Info

improving water quality conditions for these historical problems of the 1950s, 1960s, and 1970s in the Upper Mississippi River. During the 1980s and 1990s, water quality and comprehensive ecological investigations in the Upper Mississippi River have identified a number of contemporary chemical and nonchemical problems in the basin. Nonchemical issues identified as threats to the ecological processes of the river and floodplain ecosystem include, for example, loss of habitat and wetlands and man-made alterations from flood control and navigation projects.

Contemporary chemical problems include inputs of nutrients, sediments, heavy metals, pesticides, and other toxic chemicals. For example, the loading of phosphorus and suspended solids influences water quality in Lake Pepin, a natural impoundment located about 50 miles downstream of St. Paul. Lake Pepin is eutrophic, with high annual mean concentrations of total phosphorus (0.16 mg/L) and soluble reactive phosphorus (SRP) (0.07 mg/L) (James et al., 1996) recorded at the inlet to the lake (UM milepoint 797) during the average flow years of 1994-1996. Eutrophic conditions in the lake are caused by excessive loading of nutrients from point and nonpoint sources in the watershed. When physical and hydrological conditions are favorable, such as during low-flow summers, nuisance algal blooms (i.e., viable chlorophyll-« greater than 30 ^g/L) occur. Concerns related to the need for controls on phosphorus loading arose after severe algal blooms and fish kills in Lake Pepin occurred under the drought conditions of 1987-1988 (Johnson, 1999).

The Lake Pepin Phosphorus Study, conducted from 1994 to 1998, compiled historical and contemporary data sets to evaluate the human impact on (1) long-term records of sediment and phosphorus loading to Lake Pepin and (2) the corresponding water quality responses to changes in loading to the lake. Since European settlement ca. 1830s, the contemporary (1990-1996) annual input of approximately 850,000 metric tons/year (mt/year) of sediment is about ten times greater than the loading rates estimated for the presettlement era. Analysis of data from the three basins included in the study, the Upper Mississippi River, the Minnesota River, and the St. Croix River, indicates that 90 percent of the increased sediment load to Lake Pepin is contributed by erosion of fine-textured soils from the Minnesota River basin. The record of sediment deposition in Lake Pepin also indicates that the most rapid rates of sediment input to the lake occurred during the 1940s and 1950s. If current sedimentation rates continue from erosion in the Minnesota River basin, Lake Pepin could be completely filled in about 340 years (Engstrom and Almendinger, 1998).

Over the past two centuries, phosphorus concentrations in the sediments of Lake Pepin have increased twofold, while water column concentrations (inferred from diatom assemblages in the sediments) appear to have increased by a factor of 4 since European settlement ca. 1830s. Increased phosphorus levels in the sediments and water column are the result of an increase in phosphorus loads to Lake Pepin by a factor of 5 to 7 since the 1830s to the contemporary estimated loading rate of approximately 4,000 to 5,000 mt/year for 1990-1996. Wastewater discharges and agricultural applications of manure and commercial fertilizer are most likely the key factors controlling historical phosphorus loads to Lake Pepin, and the statewide ban on phosphates in detergents contributed to a reduction in phosphorus loading from municipal wastewater plants by approximately 40 percent over the period from 1970 to 1980.

Since the 1830s era, the progressive increase in phosphorus loading has resulted in a shift in assemblages of diatoms from clear water benthic algae and mesotrophic water column species in the presettlement era to planktonic species exclusively characteristic of highly eutrophic conditions in the 1990s (Engstrom and Almendinger, 1998).

In evaluating strategies to reduce phosphorus loads to Lake Pepin, the significant differences in the relative contributions of point and nonpoint sources of flow, solids, and nutrient loads under a range of flow conditions need to be considered over a time scale of decades. Point and nonpoint source loading data for suspended solids and total phosphorus have been compiled for low-flow (1988), average-flow (19941996), and high-flow (1993) conditions for the Upper Mississippi River (upstream of Lock and Dam No. 1), the Minnesota River, and the St. Croix River (Meyer and Schellhaass, 1999). Based on 21 years of data (1976-1996), the mean yield of total phosphorus from the agriculturally dominated Minnesota River (0.33 lb/acre-year) is twice as great as the mean yield from the Upper Mississippi River basin upstream of Lock and Dam No. 1 (0.16 lb/acre-year) and the St. Croix River basin (0.14 lb/acre-year). Figure 12-16 presents a comparison of the magnitude of point source loads and nonpoint source loads of total phosphorus from the Upper Mississippi River, Minnesota River, and St. Croix River basins for 1988 (drought), 1993 (flood), and 19941996 (average conditions) (Meyer and Schellhaass, 1999).

