Collapse Limit State Fragilities of Wood-Framed Residences from Storm Surge and Waves during Hurricane Ike

Abstract

Posthurricane Ike surveys for almost 2,000 individual wood-framed houses were combined with high resolution numerical hindcasts to investigate the collapse limit state arising from overland waves and surge. For this storm, freeboard above wave crests, wave height, current velocity, and construction date were found to be the most important factors determining survival. Regressions show considerable skill in predicting collapse fragilities and will prove useful in assessing vulnerability. The relatively small range of wind speeds available in the data set did not allow accurate assessments of wind effects on collapse. Several ASCE and FEMA force estimates showed relatively low skill in modeling fragility, in large part because their breaking wave assumption was not accurate over much of the range. However, the use of best-estimate wave heights in force modeling greatly improved accuracy of fragility estimates. Despite the positive results of this study, numerous questions remain. It proved impossible to separate the effects of house age and construction methods. Although computed overland wave heights showed good correlation with fragility, comparison with the limited measured data available showed that wave heights were significantly overpredicted inland. Finally, more accurate standards for integrated and local loads on elevated houses will be necessary to perform detailed structural system-based analyses of fragility.