Tillage and Timing of Manure Application Impacts on Nitrate Leaching in Karst Terrains of the Upper Midwest
N. C. Wollenhaupt, and A. Bosworth
Manure application to land is a common practice in many parts of the World. Although manure addition to land provides nutrients for crops and improves soil structure, there is a growing concern about the potential contamination of surface and ground waters from manure application on agricultural land. The objectives of this research was to evaluate the impact of manure spreading and the timing of its application on surface and subsurface water quality under two conservation tillage systems in karst terrain of the upper Midwest. In this paper, I will be only discussing the results from tillage and manure impacts on subsurface water quality. If time permit, I will present the results on surface water quality.
The experiment was conducted for two years at the University of Wisconsin Agricultural Research Station, Lancaster, WI. There were a total of 24 plots each measuring 16 ft wide and 60 ft long. Only 12 plots were instrumented with pan and wick samplers. The pan and wick samplers were installed at 2 ft and 4 ft depths.
The pan and the wick samplers consisted of a 31 inches x 14 inches PVC sheet with 1 inch wide PVC strip glued along the edges. A 1 inch hole drilled in the middle of the PVC sheet allowed the percolate to move into the collection vessel. A fiber glass wick facilitated the movement of the percolate solution from the PVC sheet to the collection vessel. The top end of the fiber glass wick was unwound and its strands spread and then glued to the PVC sheet. The base of the PVC sheet was then covered with a fiber glass cloth to insure that all water collected on the sheet was channeled to the collection vessel. The fiber glass cloth was then covered with clean fine sand to filter out any clay particles that may be appearing with the percolate solution.
The difference between the pan and the wick samplers was that in the pan sampler we placed a plastic grille over the fiberglass cloth and sand. The grille was also covered with a plastic window screen to prevent soil chunks from dropping into the pan sampler. The plastic grille in the pan sampler prevented the fiber glass cloth to come in direct contact with the soil above the sampler. Therefore, the water that percolated through the pan samplers was the water that had dripped from the soil matrix when it was saturated or the water that came through the macropores.
In the wick samplers, the filtering sand was covered with fine soil from the same depth. The contact between the fine soil on the top of the wick sampler and the face of the tunnel insured the continuity of water flow through the soil matrix to the wick samplers. Because of this continuity, the fiberglass wick applied 20 inches suction to the soil matrix in the wick sampler and in turn on the soil in the tunnel. Therefore, the leachate that percolated through the wick samplers included both the matrix and the macropore flow.
Tillage treatments were no-till and chisel plow. Chisel plowing and planting were done along contour. Manure treatments were fall application before the onset of snow and winter application on the top of snow. A no manure treatment was included as a control.
For runoff measurements, plots were isolated from the surrounding area using galvanized corrugated sheet metal (16 Gage, 8 inches by 32 inches sheets). The sheet metal was pounded into the ground to a depth of ~6 inches along the border using a sledgehammer. Every year, the borders were removed before plowing, planting, and harvest and then reinstalled. At the lower end of the plots, the plot borders were tapered in order to direct the runoff to runoff collectors and then to tipping buckets. The tipping buckets were fabricated out of galvanized steel and measured 20 inches in length and 15 inches in width. In the center, the tipping bucket was 5 inches high. Each tipping bucket was equipped with a magnetic switch to record the number of tips in each runoff event. Each tip was about an equivalent of ~0.8 gallon of runoff.
Runoff was sampled by intercepting the flow from the outflow pipe with one-inch wide splitter placed in the center of the tipping bucket. The splitter lay in one of the two cells of the tipping bucket. Runoff samples from the splitter were collected in a large 6-gallon water barrel that lay flat on the ground. After each runoff event, the volume of the sample in the water barrel was measured and then a quart sample was taken for sediment, and nutrient (phosphorous and nitrate) analysis. Since the splitter volume depended upon the flow rates, the tipping buckets had been calibrated for splitter volumes at various flow rates. Runoff samples were collected starting 19 February 1994 for two years.
Results and Discussion
Subsurface Water Quality
Pan and wick samplers: Percolation data over two years (1994 and 1995) showed that there is no significant difference in water flow between wick and pan samplers at any given depth. This suggests that most of the water collected in the pan and wick samplers was due to preferential flow. The preferential flow could have been due to the presence of earthworm burrows, root channels and interpedal voids.
