106 (DRO-1117-733955) Detection of high variability in organochlorine and PCB concentrations in water of the Detroit River using a long term mussel biomonitoring data base.
Start time: 1:50 PM
Drouillard, K¹, Haffner, G¹, Leadley, T¹, ¹ Great Lakes Institute for Environmental Research, University of Windsor, Windsor, Ontario, Canada
Calibrated freshwater mussel (Elliptio complanata) biomonitors were deployed at six locations in the Detroit River during the open water season over the period of 1996 to 2003. This large temporal data set (n= 544 samples) was interpreted along with additional spatial biomonitoring and sediment quality surveys performed throughout the entire Detroit River during 1999 and 2002. Both mussel biomonitors and sediment surveys identified similar near-shore contaminated regions within the Detroit River indicating that on-going loadings of PCBs to water continue to occur within this system. A toxicokinetic model was used to interpret time-accumulated residues measured in mussel tissues at individual sites and years. Although lipid normalization removed a portion of the site- and time-specific individual variation in mussel exposures, steady state correction of time-accumulated mussel residues failed to remove within site and within-year variation for the majority of cases. There were only a few exceptions (sites and years) where non-steady state, lipid-normalized contaminant residues in mussels followed the expected uptake trajectory predicted by the bioaccumulation model under a condition of constant water concentrations. The above variation was not likely due to artifacts in field deployed mussel toxicokinetic parameters since field validation of mussel PCB elimination kinetics showed comparable results with laboratory toxicokinetic data. These observations implicate considerable variation in PCB water concentrations in the Detroit River during the open water season. Thus, monthly water sampling as routinely performed by high-volume water extraction monitoring programs in this system may be inadequate to characterize PCB loads in the Detroit River

107 (LET-1117-815179) Point versus non-point sources and fate of major PPCPs and herbicides in the Detroit River: Wasterwater effluent to surface water to drinking water.
Start time: 2:10 PM
Letcher, R.^{1, 2}, Hua, W.², Bennett, E.², Metcalfe, C.³, ¹ Environment Canada, Canadian Wildlife Service, National Wildlife Research Centre, Ottawa, Ontario, Canada² University of Windsor, Great Lakes Institute for Environmental Research, Windsor, Ontario, Canada³ Trent University, Water Quality Centre, Peterborough, Ontario, Canada
Pharmaceuticals (PhACs) and personal care products are collectively referred to as PPCPs, and are used in considerable quantities for, e.g. therapeutic, clinical and disease alleviation purposes by humans and in veterinary science. Triclosan (5-chloro-2-(2,4-dichlorophenoxy) phenol) is a broad-spectrum antibacterial and used heavily as an antimicrobial agent. The Detroit River passes through one of the most densely populated areas in the Great Lakes basin, accepting significant urban/agricultural runoff and wastewater (WSTP) discharge from the U.S and Canadian sides. Surface water of the upper Detroit River is also the important source of drinking water for the surrounding urban complex. We report on point versus non-point source input, fate, drinking water treatment effects and annual variations of seventeen major PhACs, triclosan and the herbicide class of chloro-s-triazines (atrazine), in waters from the Canadian side of the upper Detroit River. Fifteen out of seventeen PhACs, triclosan and atrazine were detected in all WSTP effluent samples, and concentrations were highly variable (1.7 to 1244 ng/L) for three annual sampling periods. A reduction of >90% for all concentrations was found in surface waters from shoreline sites along the Canadian side of the upper Detroit River. For PhACs and triclosan, large-scale dilution upon entering river waters occurs from point source inputs from WSTPs. This was not true for atrazine, where surface water concentrations were higher than in WSTP effluent. Regardless of the season of sampling, and despite large scale dilution, trimethoprim, caffeine, carbamazepine and cotinine were consistently measurable at sites downstream of the Detroit River/Little River confluence. Only atrazine carbamazepine, cotinine and caffeine were quantifiable with any frequency near the raw water intake for City of Windsor, drinking water treatment facility. For these PhACs and atrazine, conventional treatment processes that included ozonation had a dramatic reductive effect on water concentrations, whereas conventional treatment alone had a more minor reductive effect.

