ASSUMED BACKGROUND SEISMICITY AS A PARTIAL EXPLANATION FOR DISCREPANCIES BETWEEN PRECARIOUSLY BALANCED ROCKS AND 2008 CALIFORNIA HAZARD MAPS

Precariously balanced rocks (PBRs) which have been in place thousands of years provide upper bounds on seismic hazard at a site. At a number of sites in California, tested measured and estimated toppling ground motions are inconsistent with 2008 2% in 50 yr hazard maps. The number and distribution of sites make it unlikely that the discrepancies are due to a statistical accident. Other possible explanations for the discrepancy include unrecognized site, path, and attenuation contributions to the hazard, or over-estimation of rates either of known faults or of random background seismicity. In this presentation we discuss the possibility of incorrect assignment of random background seismicity. In the recent UCERF 2 model of seismicity in California relatively large earthquakes, M=6.5-7, are randomly assigned to locations within 20 km, in many cases 5 km, of PBR sites. For the 2% in 50 yr (2475 yr recurrence time) maps there are several such events, at some sites accounting for half or more of the hazard and apparently making the hazard maps inconsistent with the PBRs. Random background earthquakes are supposed to represent the possibility of such large events occurring at sites where there are no mapped candidate faults, but where blind or buried structures could be active. However many PBR sites the large background events are assigned where exposures are good and there are no mapped faults nearby capable of such large earthquakes. Furthermore, most, if not all the “background” earthquakes cited in UCERF2 as evidence for the rate of occurrence of background earthquakes actually occurred in known active faulting areas. In our presentation we consider individual site discrepancies and possible explanations for each case, possible geologic criteria for realistically designating areas for potential for large background earthquakes, and statistical implications for earthquake hazard.