Turbulent Times – Development of an Accessible Decision-Support Methodology to Aid Uncertain Flood-Based Decision-Making

Abstract

Assessing the possible future behaviour of river systems is complicated by deep climatic and socio-environmental uncertainties. However, traditional flood protection strategies have generally ignored such uncertainties, preferring to implement extreme "hard engineering" solutions. The perceived inadequacy of these strategies has resulted in the development of decision-support methods which seek to quantitatively investigate the suitability of different interventions using flood impact assessments that take into account different uncertainty scenarios. The success of decision-support methods overseas and in New Zealand has led to the Ministry for the Environment recommending the Dynamic Adaptive Policy Pathways approach for national implementation in coastal hazard management. A similar recommendation for the widespread use of Dynamic Adaptive Policy Pathways to address river management problems is expected in the coming years. However, it is unclear whether widespread uptake of decision-support methods is feasible for New Zealand local government given the national models and software currently available to councils. In addition, communicating the outcomes of such modelling can be difficult. Serious games based on water management have proved successful in engaging stakeholders with deep uncertainty problems, but have been so far underutilised in the New Zealand context, possibly due to difficulties in calibrating them to specific management problems. This thesis therefore seeks to develop a Dynamic Adaptive Policy Pathways implementation methodology for the Lower Whanganui River flood management problem using only limited resources and readily available software. In particular, the methodology addresses how climate uncertainties might be accounted for in flood impact assessments, how asset databases might be constructed to assess vulnerability and how possible interventions might be compared within a developed impact model. A serious game session based on the impact model is then carried out to gain insights into the model and its utility for local government stakeholders. This thesis finds that carrying out decision-support investigations and creating serious games for New Zealand flood management problems is possible, but can be complex, time consuming and provide limited accuracy. It therefore concludes that widespread implementation of these methods is unlikely until national models have been improved and the process streamlined.