A Graph Neural Network for Global Daily Fire Radiative Power Prediction at Medium-Range Lead Times

📄 arXiv: 2610.11022v1 📥 PDF

作者: Li Zhang, Jun Wang, Isidora Jankov, Yongxin Liu, Gonzalo A. Ferrada, Ravan Ahmadov, Ligia Bernardet, Haonan Chen, Shobha Kondragunta

分类: physics.ao-ph, cs.LG

发布日期: 2026-10-08

备注: Submitted to Artificial Intelligence for the Earth Systems


💡 一句话要点

提出图神经网络以解决全球火灾辐射功率预测问题

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

关键词: 图神经网络 火灾预测 生物质燃烧 时空建模 环境监测

📋 核心要点

  1. 现有方法依赖延迟的火灾观测数据,导致预测准确性不足,无法有效反映火灾活动的演变。
  2. 本文提出了一种基于时空图神经网络的模型,结合气象、土地覆盖、植被信息及火灾历史进行FRP预测。
  3. 实验结果表明,该模型在1天和7天的预测中均显著降低了均方误差,且在火灾检测和强度预测上表现优异。

📝 摘要(中文)

有效预测生物质燃烧活动几天后的情况对空气质量和气溶胶预测至关重要。现有方法依赖于延迟约1.5天的卫星火灾辐射功率(FRP)产品,导致预测周期内火灾活动假设不变。本文提出了一种数据驱动模型,利用时空图神经网络,从最新观察数据中预测全球FRP,预测范围为1至7天。该模型在2020-2022年数据上训练,并在2023-2024年进行评估,结果显示其在全球季节性周期的再现性上显著优于传统方法,且在不同分辨率下均表现出较低的均方误差。

🔬 方法详解

问题定义:本文旨在解决现有火灾辐射功率(FRP)预测方法的不足,特别是由于依赖延迟数据而导致的预测准确性低下和火灾活动假设不变的问题。

核心思路:通过构建一个时空图神经网络模型,利用最新的火灾观察数据和相关气象、土地覆盖及植被信息,进行未来1至7天的FRP预测,以提高预测的时效性和准确性。

技术框架:模型的整体架构包括数据预处理、特征提取、图神经网络训练和预测输出四个主要模块。数据预处理阶段整合了气象和火灾历史数据,特征提取阶段则提取了与火灾活动相关的时空特征。

关键创新:该研究的主要创新在于将图神经网络应用于火灾辐射功率预测,能够有效捕捉时空依赖关系,区别于传统的基于时间序列的预测方法。

关键设计:模型采用了多层图神经网络结构,损失函数设计为均方误差,优化过程中使用Adam优化器,确保模型在训练过程中的收敛性和稳定性。

🖼️ 关键图片

img_0
img_1
img_2

📊 实验亮点

实验结果显示,在2023年,模型在1天和7天的预测中均降低了均方误差32%和43%;在2024年,分别降低了24%和40%。在1°分辨率下,关键成功指数范围为0.32至0.60,表明模型在火灾检测和强度预测上具有良好的性能。

🎯 应用场景

该研究的成果可广泛应用于气象预报、环境监测和灾害管理等领域,帮助相关机构提前预测火灾活动,优化应急响应和资源配置,进而提升空气质量和生态环境保护的效率。

📄 摘要(原文)

Skillful prediction of biomass-burning activity several days in advance is important for air-quality forecasting and aerosol prediction. Two operational constraints motivate this work. First, the GBBEPx satellite fire radiative power (FRP) product used to initialize NOAA's GEFS-Aerosols is available with about a 1.5-day latency, so each forecast cycle relies on the most recently available, but already outdated, fire observations. Second, these fire inputs are then held fixed throughout the subsequent 5-day operational forecast, or 7 days in the GSL experimental system, effectively assuming no evolution in fire activity. We develop a data-driven model that predicts global FRP one to seven days ahead from the most recent available observations. The model adapts a spatiotemporal graph neural network using reanalysis meteorology, land-cover and vegetation information, recent fire history, and GBBEPx FRP as the training target. It is trained on 2020-2022 data and evaluated for 2023-2024. The model reproduces the global seasonal cycle and substantially outperforms persistence. At 0.1$^\circ$ resolution, mean squared error is reduced by 32% at one-day lead and 43% at seven days in 2023, and by 24% and 40% in 2024. At 1$^\circ$ resolution, the critical success index ranges from 0.32 to 0.60. Detection skill declines only modestly with lead time, whereas intensity skill degrades more rapidly. Large fires are detected reliably, but their radiative power is systematically underestimated. These results demonstrate useful predictability of fire activity several days ahead and identify intensity calibration and small-fire placement as the main remaining challenges before predicted FRP can support operational aerosol forecasts.