For any given treatment, percolation at 4 ft depth was about twice the amount at 2 ft depth. However, there was no difference in nitrate concentration of the leachate collected from 2 ft and 4 ft depths pan and wick samplers. This suggests that percolation at 4 ft depth was also from the areas surrounding the pan and wick samplers. Supplementary studies showed that most of the water percolation at 2 ft depth was from macropores whereas at 4 ft depth there was a combination of macropores as well as lateral transport. The lateral flow at 4 ft depth was due to water flow limiting conditions just above this depth and also because of the presence of pan and wick samplers that provided an opening for flow. The following discussion is based on average of measurements taken from both pan and wick samplers at 2 ft depth.
Manure impacts: Manure application increased water percolation through the soil profile (Table 1). About 34-39% of the precipitation percolated through the soil to 2 ft depth over the two year period (1994,1995). Out of this percolation, 62 to 74% of the percolation occurred during the non-growing season. Water percolation was also higher from winter application than from fall application of manure during the non-growing season (November through April). However during the growing season (May through October), timing of manure application had no significant effect on water percolation. Increased percolation in manure applied plots was mainly due to an increase in earthworm activity. Hanewell (1996) showed that 3 years after the start of this experiment, the total worm population was 94,326; 227,516; and 267,999 worms/acre for no manure, fall manure and winter manure treatments, respectively.
Nitrate concentrations in the percolating water were above the drinking water standard of 10 mg/L, a majority of the time. As expected, nitrate concentration was higher during the growing than the non-growing season. Nitrate leaching was higher from the manure than the no manure treatment both during the non-growing and growing seasons. Furthermore, nitrate leaching was higher from the fall than from the winter applied manure. This difference in nitrate leaching between the two manure treatments was assumed to be due to greater availability of nitrate as a result of greater mineralization in the fall applied manure than in the winter applied manure treatment. Averaged over two years, nitrate leaching from no manure, fall manure and winter manure treatments were 33, 46, 40 lbs/acre, respectively (Table 2). Out of this, about 63 to 73% of the nitrate leaching occurred during the non-growing season.
The corn grain yield data over two years showed that there was a significant increase in corn yield due to manure application. However, timing of manure application had no significant effect on corn yield.
Tillage impacts: Over the two year period, 31% and 43% of the precipitation percolated through the soil in the chisel plow and no-till treatments, respectively (Table 3). About 64 to 68% of this percolation occurred during the non-growing season. Water percolation was higher from the no-till than the chisel plow treatment (Table 3). This is expected considering that macropore channels are not disturbed by cultivation in the no-till treatment. Hanewell (1996) showed that three years after the start of the experiment, the total worm population at the experimental site was 63,559 and 329,534 worms/ac for the chisel plow and the no-till treatments, respectively. Out of this population, night crawlers (Lumbricus terrestris ) comprised 19,432 and 38,459 worms/acre for the chisel plow and no-till treatments, respectively.
There was no impact of tillage practices on nitrate concentration in the percolating water. Again, nitrate concentrations were higher during the growing season than during the non-growing season. There was no statistical difference in nitrate leaching between the chisel plow (38 lbs/acre) and the no-till (42 lbs/acre) treatments (Table 4). A majority of this nitrate leaching occurred during the non-growing season.
Corn grain yield was slightly higher under chisel plow than no-till systems. This is probably due to greater mineralization of soil organic nitrogen due to cultivation.
The results of this study showed that a majority of the water percolation and N leaching in soils of the karst terrain in the upper Midwest occurs during spring recharge. Although there is a significant differences in N leaching between the manure and no manure treatments, a majority of this N leaching occurs even without the addition of N fertilizer. There was no significant effect of tillage on water percolation and N leaching either during the snowmelt period or during the growing season.
Hanewell, A. M. 1996. The effect of tillage and manure on earthworms and water infiltration in fine silty mixed mesic typic hapludalf. M. S. Thesis. University of Wisconsin, Madison.
Table 1: Manure impacts on water percolation at 2 ft depth.
Table 2: Manure impacts on N leaching at 2 ft depth.
Table 3: Tillage impact on water percolation at 2 ft depth.
Table 4: Tillage impact on N leaching at 2 ft depth.