Under the extreme flood conditions of 1993, nonpoint source loadings of total phosphorus from the Minnesota River and the Upper Mississippi River watersheds have been shown to account for 58 percent and 15 percent, respectively, of the total phosphorus load of 6,030 mt/year estimated for 1993, while point sources from the Metro plant accounted for 15 percent of the total phosphorus load. During the severe

600Cb

600Cb

1988 1993 1994-96

Figure 12-16 Comparison of total phosphorus loadings from nonpoint sources (NPS) in the Upper Mississippi River (UM, to Lock & Dam No. 1), Minnesota River (MI), and St. Croix River (SC) basins and point source (PS) loadings from the Metro plant and other facilities in the three river basins. Source: Meyer and Schellhaass, 1999.

1988 1993 1994-96

Figure 12-16 Comparison of total phosphorus loadings from nonpoint sources (NPS) in the Upper Mississippi River (UM, to Lock & Dam No. 1), Minnesota River (MI), and St. Croix River (SC) basins and point source (PS) loadings from the Metro plant and other facilities in the three river basins. Source: Meyer and Schellhaass, 1999.

drought conditions of 1988, the total phosphorus load of 1,900 mt/year was only about one-third of the 1993 load. Under the drought conditions, the Metro plant accounted for 47 percent of the total phosphorus load, and nonpoint source loading from the Minnesota River and the Upper Mississippi River contributed only 6 percent and 3 percent, respectively, of the total phosphorus load of 1,900 mt/year. During the average flow conditions of 1994-1996, the total phosphorus load of 3,800 mt/year was two times greater than the 1988 drought load. Under average flow conditions, the Metro plant accounted for 28 percent of the total phosphorus load, and nonpoint source loading from the Minnesota River and the Upper Mississippi River contributed 38 percent and 14 percent, respectively, of the total phosphorus load of 3,800 mt/year.

Meyer and Schellhaass (1999) have used this data set to develop summary budgets of the relative contributions of point source and nonpoint source loadings of total phosphorus to the three river basins during 1988, 1993, and 1994-1996. Under the drought conditions of 1988, the contribution from point sources (88.5 percent) dominated the total inputs of phosphorus, compared to the 11.5 percent accounted for by nonpoint sources. During the extreme flood conditions of 1993, nonpoint source loads accounted for about three-quarters (74.5 percent) of the total input of phosphorus, with point sources accounting for about one-quarter (25.5 percent). During the average flow conditions of 1994-1996, the relative contribution of point sources (56.2 percent) and nonpoint sources (43.8 percent) was almost comparable.

These point source loading and nonpoint source loading data sets for suspended sediments and phosphorus and a number of other field studies (e.g., James et al., 1999) have been used to support the development of an advanced model of hydrodynamics, sediment transport, and eutrophication for the Upper Mississippi River and Lake Pepin (HydroQual, 1999a, 1999b; Garland et al., 1999). As of 2000, the MCES is using the model to evaluate the effectiveness of alternative strategies to control point and nonpoint phosphorus loading to the Upper Mississippi River to achieve the water quality objectives established for Lake Pepin. Evaluations of sediment loading contributed primarily from agricultural runoff in the Minnesota River basin have also been a key issue in the Minnesota River Assessment Project (MPCA, 1994).

On the much larger scale of the entire Mississippi River basin, nitrogen loading from the Mississippi River has been identified as a major cause of the algal blooms and hypoxia that occur over a 16,000-square-kilometer area of the inner Gulf of Mexico known as the "Dead Zone" (Christen, 1999; Malakoff, 1998; Moffat, 1998; Rabelais et al., 1996; Vitousek et al., 1997). Based on technical assessments of the "Dead Zone" problem, a USEPA and NOAA Action Plan, released in January 2001, recommended that efforts be undertaken to reduce inputs of nitrogen from waste-water treatment plants and agricultural land uses (e.g., fertilizer applications and confined animal feedlots) over the entire Mississippi River basin, which drains 40 percent of the land area of the continental United States (Christen, 1999; Mississippi River/Gulf of Mexico Watershed Nutrient Task Force, 2001).

