The rapid classification of building damage states or placards after an earthquake is vital for enabling an efficient emergency response and informed decision-making for rehabilitation and recovery purposes. Traditional methods rely heavily on inspector-led on-site surveys, which are often time-consuming, resource-intensive, and susceptible to human error. This study introduces a machine learning-supported surrogate model designed to streamline the assessment of building damage, focusing on the automated assignment of damage placards within the context of New Zealand's post-earthquake evaluation frameworks. The study evaluates two key safety evaluation protocols—Rapid Building Assessment (RBA) and Detailed Damage Evaluation (DDE)—and integrates corresponding databases derived from the 2010–2011 Canterbury Earthquake Sequence (CES) in Christchurch. Six ML classifiers—Multilayer Perceptron (MLP), Random Forest (RF), Support Vector Machine (SVM), K-Nearest Neighbours (KNN), Gradient Boosting Classifier (GBC), and Gradient Bagging (GBag)—were rigorously tested across both databases. The results indicate that the RF-based surrogate model outperforms the other classifiers across both RBA and DDE protocols. Two distinct sets of critical predictors have been further identified for each protocol, allowing for the rapid retrieval of essential data for future on-site surveys, while retaining the RF model's predictive accuracy. The developed surrogate model provides a pragmatic tool for practising engineers to rapidly assign placards to damaged structures and for policymakers and building owners to make informed recovery decisions for earthquake-affected buildings
Surface-rupturing earthquakes can trigger the sudden avulsion of river channels, causing rapid and persistent coseismic flooding of previously unaffected areas. This phenomenon, known as fault-rupture-induced river avulsion (FIRA), occurs when fault displacement significantly alters river channel topography. The importance of understanding FIRA as a secondary seismic hazard was highlighted by events during the 2010 Darfield and 2016 Kaikoura earthquakes in New Zealand. This thesis develops a national model to identify and quantify FIRA susceptibility across New Zealand by integrating hydrological datasets (NIWA RiverMaps and Flood Statistics) with active fault information (NZ Active Faults Database and RSQSim earthquake simulations). The methodology applies the F-index framework proposed by McEwan et al. (2023), which quantifies FIRA potential based on the ratio of fault throw plus discharge-dependent depth to bank full depth at each fault-river intersection. The model successfully identified 3,796 potential FIRA-susceptible fault-river intersections nationwide, with 451 involving waterways equal to or larger than the Hororata River. Regional analysis revealed higher concentrations of FIRA-susceptible sites in the Bay of Plenty, Canterbury, and Marlborough regions. Validation against historical events showed the model effectively located known FIRA occurrences from the Kaikoura and Darfield earthquakes, though with some limitations in accurately predicting F-index values due to complex fault displacement patterns and challenges in modelling bank full depths of large, braided rivers. This research establishes New Zealand's first nationwide assessment of fault-induced river avulsion susceptibility. The approach creates a structured methodology for identifying high-risk fault-river intersections and determining which sites require thorough localised examination. The methodology developed offers a template for similar assessments in other tectonically active regions and contributes to improving earthquake hazard assessment and disaster preparedness planning.
