An ensemble random forest model for seismic energy forecasting

Abstract

Seismic energy forecasting is critical for hazard preparedness, but current models have limits in accurately predicting seismic energy changes. This paper fills that gap by introducing a novel ensemble-based random forest framework for seismic energy forecasting. Building on a previously established methodology, the global energy time series is decomposed into intrinsic mode functions (IMFs) using ensemble empirical mode decomposition for better representation. Following this approach, we split the data into stationary (IMF1) and non-stationary (sum of IMF2–IMF6) components for modelling. We acknowledge the inadequacy of IMFs in capturing seismic energy dynamics, notably in anticipating the final values of the time series. To overcome this limitation, the yearly seismic energy time series and the stationary and non-stationary parts are also fed as inputs to the developed models. In this study, we employ the support vector machine (SVM), random forest (RF), instance-based learning (IBk), ridge regression (RR), and multi-layer perceptron (MLP) algorithms for the modelling. Furthermore, the five models discussed above are suitably employed in a stacked regression ensemble using random forest as the meta-learner to arrive at the final predictions. The root mean squared error (RMSE) obtained in the training and testing phases of the validation model is 0.127 and 0.134, respectively. It is observed that the performance of the developed ensemble model is superior to those existing in the literature. Further, the developed algorithm is employed for the seismic energy prediction in the active Western Himalayan region for a comprehensively compiled catalogue, and the mean forecasted seismic energy for year 2024 is 7.21 × 1014 J. This work is a pilot project that aims to create a robust, scalable framework for forecasting seismic energy release globally and regionally. The findings of our investigation demonstrate the promise of the ensemble approach in delivering reliable seismic energy forecast, which can help with appropriate hazard preparedness.