HR: 1340h
AN: S53B-1095    [Abstracts]
TI: Impact of non-Homogenous Spatial Seismicity Distributions on Probabilistic Seismic Hazard
AU: Beauval, C
EM: beauval@geo.uni-potsdam.de
AF: Potsdam University, Institut fur Geowissenschaften Postfach 60 15 53, Potsdam, 14415 Germany
AU: * Hainzl, S
EM: hainzl@geo.uni-potsdam.de
AF: Potsdam University, Institut fur Geowissenschaften Postfach 60 15 53, Potsdam, 14415 Germany
AU: Scherbaum, F
EM: fs@geo.uni-potsdam.de
AF: Potsdam University, Institut fur Geowissenschaften Postfach 60 15 53, Potsdam, 14415 Germany
AB: For the estimation of probabilistic seismic hazard at a site, seismic sources must be characterised through the determination of the locations and magnitudes of potentially destructive earthquakes. In low seismicity regions, such as France or Germany, the seismicity is rather diffused and significant magnitudes are rare; both the identification of faults and the modelling of the magnitude recurrence bear high uncertainties. We focus here on the uncertainties due to the lack of knowledge on the spatial distribution. Indeed, although seismic clustering in space is now well-established, in most probabilistic hazard studies, seismic sources used are large aerial zones where seismic rates are assumed to be uniformly distributed. Based on the fractal properties of seismic occurrences in space, we quantify the uncertainty on the probabilistic hazard due to the uniform assumption, for a site located inside the source zone. We use the fractal D-value to measure the degree of clustering of earthquakes and we quantify the uncertainty on hazard as the normalised difference (in %) of hazard values with respect to the uniform value. If no information on the D-value of the source zone is available, our results show that, for a return period of 475yrs, the uncertainty on the probabilistic hazard value can be as high as 60% for the 85% percentile. Moreover, we show that the higher the "true" D-value is (i.e. the lower the clustering of seismicity is), the lower is the uncertainty on hazard due to the uniform hypothesis. Thus, if minimum and maximum bounds for D can be estimated from the seismic catalogs, the uncertainties on probabilistic hazard due to the uniform assumption can be also bounded accordingly.
DE: 7212 Earthquake ground motions and engineering seismology
DE: 7223 Earthquake interaction, forecasting, and prediction (1217, 1242)
DE: 7230 Seismicity and tectonics (1207, 1217, 1240, 1242)
SC: Seismology [S]
MN: Fall Meeting 2005