Financial Quantification of Future Climate Extremes: Three Applications in Aotearoa New Zealand

As climate change accelerates, escalating extreme weather events and rising sea levels present increasingly material threats to infrastructure and real estate investments. Despite this, few tools exist to quantify the financial implications of future climate extremes on these long-lived assets. This thesis pioneers several novel methodologies that directly translate physical climate science outputs into inputs for quantifying financial risk on these assets. First, it combines Generalised Extreme Value (GEV) analysis with sea level rise projections and property-level data to estimate the timing and extent of future insurance retreat from coastal properties in Aotearoa, New Zealand. Second, an innovative atmospheric-hydrological modelling chain is developed to generate probability distributions that reliably reflect changes in climate extremes. This framework is then used to project how the Annual Exceedance Probabilities (AEPs) of major riverine floods evolve as global temperatures increase. The analysis demonstrates how different hydrological responses to climate change can either amplify or dissipate the atmospheric climate signal. Third, evolving probabilities of exceedance are deployed in the creation of two novel metrics: Loss of Design Life and its counterpart, Time to Design Failure. These metrics quantify the reduction in infrastructure design life resulting from climate change-induced increases in the frequency of extreme precipitation and discharge events. They are used to estimate the time until design failure of riverine infrastructure, such as bridges and levees, under climate change. By doing so, these metrics enable direct translation of physical climate science into the financial valuation of long-lived assets, including infrastructure and real estate. Fourth, the thesis investigates how evolving probabilities of exceedance due to sea level rise affect residential properties in coastal inundation zones, revealing a near-certain likelihood of damaging inundation within the next few decades, often after private insurers have retreated.