Graduation of Soesja Brunink

When developing a Nature-based Solution, a stakeholder analysis is carried out to gain insight into the parties involved and what benefits it could have. Nature should be taken into account as a stakeholder, to analyse the possible benefits the solution could have for nature. Therefore, ecological assessment is necessary. This is still considered challenging due to the complexity of ecosystems, the dependency on many variables, the difference in spatial and temporal scales, the limitations in available information, and the non-linearity of nature. This thesis aims to develop a method to analyse the ecological development of an estuarine system when modifications are applied.

Ecology can be classified using ecotopes. Ecotopes are more or less homogeneous, identifiable by their similarities and differences in geo-morphological and hydrological characteristics. They are characterised by a vegetation structure linked to abiotic conditions and land use. A classification system for ecotopes, `A Dutch Ecotope System for Coastal Waters' (ZES.1), was developed by Rijkswaterstaat. It is a hierarchical system, with thresholds determining ecological categories.

In this thesis, a model was developed that maps ecotopes based on the ZES.1. It uses salinity, water-depth, and flow velocity from a hydrodynamic Delft3D-FM model. The data is labeled based on the salinity, whether the grid-point is always, never, or sometimes sub-merged, the flooding duration, the flooding frequency, flow velocity, and substrate composition. Ecotope labels are composed by combining these labels.

A validation study was done by applying the developed model to data of a Delft3D-FM model of the Western Scheldt. The resulting ecotope-map was compared per category with an ecotope-map of the Western Scheldt developed by Rijkswaterstaat, that is based on aerial photographs, laser altimetry, soundings, field measurements, and several models. The equality between these maps was 63 %. It was found that the majority of errors occurred near the thresholds. Therefore, it was chosen to perform an optimisation study on the thresholds. The optimisation study was done by setting threshold ranges and determining the performance for every combination within these ranges. New combinations of thresholds were selected based on the maximum performance. After optimisation, the average performance was 85 %.

The main advantage of the developed model is that aerial photographs and other real-time data are no longer necessary to obtain an ecotope-map. It can be applied in numerous hypothetical case studies. In addition, this reduces the costs of developing an ecotope-map significantly. There are many additional features possible, such as a translation between ecotopes and ecosystem services, which could contribute to the quantification of ecological development, which is one of the current major challenges in the development of Nature-based Solutions.