The series of locks and dams and maintained navigation channel have been an integral physical feature of the Upper Mississippi River since the early 1930s when the U.S. Congress authorized the U.S. Army Corps of Engineers to maintain the river for navigation purposes. Concerns have been raised about the disposal of dredged sedi ments, often contaminated with heavy metals and toxic chemicals, to maintain the navigation channel and the loss of ecologically critical backwater habitats to sediment deposits. The devastation caused by the Great Flood of 1993 (Wahl et al., 1993) in the upper Midwest has also triggered debates about the failure of flood control measures intended to protect river communities from floods. As the key federal agency responsible for inland waterways, the U.S. Army Corps of Engineers has initiated controversial studies to evaluate the ecological impact of maintenance dredging, flood control structures, and widening the series of locks and dams (Phillips, 1999).

Evaluation of Water Quality Benefits Following Treatment Plant Upgrades

From a policy and planning perspective, the central question related to the effectiveness of the secondary treatment requirement of the 1972 CWA is simply: Would water quality standards for DO be attained if primary treatment levels were considered acceptable? In addition to the qualitative assessment of historical data, water quality models can provide a quantitative approach to evaluate improvements in dissolved oxygen and other water quality parameters achieved as a result of upgrades to secondary and greater levels of wastewater treatment. Since the 1970s, increasingly complex models have been developed to determine wasteload allocation requirements for municipal and industrial dischargers to meet the needs of decision-makers for the Upper Mississippi River.

During the mid-1970s, the National Commission on Water Quality (see Appendix A-6) funded Water Resources Engineers (WRE, 1975) to develop a steady-state, one-dimensional water quality model (QUAL-II) of DO, BOD5, nutrients, and fecal coliform bacteria, using data collected in the Upper Mississippi River in 1964-1965 (FWPCA, 1966). The model was applied to evaluate the effectiveness of the technology-based requirements of the 1972 Clean Water Act for municipal and industrial dischargers. With funding available from the CWA Section 208 program, Hydroscience (1979) developed a water quality model (AESOP) of DO, BOD5, nutrients, algae, and bacteria using data collected in 1973, 1976, and 1977. The model, further validated by the Minnesota Pollution Control Agency using data obtained in 1980, was used for a wasteload allocation study of the Metro plant's impact on DO and un-ionized ammonia in Pool 2 (MPCA, 1981).

As a result of the severe algal blooms and fish kills that occurred in Lake Pepin during the extreme drought of 1988, a time-variable water quality model (WASP5-EUTRO5) of DO, BOD5, nutrients, and algae was developed using data collected during 1988 (MWCC, 1989), 1990, and 1991 (EnviroTech, 1992, 1993; Lung and Larson, 1995). The validated model was used to evaluate alternatives for phosphorus controls at the Metro plant and to perform a postaudit of the Hydroscience (1979) AESOP model using low-flow data collected during 1988 (Lung, 1996a). The model was also applied to track the fate and transport of phosphorus and the relative impact of the point and nonpoint sources on eutrophication in Lake Pepin (Lung, 1996b).

Following completion of the model by EnviroTech (1992, 1993), a number of uncertainty issues were identified related to: (1) fate and transport of phosphorus from point and nonpoint sources; (2) interaction of suspended solids with phosphorus transport; and (3) interaction of nonpoint source phosphorus inputs generated under low-flow and high-flow hydrologic conditions with interannual variation in the benthic release of phosphorus. To address these issues, a three-dimensional hydro-dynamic, sediment transport, and advanced eutrophication model was developed and calibrated using data collected over 12 years from 1985 through 1996 (Garland et al., 1999; HydroQual, 1999a, 1999b). The calibrated model was used to simulate the long-term (24-year) water quality response in the Upper Mississippi River and Lake Pepin to a number of alternative control scenarios over a range of hydrologic (e.g., dry and wet years) and loading conditions for point source and nonpoint source discharges of phosphorus.

