By W. Harry Doyle, Jr., Bob G. Whitworth,
George F. Smith, and Tom D. Byl
Prepared by the
U.S. GEOLOGICAL SURVEY
in cooperation with
THE UNIVERSITY OF TENNESSEE
AGRICULTURAL EXTENSION SERVICE
Beaver Creek watershed in West Tennessee covers about 95,000 acres and includes some of the Nation's most productive farmland and most highly erodible soils. A preliminary assessment of the Beaver Creek watershed, conducted by the Tennessee Department of Environment and Conservation, indicated that agricultural activity was degrading water quality. In 1989, the Tennessee Department of Agriculture (TDA) and the U.S. Geological Survey (USGS) began a cooperative study (1) to evaluate the effect of agricultural activities on water quality in the watershed and (2) to determine the effectiveness of agricultural best management practices (BMP's) designed to improve water quality by reducing agricultural nonpoint-source (NPS) pollution.
The project gained momentum in 1990, when the Beaver Creek watershed was added to the U.S. Department of Agriculture's list of Hydrologic Unit Areas, qualifying it for educational, technical, and financial resources to improve water quality. A local coordinating committee was formed consisting of county, State, and Federal agencies, and, most importantly, local farmers. Each group had some responsibility for the Beaver Creek project, such as technical information about conservation methods, financial assistance, education, BMP implementation, or water-quality evaluation.
The following questions were considered before conducting this study:
These questions have been answered. The results from these studies are summarized in this document.
Agrichemical monitoring included testing the soils, ground water, and
streams. Four farm sites ranging in size from 27 to 420 acres were
monitored. Monitoring stations were operated downstream to gain a
better understanding of the water chemistry as runoff moved from small
ditches into larger streams and, eventually, to the outlet of the
Beaver Creek watershed.
Soil loss (suspended sediment), nutrients (fertilizers), and pesticides were monitored to determine if they were getting into the stream. Pesticides and nutrients also can percolate into the soils, therefore, agrichemical migration in the soil was evaluated to determine if ground water was affected. The biological health of the stream was measured by collecting and identifying aquatic organisms at numerous stream sites.
Information was needed to define the migration of agrichemicals in the
soil and streams. The number of soil and water-quality samples was
established from preliminary studies so that an adequate representation
of the concentrations and loads was obtained.
Agrichemicals can be flushed from the fields during storm runoff. Thus,
stream sampling took place during rain events. Soil and ground-water
samples were collected after storm infiltration to see if the
agrichemicals moved downward to the water table.
Five soil samples were collected between two rows and mixed together to represent a single location in the field (fig. 1). These soil composites were duplicated at several locations in a field to establish a representative distribution of pesticides over time and space. Samples were collected at different depths each time to determine if the pesticides were moving into the ground water.
An optimal sampling strategy for characterizing chemicals and suspended sediment in agricultural runoff includes frequent sampling during stormflow and less frequent sampling after a storm and during periods of dry weather. A sampling interval equal to 5 percent of stormflow duration was determined to be adequate to characterize concentrations of constituents during stormflow with an error of less than 5 percent. Thus, for stormflow durations of 100 and 300 minutes, sampling intervals were 5 and 15 minutes, respectively.
The health of a stream is reflected by the number and types of bottom-dwelling organisms living in the stream (biomonitoring). Biomonitoring results varied greatly, depending on the sampling technique used. Potential biases resulting from the various sampling techniques were reduced by using several different sampling techniques.
No-till farming techniques did not increase ground-water contamination in the Beaver Creek watershed. Horizontal dispersion of aldicarb sulfoxide, a metabolite of aldicarb, in the soil profile was negligible. The transport of aldicarb sulfoxide downward was minimal in the conventional-till and in the no-till fields. About 85 percent of this metabolite was found within the first half foot below the soil surface; none was detected below 2.5 feet. The shallow depth of transport is attributed to the low permeability of the soils, decreased amounts of rainfall, and high rates of evaporation and plant transpiration during the summer months.
Natural, nonbiological reactions appear to be the primary mechanism for degrading aldicarb and its metabolites in the soil profile. The calculated half-lives of aldicarb metabolites in the field were much less than those derived from laboratory experiments. In the laboratory, the toxic by-product of aldicarb can last almost 3 years, but in cotton fields, it degrades to nontoxic compounds in about 2 weeks.
