“All models are wrong. Some models are useful.”
George Box, statistics pioneer1
Researchers interested in various natural hazards have spent a lot of time trying to understand the probabilities of events. Although the specific approaches vary for different hazards, the key result is that the biggest and potentially most destructive events are the rarest. As a result, their probabilities are the hardest to estimate from the limited historical records available. Doing this involves a lot of uncertainty, which we need to keep in mind. Let's explore these issues, building on the basic ideas about probability introduced in Chapter 4.
Figure 8.1 shows an example for floods of the Red River of the North ...