Spatial Analysis of Damage Distribution in the 2001 Southern Peru Earthquake

Damage distribution surveys have long been used as a means for estimating relative ground shaking intensity. In the absence of dense instrumental arrays, these surveys can be used to identify and delineate geotechnical earthquake hazard zones. The traditional means of analyzing spatially distributed damage survey data is qualitative or semi-quantitative in nature. While this may be a satisfactory approach in situations where there is a marked contrast in building performance, its effectiveness is limited when damage contrasts are subtle, and spatial patterns are less obvious. In this paper we present a quantitative approach [Getis-Ord (Gi*) statistic] for interpreting and analyzing earthquake damage surveys. By implementing neighborhood-scale spatial analysis within a Geographical Information Systems (GIS) framework, statistically significant concentrations of elevated (and/or low) damage concentrations can be identified and quantified. An example application of this method is applied to a damage survey from the 2001 Southern Peru Earthquake (M_w8.4). In contrast to preliminary studies that suggested damage concentrations were a consequence of topographic amplification, the spatial analyses demonstrated that hotspots of damage are generally correlated with flatter ground in the study area.