Introduction

Slow earthquakes have been discovered in a number of different tectonic environments (subduction zones, the San Andreas fault, Kilauea volcano) and are one of the most exciting current topics in seismology and geodesy. Slow earthquakes have been especially well studied in subduction zone environments including the Nankai trough in Japan, Cascadia, Guerrero, Mexico, and Alaska [Dragert et al., 2001; Freymueller et al., 2001; Heki et al., 1997; Kostoglodov et al., 2003; Lowry et al., 2001; Miller et al., 2002; Rogers and Dragert, 2003]. Slow slip in subduction zones is often associated with non-volcanic tremor, although there is debate over whether the tremor occurs along the subduction zone interface or within the upper plate. One of the most interesting features of many slow slip events is their apparent periodicity: for example, in the southern Vancouver Island - northern Puget Sound region slow slip episodes repeat every 13 months. The physical processes giving rise to the slow slip and associated tremor are the subject of intensive ongoing research by the community of earthquake scientists. The observations and search for theoretical explanations have stimulated some provocative and potentially important hypotheses: for example, Lowry [2006] has recently proposed that periodicity of subduction zone SEs could arise as a resonant response to climate-driven stress perturbations [Lowry, 2006].

In the past decade, Kilauea volcano's south flank has been the site of at least 7 slow (or "silent") earthquakes (SEs) identified with continuous GPS (CGPS) data [Brooks et al., 2005; Brooks et al., 2006; Cervelli et al., 2002; Desmarais et al., 2006; Desmarais et al., 2005; Foster et al., 2006; Segall et al., 2006]. The SEs can be divided into "Western" and "Eastern" families defined by similar patterns of horizontal motions [Brooks et al., 2005; Desmarais et al., 2006] and, of these, the Western family are periodic and separated by 774 +/-7 day periods [Brooks et al., 2006].

The Kilauea SEs are unusual in that they are associated with triggered microearthquakes, although it is not yet known whether nonvolcanic tremor is also present. A distinct swarm of micoearthquakes is triggered in the same place by the periodic SEs [Brooks et al., 2006; Segall et al., 2006]. There is currently some disagreement about whether microearthquakes occur on the regional decollement (~8 +/- 1km; Segall et al. [2006]) or, potentially, at shallower structural levels [Wolfe et al., In Prep.]. Resolving the discrepancy is important because the location of the microseismicity could be used to help constrain the location and further our mechanistic understanding of the SEs themselves [Segall et al., 2006]. If periodicity is maintained, then the next Western family SE and its accompanying microearthquakes are predicted to occur on, or about, March 17, 2007.