Evolution of Tremor's Tidal Sensitivity Through the Slow Slip Cycle

Solid Earth and ocean loading tides generate stresses that modulate slow slip and tremor on deep plate boundaries. Analysis of the influence of tidal stress on 35000 tremors located in a northern Cascadia region that ruptures repeatedly in M6.5 to 6.8 slow earthquakes has yielded several new findings (Houston, 2014, subm.). Tremor sensitivity to tidal stress is low for the first day or so of activity at a spot, but increases progressively over a few days of slip at that spot, constraining the pace of weakening during ETS. After the first day or so, sensitivity varies exponentially with stress. Consideration of an isotropic pore pressure model allows an in-situ estimate of intrinsic friction on the deep subduction interface, yielding values far below lab values (i.e., mu < 0.2). A model of threshold failure strength, in which tremor occurs when fault stress exceeds the strength, explains the evolving, increasing, exponential sensitivity to tides.

By applying a similar approach to the tremor that occurs between the major ETSs, I constrain the evolution of tidal sensitivity through the slow slip ‘seismic cycle’. About 40% of tremor in Cascadia occurs between major ETSs (Wech et al, 2010). Inter-ETS tremor groups of several months duration show some sensitivity to tidal stress, although not as much as tremor during the late portions of ETS. Interestingly, I see an evolution during the 12-15 month interval between major ETSs. In particular, tidal sensitivity increases moderately toward the end of the interval as the fault is reloaded and the next large ETS is imminent. First-order features are consistent with extension of the above-mentioned model of threshold failure strength versus stress. The concept is that during the inter-ETS period, strength rebuilds as the logarithm of time (e.g., Vidale et al, 1994) since the last major ETS, while average shear or Coulomb stress on the plate interface rebuilds roughly linearly due to plate convergence. Tidal stresses become more effective in triggering tremor later in the cycle as the linearly-growing stress approaches the logarithmically-growing strength. This approach illuminates the competition between healing on the plate interface and reloading with tectonic stress, and can help constrain healing rate and conditions on the deep subduction interface.