Worldwide, flood and drought events cause severe disruption to essential services. Climate change, population growth and urbanisation are due to cause further infrastructure pressures. Traditionally, hydro-hazard risk is addressed with technical interventions which loosely couple social impacts and focus on either end of the hydrological cycle but not both concurrently. Such methods are now reaching the boundaries of their performance envelope – necessitating novel planning approaches that acknowledge complexity science. This work makes two contributions in this endeavor: the introduction of a sociotechnical systems method for urban planning, and the development of metrics to illustrate cascading effects of hazards throughout modelled cities.
The first contribution of this work is to characterise cities as complex socio-technical systems, using the ‘abstraction hierarchy’. This method originated in high-risk domains after deterministic methods failed to prevent major incidents. Rather than modeling a series of control loops, the hierarchy models what a city’s physical objects can afford, the city functions they contribute to, and the values and priority measures that track city performance. The result is a hierarchical network incorporating social and technical aspects of cities, as well as their complex interrelationships, across different levels of scale. Figure 1 shows an abstraction hierarchy excerpt for the case study town of Dumbarton, UK.
Abstraction hierarchies are information-rich, and the second contribution of this work is to provide a theoretically rooted (yet tractable) means of interpretation for decision-makers. By application of network centrality metrics, critical components at each level of scale are identified. Metrics are applied to a baseline model to inform resource allocation for everyday operations – but are also used to show the relative effects of a specified hazard. By degrading ‘source’ nodes where any hazard is introduced, cascading effects throughout the hierarchy are highlighted. As a demonstrative example, a 1:100 year flood is applied to the ‘physical objects’ level of a baseline Dumbarton model. While some results are unsurprising (e.g. increased importance of emergency response), subtler community stresses are also exposed (e.g. severe reduction in vehicle repair capabilities). This estimation of relative impacts enables the effective, dynamic prioritisation of resources for a wide range of scenarios and risks. Overall this work offers a flexible way to assess complex socio-technical systems, leaving behind the reductionism of existing approaches, and meeting an urgent practical need for urban planning.