Assessment of China’s forest fire occurrence with deep learning, geographic information and multisource data
Yakui Shao,Zhichao Wang,Zhongke Feng,Linhao Sun,Xuanhan Yang,Jun Zheng,Tiantian Ma
DOI: https://doi.org/10.1007/s11676-022-01559-1
2022-11-06
Journal of Forestry Research
Abstract:Considerable economic losses and ecological damage can be caused by forest fires, and compared to suppression, prevention is a much smarter strategy. Accordingly, this study focuses on developing a novel framework to assess forest fire risks and policy decisions on forest fire management in China. This framework integrated deep learning algorithms, geographic information, and multisource data. Compared to conventional approaches, our framework featured timesaving, easy implementation, and importantly, the use of deep learning that vividly integrates various factors from the environment and human activities. Information on 96,594 forest fire points from 2001 to 2019 was collected on Moderate Resolution Imaging Spectroradiometer (MODIS) fire hotspots from 2001 to 2019 from NASA's Fire Information Resource Management System. The information was classified into factors such as topography, climate, vegetation, and society. The prediction of forest fire risk was generated using a fully connected network model, and spatial autocorrelation used to analyze the spatial aggregation correlation of active fire hotspots in the whole area of China. The results show that high accuracy prediction of fire risks was achieved (accuracy 87.4%, positive predictive value 87.1%, sensitivity 88.9%, area under curve (AUC) 94.1%). Based on this, it was found that Chinese forest fire risk shows significant autocorrelation and agglomeration both in seasons and regions. For example, forest fire risk usually raises dramatically in spring and winter, and decreases in autumn and summer. Compared to the national average, Yunnan Province, Guangdong Province, and the Greater Hinggan Mountains region of Heilongjiang Province have higher fire risks. In contrast, a large region in central China has been recognized as having a long-term, low risk of forest fires. All forest risks in each region were recorded into the database and could contribute to the forest fire prevention. The successful assessment of forest fire risks in this study provides a comprehensive knowledge of fire risks in China over the last 20 years. Deep learning showed its advantage in integrating multiple factors in predicting forest fire risks. This technical framework is expected to be a feasible evaluation tool for the occurrence of forest fires in China.
forestry