Submitted to: Journal of Soil and Water Conservation
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 3/21/2008
Publication Date: 9/1/2008
Citation: Owens, L.B., Shipitalo, M.J., Bonta, J.V. 2008. Water Quality Response Times to Pasture Management Changes in Small and Large Watersheds. Journal of Soil and Water Conservation. 63(5):292-299. Interpretive Summary: Over the years, the federal government has spent great sums of money through various conservation programs to implement best management practices (BMPs) to conserve water and soil. To assess the effectiveness of these programs, measurements need to be made to determine how the implementation of BMPs affect water quality and soil loss from the areas receiving the practices. For large watersheds (multiple square miles in area) this is a difficult, expensive, and multiple-year task. Although several years may be involved in assessments on small watersheds (a few acres), such measurements may give valuable insights for responses of BMPs on large watersheds. (Of course, “scaling up” is also a challenge.) Small watersheds were used at the North Appalachian Experimental Watershed near Coshocton, Ohio to measure response time to changes in pasture management practices. Surface runoff was sampled on an event basis; subsurface flow was sampled monthly. Because of underlying clay layers, subsurface flow returned to the surface to contribute to continuously flowing streams. With different levels of nitrogen fertility being maintained for multi-year periods, changes in nitrogen movement could be measured when the N application rate was changed. These changes were also observed downstream at larger watershed sites (multiple acres). The response times indicated that several years are necessary to measure changes in management, and that response times in multi-square miles would be at least as long, probably longer. These results are of use to scientists and land managers who are trying to evaluate changes in management practices on soil loss and stream water quality.
Technical Abstract: To interpret the effects of best management practices on water quality at a regional or large watershed scale likely response times at various scales must be known. Therefore, 4 small (<1 ha, 2.5 ac) watersheds, in rotational grazing studies at the North Appalachian Experimental Watershed (NAEW) near Coshocton, OH, were used to study management impacts on water quality and response times. Surface runoff was sampled on an event basis; subsurface flow was sampled monthly. An underlying clay layer created a perched water table in these watersheds. Samples of shallow groundwater were taken from springs developed where the clay outcropped at the soil surface. Such return flow feeds base flow in watersheds large enough to have continuously flowing streams. In 4 large watersheds, ranging from 18 to 123 ha (44 to 303 ac), base flow was over 50% of total annual stream flow and approximately 20% of annual precipitation. Nitrate-N transport in base flow was 31 to 59% of total annual NO3-N transport in stream flow. When the N fertilization rate in a “medium fertility” area, that contained 2 small watersheds, was increased from 56 kg ha-1 to 168 kg ha-1 (50 lbs ac-1 to 150 lbs ac-1) per year there was very little change in NO3-N concentrations in shallow groundwater for 4 years. Then NO3-N levels in groundwater began to increase and were still increasing after 10 years at the higher rate. With discontinuation of N fertilization, NO3-N concentrations in groundwater returned to pre-N increase levels after 6 years. In a “high fertility” grazing area with a similar perched water table, 224 kg N ha-1 (200 lbs ac-1) was applied annually. Concentrations of NO3-N increased to >10 mg L-1 (ppm) after 5 years. Legumes were then interseeded into the grass forage, and mineral N fertilization was discontinued. Nitrate-N concentrations in groundwater returned to their pre-fertilization levels after about 5 years. This multi-year response of subsurface water quality to management change in small watersheds, shown by this 25-yr data set, indicates that the response time for measurable change in multi-square-mile watersheds will be equally long, if not longer, and trends will be muted. The percentage of the large watershed that experiences a management change and spatial distribution of treated areas will affect the extent to which trends will be muted. Trends in the NAEW watershed with 23% treated area were less muted than in a watershed with 14% of the area treated.