Predicting Forest Fire Occurrence in China Using a Deep Learning Model

The annual number of forest fires at the provincial scale in China was predicted to support regional resource allocation and preventive mitigation planning. Drawing on climate, forest resource, and socioeconomic factors together with forest fire incident records from 2003 to 2023, we construct a deep learning model that integrates a Convolutional Neural Network—Gated Recurrent Unit (CNN-GRU) with a Multi-Head Attention (MHA) mechanism and a Kepler Optimization Algorithm (KOA). The CNN-GRU captures spatiotemporal features, MHA enhances the recognition of intrinsic data relationships, and KOA automatically optimizes the hyperparameters of the hybrid model. Our KOA-CNN-GRU-MHA model outperformed all traditional machine learning baselines. Compared with the native CNN-GRU model, the proposed model reduced the mean absolute error (MAE), mean absolute percentage error (MAPE), and root mean square error (RMSE) by 26.75%, 64.35%, and 16.47%, respectively, for the total annual Number of Forest Fires (NFF) and by 39.07%, 76.03%, and 32.82% for the total annual Number of Small Forest Fires (NSFF). Ranking the influence of predictor variables further supports the identification of key factors associated with provincial fire occurrence trends, which can assist region-level planning.