Environmental impacts associated with human activities have typically been considered on a project–specific basis, and with respect to water quality, those impacts have often been regulated by enforcing local end-of-the-pipe discharge limitations. Recently, policy mandated through the National Environmental Policy Act (NEPA) recommends that regulatory agencies responsible for licensing industrial projects require that environmental impacts be viewed more holistically, and assessed cumulatively, both in time and space (CEQ 1997). Spaling (1997) broadly defines cumulative environmental impacts (CEIs) as the accumulation of man-induced changes in valued environmental components across space and over time. Irving et al. (1986) characterized CEIs by type and pathway, noting that specific impacts may result in additive, synergistic or countervailing (less than the sum of individual impacts) effects.

Currently, no standard protocol for assessing CEIs has been adopted or proposed, although several papers are available that outline possible approaches and strategies for addressing cumulative effects (Barrow, 1997; Court et al., 1994; and Vestel et al., 1995). Often, the primary challenges of assessing CEIs are (1) identifying the anticipated environmental concern or issue, and (2) developing a systematic and technically sound assessment strategy.

The approach employed in this analysis to address the issue of CEIs in the Catawba River Basin involves investigating the relationship between population statistics and water quality information, at the watershed level. The Catawba River Basin, like most areas in the Southeast, has experienced unprecedented population growth over the last decade resulting in an increased demand on the region’s natural resources, including its aquatic resources. The approach employed here is not new. Variations of it have been employed by water resource managers throughout the world to identify, quantify, and combat water quality problems associated with human activities (Horne and Goldman, 1994).

The specific strategy involves two components: (1) assessment of pertinent historical data, and (2)application of a numerical water quality model. The first approach involves applying descriptive and statistical techniques to historical water quality data collected in each of the 11 Catawba reservoirs over the last three decades, and relating spatial and temporal trends in water quality parameters to various environmental and population metrics. The second component involves application of a numerical watershed-based water quality model that allows for the simulation of both historical and future water quality conditions based on various environmental and population scenarios.

Back to Top

Cultural eutrophication, the accelerated nutrient enrichment of aquatic ecosystems due to human activities, is generally considered the most evident and pervasive water quality problem facing the world today (Welch, 1980, Cooke et al., 1986, Horne and Goldman, 1994). Excessive nutrient enrichment of freshwaters often stimulates aquatic plant growth which, in turn, leads to a series of interrelated physical, chemical, and biological reactions that ultimately culminate in impaired recreational, industrial, and/or domestic use of that waterbody. Anthropogenic sources of nutrients, especially phosphorus, which is most often the primary element limiting plant growth in freshwaters, has been shown by numerous studies to be tightly linked to human growth and distribution (Wetzel, 1975; Welch, 1980). Cultural eutrophication is considered one of the primary water quality issues facing water resource management agencies in both North Carolina (NCDEHNR, 1992) and South Carolina (SCDHEC, 1991).

This study focuses on those water quality constituents that have been shown to be linked, either directly or indirectly, to population attributes, and therefore serve as indicators of the degree and extent of cultural effects (Wetzel, 1975; Welch,1980). Several water quality constituents could have been used in this approach, but only those constituents that have been consistently collected over several decades by one or more investigating groups in the Catawba River corridor were selected for analysis. These include total phosphorus and chlorophyll a, often called non-conservative parameters because they are strongly influenced by biological metabolic activity, and conductivity and chloride, often classified as conservative constituents because biological metabolism generally exerts a minimal impact on their levels. In freshwater ecosystems, increases in conductivity and chloride levels generally pose minimal environmental concern but can serve as reliable indicators of human influences, especially for point source discharges from industrial sources or municipalities. Conversely, excessive increases in nutrients and aquatic plant levels typically lead to water quality degradation.

Historical population and water quality data for the Catawba River reservoirs are available from several agencies, including private, government and academia. Population information and projections for the period 1960 through 2015 were assembled from several sources, including US census data (CCC, 1999; NCDENR, 1992; NCDENR, 1999). The most comprehensive water quality data set consists of physical, chemical and biological data collected by Duke Power on the Catawba River reservoirs since the early 1970’s. This program consists of quarterly or bi-annual water quality measurements taken at several mainstem locations in each of the eleven reservoirs. Other comparable long-term water quality data sets for the Catawba River reservoirs are available from water resources agencies in both North Carolina and South Carolina (SCDHEC, 1991; NCDENR, 1992; NCDENR, 1999). These data sets, although generally not as extensive in time and space as Duke Power's database, date back to the 1980’s and therefore serve as excellent reference sets. Another valuable data set that has application for this analysis originates from an Environmental Protection Agency (EPA) water quality study performed in 1973-1974. The focus of this investigation was to assess the trophic status of the Catawba River reservoirs, and included information on algal pigments and nutrient levels and sources (USEPA, 1975).

