Abstract. We studied the complex and non-stationary records of global earthquake employing the robust statistical and spectral techniques to understand the patterns, processes and periodicity. Singular Spectral Analysis (SSA) and correlation methods are used to quantify the nature of principle dynamical processes of global annual earthquake rates. The SSA decomposes the principle component of earthquake rates (first mode), which suggests that there is a linear increase in the yearly earthquake number from 1975 to 2005 accounting for 93% variance and may be identified with the earth's internal dynamical processes. Superimposed on this monotonic trend, there is an 11 years cyclic variation (second and third modes) accounting for 5% variance, which may corresponds to the well-known solar cycle. The remaining 2% higher order fluctuating components appears to be associated with artificial recharge and natural triggering forces (reservoir, tidal triggering etc.). The correlation study indicates that there is strong positive and negative correlation among the global earthquake rates with surface air temperature and sunspot numbers respectively. Interesting coupling mechanisms do exist, in which atmospheric circulations perturbed by the abrupt temperature variability might change the torques/momentum of inertia (earth's angular momentum) of the earth and thereby may offer the required inputs to trigger earthquake activities at the "critical phases".