S33E-08
Time-Dependent characteristics of Slow Slip Events beneath the Nicoya Peninsula, Costa Rica

Wednesday, 16 December 2015: 15:25
305 (Moscone South)
Nicholas K Voss1, Zhen Liu2, Rocco Malservisi1, Timothy H Dixon1, Yan Jiang3, Susan Y Schwartz4 and Marino Protti5, (1)University of South Florida Tampa, Tampa, FL, United States, (2)Jet Propulsion Laboratory, Pasadena, CA, United States, (3)Geological Survey of Canada Pacific, Vancouver, BC, Canada, (4)University of California-Santa Cruz, Santa Cruz, CA, United States, (5)Observatorio Vulcanológico y Sismológico de Costa Rica, Heredia, Costa Rica
Abstract:
Large geodetically resolved Slow Slip Events (SSE) beneath the Nicoya Peninsula in Costa Rica have been occurring every 21 months +/- 6 months since the installation of continuous GPS stations in 2003, with smaller shallow events occurring irregularly in between. This recurrence interval appears to continue after the 2012 Nicoya M 7.6 earthquake that ruptured a strongly locked patch [Protti et al., 2014, Xue et al., 2015] located between two recurring and well-imaged regions of shallow and deep slow slip. The most recent SSE began in February of 2014, ~17 months after the Nicoya earthquake and 21 months after the start of an earthquake preceding SSE. Despite the observed regularity in time, the characteristics of the individual SSEs vary greatly, with some having shallow or deep slip dominate and others having a more equal distribution of both shallow and deep slip.

We use a modified version of the Extended Network Inversion Filter [e.g. McGuire and Segall, 2003] (ENIF) to identify time dependent characteristics of SSEs before and after the 2012 Nicoya earthquake. Slip in the distinct shallow and deep patches appears to be closely related although not coincident in time, deep slip is slightly delayed compared to the onset of shallow slip for events in 2007 and 2009. Distinct migration both along dip and strike are observed. Further, the ability of the filter to distinguish signal from noise, has allowed for resolution of subtle slip rate variations during the SSEs that correlate reasonably well with previously identified tremor rate [Walter et al , 2011]. Such correlation is less clear in space, possibly due to a combination of uncertainty in tremor locations and inversion regularization. Some coincident migration of slip and tremor did occur during the 2007 event. Investigation of time-dependent slip behaviors of other events is still in progress. A preliminary static inversion of the 2014 event indicates an absence of slip in the shallow slow slip patch that was active during the 2007, 2009 and 2012 SSEs, suggesting a possible change in slip behavior following the 2012 earthquake although not in recurrence interval.