Gregory D. Jennings

A Field Study of Green Roof Hydrologic and Water Quality Performance

Recent regulations intended to minimize the amount of nitrogen and phosphorus in stormwater runoff have resulted in interest in stormwater treatment practices to reduce nutrient loadings. In ultra-urban areas where typical treatment practices are not optimal because of large surface area requirements, green roofs are an option to improve stormwater runoff. The hydrologic and water quality performance of two extensive green roofs in eastern North Carolinas Neuse River basin were investigated. The two green roof designs were a flat 70 m2 area with an average media depth of 75 mm, and a 3% pitch 27 m2 surface area roof with an average media depth of 100 mm. Extensive green roofs typically have shallow media depths (less than 150 mm) with vegetation that requires minimal irrigation and maintenance. Each green roof retained a significant (p < 0.05) proportion of the rainfall observed, 64% of the total rainfall measured at each site. Peak outflow of runoff was significantly reduced (p < 0.05) from the green roof (average peak flow reductions of more than 75% were observed from each green roof), and each green roof substantially delayed runoff. On average, the total nitrogen (TN) concentrations in the green roof outflow were 2.7 mg L-1 higher than the rainfall (significant at p < 0.05) and 1.3 mg L-1 higher than the control roof outflow; TN amounts in the green roof outflow were 0.02 g higher than the rainfall and 0.12 g lower than the control roof outflow. On average, total phosphorus (TP) concentrations in the green roof outflow were 1 mg L-1 higher than the rainfall and 0.8 mg L-1 higher than the control roof outflow (both significant at p < 0.05); TP amounts in the green roof outflow were 0.07 g higher than the rainfall (significant at p < 0.05) and 0.05 g higher than the control roof outflow. It was determined that the media, composed of 15% compost, was leaching TN and TP into the green roof outflow. This field study demonstrated the importance of green roof media selection in locations where nutrient removal is a concern. Results from this study serve as a benchmark for the development of an optimal media that contains fewer nutrients initially within the media mix, yet provides adequate plant growth.

Analysis of functional traits in reconfigured channels: implications for the bioassessment and disturbance of river restoration

Abstract Channel reconfiguration is a popular but controversial approach to river restoration, and ecological responses to channel reconfiguration have not been rigorously assessed. We compared physical-habitat variables, taxonomic and functional-trait diversities, taxonomic composition, and functional-trait abundances between 24 pairs of upstream (control) and downstream reconfigured (restored) reaches in 3 catchment land uses (urban, agricultural, rural) across the North Carolina Piedmont. We asked how environmental filters and functional species traits might provide insight to biological responses to restoration. Taxonomic and functional-trait differences between control and restored reaches suggest that restoration affected aquatic assemblages only in agricultural and rural catchments. Our results highlight 2 important aspects of channel reconfiguration as a restoration practice. First, responses to restoration differ between agricultural/rural and urban catchments, possibly because of modified hydrological regimes caused by urbanization. Second, we find evidence that channel reconfiguration disturbs food and habitat resources in stream ecosystems. Taxa sensitive to disturbance were characteristic of control reaches, whereas insensitive taxa were characteristic of restored reaches. Abundances of traits related to reproduction (voltinism, development, synchronization of emergence, adult life span), mobility (occurrence in drift, maximum crawling rate, swimming ability), and use of resources (trophic and habitat preferences) differed significantly between control and recently restored reaches. Our results suggest that taxa in restored habitats are environmentally selected for traits favored in disturbed environments. Our work suggests how functional-trait approaches could benefit the practice of river restoration when used to target restoration activities and to develop informed expectations regarding recovery following restoration activities.

