A CNN-BiLSTM Model with Attention Mechanism for Earthquake Prediction

TL;DR

The paper tackles the challenge of predicting earthquake activity by introducing a CNN-BiLSTM with an attention mechanism to jointly forecast the monthly count and maximum magnitude of earthquakes in Mainland China. It leverages zero-order hold preprocessing to stabilize inputs, a CNN to extract spatial features, a BiLSTM to model temporal dynamics, and an attention layer to emphasize influential features, achieving superior generalization over multiple baselines. Across nine regional divisions, the method consistently outperforms shallow and deep learning models in RMSE, MAE, and $R^2$ metrics for both prediction tasks. The work provides a robust, region-aware framework with practical relevance for seismic risk assessment and disaster preparedness, supported by extensive experiments on USGS/NSC catalogs.

Abstract

Earthquakes, as natural phenomena, have continuously caused damage and loss of human life historically. Earthquake prediction is an essential aspect of any society's plans and can increase public preparedness and reduce damage to a great extent. Nevertheless, due to the stochastic character of earthquakes and the challenge of achieving an efficient and dependable model for earthquake prediction, efforts have been insufficient thus far, and new methods are required to solve this problem. Aware of these issues, this paper proposes a novel prediction method based on attention mechanism (AM), convolution neural network (CNN), and bi-directional long short-term memory (BiLSTM) models, which can predict the number and maximum magnitude of earthquakes in each area of mainland China-based on the earthquake catalog of the region. This model takes advantage of LSTM and CNN with an attention mechanism to better focus on effective earthquake characteristics and produce more accurate predictions. Firstly, the zero-order hold technique is applied as pre-processing on earthquake data, making the model's input data more proper. Secondly, to effectively use spatial information and reduce dimensions of input data, the CNN is used to capture the spatial dependencies between earthquake data. Thirdly, the Bi-LSTM layer is employed to capture the temporal dependencies. Fourthly, the AM layer is introduced to highlight its important features to achieve better prediction performance. The results show that the proposed method has better performance and generalize ability than other prediction methods.

Figures (9)

• Figure 1: schematic diagram of bidirectional LSTM
• Figure 2: The step in determining AM
• Figure 3: The architecture of proposed method
• Figure 4: The flow chart of the proposed method
• Figure 5: A visual representation of the division of China's nine study areas