108 (WEB-1117-848783) Investigative Urban Surface Water Quality Monitoring Program Integrating Land Use and Watershed Topography.
Start time: 2:30 PM
Weber, G^{1, 2, 3}, Uwins, J^{2, 3}, Lancy, T^{2, 3}, Aguinaga, G^{2, 3}, Phillips, J^{2, 3}, Williams, D^{2, 3}, ¹ Blasland Bouck & Lee Inc, Carpinteria, CA, USA² University of California, Santa Barbara, Santa Barbara, CA, USA³ The City of Santa Barbara, Santa Barbara, CA
The last 15 years have shown a heightened awareness to general water quality impacts of urban runoff as a non-point pollution source; particularly since the release of NPDES Phase II MS4 designations, and the beginning of CWA 303(d) TMDL enforcement. In order to improve water quality and reduce the potential risk of ecological or human exposure to hazardous constituents, communities frequently utilize Best Management Practices (BMPs) in conjunction with monitoring. Commonly, monitoring is used for reporting compliance, and in gauging the effectiveness of BMPs. Less common is the use of monitoring as an investigative tool to help identify source(s) contributing to non-point source pollution. A monitoring program was developed to investigate both sub-watershed drainages and land use classifications as potential non-point sources, thus providing both a qualitative and geographic dimension of analysis. Analytes were selected using historical city monitoring data, peer review literature investigating land use - contaminant relationships, and grey literature performed by other municipalities with similar climates and cityscapes. Sample collection sites were selected by integrating Geographic Information Systems (GIS) analysis and historical sampling locations. Drawing from the U.S. Geological Survey National Ambient Water Quality Assessment (NAWQA) program, sampling sites were divided into integrator sites and indicator sites. Integrator sites were selected to characterize overall watershed conditions. Indicator sites were chosen to examine inputs from individual drainage basins that consisted of a homogeneous land use. Results from indicator sites were intended to provide insight into the impacts from specific land uses on the receiving waters in the area. If successful, the city could better identify land use types and sub-watersheds as targets for future BMPs and policy actions. This investigative monitoring methodology could potentially serve as a template for developing other watershed management programs aimed at improving urban surface water quality and/or mitigating exposure to non-point, anthropogenic contaminants.

109 (WHI-1117-731502) Managing phosphorus in a built environment.
Start time: 2:50 PM
White, Sarah¹, Taylor, Milton¹, Polomski, Robert², Whitwell, Ted², Klaine, Stephen¹, ¹ Clemson Institute of Environmental Toxicology, Clemson University, Pendleton, SC, USA² Department of Horticulture, Clemson University, Clemson, SC, USA
Nutrient-rich runoff is commonly associated with hyper-eutrophication leading to ecosystem stress. Phytoremediation in a created wetland setting is an effective way to filter nutrients, especially nitrogen (N), from runoff water. Phosphorus (P), however, is not efficiently managed in a wetland setting by plants. Our goal is to optimize wetland design to maximize P attenuation. To accomplish this we have screened ornamental wetland plants for their ability to accumulate N and P and have characterized the ability of soil substrates for P adsorption. Two woody species (Cornus amomum and Itea virginica) and several herbaceous plant species (Canna 'Bengal Tiger' and 'King Humbert', Iris 'Full Eclipse', Typha minima, Pontaderia cordata, and Peltandra virginica) were able to absorb large quantities of N and P. Two media were found to efficiently sorb large quantities of P from aqueous environments. Compiling the results of these two studies, we have formulated a wetland design that maximizes removal of both N and P from nutrient rich run-off water.

