[论文解读] Deep graphical regression for jointly moderate and extreme Australian wildfires
本文提出一个两阶段深度图回归模型,使用图卷积神经网络和在不规则空间域上的扩展广义帕累托分布,联合建模澳大利亚野火的发生与径向扩散。
Recent wildfires in Australia have led to considerable economic loss and property destruction, and there is increasing concern that climate change may exacerbate their intensity, duration, and frequency. Hazard quantification for extreme wildfires is an important component of wildfire management, as it facilitates efficient resource distribution, adverse effect mitigation, and recovery efforts. However, although extreme wildfires are typically the most impactful, both small and moderate fires can still be devastating to local communities and ecosystems. Therefore, it is imperative to develop robust statistical methods to reliably model the full distribution of wildfire spread. We do so for a novel dataset of Australian wildfires from 1999 to 2019, and analyse monthly spread over areas approximately corresponding to Statistical Areas Level~1 and~2 (SA1/SA2) regions. Given the complex nature of wildfire ignition and spread, we exploit recent advances in statistical deep learning and extreme value theory to construct a parametric regression model using graph convolutional neural networks and the extended generalized Pareto distribution, which allows us to model wildfire spread observed on an irregular spatial domain. We highlight the efficacy of our newly proposed model and perform a wildfire hazard assessment for Australia and population-dense communities, namely Tasmania, Sydney, Melbourne, and Perth.
研究动机与目标
- 通过对中等到极端蔓延的完整分布进行建模,推动对野火风险的稳健量化。
- 构建一个针对澳大利亚的新颖不规则空间聚合烧毁面积数据集(SA1/SA2)。
- 开发一个混合模型,将基于 GCNN 的发生预测器与基于 eGPD 的扩散模型结合起来。
- 提供大陆级危险地图并评估人口密集社区的风险。
- 识别驱动野火频率与强度在时空上变化的关键预测变量。
提出的方法
- 将 Y(s,t) 建模为月度烧毁面积的平方根,采用两部分分布:发生概率 p0(s,t) 和给定发生的条件扩散 F+(y)。
- 使用具有图卷积神经网络的深度逻辑回归来估计 p0(s,t) 的对数几率 logit。
- 用扩展广义帕累托分布 eGPD(κ, σ, ξ) 对 F+ 建模,其中 σ(s,t) 作为预测变量的函数建模,且 σ(s,t) 与 sqrt(area(s)) 成正比。
- 将空间数据表示为一个图,顶点为不规则区域,基于大圆距离并带截断 Delta 的加权邻接矩阵 A。
- 采用带跳跃连接的 GCNN 层来捕捉空间依赖性,结合区域特定信息与邻域信息。
- 通过对负对数似然的 Adam 优化来拟合模型,使用时间自助法评估不确定性,并采用稳健的交叉验证训练/验证划分。
![Figure 1 : Maps of the monthly radial wildfire spread [ $\sqrt{\mbox{BA}}$ ; km] (top-left), polygon area [km 2 ] (top-right), monthly average air temperature [K] (bottom-left), and NDVI [unitless] (bottom-right) for January 2002. Grey regions in the top-left panel are Statistical Areas Level 1 with](https://ar5iv.labs.arxiv.org/html/2308.14547/assets/Maps/Spread_32_Australia_C100.png)
实验结果
研究问题
- RQ1在不规则空间域上,野火发生与扩散的联合模型能否捕捉从中等到极端烧毁面积的完整分布?
- RQ2在利用预测变量的空间结构时,与密集神经网络相比,基于 GCNN 的预测器表现如何?
- RQ3哪些关键协变量在时空上驱动澳大利亚野火的频率与严重性?
- RQ4如何将扩展广义帕累托分布框架与深度学习结合,以建模非极端和极端的野火扩散?
- RQ5使用所提出框架的情况下,人口密集的澳大利亚社区的空间趋势和危险特征是什么?
主要发现
- 该框架实现了对零烧毁面积与正值烧毁面积的联合建模,以及正值在时空上的分布。
- 基于 eGPD 的分量提供了一个灵活的野火扩散模型,覆盖大量值和尾部行为,具有渐进上界尾部的理论依据。
- 在不规则空间图上的 GCNN 通过利用协变量的空间结构,提升拟合优度,相较于标准密集网络。
- 澳大利亚及主要人口中心(如悉尼、墨尔本、珀斯、塔斯马尼亚)的危险地图揭示野火频率和严重性的时空趋势。
- 该方法识别出在整个域上驱动野火发生与扩散的重要预测因子。
![Figure 2 : Top panels: region-wise $5\%$ (left) and $95\%$ (right) quantiles of the monthly radial wildfire spread [ $\sqrt{\mbox{BA}}$ ; km] pooled across the observation period. Grey regions are Statistical Areas Level 1 within the Northern Territory where the response, BA, is not available. Botto](https://ar5iv.labs.arxiv.org/html/2308.14547/assets/Maps/Spread_025_Australia_C100.png)
更好的研究,从现在开始
从阅读论文到最终审阅,大幅缩短您的研究时间。
无需绑定信用卡
本解读由 AI 生成,并经人工编辑审核。