Abstract Repeating earthquakes (REs), which intermittently rupture the same fault area, offer unique constraints on aseismic slip and stress accumulation processes. While REs are typically observed at plate boundaries, recent studies have suggested their potential occurrence within slabs, providing important constraints on intraslab processes; however, detection methods involve arbitrary thresholds whose physical meaning is ambiguous. Here, we systematically examined earthquake repetition in both interplate and intraslab environments using 3,557 M > 2 earthquakes near the 2003 M w 7.0 Miyagi‐oki, Japan, intraslab earthquake. By relocating events and applying source‐overlap criteria, we identified 400 RE pairs—all located on the plate boundary above the intraslab seismicity. No reliable REs were detected within the slab. Our analyses further indicate that earthquakes in this depth range (50–60 km) had shorter rupture durations than expected from a simple circular slip model with 3–10 MPa stress drops, suggesting higher stress drops and smaller fault sizes. These observations highlight the importance of careful RE identification in complex regimes. For comparable interevent distances and frequency bands, waveform correlations were lower for intraslab earthquakes, implying more diverse focal mechanisms and/or stronger seismic scattering. Interevent distance distributions also showed a marked contrast: interplate events formed concentrated clusters, whereas intraslab events were diffusely distributed. These differences suggest fundamental variations in seismogenic environments, including seismic patch isolation, aseismic slip contribution, and fault distribution. Such differences may, in turn, reflect distinctions in temperature, alteration, fault maturity, strain localization, and stress field, which together influence the dominance of either seismic or aseismic slip.
Keisuke Yoshida (Fri,) studied this question.