Annual modulation of triggered seismicity following the 1992 Landers earthquake in California

Abstract

The mechanism responsible for the triggering of earthquakes remains one of the least-understood aspects of the earthquake process. The magnitude-7.3 Landers, California earthquake of 28 June 1992 was followed for several weeks by triggered seismic activity over a large area, encompassing much of the western United States. Here we show that this triggered seismicity marked the beginning of a five-year trend, consisting of an elevated microearthquake rate that was modulated by an annual cycle, decaying with time. The annual cycle is mainly associated with several hydrothermal or volcanic regions where short-term triggering was also observed. These data indicate that the Landers earthquake produced long-term physical changes in these areas, and that an environmental source of stress--plausibly barometric pressure--might be responsible for the annual variation.

Annual modulation of seismicity along the San Andreas Fault near Parkfield, CA

“…[2] Microearthquakes may be triggered or modulated by climatic forces that have an annual period, implying a causal link between the hydrologic cycle and the mechanical behavior of the upper crust [Gao et al, 2000;Saar and Manga, 2003;Christiansen et al, 2005;Kraft et al, 2006]. This relation implies that stresses induced by the annual hydrologic cycle are sufficient to fracture near-critically stressed rock either through pore-pressure diffusion [Talwani and Acree, 1984;Shapiro et al, 2003;Hainzl et al, 2006] or loading/unloading of the elastic crust [Heki, 2003].…”

Semiannual Earthquake Periodicity

“…Along the SAF system, small earthquakes are distributed throughout the thickness of the seismogenic crust, they undoubtedly sample a wide range in the values of physical properties along the fault system (pore pressure, rock type, temperature, pressure, fluid chemistry, mineral hydration, fault-gouge thickness and composition, crack size and orientation, etc. ); and, except in special environments (e.g., hydrothermal and volcanic centers; Gao et al, 2000), the diversity of frictional and sliding conditions at the subkilometer scale is broad and likely chaotic and may conceal a subtle tidal signal in shallow seismicity.…”

“…First, I excluded only those foreshocks and aftershocks denoted in the UCERF catalog, all of which were identified through a consistent declustering process (Felzer, 2013) irrespective of season. It is known that, at the microearthquake level, aftershock rates can be biased by natural annual periodicity (Gao et al, 2000), and a declustering filter applied uniformly to such cases could remove more aftershocks in one half of the year than in the other. However, such cases are rare or limited to hydrothermal or volcanic centers and as yet have not been recognized in large-magnitude aftershock sequences.…”

Decomposing spatio-temporal seismicity patterns

“…short-vs. longterm), but numerical contraints also exist in that, for example, the choice of l and t is a compromise between resolution and a sufficient population of the grid cells. t also restricts the ability to detect short-term transient changes or periodic signals (Gao et al (2000) present an example for annually modulated seismicity rates). The time shift t 0 may become crucial in the case of fluctuations whose lifetime or onset should not be spread over adjacent time slices or if one wants a time window to end prior to the occurence of a large event in order to detect possible precursors (see Tiampo et al (2000) for such an application).…”