Automatic Extraction of Ridge Structures from Digital Elevation Models for the modelling software D-HYDRO
Automatic Extraction of Ridge Structures from Digital Elevation Models for the modelling software D-HYDRO
By Thirza van Noppen with supervisor: Ruud van der Ent
In order to avoid substantial damage as a result of flooding and extreme water discharges, it is of considerable importance to model flood events as it can help in flood risk reduction and mitigation. Numerous hydrodynamic simulation models have been designed for the purpose of modelling the movement of water and are widely used to assess flooding risk. The simplest and most common practice in these models is to use one-dimensional (1D) models that treat flow one-dimensionally along the river channel. Alternatively, two-dimensional (2D) models can provide more detailed results but they remain computationally demanding and data intensive. Recent advances and software developments resulted in novel mechanisms that can reduce computation costs.
One of these mechanisms is the incorporation of fixed weirs in a hydrodynamic model. Fixed weirs represent abrupt changes in altitude that have an impact on the local flow during flooding. The fixed weirs can be incorporated into modelling software such as 3Di, HEC-RAS and D-HYDRO Suite. The weir elements are aligned on the 2D-grid network, hence the resolution of the model can be kept large while sudden changes in depth are taken into account. Until now, fixed weirs are still drawn manually or with the aid of existing data of roads or railways, which is a labor-intensive and time consuming process. In addition, the process of manually drawing fixed weirs is a subjective process and is therefore largely dependent on the interpretation of the modeler.
The main objective of this study is to develop a tool that can automatically detect ridge structures based on a digital elevation model (DEM). Subsequently, the results of the tool will be evaluated in a D-HYDRO model developed for the river Roer (which is located in the province Limburg in The Netherlands). In order to do so, three approaches will be compared: a model with fixed weirs detected by the tool, a model containing no fixed weirs, and one with manually drawn fixed weirs.