Pseudoprospective Evaluation of the Foreshock Traffic-Light System in Ridgecrest and Implications for Aftershock Hazard Assessment

Abstract

The Mw 7.1 Ridgecrest earthquake sequence in California in July 2019 offered an opportunity to evaluate in near‐real time the temporal and spatial variations in the average earthquake size distribution (the b‐value) and the performance of the newly introduced foreshock traffic‐light system. In normally decaying aftershock sequences, in the past studies, the b‐value of the aftershocks was found, on average, to be 10%–30% higher than the background b‐value. A drop of 10% or more in “aftershock” b‐values was postulated to indicate that the region is still highly stressed and that a subsequent larger event is likely. In this Ridgecrest case study, after analyzing the magnitude of completeness of the sequences, we find that the quality of the monitoring network is excellent, which allows us to determine reliable b‐values over a large range of magnitudes within hours of the two mainshocks. We then find that in the hours after the first Mw 6.4 Ridgecrest event, the b‐value drops by 23% on average, compared to the background value, triggering a red foreshock traffic light. Spatially mapping the changes in b-values, we identify an area to the north of the rupture plane as the most likely location of a subsequent event. After the second, magnitude 7.1 mainshock, which did occur in that location as anticipated, the b‐value increased by 26% over the background value, triggering a green traffic light. Finally, comparing the 2019 sequence with the Mw 5.8 sequence in 1995, in which no mainshock followed, we find a b‐value increase of 29% after the mainshock. Our results suggest that the real‐time monitoring of b‐values is feasible in California and may add important information for aftershock hazard assessment.