Integration of models for better quantification of urban hydrology: Coupling urban drainage system models and distributed groundwater models

There is a growing need to simulate and predict hydrological system responses in urban environments as infrastructure becomes increasingly threatened by the effects of climate change (e.g. more frequent and severe flooding). Urban hydrology can cause flooding by drainage systems reaching maximum capacity, precipitation exceeding soil infiltration capacity, groundwater rising above surface, or a combination of these processes. The high complexity of urban hydrology is strained by the interaction between natural hydrological systems with precipitation, overland flow, recharge, groundwater flow, and drainage systems with routing of water from paved areas toward finite drainage networks. In order to asses flood risks in the urban environment, a modelling system integrating the different interacting parts in the urban hydrology should be applied. In the presented study a well-known drainage water and sewage water model, MIKE URBAN based on MOUSE is coupled with a spatially distributed groundwater - surface water model including vadoze zone processes. The primary objective for the study is to improve subsurface urban hydrological predictability because previous studies have illustrated the incapability of a stand-alone groundwater model to simulate measured groundwater head variability beneath an urbanized area. The investigated urban area is located in the western part of Denmark where urban infrastructure is vulnerable against high groundwater levels. Groundwater head predictions benefit from the coupled modelling approach where paved urban zones (with drainage runoff) and green areas (with more naturally occurring recharge) are separated, thereby affecting dynamic groundwater levels.