Aside from non-climatic events such as tsunamis, extremes in sea level (i.e., coastal flooding, storm surge, high water events, etc.) tend to be caused by large storms, especially when they occur at times of high tide. However, any low-pressure system offshore with associated high winds can cause a coastal flooding event depending on the duration and direction of the winds. Evaluation of changes in frequency and intensity of storms have been treated in Sections 2.6.3 and 2.6.4, as well as SREX Chapter 3 (Section 3.5.2). The main conclusions from both are that there is low confidence of any trend or long term change in tropical or extratropic storm frequency or intensity in any ocean basin, although there is robust evidence for an increase in the most intense tropical cyclones in the North Atlantic basin since the 1970s. The magnitude and frequency of extreme events can still increase without a change in storm intensity, however, if the mean water level is also increasing. AR4 concluded that the highest water levels have been increasing since the 1950s in most regions of the world, caused mainly by increasing mean sea level. Studies published since AR4 continue to support this conclusion, although higher regional extremes are also caused by large interannual and multi-decadal variations in sea level associated with climate fluctuations such as ENSO, the North Atlantic Oscillation and the Atlantic Multi-decadal Oscillation, among others (e.g., Abeysirigunawardena and Walker, 2008; Haigh et al., 2010; Menéndez and Woodworth, 2010; Park et al., 2011).
Global analyses of the changes in extreme sea level are limited, and most reports are based on analysis of regional data (see Lowe et al., 2010 for a review). Estimates of changes in extremes rely either on the analysis of local tide gauge data, or on multi-decadal hindcasts of a dynamical model (WASA-Group, 1998). Most analyses have focused on specific regions and find that extreme values have been increasing since the 1950s, using various statistical measures such as annual maximum surge, annual maximum surge-at-high-water, monthly mean high water level, changes in number of high storm surge events, or changes in 99th percentile events (e.g., Church et al., 2006; D’Onofrio et al., 2008; Marcos et al., 2009; Haigh et al., 2010; Letetrel et al., 2010; Tsimplis and Shaw, 2010; Vilibic and Sepic, 2010; Grinsted et al., 2012). A global analysis of tide gauge records has been performed for data from the 1970s onwards when the global data sampling has been robust and finds that the magnitude of extreme sea level events has increased in all regions studied since that time (Woodworth and Blackman, 2004; Menéndez and Woodworth, 2010; Woodworth et al., 2011).
The height of a 50-year flood event has increased anywhere from 2 to more than 10 cm per decade since 1970 (Figure 3.15a), although some areas have seen a negative rate because vertical land motion is much larger than the rate of mean sea level rise. However, when the annual median height at each gauge is removed to reduce the effect of local mean sea level rise, interannual and decadal fluctuations, and vertical land motion, the rate of extreme sea level change drops in 49% of the gauges to below significance (Figure 3.15b), while at 45% it fell to less than 5 mm yr –1 . Only 6% of tide gauge records evaluated had a change in the amplitude of more than 5 mm yr –1 after removing mean sea level variations, mainly in the southeast United States, the western Pacific, Southeast Asia and a few locations in Northern Europe. The higher rates in the southeastern United States have been linked to larger storm surge events unconnected to global sea level rise (Grinsted et al., 2012).