Search

A novel modeling methodology is presented for cascading disasters triggered by tsunami hazards considering uncertainties. The proposed methodology focuses on tsunami-triggered oil spills and subsequent fires, a type of natural hazard-triggered technological (Natech) event. The methodology numerically simulates the time-varying behavior of tsunami-triggered oil spill fires for numerous stochastically generated scenarios and performs a probabilistic mapping of the maximum radiative heat flux as a quantitative measure of the fire hazard. To enable these assessments, probabilistic tsunami hazard assessments are extended to include the tsunami-induced movement of oil storage tanks, resulting oil spills, tsunami-driven oil fire spread, and thermal radiation from fires. The uncertainty of the earthquake fault slip distribution, oil filling level of storage tanks, and fire starting time and position is incorporated into the new assessments. To demonstrate the methodology, a realistic case study is conducted for a coastal petrochemical industrial park in Japan conditioned on possible offshore moment magnitude 9.1 earthquakes. Contrary to typical tsunami direct impact assessments, the results highlight the cascading effects of tsunamis and large variability in key output variables concerning oil spills and fires. This indicates that the methodology is useful for deepening stakeholders’ understanding of tsunami-triggered cascading disasters and improving risk reduction plans.

The spatial-temporal characteristics of wildfires were summarized based on fire reports covering 55, 863 wildfires with a total burned area of 289.91 km² from 1995 to 2020 in Japan. The long-term decreasing trend was largely influenced by changes in social aspects, while the annual fluctuation was determined by large-scale events. The wildfire season started around early January in the southern part and gradually shifted northward over approximately three months. A total of 98.77% of cases were human-induced, and 1.23% were caused by lightning. Regarding response time after receiving the call, firefighting was started within 30 min in 89% of cases and controlled within 1 h in 85% of cases, where spatial characteristics were insignificant. Full-time fire departments utilized artificial water sources primarily in 72% of cases and natural water sources in 28%, while the rate using natural water sources increased as the burned area increased. Natural water sources from ponds and rivers exhibited regionalities. Cluster analysis revealed five wildfire regions reflecting natural and anthropogenic factors. Furthermore, the Mann–Kendall trend test concluded that the initiation process decreased almost uniformly across the country, while regional differences existed regarding the expansion process. Additionally, only human-induced wildfires had a significant decreasing trend.

Post-earthquake fire safety is an important functional requirement of seismic-resilient buildings and depends on seismic damage to fire protection systems. Probabilistic fire risk assessments are useful in ensuring safety and consider the post-earthquake reliability of fire protection systems. This study therefore investigated damage to fire protection systems due to the 2016 Kumamoto earthquake through a questionnaire survey among hospitals to collect data available for modeling the post-earthquake reliability. Ten types of component of a fire protection system were surveyed: emergency generators, fire detectors, water tanks used for firefighting, fire pumps, sprinkler heads, indoor fire hydrant boxes, fixed glass smoke curtains, fire doors, fire shutters, and fire exit signs. The data show that there was much damage to fire detectors, sprinkler heads, fixed glass smoke curtains, and fire doors. Zero-inflated Poisson regression analysis was conducted to derive statistical damage models that describe the association of building response quantities with the number of instances of damage. Results of a case study using the models show that building shaking exceeding a peak response acceleration of 10 m/s² may reduce the probability of successful functioning to below 40% for a sprinkler system and to below 90% for a smoke-activated door system.

Fires following earthquakes have significant impact on urban communities in earthquake-prone countries with wooden buildings, in addition to the main earthquake hazard of ground shaking. Regional-scale post-earthquake fire risk assessments are important for developing effective risk reduction strategies. Such risk assessments require the consideration of epistemic uncertainty, that is, uncertainty caused by lack of knowledge concerning the best models, in addition to aleatoric uncertainty. This study focused on epistemic uncertainty in post-earthquake fire ignition estimations, and incorporated it into risk assessments by using alternative empirical ignition models. The key objectives of this study were as follows: (1) correlating the probability of ignition per person to the ground motion intensity based on data for different past large earthquakes in Japan; (2) investigating the impact of using ignition models calibrated on different earthquake events in risk assessments through a realistic case study considering possible large earthquakes. A hierarchical Bayesian Poisson regression analysis revealed that the empirical relationship between the ignition probability and ground motion intensity greatly varies across six major large earthquakes in Japan from 1995 to 2022. Particularly, the effects of ignition prevention measures that have been widely implemented in households since the 1995 Kobe earthquake were inferred from much smaller ignition probabilities in approximately the last 10 years. The effects of this high epistemic uncertainty manifested as the large variability of the loss exceedance curve for a portfolio of buildings. These results indicate that the new ignition models can help to foster risk-informed decision-making under epistemic uncertainty.

A probabilistic methodology is presented for assessing cascading multi-hazard risk for ground shaking and post-earthquake fires at a regional scale. The proposed methodology focuses on direct economic losses to buildings caused by the combined effect of ground shaking and post-earthquake fires and evaluates the exceedance probability of the regional shaking–fire losses in a predefined future time period by comprehensively considering the effects of various uncertain factors on the losses via Monte Carlo simulations. Probabilistic seismic risk assessments are extended by integrating post-earthquake fire models with seismic activity models, ground motion prediction equations, and seismic fragility functions. The fire models include post-earthquake ignition models, a weather model, a physics-based urban fire spread model, and a fire brigade response model. This integrated modeling enables the incorporation of the following uncertain factors with causal relationships into the assessments: earthquake occurrence, ground motion intensity distribution, damage to buildings resulting from ground shaking, post-earthquake ignition occurrence and occupant firefighting, weather condition, fire brigade response time including time to detection, and damage to buildings resulting from post-earthquake urban fire spread. To demonstrate the methodology, a realistic case study is conducted for a historical urban area with closely spaced wooden buildings in Kyoto, Japan, focusing on possible large earthquakes along major active faults. Contrary to conventional single-hazard approaches, the results highlight the impact of multi-hazard consideration on risk assessments. This indicates that the methodology can be a useful tool for more appropriately understanding earthquake risk and promoting risk-informed decision-making in urban communities for risk reduction.