sediments, nutrients, agricultural runoff, non-point pollution, primary productivity


For the past few decades, environmental concern for agricultural watersheds has risen following the mechanization of farming in the mid-twentieth century. The intensively farmed Mississippi Delta has been the subject of many research studies, revealing the negative impact of agricultural practices on oxbow lakes in its watersheds. Nonpoint source pollutants, such as sediment, nutrients, and pesticides are linked to eutrophication, decline in fish population, and reduced depth in oxbow lakes, leading to a decrease in recreational and aesthetic value. The primary source (80%) of nutrient loading in Delta’s waterbodies are from agriculture runoff [1]. Agricultural practices such as soil tillage and row-crop farming have caused a myriad of water quality problems within their surrounding oxbow lakes. The mean nitrate concentration in runoff alone is a result of the increasing use of nitrogen fertilizers in the watershed and row-crop farming [2]. These practices have also contributed to the high levels of sediment in the water that diminish water clarity and depth [3]. In a natural environment, phosphorus is the limiting nutrient in waterbodies, as it generally exists in low amounts and thus restricts the growth of algal blooms and/or aquatic plants. However, with excessive phosphorus loadings from nutrient rich runoff into freshwater systems, a corresponding increase in eutrophication and algae growth occurs [4,5]. While unusual given this importance of phosphorus in freshwater productivity, secondary nitrogen limitation may occur due to the excess phosphorus [6,7,8,9]. In order to lessen the amount of nonpoint source pollution in agricultural watersheds, the 1987 amendments to the Clean Water Act (CWA) created the Section 319 Nonpoint Source Management Program [10]. Section 319 instructs for Best Management Practices (BMPs), such as no-tillage and grade stabilization structures, to be implemented [3, 10]. With the adoption of these conservation programs, both nutrient runoff and sediment levels have diminished in oxbow lakes, resulting in improved water clarity and depth, plankton growth, and fish stocks [11] Though the impact and ways to decrease the amount of runoff in agricultural waterbodies has been heavily researched and documented, the intricacies of the relationship between runoff and ecosystem components have not been thoroughly investigated [12]. With the excessive addition of nutrients and sediment, the ways in which the ecosystem interacts with its environment have changed. The findings of Miranda et al. [13] suggest that hierarchal relationships exist among ecosystem components in an aquatic watershed. They conclude that systems are linked to their watersheds through nutrient inputs, that nutrients determine algal composition and biomass, and that algal biomass which in turn affects the structure and function of communities of aquatic organisms.” Three years of data was analyzed from the mid-eighties, before common use of BMPs were implemented, in order to explore these various, complex relationships between nitrogen and phosphorus loadings, sediment, chlorophyll a, phytoplankton, radiance, and other variables affecting the water quality of a Delta oxbow lake in a cultivated watershed.

Figure 1:Map Of Moon Lake, Mississippi, Usa With Sample Sites

Materials and methods

Water quality samples were collected from Moon Lake, an oxbow lake watershed of the Mississippi River located in Coahoma County, MS (Figure 1). The 947 ha. Mississippi Delta oxbow was once home to a popular recreational scene and a thriving ecosystem of fish and wildlife [1]. By the early 1980s, however, Moon Lake suffered tremendously as a result of the Mississippi Delta’s mechanization and subsequent excess of nonpoint source pollution [12]. A major inflow to the lake enters from Phillips Bayou in Tunica County where the primary land use is row crop agriculture [3]. The lake’s elevated turbidity levels resulted in reduced water clarity, quality, and depth, which then lead to a decline in fish population [1]. Moon Lake was sampled biweekly from 1982 through 1985. Surface and profile water samples were collected from six sites, five on the lake proper and one on the inflowing channel. For this study only surface water quality samples were analyzed. Water flowed generally north to south entering downstream from site A-1 and exiting Yazoo Pass near site A-7 (Figure 1). Water samples were chilled and transported to the National Sedimentation Laboratory for analysis. Additionally, temperature, conductivity, pH and dissolved oxygen were measured in situ at the surface and at one meter increments using Martek water quality meter. Secchi depth and water depth were also measured at each site. Radiance was measured with a LI-COR light meter (model LI-185A) at 0.1 meter increments at sites A-3 and A-44. Water was analyzed according to methods described in [14] for dissolved, suspended and total solids, total phosphorus, filtered and unfiltered ortho-phosphate, nitrate, nitrite, ammonium, carbonate and bicarbonate alkalinity, and chlorophyll a, b, c, and total. Pearson’s correlation coefficients, and t-tests for zero correlation were calculated according to [15] using SYSTAT 13 software [16].

Table 1

Correlation coefficients, t values and probabilities of a larger t for total sediment, total nitrogen, total phosphorus and chlorophyll a relationships.

