Coeval rupture of imbricate reverse faults increases the moment magnitude (Mw) of the resulting earthquake. Detailed mapping and paleoseismic data can yield useful insights into the probability and Mw potential of multifault ruptures. We present a paleoseismic study of two active imbricate reverse faults, the Fox Peak and Forest Creek faults, in the central South Island of New Zealand. Both faults have recurrence intervals of ∼3000 years, most recent events with overlapping age distributions, and sole into the same structure at depth. Surface and subsurface data indicate average single event displacements of ∼2 m for the Fox Peak fault and 1 m for the Forest Creek fault. Monte Carlo simulations provide Mw estimates for a range of rupture scenarios (independent and combined), fault geometries, and coseismic displacements. The exponential fault‐to‐fault jump probability depends on the shortest distance between two faults, which is allowed to vary in the model based on regional hypocentral depths and the modeled fault geometries. Coulomb stress modeling is used to analyze stresses induced on the receiver fault plane, the Forest Creek fault, as a semiquantitative test of triggered rupture feasibility and to determine credible Mw distributions. The results suggest a maximum credible event (MCE) of Mw∼7.5–7.6 for listric geometries on the Fox Peak and Forest Creek faults. These estimates represent a 0.2–0.5 magnitude increase over most models, which show averages of Mw∼7.1–7.3 for rupture scenarios on planar faults. The Monte Carlo approach employed herein is an improvement over simple empirical relationships for estimating Mw for surface‐rupturing earthquakes and MCEs for reverse‐fault systems, because it provides realistic uncertainty estimates and can be readily applied to other fault systems around the world.
Online Material: Digital elevation model and sedimentation model of the trench 1 area, color trench logs with photomosaics, and detailed trench unit descriptions.