Field Evaluation of Level Spreader – Vegetated Filter Strip Systems for Improvement of Urban Hydrology and Water Quality

Show full item record

Title: Field Evaluation of Level Spreader – Vegetated Filter Strip Systems for Improvement of Urban Hydrology and Water Quality
Author: Winston, Ryan Joseph
Advisors: Deanna Osmond, Committee Member
Francois Birgand, Committee Member
William Hunt, Committee Chair
Abstract: ABSTRACT WINSTON, RYAN JOSEPH. Field Evaluation of Level Spreader – Vegetative Filter Strip Systems for Improvement of Urban Hydrology and Water Quality (Under the direction of Dr. William F. Hunt). Level spreaders have been used in North Carolina to produce diffuse flow from stormwater outfalls, usually through a riparian zone. However, little research exists to document their performance for improving urban hydrology and water quality. To further test their utility, four level spreader – vegetative filter strip (LS-VFS) systems were constructed, two each in Apex and Louisburg, North Carolina. At each site, a small, highly impervious watershed was retrofitted with two LS-VFS Best Management Practices (BMPs). At the outlet of the watershed, flow was split proportionally to two 3.96 m (13 ft) long level spreaders. Flow was then dispersed downslope from each level spreader through one of two treatments: a 7.6 m long grassed filter strip, or a 15.2 m long filter strip, of which the first 7.6 m was grassed and the second 7.6 m was forested. Each site was equipped with rainfall, flow, and water quality measuring devices. Precipitation events were monitored from March 15, 2008 to March 4, 2009. Hydrologic parameters, including flow volume, peak flow rate, and lag in time-to-peak, were evaluated for 27 events in Apex and 58 events in Louisburg. Hydrologic evaluation of the 7.6 m VFS in Apex was unsuccessful due to difficulties with data collection. For the other three LS-VFSs, significant (p<0.01) reduction in flow volumes was observed. Over the study period, outflow volumes were 47%, 51%, and 56% of inflow volumes for the Apex 15.2 m VFS and the Louisburg 7.6 and 15.2 m VFSs, respectively. Differences in outflow volume for the Louisburg buffers were not statistically significant. Significant reductions in peak flow rate were observed between the inlet and outlet of each LS-VFS (p<0.01). Reductions in peak flow rate were greater than 60% for all VFSs studied. The lag in time-to-peak between the inlet and outlet of the BMP was significantly different from zero minutes for all three VFSs (p<0.05). However, increased lag time was not large enough to be useful for engineering designs. Inflow and outflow runoff quality from each VFS were compared. Flow proportional water quality samples were taken and analyzed for total kjeldahl nitrogen (TKN), nitrate-nitrite nitrogen (NO2-3-N), total nitrogen (TN), ammonia nitrogen (NH3-N), organic nitrogen (Org-N), total phosphorus (TP), orthophosphate (Ortho-P), particle bound phosphorus (PBP), and total suspended solids (TSS). Median inlet concentrations were higher at Apex for 6 of the 9 pollutants studied. All LS-VFS systems studied were effective in reducing median TSS concentrations (p<0.05), with the 7.6 m buffers reducing TSS by at least 35% and the 15.2 m buffers reducing TSS by at least 63%. The 15.2 m VFSs both significantly reduced concentrations of TKN, TN, NH3-N, and Org-N (p<0.05), while results were mixed for the 7.6 m VFSs. The effects of VFS length are very important for pollutant removal, as pollutant concentration reduction was nearly always greater for the longer VFSs. Pollutant mass loadings were reduced for every constituent in Louisburg and all but Ortho-P in Apex, due to large infiltrated volumes. Temperature of stormwater was significantly reduced across the Louisburg VFSs (p<0.05), as was thermal load (p<0.05). Overall, properly sited and designed LS-VFS systems can improve both urban hydrology and water quality. System performance depends on soil texture, vegetation type, vegetation density, VFS length, construction of the level spreader, surface infiltration rates, relative size of contributing watershed, etc. Future work should include prediction of LS-VFS performance based upon these factors.
Date: 2009-06-18
Degree: MS
Discipline: Biological and Agricultural Engineering

Files in this item

Files Size Format View
etd.pdf 12.73Mb PDF View/Open

This item appears in the following Collection(s)

Show full item record