CALIFORNIA'S EARTHQUAKE ADVISORY PLAN: UNIQUE EXPERIENCE ISSUING UNCERTAIN EARTHQUAKE FORECASTS

Statistical studies in several active seismic areas worldwide show that a significant proportion of large earthquakes are preceded, within several days and a 10-50 km radius, by foreshocks as much as three magnitude units smaller. Since 1985 this principle has motivated a policy in California whereby certain M≥5 earthquakes are evaluated by the California Earthquake Prediction Evaluation Council (CEPEC). The Council may recommend to the state emergency management agency that local emergency managers, and the public, be informed of a temporarily elevated short-term (3-5 days) chance of a dangerous earthquake. Such advisories have been issued eight times since 1985.

Research by Lucy Jones in 1985 (and later with Duncan Agnew and Paul Reasenberg) found the probability that a southern California earthquake of M=5 would be followed in 3 to 5 days by another M≥5 event to be 6.5% (plus or minus 2.5%). Though low and uncertain, this probability is orders of magnitude above background. Based on this work, four earthquakes in California prompted advisories between 1985 and 1989. In 1990, the practice was formalized in the "California Earthquake Advisory Plan", which remains in effect. Developed by consultants for the Governor’s Office of Emergency Services, the Plan specifies actions by the California Emergency Management Agency during an advisory, including lists of agencies to notify, wording for a news release, checklists of mitigation actions, and instructions for ending an advisory.

Advisories are issued when earthquake probability is temporarily magnified, but remains low in an absolute sense. It is therefore unsurprising that no advisories have been followed by mainshocks in the specified space-time-magnitude window, since statistics show that only one in every 20 candidate foreshocks actually is a foreshock. In two cases - the 1989 Loma Prieta and 1992 Landers earthquakes - foreshock-based advisories issued beforehand had expired prior to the mainshocks. For several M≥5 earthquakes, CEPEC has convened but not recommended an advisory, and no mainshocks followed any of these events.

Whether the advisories promote public safety remains unstudied, but California’s earthquake advisory experience is a unique precedent that should inform planning for possible earthquake forecasts elsewhere in the United States.