Abstract

Bioretention is a relatively new stormwater management practice that relies on physical, chemical and biological processes within a terrestrial ecosystem to provide stormwater retention and treatment. Bioretention systems, also referred to as rain gardens, include a layer of high permeability soil filtration media, covered by an optional thin layer of mulch, and planted with woody and herbaceous plants. In areas with low permeability native soils, an underdrain structure is installed below the soil filtration media to prevent water from standing for excessive periods of time. An overflow structure is also incorporated into the system to drain excess water when the ponding capacity of the system is exceeded.

Field monitoring and laboratory testing performed to date have demonstrated the ability of bioretention systems to significantly decrease runoff flows and to efficiently reduce a number of pollutant loads. However, large discrepancies in phosphorus removal efficiencies have been reported from the field monitoring of bioretention systems. Two bioretention cells on the University of Maryland campus, monitored by Davis (2007), achieved 79% and 77% total phosphorus mean mass removals, respectively, over 12 storm events. Conversely, Hunt et al. (2006) noted an increase of 240% in total phosphorus, on a mass basis, in the outflow of a bioretention cell in North Carolina over a 12-month monitoring period.

Some theories have been put forward to explain the wide range of phosphorus loadings at the outflow of field-scale bioretention systems. Dietz and Clausen (2005) noted that outflow phosphorus concentrations were consistently above inflow phosphorus concentrations for two similar rain gardens monitored over a period of 12 months. After observing an exponentially decreasing trend over time in phosphorus concentrations from both the rain garden inlets and outlets, they suggested that the increased outflow phosphorus concentrations may have been caused by a disturbance in the soil. Alternatively, Hunt et al. (2006, 2008) suggested that phosphorus sorption to bioretention soils, which is maximized in low phosphorus content soils, greatly influences cycling in bioretention systems.

Current bioretention design guidelines focus mainly on peak flow reduction, which may not provide adequate treatment in areas sensitive to pollutant loading. In particular, most freshwater environments are sensitive to high phosphorus loadings, which can lead to the eutrophication of receiving water bodies. This chapter is presents the preliminary conceptual and mathematical modelling framework that will be incorporated in a user-friendly tool for designers to predict phosphorus removal in a bioretention system. This tool will also assist researchers in identifying the main factors that influence phosphorus removal in bioretention systems and in understanding the high variability in phosphorus removal observed in the field.

A comprehensive review of the current literature on bioretention indicates that bioretention system modelling is still in its infancy. Only a few models have been developed to simulate hydrologic processes inside bioretention systems. The most comprehensive of these are the RECHARGE and RECARGA models, developed by researchers at the University of Wisconsin-Madison (Dussaillant et al. 2003, Dussaillant et al. 2004), which use the Richard’s and the Green-Ampt infiltration equations to model groundwater recharge through bioretention systems. Moreover, the bioretention water quality model developed by Li and Davis (2008) simulates one-dimensional suspended solids filtration through bioretention media.

Building on previous research, the authors are currently developing the Bioretention Phosphorus Removal Model (BPRM), an event-based one-dimensional finite difference model to simulate total phosphorus removal in a bioretention system. The model comprises four layers which act as completely-mixed reactors: the ponding layer, the mulch layer (optional in bioretention systems), the soil root zone, and the deep soil zone. Model subcomponents estimate water volumes and phosphorus mass within the model layers. Total phosphorus is divided into particulate and soluble phosphorus. The model requires input time series for rainfall and runoff inflows, as well as soluble and

particulate phosphorus inflow concentrations. Processes modelled inside the bioretention system include: evapotranspiration and overflow from the ponding layer of water; sedimentation of particulate phosphorus; infiltration of water and soluble phosphorus across model layers; vegetative uptake of soluble phosphorus; mulch and soil sorption and desorption of soluble phosphorus; exfiltration of water and soluble phosphorus from the system to the surrounding native soils; as well as underdrain discharge of water and soluble phosphorus.

Preliminary model processes, input parameter requirements and model limitations will be discussed in detail in this chapter. The first section of this chapter describes the structure of the model being developed, along with the modelling assumptions made in developing BPRM. Then, the equations selected to represent each process in BPRM and the reasons behind these choices are detailed in Section XX.2. Section XX.3 includes a list of the input parameters required by BPRM for a simulation, including unit requirements and a range of recommended values for each input parameter. In the final section, the capabilities and the limitations associated to the model being developed are discussed. The work presented in this chapter is in progress and future work will focus on improving the mathematical framework presented and addressing some of the limitations described. The model developed will also be evaluated against field data and the sensitivity of modelling predictions to input parameter selection will be assessed.

This paper is only available in PDF Format:

  View full text PDF