Internship Colloquium ANNELOU HOOGERWERF

This research is based on a GIS-based overland flow model that indicates inundated areas and flow paths of overland flow after extreme rainfall. The model is improved, additional tools are added, the entire model is validated and automated during my internship at HydroLogic.

Organised by Hydrology and Quantitative Water Management

Wed 28 June 2017 10:30 to 11:00

Venue Lumen, gebouwnummer 100

Optimising a GIS-based urban overland flow model

Urban floods are receiving increasingly more attention and occur more frequent, due to climate change, urbanisation, and human actions. Water on the streets is considered inconvenient by municipalities, when buildings are flooded, roads are inaccessible or health is at risk. An overland flow model already exists to enable municipalities pinpoint possible vulnerable areas, however, to improve the model output this model required improvements, additional tools, validation and automatisation.

The overland flow model is a GIS based model, based on open source input data of elevation, buildings, urban border and waterbodies, and assumes no interaction with the sewage system. In three steps: It prepares the Digital Elevation Model (DEM), calculates the possible areas of water on the streets (WOS), computes the flow paths (streamlines) of the overland flow. The model is improved by adding a different interpolation method, more logical order of interpolation, alternative method of adding additional waterways, and additional streamlines. The model gained model additions that give an indication on the potentially damaged buildings and blocked roads. Also sub-catchments are  calculated, which led to a tool which fills the WOS stepwise based on a chosen design storm. A case study is performed in Albergen to compare the model results to actual inconvenience and other model results. In general the model performs well, but the case study is not a scientific validation, since detailed information is lacking. Finally, the model is made semi-automatic to increase workability and speed.

Although the model remains simple and based on many assumptions, it is able to act as a first tool for municipalities to indicate potentially flooded areas.