Nature of earthquake dynamics in the central Himalayan region: a nonlinear forecasting analysis

Modern nonlinear dynamical system theory has provided a novel perspective for quantifying the predictability behavior of earthquake processes. Here, we employ a nonlinear forecasting approach to analyze a time series of earthquake frequency (earthquake/month M>5) covering a time window 1960–1988. The main objectives of the present study are to (i) characterize the nature of earthquake generating mechanism and common features of temporal evolution of Himalayan earthquake dynamics in terms of deterministic/chaotic, stochastic and random components; (ii) estimate the limits of predictability using the second order Kolmogorov–Sinai entropy (K2). The nonlinear predictive analyses reveal low positive correlation between predicted and observed earthquake frequency values suggesting that the earthquake dynamics in the Central Himalayan is composed of a mixed behavior of stochastic and chaotic nature. This complex and heterogeneous dynamics can be modeled as a nonlinear iterative process with higher degree of freedom. The K2 entropy reveals a low nonzero value (0.2–0.4) suggesting that the earthquake data structure possesses a weak "memory" and hence some predictability may be expected. These results will provide significant constraints for earthquake modeling and hazard analyses in the Himalayan region.