To evaluate the incremental improvements in water quality conditions that have been achieved by upgrading municipal wastewater plants from primary to secondary and from secondary to advanced secondary levels of wastewater treatment, Lung (1998) used the WASP5-EUTRO5 model developed by EnviroTech (1992, 1993) to demonstrate the water quality benefits attained by the secondary treatment requirements of the 1972 CWA. Using the model, municipal and industrial wastewater flow and effluent loading data were used with boundary flow and loading data describing the Upper Mississippi River and Minnesota River to compare water quality conditions for three summers (1964, 1976, and 1988) characterized by comparable low-flow conditions and primary (1964), secondary (1976), and advanced secondary (1988) levels of wastewater treatment at the Metro plant. The model was applied to evaluate the water quality impact of three different treatment levels for Metro and the other municipal plants: (1) primary, (2) secondary, and (3) advanced secondary with nitrification. CBOD oxidation rates were calibrated for each of these three different data sets to reflect differences in the proportion of labile and refractory oxidizable material discharged from the Metro plant.

A comparison of the results of the model runs and observed data sets is presented in Figure 12-17. Spatial distributions of CBOD-ultimate, ammonia-N, nitrate + ni-trite-N, algal chlorophyll-a, and DO are presented from St. Paul (UM milepoint 840) to Lock & Dam No. 2 (UM milepoint 815) for 1964 (primary), 1976 (secondary), and 1988 (advanced secondary with nitrification) loading and flow conditions. The upgrade of the Metro plant from primary to secondary and the corresponding reduction of effluent BOD5 loading (see Figure 12-9) is reflected in the decrease in ambient CBOD from a peak of approximately 20 mg/L in 1964 to approximately 7 to 8 mg/L in 1976 at UM milepoint 835 near the Metro plant. As shown in the simulation results, the distributions of ammonia-N and nitrate + nitrite-N are similar under the primary and secondary treatment scenarios because upgrading from primary to secondary treatment does not change the effluent concentration of ammonia. The progressive reduction in ambient ammonia-N and corresponding increase in ambient nitrate + nitrite-N for the 1988 simulation, however, reflect the impact of the upgrade from secondary to advanced secondary with nitrification and the drop in effluent loading of ammonia-N at the Metro plant (see Figure 12-11).

During the 1960s, when Metro discharged primary effluent, a large section of the river was hypoxic or anoxic, with the worst conditions (< 2 mg/L) observed over ap-

Figure 12-17 Improvement in ultimate CBOD, ammonia-N, and DO levels in the Upper Mississippi River related to Metro treatment plant upgrades from primary to secondary and advanced secondary with nitrification. Source: Lung, 1998. Reprinted with permission of the American Society of Civil Engineers. Copyright 1998.

Figure 12-17 Improvement in ultimate CBOD, ammonia-N, and DO levels in the Upper Mississippi River related to Metro treatment plant upgrades from primary to secondary and advanced secondary with nitrification. Source: Lung, 1998. Reprinted with permission of the American Society of Civil Engineers. Copyright 1998.

proximately 15 miles from UM milepoint 820 to UM milepoint 835. The observed data and the model results indicate the elimination of anoxic conditions and a nominal improvement in DO conditions under the extreme low-flow conditions of August 1976. The minimum DO level is increased from approximately 0.5 mg/L in 1964 to approximately 2 mg/L in 1976 as a result of the upgrade from primary to secondary treatment. Even with secondary treatment at Metro, however, compliance with the water quality standard for dissolved oxygen of 5 mg/L was not achieved, and a distinct oxygen sag is observed in the 1976 data set. Compliance with the DO standard was finally achieved, even under the extreme drought conditions of 1988, after Metro was upgraded from secondary to advanced secondary treatment with nitrification.

The model results demonstrate very clearly the progressive increase in DO levels in the river following the upgrades at Metro to secondary and advanced secondary treatment. The model results also demonstrate the ability of a well-calibrated model to match observed water quality distributions that are directly related to changes in effluent loading from Metro under the three different treatment levels. The data used to define the effluent flow and loading characteristics for the primary, secondary, and advanced secondary treatment levels for the 1964, 1976, and 1988 simulations are given in Lung (1998). The data used to define effluent flow and loads from the other municipal and industrial point sources and the boundary inputs from the Upper Mississippi River and the Minnesota River are summarized by WRE (1975) for 1964 and by EnviroTech (1992, 1993) for the 1976 and 1988 simulations.

