Considerable evidence indicates that one earthquake can trigger another by stress transfer. The textbook example is a pair of magnitude 6 earthquakes that shook the Superstition Hills, California in 1987. First, the left-lateral Elmore Ranch fault ruptured. Some 12 hours later, the right-lateral Superstition Hills fault ruptured in a second earthquake. The distance between the first (source) epicenter and second (receiver) epicenter is about 10 km. Similar sequences of large (magnitude 6 at least) earthquakes occurred in 1784, 1896, and 2000 in the South Iceland Seismic Zone (SISZ). The research seeks to explain the separation in both time and in space between such triggered earthquakes, focusing in particular on the 2000 SISZ earthquake sequence. The triggering appears to be a two-step process. In the first step, the seismic waves transmit a dynamic stress over the distance from the triggering ?source? event to the (future) location of the triggered ?receiver? event. In the second, slower step, another local process is involved that leads to the second event. In this heuristic model, the first step produces the large separation in space (distance) between the source and receiver events, while the second step produces the separation in time (delay). This is one of the hypotheses that the proposed research is testing. The other hypotheses include: (1) dynamic friction that depends on the fault?s slip rate and a physical state variable; (2) bubbles produced as the seismic wave releases gas from volatile magma; (3) aseismic fault slip; and much more speculatively, (4) some kind of stress pulse that propagates with the observed apparent velocity. To test these competing hypotheses, the research is employing state-of-the-art techniques for measuring and modeling stress transfer in Iceland. The measurements include: (1) geodetic time series from continuously-recording Global Positioning System receivers (CGPS), (2) interferometry using satellite radar images (INSAR), (3) hydrological recordings of water level in boreholes, (4) meteorological recordings of barometric pressure, rainfall and temperature, and (5) precise estimations of earthquake locations and focal mechanisms. The numerical models employ 3-dimensional finite element modeling (FEM) to account for poro-elastic and visco-elastic processes in a fault zone. The model calculates three quantities: (1) the displacement field, which can be compared to INSAR and CGPS measurements; (2) the fluid pressure field, which can be compared to the water level measured in boreholes; and (3) the stress tensor, which can be resolved onto known fault planes to evaluate failure by the Coulomb criterion. Intellectual merit of the activity: The research seeks to delineate the processes causing the triggering over distances longer than several fault lengths and time scales longer than the travel time of seismic waves. It involves both measurements and models in the traditionally separate disciplines of geodesy, seismology, and hydrology. This interdisciplinary research seems poised to add to fundamental understanding of earthquake processes. Broader impact of the activity: The investigators, including colleagues in Iceland, posses a detailed familiarity with the strengths and weaknesses of the observations that places their team in an ideal position to undertake these modeling studies. They are fostering the career of a junior scientist at the intersection of three disciplines: seismology, geodesy and hydrology. By the end of the two-year project, the junior scientist will have mastered three techniques of modern geophysics (CGPS, INSAR and FEM) with applications beyond the university setting. The new information coming from this study is likely to be directly relevant to understanding time-dependent earthquake hazards in Iceland. Improved understanding of earthquake triggering also contributes to the objectives of the National Earthquake Hazard Reduction Program.
