Using Novel Earthquake Early Warning (EEW) with Optimized Sensor Model to Determine How Establishments Will Be Affected in a 7.0 Hayward Earthquake Scenario

Abstract:

The Bay Area is one of the world’s most vulnerable places to earthquakes, and being ready is vital to survival. The purpose of this study was to determine the distribution of places affected in a 7.0 Hayward Earthquake and the effectiveness of earthquake early warning (EEW) in this scenario. We manipulated three variables: the location of the epicenter, the station placement, and algorithm used for early warning. To compute the blind zone and warning times, we calculated the P and S wave velocities by using data from the Northern California Earthquake Catalog and the radius of the blind zone using appropriate statistical models. We came up with a linear regression model directly relating warning time and distance from the epicenter. We used Google Earth to plot three hypothetical epicenters on the Hayward Fault and determine which establishments would be affected. By varying the locations, the blind zones and warning times changed. As the radius from the epicenter increased, the warning times also increased. The intensity decreased as the distance from the epicenter grew. We determined which cities were most vulnerable. We came up with a list of cities and their predicted warning times in this hypothetical scenario. For example, for the epicenter in northern Hayward, the cities at most risk were San Pablo, Richmond, and surrounding cities, while the cities at least risk were Gilroy, Modesto, Lincoln, and other cities within that radius. To find optimal station placement, we chose two cities with stations placed variable distances apart from each other. There was more variability in scattered stations than dense stations, suggesting stations placed closer together are more effective since they provide precise warnings. We compared the algorithms ElarmS, which is currently used in the California Integrated Seismic Network (CISN) and Onsite, which is a single-sensor approach that uses one to two stations, by calculating the blind zone and warning times for each. Onsite was better issuing quick warnings, but less reliable due to its one station dependency. ElarmS had more delay, but was more accurate since it used four stations. Knowing how the different variables affect EEW can provide citizens with an outlook of how to be prepared for a big earthquake in the Bay Area based on their location and proximity to faults and seismic stations.