EVALUATING THE EFFECTIVENESS OF TWO STORMWATER WETLANDS IN NORTH CAROLINA

Two constructed stormwater wetlands in North Carolina were monitored to document their efficiencies at reducing nitrogen, phosphorus, and sediment loads in runoff from urban areas. Inflow to and outflow from the two wetlands were monitored during at least ten storm events, and flow-proportional samples of each were collected and analyzed. The first wetland (CMS), located in the Piedmont region, had a relatively low wetland surface to drainage area ratio of 2.2%, while the second (UNC), located in the mountains region, had less storage and a wetland surface to drainage area ratio of 4.7%. The UNC wetland had a much more diverse and extensive plant community. For the CMS wetland, the median load reduction efficiencies of nitrogen forms ranged from 47% to 54%, while the corresponding reductions for the UNC wetland ranged from 57% to 71%. The median reduction efficiencies for phosphorus forms for the CMS wetland ranged from 59% to 76%, while those for the UNC wetland ranged from -95% to 70%. The median reduction efficiencies for TSS were 72% and 88% for the CMS and UNC wetlands, respectively. The median load reduction efficiencies of nitrogen forms, total phosphorus, and sediment for the UNC wetland were greater than for the CMS wetland and were considerably greater than corresponding median efficiencies of other completed studies. These results indicate that both of these wetlands were, in general, more efficient at reducing pollutant loading than many others reported in the literature. The greater efficiencies may be the result of a relatively high surface area to drainage area ratio and a diverse plant community in the UNC wetland and a considerable storage capacity of the CMS wetland.

Evaluating the Effectiveness of Two Stormwater Wetlands in North Carolina

Two constructed stormwater wetlands in North Carolina were monitored to document their efficiencies at reducing nitrogen, phosphorus, and sediment loads in runoff from urban areas. Inflow to and outflow from the two wetlands were monitored during at least ten storm events, and flow-proportional samples of each were collected and analyzed. The first wetland (CMS), located in the Piedmont region, had a relatively low wetland surface to drainage area ratio of 2.2%, while the second (UNC), located in the mountains region, had less storage and a wetland surface to drainage area ratio of 4.7%. The UNC wetland had a much more diverse and extensive plant community. For the CMS wetland, the median load reduction efficiencies of nitrogen forms ranged from 47% to 54%, while the corresponding reductions for the UNC wetland ranged from 57% to 71%. The median reduction efficiencies for phosphorus forms for the CMS wetland ranged from 59% to 76%, while those for the UNC wetland ranged from -95% to 70%. The median reduction efficiencies for TSS were 72% and 88% for the CMS and UNC wetlands, respectively. The median load reduction efficiencies of nitrogen forms, total phosphorus, and sediment for the UNC wetland were greater than for the CMS wetland and were considerably greater than corresponding median efficiencies of other completed studies. These results indicate that both of these wetlands were, in general, more efficient at reducing pollutant loading than many others reported in the literature. The greater efficiencies may be the result of a relatively high surface area to drainage area ratio and a diverse plant community in the UNC wetland and a considerable storage capacity of the CMS wetland.

Improving construction site runoff quality with fiber check dams and polyacrylamide

Sediment and turbidity are among the most common pollutants affecting surface waters, resulting in reduced reservoir capacity, degradation of aquatic organism habitat, and decreased aesthetic value. Construction activities, including roadway projects, can be significant contributors to sediment loading in streams and lakes. We studied water quality in stormwater runoff from three systems for erosion and sediment control on two roadway projects in the North Carolina mountains. The first roadway project was divided into three experimental sections, each with one the following treatments installed in the adjacent drainage ditch: (1) the standard best management practice (BMP) consisting of narrow sediment traps in the ditch along with rock check dams, (2) fiber check dams (FCDs) consisting of a mix of straw wattles and coir logs, or (3) FCDs with granulated, anionic polyacrylamide (PAM) added to each. The second project was smaller and included only two of the experimental sections described above: (1) the standard BMPs and (2) FCDs with PAM. Significant reductions in turbidity and total suspended solids were obtained using the FCDs, particularly those with PAM added. At site 1, from June 2006 to March 2007, the average turbidity values for the stormwater runoff were 3,813 nephelometric turbidity units (NTU) for the standard BMPs, 202 NTU for the FCDs-only, and 34 NTU for the FCDs with PAM. Average turbidity in discharges at site 2 was reduced from 867 NTU for the standard BMPs to 115 NTU for the FCDs with PAM. Sediment loading at both sites was similarly reduced with the use of FCDs. At site 1, the standard BMPs lost an average of 428 kg (944 lb) of sediment per storm event compared to just 2.1 kg (4.6 lb) for the FCDs-only and 0.9 kg (2.0 lb) for the FCDs with PAM. At site 2, the standard BMPs lost an average of 3.3 kg (7.3 lb) per storm event compared with 0.8 kg (1.8 lb) for the FCDs with PAM. A conservative economic analysis suggests that the costs of the FCDs are lower than the standard BMPs. This study suggests that the use of FCDs with PAM can bring discharges from similar linear construction projects much closer to the regulatory guidelines for non-point source discharges than the current standard practices.

