Journal of Hydrologic Engineering

Bioretention cells are incorporated as part of low impact development (LID) because of their ability to release influent runoff as exfiltration to the soil or evapotranspiration to the atmosphere. However, little care is taken as to the techniques used to excavate bioretention cells, and there is little concern as to the soil-moisture condition during excavation. Certain excavation techniques and soil-moisture conditions create higher levels of compaction which consequently reduce infiltration capacity. Two excavation techniques, the conventional “scoop” method which purposefully smears the underlying soil surface and the “rake” method which uses the teeth of an excavator’s bucket to scarify the underlying soil surface, were tested. Field tests were conducted on three soil types (sand, loamy sand, and clay) under a variety of antecedent soil-moisture conditions. Multiple hydraulic conductivity, surface infiltration, and soil compaction measurements were taken for each excavated condition. In all cases, the rake method of excavation tended to yield more permeable, less compacted soils than the scoop method. The difference of infiltration and hydraulic conductivity between the two excavation techniques was statistically significant $\left(p<0.05\right)$ when tests were conducted in wet soil conditions. Also, the infiltration rate at the clay site was significantly lower $\left(p<0.05\right)$ , and the hydraulic conductivity at the sandy site was significantly lower $\left(p<0.05\right)$ when the scoop methodology was used. Based on results of the experiment and because essentially no extra cost is associated with the rake method of excavation, it is recommended over the conventional scoop method. Another recommendation is to excavate under relatively dry soil conditions. The use of the rake method under dry soil conditions is expected to increase long-term exfiltration from bioretention cells.

Green roofs are an emerging storm-water management tool that has predominantly been analyzed for runoff volume reduction and peak flow mitigation. Little research has been completed on evapotranspiration (ET) in green roofs. Sedum mexicanum (Mexican stonecrop) and Disphyma australe (New Zealand iceplant) in a New Zealand designed, pumice- and zeolite-based substrate were analyzed to determine daily and hourly ET rates under both water-abundant and drought-stressed conditions. Water loss, and thus storage recovery of the substrate, was greatest in the first 9 days. Transpiration $\left(T\right)$ by S. mexicanum contributed up to 48% of total ET (2.19 mm/day) and D. australe contributed up to 47% of total ET (2.21 mm/day). After the initial rate of rapid water loss, plants conserved water and ET was not significantly different from evaporation $\left(E\right)$ from unplanted substrate. S. mexicanum had a greater ability to conserve water and thus a greater longevity of life than D. australe under harsh drought conditions.

Municipalities are increasingly interested in using rain gardens, or bioretention cells, as a “best management practice” (BMP) for storm-water management and as a component of low impact development. However, improved technical reference materials are needed to determine the influence of planting soil mix design on the hydrologic and water quality performance of bioretention cells. This article describes three investigations undertaken to determine the influence of planting soil mix characteristics. The first was to identify the bioretention cell design characteristics and methodologies used by different agencies. The second was a laboratory investigation on the soil characteristics that influence performance including field capacity, saturated hydraulic conductivity, soil moisture content, organic content, dry bulk density, porosity, and sediment grain size distribution. The third was to quantify the effectiveness of bioretention cells as a storm-water BMP by designing, constructing, and monitoring two full-scale bioretention cells on the campus of Lawrence Technological University in Southfield, Mich. Analysis included quantifying volume retention, peak discharge reduction, and removal of storm-water pollutants including total suspended solids, total phosphorus, and total nitrogen.

Low-impact development (LID), such as bioretention, is increasingly used as a best management practice (BMP) to manage storm-water runoff. With LID becoming an integral part of storm-water management plans, it is critical to understand the hydrologic performance of these devices and their positive impacts on downstream hydrology, geomorphology, and ecology. This paper focuses on presenting a model evaluation of bioretention using calibrated and verified algorithms and describes the application of flow duration control (FDC) as a design strategy. The goal of bioretention, like many LID-based controls, is to replicate natural hydrologic processes. Storm-water controls designed to match preproject flow duration characteristics demonstrate that they come close to mimicking the natural hydrologic cycle, thereby protecting beneficial uses and promoting long-term sustainable solutions. This work presents the FDC approach for sizing bioretention and other similar flow control BMPs. This paper concludes with a summary of the sizing requirements to meet the flow duration criteria. Results from this evaluation suggest that typical water quality design criteria can be effective for individual sites with 20–30% imperviousness. For areas draining 100% imperviousness, typical of LID strategies, storage requirements should be 5–10 cm (2–4 in.) with areas of 12–25% the catchment draining to the bioretention facility. However, storage and land area requirements should be sized based on watershed and site-specific characteristics.

