A Case Study—Extreme Water Levels

As part of the efforts to reduce the vulnerability to flooding, it is of paramount importance to have available flooding charts that spell out the risk of flooding for a given location. This risk is typically expressed as the flooding levels associated with various return periods, e.g., the 100 year flooding level. In Norway the concerns about flooding events are primarily connected with two types of flooding: On the one hand, in rivers and lakes, which are mainly related to melting snow; and on the other, along the coast, which are mainly due to a combination of offshore storm surges and tides. In this chapter the attention is limited to coastal areas. Specifically, the extreme value statistics of sea levels measured at three stations along the Norwegian coastline will be investigated: Oslo, Heimsjø and Honningsvåg, see Fig. 12.1. This provides an excellent opportunity to compare the performance of the following four methods for extreme value estimation: The Annual Maxima (AM) method (Chap. 2 ), the Peaks-Over-Threshold (POT) method (Chap. 3 ), the ACER method (Chap. 5 ), and also the Revised Joint Probabilities (RJP) method (Tawn and Vassie, Extreme sea levels: The joint probabilities method revisited and revised. In Proceedings of the Institution of Civil Engineers, Part 2(87), 429–442; 1989, Tawn, Applied Statistics 41(1):77–93; 1992) (see also Batstone et al. (Calculating extreme sea level probabilities around complex coastlines: A best practice approach. In Proceedings Irish National Hydrology Conference, 24–32. Hydrology Ireland; 2009) for modifications of the RJP method). Since the RJP method has not been discussed previously in this book, it will be explained in some detail here. This chapter largely follows the work presented by Skjong et al. (Journal of Coastal Research 29(5):1029–1048; 2013).