Multi-scenario approach for liquefaction exposure assessments using a geospatial liquefaction model

Abstract

Events like the 2010–2011 Canterbury Earthquake Sequence demonstrated the impacts of liquefaction and lateral spreading across New Zealand infrastructure networks. Statistical liquefaction models based on geospatial data can be used to rapidly calculate the probability of liquefaction (manifestation) for large-scale networks. Using the simulated ground shaking intensity of 478 earthquake scenarios, the liquefaction probability is estimated across the New Zealand State Highway network. The overall exposure is measured by the number of events (NoE) that are expected to cause liquefaction manifestation along a road section. Compared to the assessment of a specific earthquake scenario or a specific return period, the multi-scenario approach helps to identify State Highways that might be affected by liquefaction manifestation during a range of different earthquakes, therefore, might lead to a recurrent disruption of transport services. The results are presented in a hazard map, highlighting State Highway sections with increased NoE, indicating that they might repeatedly experience liquefaction manifestation during seismic events. Limitations arise from the fact that the geospatial model does not account for subsurface soil characteristics and that vulnerability aspects of the State Highways are not considered in the assessment. Future research should also investigate combining the exposure results with indicators for network criticality (e.g., number of vehicles, or freight value) to better quantify the impact of liquefaction manifestation. The framework is adaptable to other infrastructure networks and can be used to support decision making processes regarding hazard mitigation or preparedness for future earthquakes.