Pervious concrete pavements in low-traffic urban areas such as parking lots reduce storm water runoff and also minimize water pollution. However, there are concerns about their expected clogging and consequential reduction of hydraulic performance in the long run. The pervious concrete pavements can be declogged using rejuvenation methods such as vacuum sweeping, or pressure washing, or a combination of both. This paper presents the results of our study that focused on the hydraulic performance of pervious concrete pavements. The study included field and laboratory investigations to evaluate the infiltration capacities of the pervious concrete cores and the underlying soils and the usefulness of rejuvenation methods in restoring their hydraulic performance. As a result of this research program, a new field test device, called the embedded ring infiltrometer, was developed for evaluating the infiltration rates of newly installed pervious concrete pavements. The results of this study indicate that the rejuvenation methods can substantially restore the performance of pervious concrete pavements for better management of storm water.

It is well established that the excess storm-water runoff volume from impervious areas can lead to impairments and water pollution originating from the sewer system overflow and combined sewer systems overflow. This redirection of the runoff into wastewater treatment plants and stream channels can also deprive shallow groundwater tables with recharge, as an impervious surface prevents water from infiltrating to aquifers. The runoff from impervious areas and, in particular, directly connected impervious areas, has been proven to cause the majority of the problem. Controlling the runoff at its source and disconnecting the impervious area from the sewer system is a way to resolve and reduce the impact of excess runoff. This is achieved by implementing specialized detention technologies for runoff reduction. This paper builds on new modeling techniques of two best management practices, rain gardens and rain barrels, implemented in the EPA storm-water management model Version 5 (SWMM-5). The behaviors of a continuously draining rain barrel and an overflowing rain barrel were studied under steady state and unsteady state conditions using C++ and MATLAB programs. The models obtained were compared to a rain barrel conceived within the EPA SWMM-5 subcatchment architecture. Next, a model input was derived to best describe the behavior of a treatment train for water quantity composed of a rain garden, the overflowing rain barrel, and/or the continuously draining rain barrel. A simulated rainfall event in EPA SWMM-5 assesses the results of each subcatchment’s model input and estimates the potential percentage of runoff reduction and the potential reduction in the peak flow and timing of outflow.

With continued urbanization pressure, regulators and developers alike are increasingly looking to new forms of green infrastructure and low-impact development technologies as a means of appropriately integrating built infrastructure into the landscape. This paper describes the results of a series of experiments designed to simulate the hydrologic performance of green roofs under variable precipitation conditions. The experiments were designed in order to test performance under both steady, low-intensity rainfall, as well as under short duration, high-intensity rainfall conditions. A control membrane roof and prototype green roofs of 2.5-, 6.3-, 10.1-cm depths were subjected to simulated precipitation in a laboratory setting. The green roofs delayed, prolonged, and reduced the peak rates of green roof discharge to 22–70% that of a standard roof surface, with greater percent reductions associated with deeper roofs. Negligible discharge was observed from all of the prototypes during the first 10 min of simulated precipitation. Although the fate of the 0.35 cm of precipitation that were applied over this time period can only be determined through additional controlled testing of the prototypes with shorter duration rain events, the potential significance of green roofs that retain this quantity of water is discussed in the context of the historical New York City precipitation record. The results also indicated that nearly all of the precipitation applied was discharged as drainage over the 24 h period immediately following the experiment, suggesting that the percentage of large storms that are retained in green roofs may be insignificant. Green roof runoff coefficients computed from an analysis of the discharge hydrographs ranged from 0.2–0.7, consistent with other studies. Two approaches to predicting the observed discharge using the U.S. EPA’s Stormwater Management Model (SWMM) are also presented. The “storage node” approach achieves better overall predictions than the “curve number” approach, which itself tends to significantly underpredict discharge from these systems. Although reasonable sets of predictions were eventually obtained, the selection of appropriate model parameters would not have been possible without the availability of experimental data with which to calibrate the models. The experimental results support the argument that the storm water benefits of green roofs could be significant. However, the writers urge caution in interpreting the results of green roof drainage discharge calculations made using SWMM until additional calibration and validation attempts have been performed.

A green roof is a multilayered vegetated roof covering consisting of a root barrier, a waterproof membrane, an under drainage system, and a layer of growing medium covered by plants. The physical principles governing the primary hydrologic and hydraulic processes in the growing medium by infiltration, drainage and recession are described. Contrary to conventional infiltration modeling approaches in which all rain water is used to advance the wetting front, the model suggests that a portion of rain water will directly drain through the growing medium to the under-drain after the field capacity is exceeded but before the growing medium is completely saturated. The model has been tested successfully using more than three years flow survey data from the green roof of Hamilton Building of Portland, Oregon.

