NCSU Institutional Repository >
NC State Theses and Dissertations >
Please use this identifier to cite or link to this item:
|Title: ||Watershed-Scale Stormwater Monitoring of a Mixed Land Use Watershed in the North Carolina Piedmont|
|Authors: ||Skipper, Gabrielle Marie|
|Advisors: ||William F. Hunt, Committee Chair|
Gregory D. Jennings, Committee Member
James D. Gregory, Committee Member
|Keywords: ||Neuse River|
|Issue Date: ||8-Dec-2008|
|Discipline: ||Biological and Agricultural Engineering|
|Abstract: ||The Neuse River Basin in North Carolina has a history of eutrophication and is subject to stormwater regulations regarding nutrients. House Creek watershed is a 217.5 ha, mixed land use watershed located in Raleigh, North Carolina in the upper Neuse River Basin. The watershed consists of six land use areas: urban (22.8 ha), golf course (19.6 ha), highway (17.3 ha), agricultural/pasture (74.2 ha), residential (29.9 ha) and wooded (53.3 ha). The objectives of this research were to compare runoff volumes, peak flowrates, pollutant concentrations, loads and exports among land uses, as well as compare upstream subwatersheds to the downstream outlet. Five subwatersheds and the outlet were instrumented in a nested, upstream-downstream design. The land uses of the five subwatersheds were: urban (39% impervious), agricultural/pasture (22% impervious), golf course (26% impervious), and two highway subwatersheds differing in size and percentage of impervious surface (24% and 10%, respectively). The outlet station represented a mixed land use (22% impervious). The wooded and residential areas were not monitored. Flow measurements were obtained using a weir and bubbler or area velocity sensor. Continuous, automatic, flow proportional stormwater samples were analyzed for total kjehldahl nitrogen, nitrate-nitrite, total nitrogen, ammonia-nitrogen, total phosphorus, orthophosphate and total suspended solids. Rainfall was measured using four automatic gages located throughout the watershed and data were averaged to produce a single rainfall and intensity for each storm. Extreme drought during the summer of 2007 had a substantial impact on rainfall and runoff generation.
The highest mean runoff volumes and peak flowrates occurred at the more impervious land uses (urban, golf course, impervious highway) while the pervious highway and mixed land uses generated substantially less runoff and lower peak flowrates. Runoff coefficients were: urban, 0.26; golf course, 0.21; impervious highway, 0.20; pervious highway, 0.16; mixed, 0.11. Significantly lower volumes and peak flowrates at the outlet provided evidence of a dampening or scaling effect, whereby as watershed size increases, more opportunities for infiltration, evapotranspiration or storage decrease the runoff per unit area.
TSS concentrations were significantly higher at the agricultural/pasture land use and at the outlet, partially due to streambank erosion and construction. Nutrient concentrations were significantly higher at the pasture site, partially due to cattle access. Nutrient concentrations were significantly lower at the urban land use, likely due to pollutant removal by an upstream pond, and at the outlet, further evidence of a dampening effect. The highest mean TSS load occurred at the urban site due to high runoff generation, while the more pervious highway site was lowest. The highest nutrient loads occurred at the golf course and the lowest occurred at the urban site. Site rankings for pollutant exports generally followed the same trends as concentrations.
Volumes, peak flowrates, TSS concentrations and loads were significantly higher at the impervious highway site compared to the pervious highway site. The differences were attributed to less connected impervious surfaces and filtration in the grass median at the pervious highway site.
Benthic macroinvertebrate sampling was conducted at six sites during both years of the study. The poorest water quality indicators were found at the agricultural/pasture land use and indicators improved moving downstream into the wooded area.
Streambank erosion was monitored for 1.1 years at 14 eroding streambanks in the watershed using bank pins. Average annual retreat was 39 mm/yr. A watershed inventory found 45% of the 278 streambanks were eroding and the estimated streambank sediment export was 200,000 kg/yr. Cattle impacts contributed to erosion at the pasture site.|
|Appears in Collections:||Theses|
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.