Laserfiche WebLink
BEDAN AND CLAUSEN <br />equation (Table 4). The mass export of Cu did not <br />change significantly in the postconstruction period. <br />Metals exports of Pb and Zn most likely decreased <br />because storm flow from this watershed declined dur- <br />ing the postconstruction period. Studies of individual <br />BMPs have shown reduced loadings of heavy metals <br />as well. Grassed swales are reported to reduce Pb <br />and Zn loadings in stormwater runoff by 60 %, and <br />permeable pavement can reduce these metals by 90% <br />(USEPA, 1993a). Rushton (2001) also reported that <br />pervious pavement with grassed swales reduced Pb <br />and Zn loadings from parking lot runoff by at least <br />75 %. <br />Total Suspended Solids <br />Traditional Watershed. Mean TSS concentra- <br />tions in runoff from the traditional watershed signifi- <br />cantly decreased (p = 0.001) by 79% in the <br />postconstruction period (Table 3). Decreased TSS con- <br />centrations may be due to the primary conveyance of <br />stormwater across the asphalt road and the curb and <br />gutter system in the postconstruction period. A lawn <br />area was the primary conveyer of stormwater in pre - <br />development conditions and soils were a likely source <br />of TSS. Dilution could also explain lower TSS concen- <br />trations. Median TSS EMCs from the NURP study <br />were higher compared with this study (Table 3) <br />(USEPA, 1983b). Total suspended solids concentra- <br />tions observed in this study (Table 3) were at the <br />lower end of the range (4 to 1,223 mgA) reported by <br />Makepeace et al. (1995) in their review of urban <br />stormwater quality. <br />The mass export of TSS from the traditional <br />watershed significantly increased (p < 0.001) by 32 <br />times compared with the values predicted by the cali- <br />bration equation (Table 3). Since the TSS concentra- <br />tions decreased, the increased storm flow in the <br />postconstruction period presumably caused the <br />increase in the mass export of TSS. Increased runoff <br />from urbanization is expected to yield higher loads <br />of sediment compared with nonurbanized areas <br />(Leopold, 1968). <br />Low Impact Development Watershed. Total <br />suspended solids concentrations in runoff from the <br />LID watershed significantly increased (p < 0.001) by <br />almost three times in the postconstruction period <br />(Table 4). Increased TSS concentrations were likely <br />due to the grassed swales because stormwater was <br />primarily directed through them. The mass export of <br />TSS increased in the postconstruction period as well <br />(Table 4). However, observed TSS concentrations <br />were low. The NURP study (USEPA, 1983b) reported <br />TSS concentrations in residential runoff that were <br />ten times greater than the concentrations observed in <br />runoff from the LID watershed (Table 4). <br />Biochemical Oxygen Demand <br />Traditional Watershed. The mean BOD concen- <br />tration in runoff from the traditional watershed sig- <br />nificantly decreased (p = 0.009) during the <br />postconstruction period when compared with values <br />predicted using; the calibration equation (Table 3). <br />Low Impact Development Watershed. In the <br />LID watershed, the mean BOD concentration in runoff <br />did not change significantly (p = 0.693) during the <br />postconstruction period when compared with values <br />predicted using the calibration equation. Mean BOD <br />concentrations in runoff from the LID and traditional <br />watersheds were not significantly different in the post- <br />construction period based on a paired t -test (t = — 0.39, <br />p = 0.708). In this study, LIDs did not effectively <br />reduce BOD concentrations. Mean BOD concentra- <br />tions from this study (Tables 3 and 4) were generally <br />lower than the median EMC (10 mg/1) for residential <br />areas reported in the NURP study (USEPA, 1983b). <br />Fecal Coliform Bacteria <br />Traditional Watershed. Fecal coliform bacteria <br />abundance in runoff from the traditional watershed <br />did not significantly change from the calibration to <br />postconstruction period. <br />Low Impact Development Watershed. Fecal <br />coliform bacteria in runoff from the LID watershed <br />decreased significantly from the calibration to post- <br />construction period (Table 4). The mean FC concen- <br />trations of the study watersheds (Tables 3 and 4) <br />were low when compared with the FC concentrations <br />reported by the NURP study (USEPA, 1983b). Addi- <br />tionally, FC concentrations in this study were some- <br />what lower than the reported concentrations <br />( >800 fcu/100 ml) in residential runoff from else- <br />where in Connecticut (Dietz and Clausen, 2004). <br />CONCLUSIONS <br />Storm flow from the traditional development <br />increased dramatically during postdevelopment <br />conditions compared with predevelopment. The <br />increased runoff resulted in increased mass exports <br />of NO + NO NH TKN, TP, TSS, Cu, and Zn. <br />JAWRA <br />1006 JOURNAL OF THE AMERICAN WATER RESOURCES ASSOCIATION <br />