Eric Pitts, Kyung Yoo, Mary Miller-Goodman, and William de los Santos

As animals graze forages, they alter the vegetative cover and soil physical properties of pastureland. These alterations may decrease infiltration
of water through the soil which, in turn, increases the amount of surface runoff. On lands where poultry waste has been used as a fertilizer for perennial pastures, this increased runoff can deposit high concentrations of nitrogen into water bodies adjacent to pastures, which can be detrimental to the ecology of the water sources. The potential for negative effects to the environment surrounding the areas where the waste is applied increases the need for sound management practices. A recent AAES study was conducted to help define best management practices in these situations.

The study related grazing pressure to surface erosion, runoff, and runoff water quality comparing two grazing pressures on three pasture vegetation types (switchgrass, bermudagrass, and tall fescue) in the Tennessee Valley region of Alabama.

Runoff, sediment loss, and water quality data were recorded at the Tennessee Valley site from November 1994 through September 1996. Data gathered in the field are being compared to trends predicted by two non-point source computer simulation models (GLEAMS and WEPP).

WEPP is the USDA Water Erosion Prediction Project and was initiated in 1985 to develop a new generation of water erosion prediction technology. The model is designed for use in soil and water conservation planning and environmental assessment. It is capable of estimating spatial variability in topography, surface roughness, soil properties, hydrology, and land use capabilities. GLEAMS is a field-scale model that consists of hydrology, erosion, nutrient, and pesticide simulation components. The latest version of GLEAMS (version 2.1) has been updated to make it more user friendly and comprehensive. Mainly, changes have been made to streamline the input to reduce user dependence and allow more internal estimation of parameters.

The study was conducted at the Tennessee Valley Substation in Belle Mina on 12 0.25-acre rectangular runoff plots. Each plot was equipped with a flume for measuring surface runoff and a runoff sampler. Plots were fenced so that cattle were restricted to certain plots according to the type of management practice the plot was under. Grazing could take place only in plot area or in the travel area between plots.

The plots were divided into proper (recommended) and heavy (twice the recommended rate) grazing treatments. Grass height was used to determine the grazing pressure. Control heights of four inches for bermuda, six inches for tall fescue, and eight inches for switchgrass were used for the proper grazing treatment. Control heights of two inches, three inches, and four inches for bermudagrass, tall fescue, and switchgrass, respectively, were used for the heavy grazing treatment. Once a grass was grazed to the control height, the cattle were removed from the plot until the grass recovered enough for grazing to resume. A combination of cowcalf pairs and single cows were used throughout the grazing period.

Broiler litter was applied to the site in split applications at a rate of 3.02 tons per acre each in spring and summer for bermudagrass and switchgrass and 2.52 tons per acre each in spring and fall for tall fescue. Runoff from the plot areas was sampled after each runoff producing rainfall event and weather data was monitored at the site.

Data suggest that there is a significant interaction between forage species and grazing pressure (see the figure). Under proper grazing management, the bermudagrass plots contributed the least amount of sediment (0.088 ton per acre), followed by tall fescue (0.089 ton per acre). Switchgrass lost the most, 0.17 ton per acre. Under heavy grazing management, the bermudagrass and tall fescue plots were very close, losing 0.131 and 0.132 ton per acre respectively. Switchgrass lost the least sediment, 0.11 ton acre, when grazed heavily.

Observed average cumulative runoff and average cumulative sediment loss by grazing pressure for bermudagrass, tall fescue, and switchgrass plots, Tennessee Valley Substation, Belle Mina.

Runoff amounts observed also varied. Under proper grazing management, bermudagrass had 7.21 inches of runoff, while tall fescue had 6.91 inches, and switchgrass 11.12 inches. When grazed heavily, the bermudagrass contributed the least amount of runoff, 6.93 inches. Switchgrass was next with 8.43 inches, followed by tall fescue, which contributed 9.24 inches. The total rainfall amount recorded for the 23-month study was 92.8 inches.

Although the switchgrass plots produced significantly more cumulative runoff and sediment than the other plots under proper grazing management, nutrient analysis for total nitrogen indicated that tall fescue pasture lost significantly more than both bermudagrass and switchgrass regardless of the grazing pressure. Dissolved phosphorus and sediment phosphorus in the runoff water showed similar patterns of loss.

In the context of surface water quality, this trend becomes critical considering the influence of phosphorus on eutrophication, the extensive use of tall fescue as a winter pasture grass, and the wide use of poultry litter as a nutrient source for tall fescue pastures in areas where poultry is produced in Alabama.

The collected data was compared to the GLEAMS and WEPP models, and showed that in all cases except for switchgrass heavy grazing pressure contributed to increased amounts of sediment when compared to the proper grazing pressure. The GLEAMS model overpredicted amounts of sediment loss in all cases, while underpredicting runoff amounts. Using site-specific input parameters and observed weather data for the same time period, the WEPP model has predicted no sediment losses, yet very large amounts of runoff for all of the plots.

Compared to observed data, these predictions were not on target, and more work will be needed to make sure that adequate documentation is provided and the models perform well enough to develop reliable decisions on pasture management practices. This is especially important in the Southeast, where very little research such as described here has been done.