Rapid Rupture Mapping of Potential Tsunamigenic Earthquakes in the Southwest Pacific Using Regional Seismic Arrays

Abstract Rapid spatial characterization of rupture in tsunamigenic earthquakes (Mw≥7.5) is critical for improving tsunami early warning capabilities, particularly for sparsely instrumented regions. We investigate how regional array seismology can be optimized for such settings by enhancing real-time rupture imaging. Using the Multiple Signal Classification (MUSIC) backprojection technique, we evaluate regional array processing performance with the New Zealand broadband seismic network in the Southwest Pacific. Performance is assessed using synthetic earthquakes spanning a range of magnitudes (Mw 8.5–9.1), rupture lengths, directivities (unilateral and bilateral), and source–array geometries (epicentral distances and back azimuths). Results show that the New Zealand seismic network can reliably constrain rupture extent for large regional tsunamigenic earthquakes, with an average rupture length error of less than ∼15%. Earthquake magnitude primarily governs the optimum processing parameters (frequency band and time window length), whereas source–array geometry ultimately limits the rupture imaging quality that can be achieved. Integrating MUSIC-derived rupture extents into tsunami warning workflows generally improves agreement with reference tsunami forecasts compared with magnitude-based scaling alone, demonstrating the value of rapid spatial rupture constraints as a complement to magnitude and ocean observations. These findings provide practical guidance for parameter selection in real-time rupture imaging and highlight the potential of MUSIC-based array processing for rapid characterization of large tsunamigenic earthquakes.