Detecting Large-scale Intracontinental Slow-slip Events (SSEs) Using Geodograms

Abstract

Since the advent in the 1980s of GPS networks to monitor crustal velocity fields, interpretations of geodetic data have generally been based on maps of Earth's surface showing average horizontal site velocity over a specified period of time and plots showing velocity gradients as a function of a position coordinate (e.g., Donnellan et al. 1993; Bennett et al. 1999). For continuous networks, these plots are typically supplemented by time series of position in order to assess the importance of time-dependent or transient behavior (Bock et al. 1993; Hudnut et al. 2002). Thus far, regional transient motions have been revealed by plotting position time series from multiple sites on a common time axis. These plots have been effective in demonstrating the existence of slow-slip events (SSEs) on subduction megathrust interfaces around the globe (e.g., Miller et al. 2002; Melbourne et al. 2005; Schwartz and Rokosky 2007). A large-scale intraplate SSE in the northern Basin and Range Province that occurred between 1999 and 2005 was initially identified by plotting continuous time series with a vertical time axis, arranged according to a spatial position coordinate for each site (Davis et al. 2006, their Figure 3A). Because transient motions are by definition changes in velocity, however, the spatial coherence and magnitude of velocity changes are most directly addressed by plotting the time dependence of velocity rather than position. Here, we describe a method for calculating velocity time series, and then we use these to construct a geodogram from raw continuous GPS time series. The new time series reveal additional transient motions from 2005 to 2007 that are interpreted to reflect the onset of a new SSE beginning in late 2006.