Large-scale flood modeling and forecasting with FloodCast

📄 arXiv: 2403.12226v1 📥 PDF

作者: Qingsong Xu, Yilei Shi, Jonathan Bamber, Chaojun Ouyang, Xiao Xiang Zhu

分类: cs.LG, cs.CV, physics.flu-dyn

发布日期: 2024-03-18

备注: 40 pages, 16 figures, under review


💡 一句话要点

提出FloodCast以解决大规模洪水建模与预测问题

🎯 匹配领域: 支柱八:物理动画 (Physics-based Animation)

关键词: 洪水建模 水动力模型 物理信息神经网络 多卫星观测 几何自适应 预测精度 灾害管理

📋 核心要点

  1. 现有的大规模水动力模型依赖固定分辨率网格,导致计算成本高且预测精度不足。
  2. 本文提出FloodCast框架,结合多卫星观测与几何自适应的物理信息神经求解器GeoPINS,提升洪水预测能力。
  3. 实验结果表明,序列到序列GeoPINS在高水位情况下与传统水动力模型表现一致,且预测误差更小。

📝 摘要(中文)

大规模水动力模型通常依赖固定分辨率的空间网格和模型参数,且计算成本高,限制了其准确预测洪水峰值和发出紧急警报的能力。本文构建了一个快速、稳定、准确、分辨率不变且几何自适应的洪水建模与预测框架FloodCast。该框架包括两个主要模块:多卫星观测和水动力建模。在多卫星观测模块中,提出了一种实时无监督变化检测方法和降雨处理分析工具,以充分利用多卫星观测在大规模洪水预测中的潜力。在水动力建模模块中,引入了一种几何自适应的物理信息神经求解器GeoPINS,具有不需要训练数据的优点,并且采用快速、准确、分辨率不变的架构。GeoPINS在常规和不规则域上的表现令人印象深刻。基于GeoPINS,提出了一种序列到序列的GeoPINS模型,以处理大规模洪水建模中的长期时间序列和广泛空间域。

🔬 方法详解

问题定义:本文旨在解决现有大规模水动力模型在洪水预测中的高计算成本和精度不足的问题。传统模型依赖固定分辨率的网格,限制了其在复杂地形和大范围区域的应用。

核心思路:提出FloodCast框架,通过多卫星观测和几何自适应的物理信息神经求解器GeoPINS,克服传统模型的局限性。GeoPINS不需要训练数据,且具备快速、准确和分辨率不变的特性。

技术框架:FloodCast框架分为两个主要模块:多卫星观测模块和水动力建模模块。多卫星观测模块利用实时无监督变化检测和降雨分析工具,水动力建模模块则使用GeoPINS进行洪水模拟。

关键创新:GeoPINS是本研究的核心创新,采用物理信息神经网络架构,能够在不规则域上高效求解偏微分方程,显著提高了洪水预测的准确性和效率。

关键设计:GeoPINS的设计包括快速的傅里叶神经算子架构,能够处理复杂的几何形状和边界条件,且在高水位情况下与传统水动力模型的结果高度一致。

🖼️ 关键图片

fig_0
fig_1
fig_2

📊 实验亮点

实验结果显示,序列到序列GeoPINS在洪水预测中表现优异,尤其在高水位情况下,其预测误差显著低于传统水动力模型,验证了其在复杂环境下的有效性和可靠性。

🎯 应用场景

该研究的FloodCast框架具有广泛的应用潜力,尤其在洪水预测、灾害管理和环境监测等领域。通过提高洪水预测的准确性和时效性,能够为相关决策提供重要支持,减少灾害带来的损失,具有重要的社会和经济价值。

📄 摘要(原文)

Large-scale hydrodynamic models generally rely on fixed-resolution spatial grids and model parameters as well as incurring a high computational cost. This limits their ability to accurately forecast flood crests and issue time-critical hazard warnings. In this work, we build a fast, stable, accurate, resolution-invariant, and geometry-adaptative flood modeling and forecasting framework that can perform at large scales, namely FloodCast. The framework comprises two main modules: multi-satellite observation and hydrodynamic modeling. In the multi-satellite observation module, a real-time unsupervised change detection method and a rainfall processing and analysis tool are proposed to harness the full potential of multi-satellite observations in large-scale flood prediction. In the hydrodynamic modeling module, a geometry-adaptive physics-informed neural solver (GeoPINS) is introduced, benefiting from the absence of a requirement for training data in physics-informed neural networks and featuring a fast, accurate, and resolution-invariant architecture with Fourier neural operators. GeoPINS demonstrates impressive performance on popular PDEs across regular and irregular domains. Building upon GeoPINS, we propose a sequence-to-sequence GeoPINS model to handle long-term temporal series and extensive spatial domains in large-scale flood modeling. Next, we establish a benchmark dataset in the 2022 Pakistan flood to assess various flood prediction methods. Finally, we validate the model in three dimensions - flood inundation range, depth, and transferability of spatiotemporal downscaling. Traditional hydrodynamics and sequence-to-sequence GeoPINS exhibit exceptional agreement during high water levels, while comparative assessments with SAR-based flood depth data show that sequence-to-sequence GeoPINS outperforms traditional hydrodynamics, with smaller prediction errors.