Using spatial modelling to develop flood risk and climate adaptation capacity metrics for assessing urban community and critical electricity infrastructure vulnerability

Using spatial modelling to develop flood risk and climate
adaptation capacity metrics for assessing urban community
and critical electricity infrastructure vulnerability

The aim of this study was to develop a new spatially-explicit analytical approach for urban flood risk assessment and generation of climate adaptation capacity metrics for vulnerability assessment of critical electricity infrastructure.
Using the January 2011 flood in Queensland (Australia) with the core suburbs of Brisbane City as the study area, this study addressed the sufficiency of indicating variables and their suitability for climate risk modelling. A range of geographical variables were analysed using a) high resolution digital elevation modelling and urban morphological characterisation with 3D analysis, b) spatial analysis with fuzzy logic, c) proximity analysis, d) quadrat analysis, e) collect events analysis, f) geospatial autocorrelation techniques with global Moran’s I and Anselin Local Moran's I, and g) hot spot analysis. The issue on the sufficiency of indicating variables was addressed using the topological cluster analysis of a 2-dimension self-organising neural network (SONN) structured with 100 neurons and trained by 200 epochs. Furthermore, the suitability of flood risk modeling was addressed by aggregating the indicating variables with weighted overlay and modified fuzzy gamma overlay operations using Bayesian joint conditional probability. Variable weights were assigned to address the limitations of normative (equal weights) and deductive (expert judgment) approaches.
The outputs of the topological cluster analysis showed that 15 out of 22 indicating variables were found sufficient to spatially model the flood risk and climate adaptation capacity metrics. The analyses showed that 214ha (9%) and 255ha (11%) of the study area were very highly impacted by the January 2011 flood as indicated by the very high flood risk metrics and the very low adaptation capacity metrics, respectively. In the electricity network vulnerability assessment, a total count of 72 critical assets (zone supply substations, high voltage switching sites, and pole transformer sites) were found highly vulnerable to flood hazard. The flood damage disrupted electricity supply along 627km and 212km of transmission lines on the north western and south eastern sides of the study area, respectively.
The newly developed spatially-explicit analytical technique, identified in this study as the flood risk-adaptation capacity index/metrics-adaptation strategies (FRACIAS) linkage model, will allow the integration of flood risk and climate adaptation assessments which have been treated separately in the past. As technical support to the Queensland Floods Commission of Inquiry (QFCI) recommendations, this study also provides a tool and identifies adaptation strategies to enable urban communities and the power industry to better prepare and mitigate future flood events.
The tool can also be used to assess the physical vulnerability of other critical assets (e.g. water supply, sewerage, communication, stormwater, roads and rails) to flooding.