Increasing Stream Geomorphic Stability Using Storm Water Control Measures in a Densely Urbanized Watershed

This study used previously established unit critical discharges, annual allowable erosional hours, and annual allowable volume of eroded bed-load standards to evaluate two types of storm water control measures (SCMs): low-impact development (LID) practices and a large detention SCM (wet pond). Nine initial scenarios modeled in PC Storm Water Management Model (PCSWMM) incorporated different combinations of wet ponds, green roofs, rainwater harvesting systems, permeable pavement, and rain gardens to determine the best scenario for reducing stream erosion potential within a highly urbanized watershed in Chapel Hill, NC. The best-case scenario to reduce annual erosional hours and eroded bed load within the stream consisted of an aggressive utilization of LID practices in combination with an undersized wet pond. Although this scenario did not meet the annual erosional hour standard for rural reference streams, 0.35 h/ha/year, it did reduce erosional hours and eroded bed-load sediment by factors of 2.4 and 2.5, respectively, improving the existing condition. An alternative wet pond outlet structure that used two elevations of small drawdown orifices was explored to determine if current wet pond design practices could be improved to include stream stability. The new configuration provided a modest reduction in the number of erosional hours, a factor of 1.3, but increased the volume of eroded bed load by a factor of 1.2 when compared with the normal wet pond. However, adding widespread LID practices to the alternative outlet design reduced erosional hours and bed load transport by factors of 1.8 and 1.2, respectively, when compared with the normally configured wet pond. The failure to meet the erosional standards in all scenarios demonstrated the difficulty of requiring highly urbanized watersheds (60% impervious) to meet such strict stream protection measures.

Water quality improvements using modified sediment control systems on construction sites.

A study of the water quality of discharges from three different sediment control systems was conducted on a large construction site in North Carolina. Samples were obtained during storm events at the outlets of 11 of these systems using automatic samplers. Turbidity and total suspended solids (TSS) were measured, and a storm-weighted average (SWA) was determined for the systems. Water discharged from five standard sediment traps with rock dam outlets and unlined diversion ditches with rock check dams had an SWA turbidity of 4,320 nephelometric turbidity units (NTU) and an SWA peak of 12,640 NTU over a total of 26 storm events. The representative TSS values were 4,130 and 11,800 mg L-1, respectively. Measurements of runoff entering and exiting the traps suggested that heavy sediment was being captured, but turbidity was not reduced. Three traps with modifications including forebays, porous baffles, improved ditch stabilization (lining, additional check dams), and polyacrylamide application had SWA and peak turbidity of 990 and 1,580 NTU, respectively, over a total of 31 events. Total suspended solids were also much lower, at 740 and 1,810 mg L-1, respectively. Three basins with these same modifications, but with surface outlets, had somewhat higher average SWA values (1,560 NTU, 820 mg L-1), suggesting that the outlet type may not improve discharge water quality above the benefits of the other modifications to the standard sediment trap. However, when one of the latter systems was at optimal function, turbidity was reduced to below the receiving stream water levels (<100 NTU). These results strongly suggest that relatively simple modifications of commonly employed sediment trapping systems can dramatically improve discharge water quality and reduce the impacts on receiving waters.

