictions

Innovation

• Generative AI for land cover forecasting
• Diffusion model for decadal imperviousness prediction
• Conditioned synthesis using historical spatial data