THE TRIALS AND TRIBULATIONS OF DEVELOPING TMDLS FOR DIOXIN AND PCBS DOMINATED BY UNDEFINED, NON-POINT SOURCES IN THE OHIO RIVER

The Ohio River is a large river system spanning 981 miles in length, with a drainage basin of over 200,00 square miles covering portions of 15 states. The entire length of the Ohio River is listed as impaired due to elevated PCBs (polychlorinated biphenyls) in fish tissue, and dioxin concentrations well above the established ambient water quality standards (WQS). Due to the fish consumption use impairment, Total Maximum Daily Loads (TMDLs) for both PCBs and dioxin are required for the full length of the Ohio River. Developing TMDLs for a river with such an expansive watershed, however, poses significant challenges due to the large geographical area that must be considered. The problem is compounded when dealing with pollutants, such as dioxins and PCBs, in which the contamination is widespread and the numerous sources are poorly defined and diffuse in nature. Due to the interstate implications of these TMDLs, the Ohio River Valley Water Sanitation Commission (ORSANCO) has taken the lead role in developing the TMDLs on behalf of the affected states and the United States Environmental Protection Agency (US EPA).The main goal of this study is to develop meaningful TMDLs that result in improved water quality through the reduction of PCBs and dioxin loadings to the Ohio River. The key elements of this effort include: 1) define existing concentrations and loadings of PCBs and dioxin in the Ohio River; 2) quantify load reductions necessary to meet applicable water quality standards; 3) identify sources by reviewing databases of known contaminated sites and targeted monitoring; 4) quantify source inputs including loadings from tributaries, atmospheric deposition, and point sources when possible; and 5) assign load allocations (LA) and wasteload allocations (WLA) to meet the necessary load reductions.There are numerous challenges that must be addressed to develop large-scale TMDLs for legacy pollutants such as PCBs and dioxin. The first hurdle that must be overcome is simply being able to detect the contaminants in the water column as ambient water concentrations for the two pollutants are typically well below normal analytical detection limits. ORSANCO was able to quantify in-stream concentrations through use of a relatively new sampling technique referred to as high-volume water sampling. This method involves filtering a large volume of water (1000 liters) first through a glass fiber filter and then an XAD resin column in order to concentrate the sample to allow for significantly lower detection limits. When combined with the best high-resolution analytical methods, this sampling technique allows for contaminants to be quantified down to the low femtogram/liter (parts per quintillion) level, which is below the ambient water quality standards for dioxin (0.005 fg/L) and PCBs (64 pg/L).Source identification is also problematic for PCBs and dioxin as the contamination is widespread, and isolating specific sources is difficult. There are numerous routes for non-point source loadings to enter the Ohio River, all of which are poorly defined. In addition, very few point sources have been confirmed. ORSANCO took a multi-media approach to better characterize sources. Ambient air monitoring was conducted at 20 locations along the river in order to quantify ambient air concentrations, identify possible hot spots of air contamination, and provide a gross estimate of loadings from atmospheric deposition. A large-scale bottom sediment survey spanning the entire Ohio River was completed to measure concentrations of PCBs and dioxin in sediment and to locate hot spots of contamination, which could lead to identifying sources. High-volume water sampling was also conducted at 11 POTWs (Publicly Owned Treatment Works) to characterize potential source loadings from municipal wastewater treatment plants. Other potential sources were identified through personal communications with state personnel and queries of the CERCLIS (Comprehensive Environmental Response, Compensation, and Liability Information System), TRI (Toxic Release Inventory), and PCS (Permit Compliance System) databases.A total of 207 high-volume water sampling events were conducted at 34 Ohio River locations and 25 tributary sites. Results indicate that ambient WQS for both PCBs (64 pg/L) and dioxin (0.005 pg/L) were exceeded at all locations for all conditions sampled. Ohio River concentrations ranged from 540 to 12600 pg/L for PCBs, while dioxin concentrations ranged from 0.033 to 2.25 pg/L TEQ (Toxicity Equivalents). The most elevated concentrations occurred during high flow/high suspended solids conditions, indicating that overland runoff and resuspension of sediments play a dominant role under those conditions.Sediment sampling confirmed widespread, low-level contamination throughout the Ohio River, with several hot spots of contamination identified. A follow-up survey to target the areas with elevated concentrations was recently completed, and these results will be included in the final TMDLs, which are scheduled for completion by the end of 2005. Ambient air monitoring found concentrations ranged from 21.6 to 4940 pg/m3 for PCBs, while dioxin levels ranged from 4.63 to 3160 fg/m3 TEQ. The final TMDL will include estimated pollutant loadings from atmospheric deposition directly entering the Ohio River. High-volume water sampling conducted at POTWs revealed the presence of high levels of PCBs and dioxin relative to the ambient WQS. Concentrations for PCBs and dioxin in POTW effluent samples averaged 3660 and 0.404 pg/L TEQ, respectively. The results from the 11 facilities monitored are likely not unique to just those facilities, and may indicate that POTWs collectively discharge sizeable loads of the two pollutants to the Ohio River.In-stream monitoring revealed that PCBs and dioxin loads in the Ohio River must be reduced by as much as 99 percent in some areas. While great strides have been made to identify and quantify sources, such as atmospheric deposition and POTWs loadings, there are still significant sources undefined. Because of the sheer size of the watershed, and the diffuse nature of the sources, in the end, these TMDLs will result in load allocations to each of the major sub-basin that drain to the Ohio River. These load allocations represent the sum of all allowable loads from point and non-point sources. Implementation of the TMDLs would require follow-up studies in each sub-basin to assign LA and WLA to specific sources. The discovery of POTWs as sources of PCBs and dioxin does present a mechanism for some reductions. These TMDLs may include WLA for all 180+ POTWs that discharge to the Ohio River based on the sampling results from the 11 facilities monitored. Any solutions though involving modifications to POTWs would be extremely expensive to implement and met with much controversy. These TMDLs for the Ohio River establish the goals that must be met to achieve the water quality objectives, but meaningful reductions can only be attained if efforts are continued to better characterize the sources.