Denoising Diffusion Probabilistic Models for Coastal Inundation Forecasting

2025-05-08

AI Summary

This paper addresses the challenge of real-time, high-precision forecasting of coastal flooding by proposing DIFF-FLOOD, a spatiotemporal probabilistic forecasting method based on denoising diffusion probabilistic models. DIFF-FLOOD is the first to introduce diffusion models into coastal flood probability forecasting, innovatively integrating static spatial priors—specifically digital elevation models—with dynamic temporal context, including neighboring water levels, historical time series, and exogenous covariates, thereby enabling uncertainty-aware, high-resolution spatiotemporal predictions. The model combines convolutional neural networks with cross-attention mechanisms to effectively fuse heterogeneous, multi-source spatiotemporal data. Evaluated on real-world observational data from the Eastern Shore of Virginia, DIFF-FLOOD achieves 6–64% improvements in mean absolute error (MAE) and continuous ranked probability score (CRPS) over state-of-the-art methods, demonstrating substantial gains in prediction accuracy, reliability, and computational scalability.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Machine Learning: Time-Series/Data Streams
• Reasoning under Uncertainty: Relational Probabilistic Models

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Web Mining and Content Analysis: Large pretrained models with web data
• Social Networks and Social Media: Influence propagation, information diffusion, and the prediction on networks

Abstract

Coastal flooding poses significant risks to communities, necessitating fast and accurate forecasting methods to mitigate potential damage. To approach this problem, we present DIFF-FLOOD, a probabilistic spatiotemporal forecasting method designed based on denoising diffusion models. DIFF-FLOOD predicts inundation level at a location by taking both spatial and temporal context into account. It utilizes inundation levels at neighboring locations and digital elevation data as spatial context. Inundation history from a context time window, together with additional co-variates are used as temporal context. Convolutional neural networks and cross-attention mechanism are then employed to capture the spatiotemporal dynamics in the data. We trained and tested DIFF-FLOOD on coastal inundation data from the Eastern Shore of Virginia, a region highly impacted by coastal flooding. Our results show that, DIFF-FLOOD outperforms existing forecasting methods in terms of prediction performance (6% to 64% improvement in terms of two performance metrics) and scalability.

Problem

• Predict coastal inundation levels accurately and quickly
• Incorporate spatial and temporal data for forecasting
• Improve scalability and performance over existing methods

Innovation

• Denoising diffusion models for probabilistic forecasting
• Spatiotemporal context with CNNs and cross-attention
• Improved accuracy and scalability in flood prediction