基于因子频发模型的楚雄州森林火灾易发性研究

Susceptibility Study of Forest Fires in Chuxiong Prefecture Based on a Multi-Factor Frequency Model

摘要: 近年来，云南省楚雄彝族自治州森林火灾频发，对生态安全与社会经济构成严重威胁。为揭示火灾驱动机制与时空分异规律，本研究基于多因子频发模型，集成气象气候、地形地势、人类活动及可燃物特征四类指标，结合主成分分析(PCA)与核密度法，系统评估了火灾易发性及其空间动态。方法层面，通过层次分析法量化12项因子权重，构建线性回归模型解析多因子交互效应；利用ArcGIS 10.8平台计算欧式距离，分析2016~2023年火灾点的时空集聚特征，并生成易发性区划图。结果表明：(1) 气象气候条件为火灾主导驱动因子(权重0.61)，其中降水(贡献率0.23)与人类活动(权重0.13)分别通过干旱环境加剧与火源暴露度提升显著影响火灾发生；(2) 核密度分析显示，火灾高密度区(核密度 > 0.75)沿楚雄市–禄丰县–双柏县形成带状扩散格局，且欧式距离表明热点与城镇中心的近域关联性(<50 km)；(3) 空间易发性分布呈现“双核驱动、地形分异”特征，高易发区(指数 ≥ 3.5)集中于干热河谷与交通干线，低易发区(指数 ≤ 1.5)受地形屏障抑制。研究提出差异化防控策略，为区域火灾风险管理提供科学依据。

Abstract: In recent years, frequent forest fires in Chuxiong Yi Autonomous Prefecture, Yunnan Province, have posed severe threats to ecological security and socio-economic stability. To unravel the driving mechanisms and spatiotemporal heterogeneity of fire occurrences, this study employed a multi-factor frequency model, integrating four categories of indicators-meteorological conditions, terrain features, human activities, and combustible material characteristics. Principal Component Analysis (PCA) and kernel density estimation were combined to systematically assess fire susceptibility and spatial dynamics. Methodologically, the Analytic Hierarchy Process (AHP) was applied to quantify the weights of 12 influencing factors, while a linear regression model was constructed to analyze multi-factor interactions. Euclidean distance calculations on the ArcGIS 10.8 platform were utilized to examine the spatiotemporal clustering patterns of fire incidents from 2016 to 2023, generating a susceptibility zoning map. Key findings include: (1) Meteorological conditions dominated fire risks (weight = 0.61), with precipitation (contribution rate = 0.23) and human activities (weight = 0.13) significantly amplifying risks through drought intensification and ignition source exposure; (2) Kernel density analysis revealed a belt-shaped expansion of high-density fire zones (kernel density > 0.75) along the Chuxiong-Lufeng-Shuangbai corridor, with Euclidean distance confirming the proximity (<50 km) of hotspots to urban centers; (3) Spatial susceptibility exhibited a “dual-core driven, terrain-modulated” pattern, where high-susceptibility areas (index ≥3.5) clustered in arid valleys and transportation corridors, while low-susceptibility zones (index ≤1.5) were constrained by topographic barriers. The study proposes differentiated prevention and control strategies, offering scientific guidance for regional fire risk management and prevention.