This thesis provides a contribution to the consideration of intra-annual to multi-decadal sea level variability in coastal enginnering concepts. Coastal structures are usually designed for a lifetime of several decades. Nowadays a sustainable design requires the consideration of long-term changes in the loading factors due to enhanced greenhouse gas emissions throughout the 21st century.
Hence, policy makers and managers need to know as early as possible which climate change pathway the Earth’s climate is following and by how far regional sea levels are changing to support adequate and timely adaption. This, however, is challenging since superimposed on any deterministic long-term trend in sea level there is a considerable fraction of intra-annual to decadal variability linked to climate internal processes. Such variability patterns may be as large as the secular change observed through the 20th century and persist over at least one decade, (i) hampering an early detection of long-term changes or accelerations and (ii) increasing/decreasing the risk of coastal flooding during these periods.
This thesis therefore investigates variability patterns in two of the most important loading factors for coastal infrastructure: mean sea level and storm surges. Different filtering techniques in combination with statistical regression models and physical theory are used to characterize the sea level variability in the North Sea (German Bight) over various time scales and to discover the contribution of different oceanographic and atmospheric forcing factors. After identifying the main contributors to the sea level variability, the long-term changes are reassessed over the past 140 years and the atmospherically induced variability patterns are projected over the ongoing century up to the target year of 2100. It is demonstrated that an improved understanding and the subsequent removal of interannual to decadal variability reduces the uncertainties when estimating long-term trends and allows for earlier detection of accelerations (up to 60 years depending on the considered location scenario). This in turn increases the statistical certainty about possible future states, which can be considered in the process of decision making for possible adaption strategies.