Assessment of low impact design (LID) strategies using integrated and distributed surface water/groundwater models

J.D.C. Kassenaar, M.A. Marchildon

In the proceedings of: GeoMontréal 2013: 66th Canadian Geotechnical Conference; 11th joint with IAH-CNC

Session: Groundwater-Surface Water Interactions III

ABSTRACT: Low-Impact Development (LID) strategies have recently received significant attention due to the requirement to reduce the adverse hydrologic and water quality effects of urbanization. Many proposed LID strategies are also designed to protect natural heritage features such as headwater stream channels and wetlands. Current surface water modelling tools that simulate LID functionality are limited in their ability to simulate rainfall runoff retention for the purposes of reducing of peak flow events; however many of simulated LID solutions involve the enhancement of groundwater infiltration that in reality may not be practical given the hydrologeologic regime. For example, the application of bioswales, infiltration galleries, and soakaway pits imply a capacity to recharge excess water; bioswales and similar green infrastructure must infiltrate their retained water effectively prior to subsequent rainfall events. Demonstrating the effectiveness of a LID design requires consideration of changes to the local terrain, depth to water table, infiltration capacity of the surficial materials and an overall understanding of the hydrogeologic linkage between the proposed LID feature and the nearby stream or wetland; therefore, a groundwater modelling solution is required The purpose of this paper is to illustrate the challenges and insights that Earthfx has encountered when simulating common LID strategies using the distributed USGS GSFLOW integrated model. Examples presented in this paper illustrate how GSFLOW can be used to represent infiltration enhancement facilities and bio-swales for the purposes of mitigating impacts to groundwater recharge and increased runoff due to urbanization. Modifications to the sub-cell model processes allow representation of smaller scale features such as green roofs and roof-to-lawn down-spout disconnection strategies. An assessment of the relative contributions of surface water and groundwater entering wetlands, riparian areas, storm water management facilities and bio-swales under multiple site specific future scenarios proved useful at the preliminary design level. Current and estimated future recharge rates, ground water discharge patterns, and runoff were used to compare various designs and overall LID effectiveness. In summary, the GSFLOW analysis provided a quantitative approach to compare and demonstrate the ecological linkage and benefit of various LID strategies with consideration of groundwater feedback mechanisms. Many of the challenges encountered in the analysis were related to the complexity and differences between the various LID measures. Both small scale on-site LID measures (i.e. roof leader diversion to backyards) as well as larger neighbourhood scale end-of-pipe measures (i.e. infiltration swales) needed to be represented. Each LID measure also affects a mix of hydrologic processes, including GW infiltration, evapotranspiration, and runoff detention. Finally, while the analysis needed to represent detailed designs and LID function, data gaps related to the ongoing neighborhood design meant that some development components, such as final lot sizing, were not available. These challenges were addressed by adding a new reservoir module to GSFLOW. The new module was configured to receive a specified amount of runoff from the impervious portion of the model cell. The reservoir has a defined storage volume and can lose water through multiple processes, including evapotranspiration, drainage and spill over into the soil zone. By varying these control parameters a variety of LID measures can be represented, from green roofs to parking lot groundwater infiltration galleries. Finally, the flexible GSFLOW cascade network algorithm was modified to represent lot level water capture and transfer to end-of-pipe LID measures. A series of simulations with the new GSFLOW LID reservoir module were completed to evaluate the effect of the LID measures. Scenarios include current conditions, development without LID measures, and multiple LID scenarios. Numerous comparisons were undertaken to evaluate the effectiveness of the LID measures. Evaluating the effectiveness of the recharge LID measures at preserving groundwater levels in the underlying aquifers was relatively straightforward. Comparing the drawdown in the underlying aquifers provided a simple means of assessing LID performance. Similarly, LID measures related to preservation of baseflow and groundwater discharge to streams could be evaluated by comparing changes in groundwater discharge volumes. Assessing the effectiveness of runoff-control LID measures was considerably more difficult. First, it was necessary to differentiate between generated runoff (i.e. the runoff generated by all processes within the model cell) and net runoff that reaches a stream. LID measures that simply detain runoff were difficult to assess on a water budget basis.

Please include this code when submitting a data update: GEO2013_695

Access this article:
Canadian Geotechnical Society members can access to this article, along with all other Canadian Geotechnical Conference proceedings, in the Member Area. Conference proceedings are also available in many libraries.

Cite this article: