Insurers are sometimes considered to be boring, yet they are at the forefront of developing catastrophe models. As insurance companies, they put their money at risk by providing assurance against weather catastrophes. They have this huge incentive to model long-term climate trends and future weather catastrophes with some accuracy. You know the joke about the Chicago future change. The price of a future's contract on oranges is the best predictor of night frost in Florida where oranges are grown. The forecast of the future's market is even better than that of weather forecasters. Insurers use sophisticated catastrophe models to assess the economic losses from catastrophes. A catastrophe model consists of five components. First, the hazard component estimates the extent and intensity of the catastrophe. The catastrophe is the natural phenomenon or asset with the potential to cause damage or loss. It is expressed in hazard metric, such as peak wind speeds across a storm track. Second, the vulnerability component assesses the relative damage to the assets, like property, and their contents, and infrastructure, that are likely to be damaged by the hazards. These vulnerability models connects the hazard intensity with the estimated damage as a ratio of total value. Third, the exposure component is used in two steps in catastrophe models. In the first step, exposure data for the specific objectives are modeled. In the second step, a representative industry exposure for the region is also used for model. The outcome is an exposure value, which may be split between building values, contents values, and business interruption values. Fourth, the financial loss component translate the physical damage into total monetary loss. That's before the application of any insurance or reinsurance financial structures, it takes a damage to a building and its contents and estimates which party is responsible for paying. The results of the determination are then interpreted by the model user and applied to business decisions. Fifth and last, the platform component provides the software that implements and integrates the four model components. The platform usually also provides a structured way of inputting, validating, analyzing, visualizing, and storing the different data components, and running result reports. Catastrophe models are designed to quantify catastrophe risk, and important metric is the exceedence probability, which is the probability that the loss will exceed a certain amount in the year, only then the catastrophic insurance kicks in. The fundamental output from catastrophic models is a probability distribution of loss. Underlying drivers of weather catastrophes are global warming and related sea level rise. We use different pathways projected by the International Panel on Climate Change, the IPCC, which is the independent and academic body. The blue line reflects the most stringent carbon emission reduction scenario that aims to keep increases in global temperature to two degrees Celsius. The red line reflects the highest emission scenario. The lines present the mean of the distribution and the shaded area, the 5-95 percent confidence interval. The sea level rise in the stringent scenario, for example, has a wide range from 28-60 centimeters, with a mean of 44 centimeters in 2,100. Climate change has an impact on the frequency, severity, and geography of natural hazards. Insurance therefore, frequently update the catastrophe models to evaluate and manage catastrophe risks. The underlying climate risk show structural uncertainty and is long linear. We can illustrate the non-linearity and long-term nature of climate risk with the example of sea level rise. This is caused by thermal expansion due to warming of the ocean, since water expands as it warms, and there is increased melting of land-based ice, such as ice sheets and glaciers. The rising sea level has a major impact on coastal areas with a long lead time. Flood risks crow with sea level rise as it raises the likelihood of extreme sea levels. We give you the example of Oakland and Wellington, two coastal cities in the beautiful New Zealand. The sea level is projected to rise by about 30 centimeters between 2015 and 2065, which is the midpoint of the four IPCC scenarios for global mean sea level rise. For a 30 centimeter sea level rise, the frequency of current 100 year flood event, is expected to increase to every four years at the port of Oakland and every year at the port of Wellington. Remember, that extreme scenario is an 80 centimeter sea level rise, that increases the frequency to every week for Oakland and every tide during the day for Wellington. You probably cannot imagine such a frequency, but it is an actival mode of prediction of the extreme scenario. These numbers show very clearly that flooding risk is expected to increase at an exponential rate. They also show that we need to mitigate climate change to avoid extreme scenario, and we need to adapt to climate change with flood barriers to protect these cities. That is for another video. Thank you for listening.
