Real-Time Performance of the PLUM Earthquake Early Warning Method during the 2019 M 6.4 and 7.1 Ridgecrest, California, Earthquakes

Abstract

<jats:title>ABSTRACT</jats:title> <jats:p>We evaluate the timeliness and accuracy of ground-motion-based earthquake early warning (EEW) during the July 2019 M 6.4 and 7.1 Ridgecrest earthquakes. In 2018, we began retrospective and internal real-time testing of the propagation of local undamped motion (PLUM) method for earthquake warning in California, Oregon, and Washington, with the potential that PLUM might one day be included in the ShakeAlert EEW system. A real-time version of PLUM was running on one of the ShakeAlert EEW system’s development servers at the time of the 2019 Ridgecrest sequence, allowing us to evaluate the timeliness and accuracy of PLUM’s warnings for the M 6.4 and 7.1 mainshocks in real time with the actual data availability and latencies of the operational ShakeAlert EEW system. The latter is especially important because high-data latencies during the M 7.1 earthquake degraded ShakeAlert’s performance. PLUM proved to be largely immune to these latencies. In this article, we present a retrospective analysis of PLUM performance and explore three potential regional alerting strategies ranging from spatially large regions (counties), to moderate-size regions (National Weather Service public forecast zones), to high-spatial specificity (50 km regular geographic grid). PLUM generated initial shaking forecasts for the two mainshocks 5 and 6 s after their respective origin times, and faster than the ShakeAlert system’s first alerts. PLUM was also able to accurately forecast shaking across southern California for all three alerting strategies studied. As would be expected, a cost-benefit analysis of each approach illustrates trade-offs between increasing warning time and minimizing the area receiving unneeded alerts. Choosing an optimal alerting strategy requires knowledge of users’ false alarm tolerance and minimum required warning time for taking protective action, as well as the time required to distribute alerts to users.</jats:p>

The PLUM Earthquake Early Warning Algorithm: A Retrospective Case Study of West Coast, USA, Data

“…The results presented in our study are consistent with past analyses that use a two-station version of PLUM to study U.S. data (Cochran et al 2019;Minson et al, 2020). However, in this work we use a more conservative version of PLUM, applying a higher secondary triggering threshold of 3.0 instead of the lower 2.5 value in past work (Cochran et al 2019;Minson et al, 2020). With the higher secondary threshold, we successfully maximize the detection of M5+ events while at the same time minimizing the detection of M<5 events, many of which would be so small that timely detections would not be possible (i.e., the IMMI 4+ energy would have passed prior to the event detection).…”

“…Numerous alerting methods exist, including station-based, grid-based and spatial configurations of different sizes (Minson et al, 2020). We do not implement alert region methods in this work.…”

“…For most stations in the ShakeAlert network, the distance to its nearest neighbor is less than 45 km (based on the mean plus two standard deviations, Supplemental Figure S1). If we assume that significant shaking is carried by the S-wave, and that S-wave speeds are ~3 km/s (Minson et al, 2020), then it should take no longer than 15 seconds for the initial trigger from one station to be confirmed by a neighboring station. Although more detailed calculations could be made using the exact locations of the triggering stations, we found that was not necessary and that these approximations successfully capture suspect station pairs.…”

“…For simplicity, we use a binary metric (i.e., timely or not timely), assuming all potential warning times exceeding zero are timely warnings (Table 3). This is because warning times can vary significantly depending on the final alerting strategy employed (Minson et al, 2020).…”

“…as well as the IMMI 4+ propagation speed derived byMinson et al (2020) for the majority of the IMMI 4+ areas in the 2019 M6.4 and M7.1 Ridgecrest, California, earthquakes. To determine which event detections are timely and which are not we use the following steps.…”