Modeling and Estimating Post-Earthquake Fire Spread

Abstract

This paper describes the development of a GIS-based dynamic fire-spread model, with seven distinct modes of fire spread: direct contact, spontaneous ignition of claddings, piloted ignition of claddings, spontaneous ignition through windows, piloted ignition through broken windows, fire spread via non-fire-rated roofs and branding. All except the first two modes include in-built probabilities, but these can be selected individually and given user-defined values. Fire spread modes can be added to the model or altered to suit available building information. Critical details of buildings are obtained from an existing-buildings database, street surveys, or deduced using conditional probabilities from available data. Results show that comparison with actual fires is reasonable. The model could be extended with further development for use as a real time firefighting tool.

Estimating fire following earthquake risk for Wellington City, New Zealand

“…Finally, validation is a problem for all FFE modelling due to relatively few events having occurred in recent decades. The original fire spread models that this study are based upon have been validated against the 1931 Napier and 1995 Kobe events with good results [16]. However, the required data for validation of the models used in the current study were not readily available.…”

“…The critical separation distances signify the maximum distance a fire can spread given the wind conditions and rules to account for physical fire behaviour such as direct contact, heat radiation and branding. The values were derived from loss and wind speed data as reported from historical US earthquakes by Scawthorn [37], and the method is described in Thomas et al [16].…”