Back to Top

The Catawba River Basin is the most densely populated river basin in North Carolina, and like many regions of the Southeast, has experienced unprecedented growth over the last decade. Based on the 1990 census data, the watershed population was estimated at about 1.2 million individuals (Figure 4-1), with most of this population centered in the middle and upper-lower portions of the basin in and adjacent to major urban centers such as Charlotte, North Carolina’s largest city (Table 1, DPC unpublished). In North Carolina, nineteen percent of the State’s total population was found in the Catawba River Basin (NCDENR, 1999). Population density within the Catawba River Basin, or individuals per square mile (based on the 1990 data), was estimated (DPC unpublished) to be the greatest in the Fishing Creek sub-watershed (658) and least in the Lake Wateree sub-watershed (17) (Figure 4-1).

US Census population information for select counties in the Catawba River corridor over the period 1960 to 1990 showed an average increase of about 44 percent, with the highest growth (75-88%) found in Lincoln, Mecklenburg, and Union Counties (Table 4-1). Various growth projections indicate that regional growth will continue to progress at a rate similar to that observed over the last decade (CCC, 1999; NCDENR, 1999). Based on these projections, the Catawba River Basin will contain well over 2 million residents by 2000, with a significant portion centered in the Mecklenburg County region.

A distinct up-river to down-river trend of decreasing water quality (increasing total phosphorus, chlorophyll a, conductivity and chloride levels) is observed for the Catawba River watershed, based on Duke Power’s data set (Figure 4-2,). Generally speaking, concentrations of all constituents are lowest in the North Carolina reservoirs and increase appreciably in the South Carolina reservoirs. This spatial trend is supported by other water quality data sets for the Catawba River (USEPA, 1974; SCDHEC, 1991; NCDENR, 1992; NCDENR, 1999; DPC, 2000), and has exhibited spatial consistency for several decades. Population influences (Table 4-1), particularly those related to the magnitude of point and non-point sources of chemical inputs to these waterbodies, are believed to be the primary determinate of this spatial trend (USEPA, 1974; SCDHEC, 1991; NCDENR, 1992; NCDENR, 1999). Other factors which may modulate the influence of population effects by either increasing or decreasing the assimilation efficiency of the respective waterbody include sub-watershed area, and reservoir depth, volume and retention time. Longitudinal gradients in water quality along many of the major river systems throughout the world have been observed, and the role of human activities in influencing water quality gradients are well documented (Cooke et al., 1986; Horne and Goldman, 1994).

Temporal trends in water quality parameters within the Catawba River Basin appear to be species specific (Figure 4-3). Total phosphorus levels have not exhibited any statistically significant upward or downward trends in any of the waterbodies over the last 20 to 30 years despite appreciable population increases in almost every sub-watershed. (No comparable long-term data set for chlorophyll a was available from any agency or group. Duke Power did not include chlorophyll a as a component of their monitoring program until the late 1980s). These data are corroborated by results from other long-term water quality studies on these waterbodies (NCDENR, 1992; SCDHEC, 1991; DPC, 2000). The anticipated degradation of water quality often associated with increased population levels was not observed, at least in the mainstem locations of these waterbodies. Apparently the increased nutrient loadings expected from increased population levels were off-set by improvements in point and non-point nutrient management practices which resulted in either stabilizing or possibly decreasing total nutrient loads to the reservoirs. These improvements included such things as modifications in landuse practices aimed at minimizing non-point runoff from the adjacent terrestrial landscape, and technological upgrades to municipal wastewater treatment facilities resulting in a smaller percentage of human wastes being discharged to the receiving aquatic ecosystems (NCDENR, 1999).