Relating Stormwater Control Measure (SCM) Discharge Design Standards to Stream Erosion in Piedmont North Carolina: Case Studies in Raleigh and Chapel Hill, North Carolina

watersheds become urbanized, the percent imperviousness increases, altering the hydraulics of stormwater runoff. To restore urbanized watersheds, stormwater control measures (SCMs) are often implemented and designed through water quality measures to return peak flow attenuation to a pre-developed state; however, the affect of release rates on the geomorphic stability of streams are often neglected. The SCMs reduce peak flow rates, but increase duration of sub-bankfull flows within channels, extending periods of erosion. Seventeen rural reference streams along with PCSWMM were used to 1) develop a power regression equation controlling maximum release rate of SCMs based on critical shear stresses of bed sediment and 2) develop annual allowable erosional hours and annual allowable volume of eroded sediment standards. These standards were then applied to 2 urbanized watersheds, House Creek in Raleigh, NC and Tanyard Branch in Chapel Hill, NC to assess the practicality of using the developed standards in an urbanized watershed and determine the geomorphic impact of using LID practices instead of a wet pond on a stream.

Restoration Case Studies for Incised Rural North Carolina Streams

Many rural North Carolina streams are degraded due to historic channelization, dredging, and loss of riparian vegetation. The resulting incised channels have poor habitat and unstable bed features and streambanks. Stream restoration options for incised channels include constructing a new stable channel at the floodplain elevation, enhancing the floodplain at the existing channel elevation, or stabilizing streambanks in place. This paper describes six North Carolina stream projects completed since 1998 that make use of a variety of techniques to restore incised streams. Components of each project include channel geometry modification, in-stream structures, streambank stabilization, and riparian corridor restoration. Project objectives are to improve water quality and habitat, reduce streambank erosion, and enhance floodplain functions. The Yates Mill Tributary project is a Priority 1 restoration in which an incised stream was relocated at a higher elevation on the adjacent floodplain. The Bobs Creek, Muddy Creek, and South Fork Mitchell River projects are Priority 2 restorations in which channel geometry was modified at the existing elevation to create a new meandering stream with enhanced floodplain. The Cove Creek project is a Priority 3 restoration in which the floodplain of an incised straightened channel was widened and boulder structures were used to protect streambanks. The Shawneehaw Creek project is an example of stabilizing existing channels in place using instream structures and riparian vegetation. The planning, design, construction, and monitoring of these projects are described along with lessons learned about effective restoration techniques.

Relating Stormwater Design to Stream Health: A Case Study in Chapel Hill, NC

Stormwater control measures (SCMs) commonly use peak flow attenuation to return an urbanized, developed watershed back to its pre-developed peak flow conditions. The discharge design standards for the SCMs normally target low-frequency storms (usually 2- and/or 10-year events), but ignore the smaller, more frequent events. Recent emphasis has been placed on these smaller storms as well as the impact of SCMs on the geomorphic stability of streams. The more common sub-bankfull flows produced by SCMs are subjecting streams to longer and more frequent periods of erosion, increasing the instability of the stream. A stream experiencing extreme bank erosion in Bolin Creek Watershed, Chapel Hill, NC was modeled using continuous simulations in the Stormwater Management Model (SWMM) to determine a design discharge rate or volume control standard for a stormwater wetland that would reduce the frequency and duration of erosional events the stream was currently experiencing. The post-development erosion potential of the critical portion of the flow frequency curves was compared to the pre-developed or natural state of the watershed. Stormwater Wetlands with a variety of orifices within the outlet structure were sized for peak attenuation as well as erosion protection to determine the optimal discharge design standard for Bolin Creek. This study will hopefully influence other SCM design standards to protect stream stability.