Communicating Flood Risk Uncertainty for Decision-Making in Aotearoa-New Zealand

Flood risk uncertainty is a growing problem in New Zealand and the rest of the world. Decision-makers are facing increasing uncertainty in planning for future events. Growing population centres, increased cost of living and the resulting increased exposure to these natural hazards are just some of factors they need to consider in planning and mitigating future events. Climate change predictions represent a large part of the uncertainty present in these future flood risk assessments. Variables such as rainfall intensity and duration are likely to change significantly with increased temperatures which would result in potentially larger and more frequent flood events. To better understand how these different uncertainties could influence decision-making, a series of flood model and risk assessment output representations containing uncertainty were generated from a Monte Carlo framework. These representations were tested using an online survey and focus groups across regional councils, national response agencies and private companies that work with flood information. The results showed that traditional flood outputs such as depth and extent were still rated more useful than uncertain outputs such as confidence and exceedance probabilities. Larger AEPs (annual exceedance probabilities) such as 0.5% and 0.1% were seen as useful for long-term development planning but lower AEPs such as 1% and 5% were better suited for mitigation and emergency response plans. Across all the uncertainty outputs, respondents stressed the need for additional contextual information such as socio-economic overlays, area specific information such as land use and building types that would work in tandem with rebuild cost estimates and building damage data. From this feedback, a series of recommendations for presenting flood uncertainty information to decision-makers were created.