This thesis describes the development and the characteristics of the coastline along the Dutch coastal zone, managed by Hoogheemraadschap Hollands Noorderkwartier. The coastal zone is a rich environment with multiple stakeholders. These stakeholders use the coastal zone for a range of functions such as safety, nature, business, and recreation. Coastal dunes have been the first line of defense against the sea for many years. On decadal and intercentennial time scale, climate change (e.g. sea level rise) and human intervention (e.g. nourishment), affect the variability of the coastline in different ways. Reports have shown an increase of the mean sea level, which is expected to increase even after the year 2100. Bruun was one of the first researchers, who found a relation between sea level rise and shoreline recession. While Bruuns findings contain the fundamental adaption of the coastline due to sea level rise, the coastline remains a highly complex system. Morphological data has been collected yearly of the Dutch coast as part of the JarKus program. Meanwhile, the data collection and computational power have increased exponentially, while the computational cost has gone down. Combining the newly computational power with the extensive JarKus data set, provides new insight into the complex coastal system. The derived variables from the JarKus data set range from widths, gradients, volumes to heights. These derived variables are combined with nourishment into a high-dimensional data set. Clusters of comparable coastal profiles in the Hoogheemraadschap Hollands Noorderkwartier area, have been made using machine learning techniques. Machine learning techniques such as K-means and the Self-Organizing Map (SOM) contain an intelligence, with the capability of clustering similar high-dimensional objects or data, without knowing the desired output. While both methods have the same goal, K-means moves its centroids inside the data and the SOM moves the data to its centroids (BMU). The advantage of combining both methods, is to keep the topological preservation while having the 'hard' clustering advantage of the K-means, which provides easy interpretation and therefore computations. The coastline of the Noorderkwartier can be broken up into nine clusters. Five of these clusters are classified as main-clusters, having a large number of transects. Four clusters are classified as sub-clusters, having just a few transects. Each of the clusters contains its own characteristic variables. Each of these characteristics originates from its own long-term natural and human-induced causes. The variable dominating the general clustering, is the active profile of the coastline. The lesser dominating variables are the foreshore nourishment, depth of closure and increase in foreshore volume. Meaning that the high-dimensional data set, find their similarities due to these dominant variables. Upon further investigation, the clusters containing the highest active profiles, were also the clusters containing historical larger nourishments. Comparing the yearly change of the standard deviation, shows that the clusters containing larger historical nourishment, have a shifting depth of closure. With respect to the dunes, correlations are found between the dune foot, y-coordinate of the boundary between marine and aeolian transport, dune volume and active width of the dune. While the exact reason for this correlation is still unknown, it shows potential for further research. The results of this research contribute to another step in understanding the complex coastal zone. Hoogheemraadschap Hollands Noorderkwartier can now adjust its policy and approach for each cluster based on the results of this research. For further research, it is now possible to focus on specific clusters with their unique characteristics. Lastly, results highlight the importance of the effect of nourishment on the active profile and with this the future dynamic equilibrium of the coastline.

Dunes are the primary sea defence along the Dutch coast. This research investigated the development between 1965 and 2021 of the beaches and dunes of the Hoogheemraadschap Hollands Noorderkwartier (HHNK) area which stretches from IJmuiden to northern Texel. This research used an advanced data analysis on the annual coastal elevation data. The coastline of the HHNK area is maintained by several different measures including nourishments, planting and removal of grass, placing reed fences and building regulations. The aim of this research was to relate the investigated developments of the beaches and dunes to these measures and natural processes. A literature study was performed first, to get acquainted with the research area and the processes that drive the beach and dune development in the area. In four phases, the developments and their relationships with the drivers were investigated for several subsections of the research area that showed similar behaviour. In phase 1, the coastal profile data was collected for multiple transects spanning the research area. Characteristic parameters describing features of the dune and beach were derived from the coastal profile data. In phase 2, the collected profile data was decomposed into spatial and temporal patterns by a principal component analysis. In phase 3, the transects of the research area were categorised on the temporal development of their coastal profile using the results of the spatial and temporal decomposition. This created subsections of the coast that showed a similar morphological behaviour. In phase 4, the morphological behaviour of several subsections was investigated and related to the natural and human drivers. The relationships between the drivers and the developments were investigated for several subsections, varying in morphological behaviour and drivers, to gain insight into these relationships for the entire research area. The coastal categorisation resulted in 36 subsections which were mainly continuous in space with the exception of four clusters. Seven subsections were studied in more detail, by investigating the development of several characteristic parameters derived from the coastal profiles, like dune volume, beach slope and shoreline location among others. The shoreline location and beach width showed a strong correlation with the nourishments. The beach slope and width were expected to influence the dune volume changes, which was not supported by the results. The beach width is assumed to be larger than the critical fetch length and the variations in beach slope had a marginal effect on the transport capacity. The presence of beach pavilions limited the dune growth in both height and volume, but regularly moving the pavilions did allow for a seaward migration of the dune front. The effect of small beach houses on the dune volume change was found negligible. Maintenance works of HHNK influenced the dune development locally. The placement of reed fences caused seaward migration of the dune front and the creation of blow-outs increased the dune volume behind the most seaward dune. The strongest dune volume increase was found at coastal areas behind shoals that reduced the incoming wave energy. The coastal categorisation succeeded in grouping transects with a similar development of their modified coastal profiles. Several modifications were proposed to improve the categorisation of the original profiles and to take into account smaller scale processes. Nourishment was distinguished as most important driver for the development of the coastal area, which has maintained the shoreline location and caused a trend break in several other characteristic parameters. Human influences like buildings, coastal structures and reed fences have a strong local effect on the development of the beaches and dunes. Further research could take into account wave and wind climate, water level variations and grain size quantitatively to get a more comprehensive study of the relationships between the development and the drivers.