In generating the simulation results for the three different treatment scenarios, all model coefficients, except the CBOD oxidation rate, are based on the same numerical values for each of the three model runs. The in-stream oxidation rate for CBOD (Kd) is assigned different values for primary (0.35/day), secondary (0.25/day), and advanced secondary with nitrification (0.07/day) treatment levels, since this kinetic reaction rate is dependent upon stabilization of the effluent and the quantity of labile and refractory components of oxidizable organic matter in the effluent (Chapra, 1997; Thomann and Mueller, 1987). Using effluent loading rates that are representative of the three different treatment levels for Metro, the model results confirm that the improvement in water quality observed in the Upper Mississippi River can be attributed to investments made to upgrade the Metro plant.

Recreational and Living Resources Trends

Long-term trends in recreational uses, private investments along the riverfront, and biological resources dependent on the integrity of aquatic ecological conditions are meaningful nonchemical indicators of water quality conditions in the Upper Mississippi River. One very simple indicator is the use of the river for recreational boating. If water quality conditions are very poor, as was the case during the 1950s and 1960s, the noxious conditions are not desirable for boating as a recreational activity. If water quality is not degraded, the river might be considered desirable for boating. As shown in the long-term trend of recreational boat traffic through Locks 1 through 4 of the river (Figure 12-18), annual recreational vessel usage of the river ranged from approximately 25,000 vessels to approximately 30,000 vessels from the mid-1970s through the mid-1980s. Beginning in the mid-1980s, the improvements in water quality in the Upper Mississippi River suggest a strong correlation with the dramatic increase in annual recreational vessel traffic on the river to approximately 45,000 to approximately 53,000 boats (Erickson, 2000), with the recreational vessel traffic in Locks 1 through 4 increasing by about two-thirds between 1986 and 1998 (Figure 12-18). Note that traffic in 1993 dropped by about one-half because of the extreme flood conditions of that year.

Recreational boats require marina space, and in 1990 about 2,700 new marina slips were in various planning stages—enough to double marina capacity. The number of permit applications received by the St. Paul District U.S. Army Corps of Engineers for docks, marinas, boathouses, boat ramps, and beach and wildlife improvements soared from only 3 in 1981 to 22 by 1989 (Figure 12-19). In the late 1970s, nobody would have considered investing in a marina in Pool 2 because of poor water quality conditions in the vicinity of St. Paul. Apparently related to improvements in

Figure 12-18 Recreational vessel traffic in Locks 1 through 4 of the Upper Mississippi River. Source: Erickson, 2000.
Figure 12-19 Recreational permit applications for Wabasha, Dakota, Washington, Goodhue, Pierce, and Pepin counties along the Upper Mississippi River. Source: Erickson, 2000.

water quality conditions, several marinas were proposed and constructed for this area of the river beginning in the mid-1980s. Lake City, located on Lake Pepin, for example, obtained a permit for a new marina in 1984; within a year, several hundred spaces were added for sailboats.

Increases in recreational uses of the river prompted eight agencies to form a partnership agreement in 1990 to study recreational trends and resolve conflicts over river and parkland use. The agencies included two park services, three state departments of natural resources, the U.S. Fish and Wildlife Service, the U.S. Army Corps of Engineers, and the Minnesota-Wisconsin Boundary Area Commission. They conducted a study to sort out the issues, uses, and resource management conflicts related to the rediscovery of the delights of a cleaned-up river by boaters, fishermen, and hikers (MPCA, 1993).

Partly because of the ban on DDT, the establishment of wildlife reserves, and reduced loadings of industrial pollutants from the pretreatment program, populations of water birds have increased in the Upper Mississippi River. Peregrine falcons, bald eagles, mallard ducks, and great blue herons have been observed in the Minneapolis-Saint Paul metropolitan area and in the floodplain wetlands located on the Upper Mississippi River near the Metro plant. Black crowned night herons have been observed feeding below the Ford Dam (Galli, 1992). Animals sensitive to the bioaccumulation of PCBs in their aquatic food, such as fish-eating mink, are also making a comeback (Smith, 1992). The number of great egrets and great blue herons nesting in Pig's Eye Lake has increased since the late 1970s and early 1980s, and cormorants has been observed nesting in the lake since 1983 (Galli, 1992) (Figure 12-20).

0 0

Post a comment