Abstract:
Characterizing spatial and temporal variations of slip behavior observed along subduction faults is of great significance for understanding the dynamics of subduction zones, features of great subduction zone earthquakes and deformation patterns across the subduction plate boundary through the seismic cycle. The Alaska-Aleutian subduction zone is one of the most tectonically active margins in the world. Great earthquakes and slow slip events recorded in this area are closely related in space. An increasingly dense array of Global Positioning System (GPS) receivers measures surface deformation at sites with high accuracy and provides a perfect tool for estimating the slip distribution on the plate boundary. GPS observations show that the motion of the Earth is not entirely linear: the long-term steady motion is interrupted by events like earthquakes, slow slip events (SSEs) and deformation of volcanoes, etc. Two long-term SSEs were detected in Lower Cook Inlet, Alaska (1992.0-2004.8 and 2009.85-2011.81) by inverting the slip distributions from GPS site velocities. The occurrence of SSEs based on the estimated slip distribution patterns provides strong evidence for the transition from stick-slip behavior to episodes and continuous aseismic creep on the subduction plate interface. Coulomb stressing rate changes (CSRC) due to the two detected long-term SSEs indicate that regions in the shallow slab (30-60 km) that experience significant increase in CSRC show an increase in seismicity rate during SSE periods. The modified quantitative rate/state stress transfer model suggests that the SSEs increase stress on surrounding faults, thereby increasing the seismicity rate even though the ratio of the SSE induced stressing rate to the background stressing rate is small. The SSEs were shown to cause significant stress changes in the seismogenic zone. This highlights the importance of exploring the relationship between SSEs and earthquakes, as well as how this relationship impacts the strain accumulation in the subduction zone. A repeat survey of the existing campaign GPS sites combined with continuous GPS sites provided a > 20 year time span for estimating the interseismic velocities of the Alaska Peninsula. From this I inferred a more precise model for the location and spatial extent of the change from locked to creeping behavior across the Alaska Peninsula. Given this more detailed distribution of the slip behavior, the results suggest that slip behavior correlates with the pre-existing plate fabric on the downgoing plate, seismic behavior, the reflection character of the slab interface itself and the rupture history of past great earthquakes.