The effective use of pervious concrete in environmental site design requires consistent design procedures integrating the structural and material properties of the pervious concrete pavement with hydrologic performance of the pervious concrete system. Design procedures to size pervious concrete storm-water systems are presented based on criteria for freeze-thaw protection and drawdown reliability. Hydrologic performance criteria are quantified by an effective curve number, estimated from simulated routing of design storm hydrographs using standard storm-water computations. Combining operational design criteria with the evaluation of hydrologic performance criteria, as an effective curve number, integrates pervious concrete systems with traditional storm-water management practice and emerging standards for environmental site design.

The respective runoff from a $200{\text{-m}}^2$ permeable pavement test site and an adjacent $850{\text{-m}}^2$ conventional asphalt road catchment in Auckland, New Zealand, was monitored concurrently between 2006 and 2008. Despite installation over relatively impermeable subgrade soils, and on an atypically high slope (6.0–7.4%) and active roadway, the overall hydrologic performance of the permeable pavement was exceptional. Measured discharge from the permeable pavement underdrain demonstrated peak flow (81 storms) comparable to or below modeled predevelopment conditions for most storms, regardless of antecedent conditions, including a 10-year, 24-h annual recurrence interval event. For large events (5% exceedance), underdrain discharge volume was comparable to predevelopment conditions, but it was substantially less than asphalt runoff for all events up to approximately 70% exceedance. The distributions of peak flow and volume were statistically different between the asphalt catchment runoff and the permeable pavement underdrain discharge (0.05 level of significance). Runoff coefficients ranged from 0.29–0.67 for underdrain discharge (10th–90th percentile events), and from 0.41–0.74 when permeable pavement comprises about one-half of an otherwise impervious catchment. Underdrain lag time and hydrograph duration were reminiscent of a vegetated area. Spearman correlation indicated influences of rainfall depth, intensity, and duration on runoff parameters from both catchments, while antecedent moisture condition was correlated to underdrain discharge lag time. Surface infiltration measurement at four permeable pavement sites revealed that surrounding land uses likely have more influence on pavement clogging than does traffic load. Permeable pavements should be given strong consideration as an low impact development source control, and can also mitigate conventional large design storm flows, but care must be taken during installation to ensure proper function.

This article presents the elements involved in the design of a bioretention planter/trench infiltration-detention system as part of a very large-scale urban retrofit project. The prototype system was designed to intercept all of the runoff from a synthetic 5.08-mm 24-h rainfall event. Diverted flows were conveyed into bioretention planter for treatment. The bioretention systems were fingerprinted into areas comprising 0.8% of the contributory drainage areas, with an associated stone trench comprising another 3.4%. As layered systems, an approach that is capable of modeling vertical flows in addition to dynamic routing of outflows is used. The system was first modeled using HydroCAD, a design storm event modeling software. A four-compartment node system is used to model the dynamics of flow through the layers. The system was then modeled using SWMM 5.0.014 continuous simulation software. The resulting response to a design storm was computed by both of these models to compare the results of each method. The resulting SWMM model was then run on the 2005 design year rainfall distribution. Under existing conditions, over 60% of annual runoff volume exceeded the $3.50\text{L}\cdot {\text{s}}^{-1}\cdot {\text{ha}}^{-1}$ $\left(0.05{\text{cfs-ac}}^{-1}\right)$ threshold for initiation of combined sewer overflows (CSOs). Nearly all runoff was intercepted by the planter/trench infiltration system and even with a soil infiltration rate of only $2.54\text{mm}\cdot {\text{h}}^{-1}$ , 47% was infiltrated, and less than 6% was discharged at rates that could initiate CSOs. The number of CSO exceedance pulses was reduced from 233 to 6, a reduction of 97%. The volume of flows exceeding the CSO threshold decreased by 90% in the planter/trench system.

The premise of low impact development (LID) is to reduce runoff volumes while simultaneously matching target (often predevelopment) infiltration and evapotranspiration volumes. Many structural practices used in LID, such as bioretention, infiltration trenches, and permeable pavement, require a seasonally high water table (SHWT) to be 1–2 m from the surface. Conversely, level spreader–vegetated filter strip systems can be implemented in locations where a SHWT would restrict the use of other practices. A 19.4 m long reinforced concrete level spreader upslope of a $900{\text{m}}^2$ mixed grass/weed vegetated filter strip was monitored for runoff reduction for 23 precipitation events from November 2005 to January 2007 in Charlotte, N.C. The practice treated runoff from a 0.87 ha (2.15 acre) residential watershed. Of the 23 monitored events, only 3 produced outflow. All events which produced outflow exceeded 40 mm (1.6 in.) of precipitation. Cumulative volume reduction associated with the events was 85%, which compares favorably to other LID structural practices tested in the region. Because of its simple design, construction, and maintenance, and its ability to be sited in locations with otherwise restrictive SHWTs, level spreader–vegetated filter strips may be a valuable LID practice.

