arXiv cs.AI· Hager Radi Abdelwahed, Toluwaleke Olutayo, Chirag Goel, Jana Osea, Aldo Zaimi, Tom Swystun, Sandy Erni, Francisco Emiliano Lopez Saavedra, Yan Boulanger, Ga\'etan Marceau Caron·· 12 小时前AI 评分34
用 AI 代理模型框架加速加拿大全国野火灾害制图
Accelerating Wildfire Hazard Mapping Across Canada with an AI Surrogate Modelling Framework
AI 导读
研究团队提出 AI 代理模型框架，用多任务深度学习模型模拟 BurnP3+ 的燃烧概率、平均火强度和平均蔓延速度，将原本需数天至数周的计算缩短至秒级。在代表性区域上，该代理模型实现 806 倍端到端 CPU 加速，仅占用 BurnP3+ 13% 的主机内存，并保持下游灾害分类结果。模型还展现出对另一区域 BurnP3+ 研究的零样本迁移能力，轻量微调可显著改善区域校准与预测一致性。
正文 · 原文
Abstract:Canada's increasing wildfire activity requires frequent national hazard mapping to guide proactive mitigation, including fuel management, land-use planning, infrastructure protection, and evacuation routing. However, process-based wildfire simulation frameworks (e.g., BurnP3 and BurnP3+) rely on computationally intensive Monte Carlo simulations to model burn probability, fire intensity, and rate of spread. These workflows can take days to weeks to process a single regional modelling unit. To overcome these computational bottlenecks, we present three contributions developed through collaboration between AI researchers and wildfire modelling experts: (a) a standardized framework for training, fine-tuning, and evaluating AI surrogate models against process-based wildfire simulators; (b) a multi-task deep learning surrogate trained on national-scale data to emulate burn probability, mean fire intensity, and mean rate of spread at a fraction of the computational cost (seconds); and (c) domain-specific metrics and counterfactual evaluation protocols to probe out-of-distribution generalization. Integrating geospatial data (fuel, terrain, ignition likelihood) and regional weather, our model approximates BurnP3+ simulation outputs, achieving strong performance across agreement, rank-correlation, error, and spatial-overlap metrics. It preserves downstream hazard classifications and produces physically plausible responses to expert-informed counterfactual interventions. Furthermore, we demonstrate zero-shot transfer to a separate regional BurnP3+ study and show that lightweight fine-tuning can substantially improve regional calibration and predictive agreement. On a representative region, the surrogate achieves 806-fold end-to-end CPU acceleration and uses 13% of BurnP3+'s host memory, facilitating rapid scenario analysis.
| Comments: | Preprint - under review |
| Subjects: | Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.30170 [cs.AI] |
| (or arXiv:2610.30170v2 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2610.30170 arXiv-issued DOI via DataCite |
Submission history
From: Hager Radi Abdelwahed [view email]
[v1]
Fri, 16 Oct 2026 19:06:07 UTC (38,594 KB)
[v2]
Fri, 23 Oct 2026 01:54:02 UTC (38,597 KB)
来源：arXiv cs.AI · arxiv.org