Pesticide contamination of ground water is not a problem in the Beaver Creek watershed. Ninety-five domestic wells in the watershed were sampled for aldicarb, atrazine, and alachlor in summer 1992. Aldicarb was not detected in any of the wells. Concentrations of atrazine detected in two wells and alachlor in three wells were less than proposed primary drinking water standards.
Significantly higher nitrate concentrations were measured in water from wells near septic tanks and confined animal facilities than from wells near fertilized fields. Wells deeper than about 150 feet were not affected by surface- or near-surface nitrate sources.
Suspended sediment resulting from erosion is the major water-quality problem in the watershed. Run-off from no-till fields is cleaner than runoff from conventional-till fields. However, the cleaner water entering ditches and streams can become contaminated with sediment eroded from the stream channel. This process of sediment contamination is more accelerated in the channelized streams commonly found in the Beaver Creek watershed. Excessive suspended sediment interferes with biological habitat and normal decomposing of nutrients and pesticides in the streams.
Approximately 80 percent of the total nitrogen carried by storm runoff is organic nitrogen. About 75 percent of this organic nitrogen is flushed from the fields during late winter and early spring storms. Thus, BMP's that prevent field erosion during the non-growing season can help to reduce nitrogen runoff. Inorganic nitrogen (nitrate and nitrite) is only 20 percent of the total nitrogen runoff load. About 50 percent of this inorganic nitrogen is flushed from the fields during late spring storms. Only 10 percent of the inorganic nitrogen runoff occurs during the growing season probably because of assimilation and less precipitation.
From 1991 to 1995, aldicarb and its metabolites (aldicarb sulfoxide and aldicarb sulfone) were detected in runoff from the farm fields with the highest concentrations measured during a storm event only hours after aldicarb was applied. This result demonstrates the importance of weather conditions and farming activities on agricultural runoff. No aldicarb was detected in storm-runoff samples collected a few weeks after application. However, the metabolites were detected as late as 76 days after an aldicarb application.
Prior to the implementation of BMP's at one of the farm study sites, some storms produced an average suspended-sediment concentration of 70,000 milligrams per liter (mg/L). After the implementation of BMP's, however, the average value never exceeded 7,000 mg/L, which is a tenfold reduction.
No-till crop production was the most effective BMP for conserving soil on the farm fields tested. Other studies also have found that no-till farming preserves the structure of the soil and retains the crop residues from the previous season.
A natural bottomland hardwood wetland and a constructed wetland were evaluated as instream resource-management systems. The wetlands improved water quality downstream by acting as a filter and removing a significant amount of NPS pollution from the agricultural runoff. The constructed wetland reduced the nutrient, sediment, and pesticide load by approximately 50 to 90 percent over a 4-month period (fig. 2).
Streambanks tend to be more stable if vegetated with large trees rather than herbaceous plants; however, downcutting can still occur.
The results of the Beaver Creek watershed study have increased the understanding of the effects of agriculture on water resources. Study results also demonstrated that BMP's do protect and improve water quality. This information can help water managers, farmers, and others to protect and improve the quality of the environment for future generations.
What does all this mean to the farm community and the people that work with them?
These investigations also provide information that may be useful to other watershed monitoring projects:
Byl, T.D., and Carney, K.A., 1996, Instream investigations in the
Beaver Creek watershed in
West Tennessee, 1991-95: U.S. Geological Survey Water-Resources Investigations Report 96-4186, 45 p.
Doyle, W.H., and Baker, E.G., 1995, Collection of short papers on the Beaver Creek watershed study in West Tennessee, 1989-94: U.S. Geological Survey Open-File Report 95-156, 54 p.
For more information contact:
U.S. Geological Survey
640 Grassmere Park, Suite 100
Nashville, Tenn. 37211
Further information can be obtained by contacting your County Extension Agent. An Extension Publication Catalog can be obtained by e-mailing: SandraJ@AES.gw.utk.edu. Requests for a specific publication can also be e-mailed to this address.
In the future, Extension publications will also be available through the world wide web at http://www.utextension.utk.edu/publications/default.asp.
Copies of this report may be purchased from:
U.S. Geological Survey
Branch of Information Services
Denver, Colorado 80225-0286