Seasonal terrestrial water load modulation of seismicity at the southeastern margin of the Tibetan Plateau constrained by GNSS and GRACE data

The seasonal terrestrial water load modulation of seismicity is an important phenomenon to understand the mechanism behind earthquake triggering and nucleation. The presence of high-level seismicity and large seasonal water load amplitudes at the southeastern margin of the Tibetan Plateau provides a natural experimental environment for studying the modulation mechanism. The spatiotemporal distribution of the water load was inverted using Global Navigation Satellite System (GNSS) and Gravity Recovery and Climate Experiment (GRACE) data, and an earthquake catalogue (M >= 2.5) was declustered to obtain the background seismicity using the Epidemic Type Aftershock Sequences (ETAS) model. The Multichannel Singular-Spectrum Analysis (M-SSA) is adopted to the time-series of monthly averaged terrestrial water load and background seismicity rates, and the results show 1- and 2-yr periodicities in the seismicity rates and water load. The 1-yr periodicity in the seismicity rate is correlated with the rate of change of the monthly averaged water load. To evaluate the seasonal principal stress perturbations on the tectonic background stress orientations, the stress changes caused by the seasonal water load changes are projected onto the tectonic background stress field orientations constraining by 8 yr of earthquake focal mechanism solutions. The results show that the largest change of the seasonal principal stress perturbations is about 16 kPa. The number of excess earthquakes is evaluated with the background seismicity rates for discrete stress intervals. The results indicate a similar to 10 per cent increase in the seismicity rates that correlate with the rates of the minimum and mean principal stress perturbations. The results above can be explained by the model of harmonic stress perturbations on rate-and-state fault. Based on this model, the nucleation period of the seasonal seismicity should be less than 1 yr at the southeastern margin of the Tibetan Plateau.