Back to Top

A word of clarification is critical at this point. Most of the long-term water quality monitoring data available for these reservoirs represent information collected from what is typically referred to as "mainstem" locations. These locations are generally located in the deeper, open-water portions of the waterbody, and are selected as monitoring sites because they tend to represent the composite status of water quality in the main sections of the reservoir. Generally, the open-water or mainstem portions of the reservoir contain the majority of the waterbody’s total volume. Consequently, these sites serve as reliable indicators of the reservoir’s overall ability to assimilate or process material imported into the system from the surrounding watershed. Mainstem locations are not intended for and should not be used to characterize "worst case" conditions in the waterbody, or as a zone of "first alert." Isolated parcels of water that are found in embayments and coves are generally more responsive to localized perturbations and loading influences because of morphometric and hydraulic constraints, and therefore are more variable in time and space than mainstem locations. Soballe (1998) provided a critical review of various water quality sampling programs and outlined the components necessary for implementing a successful long-term monitoring program. He reiterated a theme often forgotten by stakeholders, and water quality managers and researchers: "Monitoring programs cannot measure all important water quality parameters at all times and places. Monitoring data represent an abstraction or sample that generally includes characteristics from a small fraction of a larger and more complex system. Consequently, data have limitations and should be used only to answer those questions that are consistent with their original purpose."

In contrast to total phosphorus, a non-conservative chemical parameter, both conductivity and chloride levels were consistently higher in the 90’s than in the 70’s and 80’s. This pattern was observed in every reservoir within the Catawba River Basin except in Lake James, the upper most waterbody. Excluding Lake James, mean conductivity increased by 30 umhos/cm and chloride increased by an average of 2.9 mg/L over the period 1975-1997. Both chloride and conductivity measurements represent individual or composite measurements of ions that are by-products of human waste and consequently have been shown to provide a useful measure of cultural attributes on water quality, especially within similar geological regions.

WARMF (Watershed Analysis Risk Management Framework) is a decision support system that has been applied to the Catawba River Basin in North and South Carolina. The Catawba River application was developed as a joint venture between Systech Engineering, Inc. of San Ramon, California, the Electric Power Research Institute (EPRI) and Duke Power Company.

WARMF is an educational tool that stakeholders can use to determine how meteorology generates hydrology and non-point loads; how land use affects non-point loads; how point and non-point loads are spatially distributed; how point and non-point loads translate to water quality in rivers and lakes; and whether the water quality is suitable for intended uses.

Back to Top

The engineering module in WARMF contains catchment, river, and reservoir models. The catchment model accepts daily meteorological data and monthly air quality data, and simulates canopy processes, infiltration into the ground, ex-filtration of ground water, surface runoff, and associated non-point source load. The river model accepts outflow from catchments and point source discharges, and routes the hydrology and water quality from one stream segment to the next until the water flows into a reservoir. The reservoir model accepts inflow from adjacent rivers and catchments, direct point source discharges, non-point sources, and direct deposition, and simulates thermal stratification and water quality in the lake and its outlet. The outflow from a reservoir typically enters a river, which may flow into the next reservoir. The interface between models is automatic and seamless. It allows WARMF to represent an entire watershed as a single, interconnected system. WARMF outputs hydrology and water quality throughout the river basin.

For this study, WARMF was used to determine the cumulative impact that point source dischargers (located within Duke Power’s project boundaries) have on nutrient loading rates. Specifically, loading rates for total phosphorous and total nitrogen were determined under three different scenarios:

1. Base Case – all point source dischargers are included in the model;
2. Duke Power Facilities with NPDES permitted outfalls are removed from the model (i.e., 3 fossil stations and 2 nuclear stations); and
3. All NPDES dischargers within Duke Power’s project boundaries are removed from the model. Note that 55 individual dischargers (some with multiple outfall locations) were removed from the Base Case for this scenario.

The cumulative loading rates for phosphorous and nitrogen from scenarios 2 and 3 were then compared with the Base Case. These comparisons allow the determination of how much nutrient loading is reduced as point source dischargers are systematically removed from within the project boundaries. The following two tables provide the percent of phosphorous and nitrogen removal associated with point sources within Duke Power’s project boundaries.

Table 4-2 compares the results of Scenario 2 (removal of Duke Power facilities) with the Base Case. Cumulative reductions in total phosphorous and total nitrogen are provided.

Removal of Duke Power facilities has very little to no impact on nutrient loading to the three reservoirs where they are located. Even in Mountain Island Lake, an increase of 3% in total phosphorous and a decrease of 5% in nitrogen are within the error range of the model.

Table 4-3 compares the results of Scenario 3 (removal of all dischargers within Duke Power’s project boundaries) with the Base Case. Cumulative reductions in total phosphorous and total nitrogen are provided.

Removing 55 point sources in the Catawba River Basin has very little impact on phosphorous reductions. The range is from 0% reduction in Lake James to an 8% reduction in Lake Wylie. Because removal of point sources has such a small impact on reducing phosphorous, these results indicate that a majority of the loading is either from non-point sources or from point sources outside of Duke Power’s project boundaries. In addition, phosphorous entering the reservoirs from non-point sources is typically bound with sediment and is not bioavailable.

