Distributed models coupling soakaways, urban drainage and groundwater

Alternative methods for stormwater management in urban areas, also called Water Sensitive Urban Design (WSUD) methods, have become increasingly important for the mitigation of urban stormwater management problems such as high runoff volumes, combined sewage overflows, poor water quality in receiving waters, urban flooding etc. WSUD structures are generally small, decentralized systems intended to manage stormwater near the source. Many of these alternative techniques are based on infiltration which can affect both the urban sewer system and urban groundwater levels if widely implemented. In order to assess these effects at local- and catchment-scale, there is a need for reliable and efficient modeling tools that can account for the interaction between the various urban water systems involved.
This thesis focuses on small-scale stormwater infiltration structures, often called soakaways, and how these can be modeled in an integrated environment with distributed urban drainage and groundwater flow models. The thesis: 1. Identifies appropriate models of soakaways for use in an integrated and distributed urban water and groundwater modeling system
2. Develops a modeling concept that is able to manage the bi-directional interaction between stormwater infiltration and groundwater 3. Develops suitable upscaling/downscaling techniques for the integrated soakaway model
4. Assesses the effects of extensive use of soakaways on sewer and groundwater flows in case studies Based on a review of the literature and on modeling studies, a new modeling concept is proposed which fulfills the need for integrated models coupling distributed urban drainage with groundwater. The suggested solution consists of a base equation for soakaway infiltration and additional components for clogging, upscaling/aggregation and groundwater interaction. The soakaway infiltration model consists of a mass balance equation for the soakaway with a depth dependent term for outflow that is based on the Darcy equation. Clogging is accounted for by modifying the hydraulic conductivity to account for low conductivity sediments deposited in the bottom of the soakaway. Upscaling considers variation in soil properties and it is shown that an upscaled geometric mean conductivity best matches a spatially variable model. Finally, the decrease in soakaway infiltration due to groundwater table rise is accounted for using an analytical expression for the local mounding under the soakaway.
These components can be combined to create a model that is best suited to the desired application. The model is tested on a case study in Harrestrup Å, Copenhagen where it is shown that in areas with high groundwater tables and low permeability soils, soakaways provide only a modest reduction of peak stormwater loads to the sewer system. The case study shows that soakaways will work best in areas with deeper groundwater and more permeable soils.
The soakaway modeling system developed in this thesis provides a simple, but complete description of soakaway behavior. The next step is to include it in commercial software and benchmark it for a broad range of case studies.