1 Introduction
Earthquake rupture forecasts rely in part on geologic deformation models that combine fault geometries and long-term fault slip rates (Field et al., 2014; Hatem et al., 2022). For subduction interfaces, deformation models can be difficult to construct because simple notions of fault slip rates do not apply to subduction systems. This is because geodetic coupling varies along strike and downdip (Chlieh et al., 2008; Freymueller & Beavan, 1999; Lay et al., 2012; Pacheco et al., 1993; Scholz & Campos, 2012; Wang, 1995) and strain release along subduction interfaces can be complicated by complex, overlapping patterns of aseismic, coseismic, and postseismic slip and upper-plate structures such as splay faults (Barnhart et al., 2016; Liberty et al., 2013). Because aseismic slip is nearly ubiquitous in the subduction setting, fault slip rates are replaced by the concept of slip deficit rates, which represent the long-term plate convergence rate times the coupling coefficient along the subduction interface determined geodetically (Pacheco et al., 1993). The coupling coefficent ranges from 0 (or 0%) when the interface is fully decoupled and the interseismic (aseismic) slip rate is equal to the local plate convergence rate, to 1 (or 100%) when the interface is fully locked and the slip deficit rate equals the local plate convergence rate (Pacheco et al., 1993).
Recurrence estimates based on paleoseismic data are especially important in subduction zones because the return times of the largest subduction earthquakes greatly exceed the length of historical seismic catalogs and slip deficit rates are not easily converted to earthquake rates (Nelson et al., 2021; Satake & Atwater, 2007). Subduction paleoseismology provides estimates of earthquake recurrence from abrupt changes in relative sea level recorded by geologic archives, such as coastal marsh stratigraphy (Atwater, 1987) and coral microatolls (Taylor et al., 1987). Indirect proxies for subduction interface rupture include turbidites (Adams, 1990; Goldfinger et al., 2012) and tsunami inundation (Witter et al., 2016).