(58061) COFFEE BREAK.
Start time: 3:10 PM

110 (TWA-1117-870630) Prevention and control of non-point source pollution of soil and ground waters using abundant natural and waste organogenic materials.
Start time: 3:50 PM
Twardowska, I.¹, Kyziol, J.¹, Janta-Koszuta, K.¹, Stefaniak, S.¹, ¹ Polish Academy of Sciences, Institute of Environmental Engineering, Zabrze, Silesia, Poland
Growing human-induced impact on soil and water quality, among others due to worldwide increase of sewage sludge (biosolids) generation and use in agriculture, as well as disposal of metal-rich sufidic waste from mining results in the need of simple, efficient and cost-effective methods of non-point pollution prevention and control through reduction of metals content in sewage sludge (SS) and prevention of metals leaching and transport from dumping sites to the aquatic environment. Source control of metals from artisan activities in municipal wastewater and application of permeable barrier layers during waste dump construction might greatly reduce a risk to the environment. Application of abundant, low-cost organogenic natural (e.g. peat) or waste materials, e.g. stabilized SS as sorbents seems to be an attractive contribution to the solution of these problems. Studies on the use of peat for metals removal from strongly acidic polymetallic electroplating wastewater (EPW) in batch and flow-through systems confirmed high efficiency of these processes. In the batch process at input concentrations of Me-SO₄ in EPW: 14985 mgFe/L>2807 mgZn/L>335 mg Cr/L>171 mgCd/L>122 mgMn/L, pH 1.47, sorption capacity of peat for metals from undiluted EPW accounted for 107330 mgFe/kg, 3520 mgZn/kg, 2002 mgCr/kg, 1053 mgCd/kg and 371 mgMn/kg. The removal efficiency with respect to the total input load was 75% Fe, 85% Cr, 61% Cd, 30% Mn and 13% Zn. Flow-through process showed considerably higher sorption capacity for all the metals, in particular for Zn (22985 mg/kg) and Cr (17431 mg/kg) that accounted for 80% and 84% of total input load of these metals. In both systems the predominant role in metal binding played strongly bound organometallic and labile exchangeable sorption "pools". Peat appeared to be an efficient sorbent to be used both in simple batch reactors and in protection layers, as well as in constructed wetland systems, also for strongly acidic wastewaters. Wetland protection requirements limit extensive use of peat. For this reason, stabilized SS is better applicable in batch reactors, where Me remowal from the EPW in the range 83->95% for Me concentrations in the input >50 mg/L, and 47-65% at lower Me input concentrations was achiewed, showing high feasibility of these materials application as efficient sorbents.

111 (MUT-1117-808282) Treatment of heavy metals in stormwater runoff using retention ponds and constructed wetlands.
Start time: 4:10 PM
Muthukrishnan, S¹, ¹ ORISE Post Doctoral Fellow, Office of Research and Development, U.S. Environmental Protection Agency, Edison, New Jersey, USA
Urban stormwater runoff is a significant source of suspended sediments and associated contaminants, including heavy metals, to receiving waterways. These metals are either dissolved or bound to particulates (coarse >75 m; fine particulates <75 – 1 m; colloids <1 m). Information on the particulate association of heavy metals in stormwater is a critical requirement prior to using pond and wetland best management practices (BMPs) as stormwater treatment controls. Heavy metals in stormwater are primarily removed by sedimentation in structural BMPs. These sediments may be toxic to benthic invertebrates and aquatic microorganisms. Research is being conducted at U.S. EPA′s Urban Watershed Research Facility in Edison, NJ to evaluate the effectiveness of retention pond and constructed wetland BMP mesocosms to remove particulate-bound heavy metals from roof and parking-lot runoff. The runoff draining from roofs and parking lots is being investigated and the following heavy metals were chosen based on an earlier screening procedure: Cu, Zn, Pb, Cr, Fe, Al, and Mn. The research objectives include: (i) characterizing the association of heavy metals with fine particulates (20 to 0.4 m) in stormwater runoff; (ii) evaluating the relative removal of particulate-bound as well as dissolved heavy metals in wet pond and cattail wetland mesocosms; and (iii) investigating the solid-phase chemical associations of heavy metals in cattail wetland sediments by selective sequential extraction procedures and thereby assessing the potential for sediment toxicity and heavy metal bioavailability. The characterization of stormwater runoff showed that Fe and Al were primarily particulate-bound; Mn was mostly in the soluble fraction; and Cu and Zn were associated with fine particulates and in the dissolved fraction. Preliminary results from the first stormwater sampling event conducted in April 2005 showed retention pond and cattail wetland mesocosms to be effective in attenuating Cu, Zn, and Al.