With the exception of Lake James, nitrogen reductions in the upper and middle portions of the basin are more significant percentage-wise than were the reductions in total phosphorous. Lake James is at 0% due to the minimal amount of development and point source dischargers in that sub-basin. Lake’s Rhodhiss, Hickory, Lookout Shoals, Norman, Mountain Island, and Wylie average a 20% reduction in nitrogen. This is primarily from the elimination of wastewater dischargers in those reservoirs. Because these lakes are nutrient limited by phosphorous, and not nitrogen, a 20% reduction in nitrogen levels is not expected to have a significant water quality impact.

Below Lake Wylie, average nitrogen reductions are only in the 8% range. For Fishing Creek Lake, this is because the larger wastewater treatment facilities in that area discharge into tributaries above Duke Power’s project boundaries. Thus, these facilities were left in the model simulation. Below Fishing Creek Lake, the small percentage reductions in nitrogen loading are due to the large amount of nitrogen entering the reservoirs from non-point sources. Non-point nitrogen loading overshadows the amount coming in from point sources.

In summary, removing only the Duke Power facilities (Scenario 2), or all of the dischargers within Duke Power’s project boundaries (Scenario 3), would have minimal impact on water quality in the Catawba River Basin. Of more significance are the contributions of dischargers located outside of Duke Power’s project boundaries and contributions from non-point sources.

Back to Top

Barrow, C.J. 1997. Environmental and Social Impact Assessment. London: Arnold Publishers.

Cooke, G. D., E.B.Welch, S.A.Peterson, and P.R. Newroth. 1986. Lake and Reservoir Restoration. Butterworth Publishers. Stoneham, MA. 395pp.

Central Carolinas Choices (CCC). 1999. Focus on the Region – 1999. Univ. North Carolina Charlotte. Charlotte, North Carolina. 12pp.

Council on Environmental Quality (CEQ). 1997. Considering Cumulative Effects Under the National Environmental Policy Act. Washington: Government Printing Office.

Court, J.D., C.J. Wright, and A.C. Guthrie. 1994. Assessment of Cumulative Impacts and Strategic Assessments in Environmental Impact Assessment. Commonwealth Environmental Protection Agency, Barton, Australia

Duke Power (DP). 2000. The Catawba: An Update on the Catawba River Basin and the Catawba Reservoirs. Duke Power Technical Report. 51 pp.

Horne, H.J., and C. R. Goldman. 1994. Limnology. McGraw-Hill, Inc. New York, New York. 576pp.

Irving, J.S., M.B.Bain, E.A. Stull, and G.W.Witmer. 1986. Cumualtive Impacts-Real or Imagined? Conference No. 8603104. United States Department of Energy, Washington, D.C.

North Carolina Department of Environment, Health, and Natural Resources (NCDEHNR). 1992. North Carolina Lake Assessment Report. Technical Report No. 92-02. 355pp.

North Carolina Department of Environment, and Natural Resources (NCDENR). 1999. Catawba River Basinwide Water Quality Plan. Division of Water Quality. 235pp.

Soballe, D.M. 1998. Successful Water Quality Monitoring: The right combination of Intent, Measurement, Interpretation, and a Cooperating Ecosystem. J. Lake and Reservoir Management 14 (1): 10-20.

South Carolina Department of Health and Environmental Control (SCDHEC). 1991. South Carolina Lake Classification Survey. Technical Report No. 006-91. 560pp.

Spaling, H. 1997. Cumulative Impacts and EIA; Concepts and Approaches. EIA Newsletter 14. University of Manchester, Manchester, England.

Vestel, B.A., A. Rieser, M.Ludwig, J. Kurlan, C. Collins, and J. Ortiz. 1995. Methodologies and Mechanisms for Management of Cumulative Coastal Environmental Impacts. Part I: Synthesis with Annotated Bibliography, and Part II: Development and Application of a cumulative Impacts Assessment Protocol. NOAA Coastal Ocean Program Decision Analysis Series. No. 6., Maryland: NOAA, US Dept.Commerce.

United States Environmental Protection Agency (USEPA). 1975. National Eutrophication Survey. Working Papers No. 380, 383, 385, 386, 387, 388, 430, 440, and 441.

Welch, E.B. 1980. Ecological Effects of Wastewater. London: Cambridge Univ. Press

Wetzel, R.G. 1975. Limnology. W.B.Saunders Company. Philadelphia, Pa. 745pp.

Back to Top

Back to Top

Next