Comparison r2 Pearson’s r t p-value
Sediment vs Chlorophyll a 0.080 -0.283 2.437 0.017
Total Sediment vs N:P 0.033 -0.183 1.536 0.129
Chlorophyll a vs total P 0.100 -0.316 2.748 0.008
Chlorophyll a vs total N 0.036 -0.190 1.596 0.115
Total Sediment vs total P 0.268 0.518 4.994 <0.001
Total Sediment vs total N 0.080 0.282 2.423 0.018
Total N vs total P 0.000 0.015 0.126 0.900
Secchi vs Chlorophyll a 0.158 0.397 3.485 0.001
Secchi vs TP 0.187 -0.432 6.196 <0.001
Secchi vs TN 0.138 -0.371 3.223 0.002

Table 2

Correlation coefficients, t values and probabilities of a larger t for suspended sediment, nitrate, phosphate and chlorophyll a relationships.

Comparison r2 Pearson’s r t p-value
Suspended vs Chlorophyll a 0.099 -0.315 2.717 0.008
Suspended vs NO3:PO4 0.043 -0.207 1.73 0.088
Chlorophyll a vs Ortho PO4 0.235 -0.485 4.77 <0.001
Chlorophyll a vs Nitrate 0.020 -0.142 1.257 0.213
Suspended vs Ortho PO4 0.111 0.333 2.892 0.005
Suspended vs Nitrate 0.000 0.013 0.103 0.918
Nitrate vs Ortho PO4 0.023 -0.152 1.147 0.255

Results and discussion

Correlations were performed on water quality data shown in Table 1 to see if relationships could be found between water quality parameters. None of the correlations were strong although some were statistically significant. The strongest correlation in the data was a positive relationship between sediment and phosphorus (Figure 2). Phosphorus is abundant in the soils of the Mississippi Delta [17, 18]. The soils have a moderate RUSLE K factor, approximately 0.32, resulting in moderate runoff [19]. In 2003, erosion rates in cropped soils in this region were estimated to be 35,888 tons/yr [20]. This is a conservative estimate for the study period (1982-1985) when application of conservation practices in the area was less prevalent. The likely major source of total phosphorus in Moon Lake was phosphorus adsorbed to sediments and transported via agricultural runoff (Figure 3). Nitrogen was also positively correlated to sediment. (Figure 4). Nitrate being water soluble and not as adsorbed to clay is still a component of contaminated runoff in the Mississippi Delta [21]. Secchi visibility was positively correlated with chlorophyll a and negatively correlated to total nitrogen and total phosphorus. Somewhat surprisingly there was a weak but statistically significant negative relationship between chlorophyll a and total phosphorus (Figure 5). This further supports the idea that the source of phosphorus is sediment in runoff. Typically primary productivity is driven by phosphorus so one would expect chlorophyll a to increase as phosphorus increases. However in Moon Lake there was a corresponding increase in sediment as phosphorus increased therefore reducing light and suppressing phytoplankton productivity. Chlorophyll a and total nitrogen were unrelated. As expected given the results related to total phosphorus and total nitrogen, soluble phosphate and nitrate were similarly correlated with sediments and chlorophyll a as total phosphorus and nitrogen (Table 2). Suspended sediment was negatively correlated to chlorophyll a and the ratio of nitrate to soluble phosphate. Suspended sediment was positively correlated to ortho phosphate. No relationship was found between nitrate and chlorophyll a or suspended sediment. Nitrate and soluble phosphate were not correlated. Radiance data indicated seasonal trends of increasing light penetration in the summer and decreasing penetration in the winter months (Figure 7). This pattern was also seen in suspended sediments with higher concentrations in the winter months accompanied with lower Secchi depths and lower values in the summer and greater Secchi visibility. Several studies have shown that phosphorus is typically limiting in freshwaters and when phosphorus is abundant nitrogen may become limiting to productivity [6, 22, 23, 24, 25]. In Moon Lake, phosphorus was negatively related to phosphorus indicating some other limiting factor. Knight [26] reported that as suspended sediments concentrations in winter and early spring increased so did total phosphorus as well as soluble phosphate however there was a negative impact on primary production as light became limiting. This study also supports the concept that sediment has an overriding impact on primary productivity at least once sediment concentration exceeds some critical threshold value.

Figure 2: Sediment vs Chlorophyll a from Moon Lake, MS

Figure 3: Total sediment versus total phosphorus from Moon Lake, MS

Figure 4: Sediment versus Total Nitrogen from Moon Lake, MS

Figure 5: Chlorophyll a vs Total phosphorus from Moon Lake, MS

Figure 6: Chlorophyll a vs Total Nitrogen from Moon Lake, MS

Figure 7: Radiance profiles for sites A3 and A4 on Moon Lake, MS


Results of this study strongly support the conclusions by Knight [26] that as suspended sediments concentrations in winter and early spring increased so did total phosphorus as well as soluble phosphate. However this same increase in sediment limited light availability thus suppressing primary productivity. Given these results and from the point of view of a resource manager the relationship between sediment, nutrients and chlorophyll resource management becomes an important one. Managing primary productivity is no longer just a case of controlling phosphorus entering an aquatic resource. If sediment is too abundant primary productivity and corresponding secondary productivity will be suppressed. Also removing sediment from the water column may uncover an underlying problem with excessive eutrophication if sufficient phosphorus remains in the system.


The authors wish to thank Terry Welch and Steve Corbin for providing field and laboratory support.