112 (REY-1117-836370) Riparian buffer width, vegetative cover, and nitrogen removal effectiveness: A review of current science and regulations.
Start time: 4:30 PM
Reynolds, S¹, Mayer, P¹, McCutchen, M², Canfield, T¹, ¹ US Environmental Protection Agency, National Risk Management Research Laboratory, Ada, Oklahoma, USA² East Central University, Ada, Oklahoma, USA
Riparian zones, the vegetated region adjacent to streams and wetlands, are thought to be effective at intercepting and controlling nitrogen loads entering water bodies. Buffer width may be related to nitrogen removal efficiency by influencing nitrogen retention through plant sequestration or removal through microbial denitrification. We surveyed peer-reviewed scientific literature containing data on the relationship between riparian buffer width and nitrogen concentration in streams and groundwater of riparian zones to identify trends in the relationship between buffer width and nitrogen removal capacity. We also examined Federal and State regulations regarding riparian buffer widths to determine if such legislation reflects the current scientific understanding of buffer effectiveness. While some narrow buffers (< 15 m) removed significant proportions of nitrogen, others contributed to nitrogen loads in riparian zones. Larger buffers (> 50 m) appeared more certain to remove significant portions of nitrogen. Subsurface removal of nitrogen was efficient but did not appear to be related to buffer width. Surface removal of nitrogen was partly related to buffer width, but was generally inefficient, removing only a small fraction of the total nitrogen flowing through soil surface layers and, in some cases, actually contributing to nitrogen loads. Type of vegetative cover was not related to nitrogen removal effectiveness in the subsurface but was in surface flow. The general lack of vegetation or width effects on nitrogen removal, especially in the subsurface, suggests that soil type, watershed hydrology and subsurface biogeochemistry may be more important factors due to their influence on denitrification. State and Federal guidelines for buffer width also varied widely but generally were consistent with the peer-reviewed literature on effective buffer width, recommending or mandating buffers 15-30 m wide.

113 (LEB-1117-832239) Paradigms lost: TMDLs and the failure of implementation plans.
Start time: 4:50 PM
Lebednik, P.¹, ¹ LFR Levine-Fricke, Emeryville, CA, USA
The advent of TMDL regulatory processes holds substantial promise for reducing impairment of water bodies, particularly those impairments associated with non-point sources, thereby leading to further rehabilitation of the nation's aquatic ecosystems. However, it has become apparent that the paradigm employed in developing technical guidance for performing TMDLs is inappropriate for contaminants with certain environmental characteristics. Use of the existing paradigm in such cases is unlikely to generate results that will accomplish reduction of water body impairments for these contaminants. This prospect is especially problematical because of the time and financial resources that may be expended on these TMDL evaluations and their resulting implementation plans. The purposes of this presentation are to identify those classes of contaminants for which the paradigm is inappropriate and to discuss an alternate TMDL approach for those contaminants that is more likely to provide the information that would be efficacious for reducing impairment. Key aspects of this presentation will be illustrated using information and evaluations associated with 303(d) list contaminants from the San Francisco Bay Estuary.

114 (ZHA-1117-780006) A Chemodynamic Sediment Flux Model in TMDLs Determination of PAHs in Patrick Bayou, Texas.
Start time: 5:10 PM
Kaulen, Mark¹, Zhang, Chunlong¹, ¹ University of Houston-Clear Lake, Houston, Texas, USA
Patrick Bayou, a shallow and tidal tributary of the Houston Ship Channel in Texas, receives treated industrial and municipal wastewater, and non-point source storm water from the surrounding area. It is currently listed as 303(d) impaired water due to sediment toxicity and is designated as a National Priorities List site under the federal Superfund program. Although sediment toxicity in Patrick Bayou remains "undefined" from several studies by local industries and partners, historical data strongly suggested heavy PAHs contamination in sediments. These contaminated sediments serve as a potential source for water column PAHs and can contribute to the overall budget and determination of Total Maximum Daily Loadings (TMDLs). In this study, a chemodynamic model based on resistance-in-series algorithm initially developed by Thibodeaux (1996) is extended and applied to estimate sediment flux under typical conditions of Patrick Bayou. By incorporating a series of compound- and site- specific parameters, this steady state model can estimate sediment flux as a function of several key parameters, such as porewater concentration, partition coefficient, molecular diffusivity, biodensity, water depth, and water flow rate. The model predicts that flux rates under typical conditions of Patrick Bayou are 87, 43, and 11 mg/m²*year for naphthalene, phenanthrene and pyrene, respectively. Converting these sediment flux rates into total mass of PAHs from Patrick Bayou water segment, the contaminated sediments are clearly shown to be a significant input relative to non-point sources that cannot be neglected in TMDLs formulation. The simplicity of this spreadsheet chemodynamic model may imply its utility as an approximate yet cost-effective water quality management tool, particularly for similar slow-moving, shallow waters overlying historically contaminated sediments where bioturbation and water-side mass transfer may become the predominant mechanisms at the water-sediment interface.