Modelling fire following earthquake in Wellington: a review of globally available methodologies

(Inc. GST)
(Ex. GST)
Write a Review

Scheele, F.R.; Horspool, N.A. 2018 Modelling fire following earthquake in Wellington : a review of globally available methodologies. Lower Hutt, N.Z.: GNS Science. GNS Science report 2017/42. 23 p.; doi: 10.21420/G27S&V

Abstract: Fires are a common secondary hazard following earthquakes, and can result in extensive damage if they spread over a large area. As part of the It’s Our Fault research programme, fire following earthquake (FFE) modelling is being updated for Wellington City, building on earlier work undertaken over a decade ago. This report presents a literature review of globally available methodologies for modelling FFE, focused on relevance to the New Zealand context and the case study of Wellington City. Since the early 2000s, there has been a significant increase in published literature on FFE modelling following earthquake events in Northridge, USA (1994) and Kobe, Japan (1995), and as computational power increased. Models are usually categorised as ignition models or fire spread models. Ignition models estimate the number and locations of ignitions based on the characteristics of an earthquake event or the attributes of an area of interest. Up until circa 2008, most ignition models estimated ignitions based only on measures of earthquake intensity. More recent models incorporate additional factors (e.g. building damage, population density) to more accurately estimate ignitions. Post-earthquake fire spread models estimate the area burnt, considering the built environment and the ignition locations. Recent FFE spread models are mostly simulations, many of which utilise GIS and/or physics-based fire spread equations. Spread models may incorporate various factors such as building attributes (e.g. combustibility, building separation, height), weather (wind speed and direction, rain), vegetation, slope, and fire suppression. As an extension to ignition and fire spread modelling, additional aspects of FFE such as risk indices and human impact assessments have been explored by some studies. Previous modelling for Wellington City has focused primarily on fire spread and loss estimation. Increasing resilience to damaging FFE events requires identification of areas of relatively high risk, which may be the focus of further modelling or risk mitigation activities by authorities. Fire spread modelling will need to be able to estimate the effect of mitigation strategies, as well as be useful for scenario exercising. Several recently developed models incorporate many important factors and approaches that are relevant for the Wellington City context, and demonstrate the increased potential for the usefulness of FFE modelling. (auth)