Constituents analyzed were total nitrogen, nitrate, organic nitrogen, total phosphorus, phosphate, sodium, chloride, iron, manganese, and zinc. Additionally, suspended solids and suspended sediment were analyzed for streams, and chlorophyll a was analyzed for lakes.
At the time of this study, seven wastewater-treatment facilities discharged effluents to Ellerbe Creek, Little Lick Creek, Knap of Reeds Creek, Eno River, Morgan Creek, New Hope Creek, and Northeast Creek. Effluent discharges were located immediately upstream from the water-quality sample-collection sites on those streams.
Total nitrogen concentrations were stable at most stream and lake sites (table 1; figs. 5, 6). A decreasing trend in total nitrogen concentration was observed in the Eno River at Hillsborough, Little River, Morgan Creek near White Cross, Swift Creek, and Cape Fear River near Brickhaven. The first three sites are in headwater basins where development has been relatively minor (less than 10 percent of land cover) and agricultural land uses comprise about 35 to 40 percent of land cover. Harned and others (1995) noted decreases in nitrogen in other parts of the Neuse River Basin from 1980 to 1989 that were attributed to changing agricultural practices. The decreasing trend in total nitrogen concentrations for the Swift Creek Basin is somewhat surprising considering the significant amount of development from 1975 to 1988 (fig. 2B) and may be due to the extensive use of impoundments to control runoff. An increase in nitrogen concentration was observed at only two sites--Neuse River at Smithfield and near Clayton (fig. 5)--and may be related to land cover changes in the Neuse River Basin between the Falls Lake dam and Smithfield. The percentage of forested and wetlands land cover in that area has decreased 19 percent from 1975 to 1988, whereas agricultural land cover has increased 7 percent, and developed land cover has increased 12 percent. Stable nitrate concentrations were reported by Harned and others (1995) at Smithfield for the period 1980-89. The shorter time period and the fact that concentrations were not adjusted for streamflow probably account for the different result.
The concentration of total nitrogen in lakes varied seasonally. Concentrations tended to peak in winter months and decline in summer months when lake productivity increases (site 2, fig. 7). Most lake sites had no trends in total nitrogen concentrations during the 1983 through 1995 water years (table 2). Declines in total nitrogen concentrations were observed at Little River Reservoir, Lake Michie, and Cane Creek Reservoir--all located in headwater areas where development is minor and agriculture is about 29 to 40 percent of land cover. In addition, Little River and Cane Creek Reservoirs are relatively new, impounded in 1988 and 1989, respectively; thus, reservoir aging may be an additional factor.
Trends in nitrate and organic nitrogen, two components of total nitrogen, also were examined. Most stream sites had stable concentrations of nitrate for the period tested (table 1). Increasing nitrate trends were detected for the Eno River near Durham, Knap of Reeds Creek, Neuse River near Clayton and at Smithfield, and New Hope Creek. For Knap of Reeds Creek there was a coinciding decrease in organic nitrogen, probably indicating improvement in wastewater treatment at the Butner facility. Although an increase in nitrate was observed at Eno River near Durham, no trend was detected at Eno River near Weaver, which is only about 4 miles downstream and received WWTP effluents from the former Durham WWTP until 1994. A decrease in nitrate and organic nitrogen occurred at Cape Fear River near Brickhaven. Land-cover data for this basin are incomplete, and the cause of the decreasing nitrate trend is unknown.
Decreasing trends in nitrate at lake sites occurred at Lake Michie, Lake Benson, Cane Creek Reservoir, University Lake, and Falls Lake near the dam (table 2). Nitrate concentrations have increased at Falls Lake at Highway NC-98. Because dissolved nutrients are taken up by algae, a productive algal population, indicated by measuring chlorophyll a concentrations, may account for the decreasing nitrate concentrations. Similar nitrate trends were not observed at the major lake tributaries.
Decreasing organic nitrogen trends occurred at Eno River near Weaver, Knap of Reeds Creek, Cape Fear River, Swift Creek, Cane Creek, Morgan Creek near White Cross, and downstream from Falls and Jordan Lakes. Various causes may account for these decreases, among these are improved nitrification in wastewater treatment (Eno River, Knap of Reeds Creek), improved agricultural practices (Cane and Morgan Creeks), and the presence of impoundments (Swift Creek, Falls and Jordan Lake outflows). Decreasing organic nitrogen trends also were observed for the period 1980-89 in parts of the Neuse River Basin (Harned and others, 1995). Organic nitrogen concentrations were stable at lake sites except at Little River and Cane Creek Reservoirs. Concentrations at these relatively young reservoirs were decreasing.
Phosphorus data were tested for two types of trends: (1) a step trend (change between two time periods) to determine if a decrease occurred following the 1988 phosphate-detergent ban, and (2) a monotonic (consistently increasing or decreasing) trend to determine if phosphorus concentrations were decreasing, increasing, or stable after the phosphate-detergent ban. The results of step-trend analysis indicate that a significant phosphorus-concentration decrease, ranging from 25 to 81 percent in 1988, occurred at all stream sites (figs. 8 and 9). Greatest decreases occurred at those stream sites that are located just downstream from WWTP effluents where concentrations have been greatest and the effects of improvements in wastewater treatment would be most pronounced. Smallest decreases occurred in outflow from Falls and Jordan Lake dams where concentrations of total phosphorus were low because of nutrient trapping that occurs in the lakes.
