Capturing detailed human dependencies on the built environment for enriched natural-hazard risk assessments

Current natural-hazard risk assessment tools fall short in capturing the complex dependencies that individuals have on the built environment. This limitation undermines their efficacy in supporting people-centred equity-aware (i.e., justice-oriented) decision-making processes related to risk management that account for the specific needs of different individuals or groups of people. Addressing this shortcoming, we propose a novel simulation-based framework for characterising relationships between people and the built environment that can be directly integrated into a conventional natural-hazard risk model. The proposed framework consists of three modules: (1) Population Synthesis, which statistically constructs a set of hypothetical individuals and households designed to reproduce distributions of important attributes across a real population; (2) Building Data Fusion, which spatially merges building-related datasets to obtain buildings’ physical and detailed occupancy characteristics; and (3) Population Spatialisation, which involves probabilistically linking household and individual socio-demographics to the physical vulnerability characteristics of buildings that people depend on for services/activities. The outcome of the framework is one (or more) synthetic population(s) with detailed asset-level spatial dependencies on the built environment, which can be used as part of a nuanced risk quantification process to determine how different individuals are affected by natural-hazard events. The approach is evaluated for Christchurch, New Zealand, considering possible (future) flooding scenarios and associated disruptions in people’s access to housing, employment, education, healthcare, and food services. The application demonstrates the added value that the proposed enriched risk modelling approach can bring to justice-oriented risk management decision-making, relative to conventional methods.