基于Logistic和GWR Logistic的雷州半岛野火驱动因素的分析

Analysis on Driving Factors of Wildfire in Leizhou Peninsula Based on Logistic and GWR Logistic

摘要: 利用全局Logistic和地理加权Logistic回归模型，分析了2001—2018年中国雷州半岛热带野火的空间分布和驱动因素。结果表明：2001—2018年雷州半岛野火的总体分布是空间聚集的。雷州半岛野火的驱动因子具有明显的空间差异，GWLR模型更好地解释了这种关系；该地区野火主要是由气象因素驱动的，特别是气温日较差和降雨量驱动的；此外，雷州的野火更多地是由人类活动驱动的，尤其是对农田的管理。本研究可为区域火灾管理战略的制定提供参考。

Abstract: The spatial distribution and drivers of tropical wildfires in Leizhou Peninsula of China from 2001 to 2018 were analyzed by using global Logistic and geographically weighted Logistic regression models. The results show that the overall distribution of wildfires in Leizhou Peninsula from 2001 to 2018 is spatial aggregation. The drivers of wildfires in Leizhou Peninsula have obvious spatial differences, and GWLR model can better explain this relationship. Wildfires in this area are mainly driven by meteorological factors, especially the diurnal temperature range and rainfall. In addition, the wildfires in Leizhou are more driven by human activities, especially the management of farmland. This study can provide reference for the establishment of regional fire management strategy.