Only three sites on Falls Lake and one on Jordan Lake had sufficient pre-1988 data for analysis of a step trend in total phosphorus. A significant downward trend was detected at Falls Lake at I-85, NC-98, and near the Falls Lake dam. Total phosphorus concentrations were stable at Jordan Lake at buoy 9.
Data also were tested for a monotonic trend in total phosphorus and phosphate concentrations during the 7-year period from 1988 to 1995. Significant decreasing trends in total phosphorus were detected for the Eno River near Weaver (site 8), Ellerbe Creek near Gorman (site 7), and Morgan Creek near Farrington (site 17). The large downward step trend in phosphorus concentration following the phosphate-detergent ban for sites downstream from WWTP's and the continuing downward trend in phosphorus concentration after 1988 indicate the effects of both the phosphate-detergent ban and improvements in wastewater-treatment processes on removal of phosphorus. A significant increasing trend was detected at Little Lick Creek near Oak Grove for 1989-95. The WWTP discharging to Little Lick Creek was taken out of service in 1994, and monitoring was discontinued after 1995. More post-1994 data are needed to determine if closing the WWTP has reversed the phosphorus trend. Significant downward trends were observed at Falls Lake (fig. 7; for example, site 5), Little River Reservoir, Lake Michie, and Jordan Lake at buoy 9.
Phosphate concentrations at the lake sites generally were below the analytical detection limit of 0.01 mg/L. This was too low for trend analysis.
Suspended-sediment concentrations were stable at the 11 stream sites with suspended-sediment data. Of these, Swift and Cane Creek Basins and Morgan Creek Basin near White Cross had substantial (greater than 20 percent) conversion of forested land cover to agricultural and developed land cover between 1975 and 1988 (fig. 2B) and might be expected to evidence an increase in suspended sediment. However, Cane and Morgan Creek Basins were still more than 50-percent forested in 1988 (fig. 2C), and suspended-sediment yields in these basins are among the smallest in the study area (Childress and Treece, 1996).
More unexpected was the lack of trend for Swift Creek Basin, which is more than 75 percent developed above the study site (fig. 2C). No yield or load data are available for this site because continuous streamflow data were not collected; however, the mean suspended-sediment concentration at this site for 1989-94 was similar to the mean for Cane and Morgan Creeks (Garrett and others, 1994). The lack of a suspended-sediment trend in the Swift Creek Basin may be due to several factors or combination of factors including the short period for suspended-sediment data (1989-95) that does not predate most development in the basin, the presence of several small reservoirs on tributaries, the use of sediment holding ponds and other erosion mitigation efforts required during building construction, or the transient nature of most construction activities in a developing urban area. More data, particularly streamflow data, are needed to evaluate suspended-sediment loads in Swift Creek.
Suspended-solids concentrations were stable during the study period at stream and lake sites. Downward trends were observed only for Knap of Reeds Creek and Haw River near Bynum. Improved wastewater treatment is likely the most important factor for Knap of Reeds Creek, which is a very small watershed that received substantial wastewater discharges. No cause could be identified for the downward trend at Haw River near Bynum, a large, mixed land-use basin.
Chloride concentration also may indicate development. Chloride is generally low in natural waters in the Triangle (Childress and Treece, 1996) but may be present in elevated concentrations in waters receiving industrial and municipal wastewater effluent. No trends were observed for chloride concentration.
Concentrations of iron and manganese were stable at all stream sites except Morgan Creek near White Cross where iron concentration was increasing. This may be from increased land disturbance; forested land cover has decreased by nearly 25 percent, and agricultural land cover has increased by about 18 percent. However, no increasing trend in suspended-sediment concentration was detected at this site.
At lake sites, decreasing iron trends were observed for Little River and Cane Creek Reservoirs and Lake Michie, and a decreasing manganese trend was observed for Cane Creek Reservoir. These trends did not coincide with decreasing suspended solids trends suggesting that the trend is for the dissolved form of iron and manganese. Both Cane Creek and Little River Reservoirs are relatively new, and these trends could be related to reservoir aging.
An increasing manganese trend occurred in Falls Lake near the dam. Here the concentration ranged from 14 to 1,000 mg/L with a median of 64 mg/L.
Zinc is a trace metal often observed in elevated concentrations near urban areas, especially in streams receiving wastewater effluents (Elder, 1988). Trends were analyzed for seven sites with concentration and streamflow data from 1983 to 1995. Most were collected by the DWQ. Decreasing zinc concentrations were observed at Little Lick Creek and Knap of Reeds Creek.
Two-thirds of all the sites analyzed had significantly increasing trends in chlorophyll a (table 2) ranging from 17 to 52 percent per year. Increasing trends for Falls Lake at sites 2, 3, and 4 (fig. 11) and Jordan Lake at sites 21A and 22 (fig. 12), where data collection began in 1992, were not affected by the sample collection method change. Sites that did not have significant trends for chlorophyll a were Falls Lake at I-85, Lake Michie, Cane Creek Reservoir, and the Haw River arm of Jordan Lake.
Chlorophyll a trends in this study increased during a period in which total phosphorus concentrations were decreasing or stable (table 2). Reservoir dynamics are complex and many factors other than phosphorus concentration affect algal populations including temperature, suspended solids concentrations (light penetration), and nitrogen concentrations. Chlorophyll a trends may indicate the effect of phosphorus stored in lake sediments. More intense study of processes affecting algal populations in these reservoirs is needed.
Return to Water-Quality Trends for Streams and Reservoirs in the Research Triangle Area of North Carolina, 1983-95