This paper presents an approach for designing storm-water controls to promote sustainable ecosystems. The approach focuses on understanding and resolving impacts from storm-water discharges on geomorphic stability of stream channels. To develop effective controls, it is critical to understand performance with links to instream hydrology and geomorphology. Comparing work curves and erosion potential in receiving channels between pre- and posturban conditions is the preferred method to test effectiveness and define management criteria. Work incorporates the combined influence of flow, sediment, geometry, slope, and resistance of the channel boundary and can be used to measure the change in erosive energy resulting from urban discharges. Flow duration control can be used to design on-site storm-water controls because flow duration is directly linked to channel processes. Flow duration control also has the potential to “mimic” predevelopment hydrology and be protective of beneficial values. Examples are presented to illustrate the application of work, the erosion potential, and flow duration control as decision making and design criteria.

A 1 year-old parking lot in eastern North Carolina consisting of four types of side-by-side permeable pavement and standard asphalt was monitored from January 2007 to July 2007 for water quality differences among pavement types. The four permeable sections were pervious concrete (PC), two different types of permeable interlocking concrete pavement (PICP) with small-sized aggregate in the joints and having 12.9% (PICP1) and 8.5% (PICP2) open surface area, and concrete grid pavers (CGP) filled with sand. The site was located in poorly drained soils, and all permeable sections were underlain by a crushed stone base with a perforated pipe underdrain. Composite, flow-weighted samples of atmospheric deposition and asphalt runoff were compared to those of permeable pavement subsurface drainage for pH, TN, ${\text{NO}}_{2,3}–\text{N}$ , TKN, ${\text{NH}}_4–\text{N}$ , and ON concentrations and loads. All pavements buffered acidic rainfall pH $\left(p<0.01\right)$ . The pH of permeable pavement subsurface drainage was higher than that of asphalt runoff $\left(p<0.01\right)$ with the PC cell having the highest pH values $\left(p<0.01\right)$ . Permeable pavement subsurface drainage had lower ${\text{NH}}_4–\text{N}$ $\left(p<0.01\right)$ and TKN concentrations than asphalt runoff and atmospheric deposition. With the exception of the CGP cell, permeable pavements had higher ${\text{NO}}_{2,3}–\text{N}$ concentrations than asphalt $\left(p<0.01\right)$ , a probable result of nitrification occurring within the permeable pavement profile. The CGP cell had the lowest mean TN concentrations; however, results were not significantly different from those of asphalt runoff. The possible nitrogen removal exhibited by the CGP cell is similar to that observed in sand filter research, not surprising considering CGP contained a 10 cm (4 in.) sand bedding layer. Overall, different permeable pavement sections performed similarly to one another with respect to water quality, with the CGP cell appearing to best improve storm-water runoff nitrogen concentrations.

In moving toward a water balance approach to storm water management and the modeling tools that can support such an approach, it is clear that there are substantial gaps in our knowledge of the groundwater recharge and evapotranspiration (ET) in urban areas. At a microscale, weighing lysimeters can be used to estimate the ET and may also be designed to also allow drainage (groundwater recharge) to be quantified. A weighing lysimeter has been designed and constructed in Guelph, Ont. to quantify the ET and groundwater recharge from a bioretention area. The results at this scale will vary with the urban surroundings, but will nevertheless contribute to the scant information available. The installation of similar systems in various urban micrometeorological conditions will help to build a database of urban ET and recharge. Other approaches, such as the eddy correlation approach for the ET, are also needed to complement microscale measurements.

Rain gardens effectively remove some stressors from storm water, but in most cases they show much smaller removal rates of nitrate, likely due to the high sand and low organic matter content of rain garden media inhibiting denitrification. A bench-scale experiment was conducted to test the drainage capability of media containing shredded newspaper layers as a carbon amendment. Storm water was introduced at low and high rates to bins containing zero, one, and two layers of newspaper at varying depths. While there were differences in effluent volumes and flow rates between control and newspaper treatments, surface ponding occurred in all three treatments, suggesting that some other factor besides the newspaper had an effect on drainage properties. Grain size and clay mineralogy analyses indicated the migration of finer particles into the deeper soils, which could have inhibited drainage.