Kansas Geological Survey, Open-file Report 2000-41

Reno County Swine Facility Report

by Margaret Townsend, Hydrogeologist
1930 Constant Avenue, Campus West
University of Kansas
Lawrence, KS 66047

KGS Open File Report 2000-41
August 2000

Abstract

Eleven samples were collected in December 1999 from supply and monitoring wells at the facility and from three domestic wells that the facility operator had requested be sampled. Additional samples were collected from the supply and monitoring wells at the facility in March 2000.

The results of the December sampling showed a fertilizer signature (<+8‰) for all of the wells at the facility except for a monitoring well in the downgradient-flow direction from the waste lagoon. The monitoring well located downgradient from the facility showed an animal-waste signature (+12.8‰) similar to the value measured from the lagoon (+18‰). Monitoring wells located to the east and northeast of the facility showed fertilizer at the northeast site and animal waste to the south. Three domestic wells sampled at this time all showed animal waste as a probable source. All three domestic wells had animal feedlots or septic tanks located near the wells and were probably influenced by these sources. All domestic wells were upgradient or away from the downgradient-flow direction of the facility.

The March 2000 sampling showed that the supply well and three of the monitoring wells had nitrate-nitrogen of fertilizer origin. The area surrounding the facility to the south and west is dry-land farming. Evaluation of rainfall records for the area indicates that there was sufficient rainfall and warm temperature to permit nitrification and movement of spring-applied nitrogen fertilizer. The monitoring well downgradient of the facility continued to show an animal waste d ¹⁵N signature of +12.8‰.

Evaluation of the water chemistry from the area showed a background level of 3 to 25 mg/L of chloride. The chloride value from the lagoon was 547 mg/L, and the monitoring well downgradient from the lagoon had a chloride value of 159 mg/L. Bicarbonate values of 437 mg/L were significantly higher for the downgradient monitoring well versus 50 to 200+ mg/L at the other monitoring and domestic wells.

Nitrate-N at the down gradient monitoring well was 4 to 5 mg/L as opposed to the other monitoring wells which always had values above the drinking-water limit of 10 mg/L. Use of the chloride ratio showed that approximately 30% of the water contributing to the well sample was from the lagoon. The d ¹⁵N and chloride values strongly support the idea that the lagoon is leaking and affecting the ground water. The bicarbonate and lower nitrate-N values suggest the possibility that the sampled water is a mixture of denitrified regional ground water plus lagoon water resulting in a high d ¹⁵N value and a lower nitrate-N value.

Introduction

Figure 1. Location map of swine facility and other sampling points for September and December, 1999 and March, 2000 sampling by KDHE and/or KGS.

Methods

The water level was measured at each monitoring well and then the well was purged using an airlift pump system provided by GMD2 personnel or a small 2" submersible pump used by KDHE personnel. The monitoring wells were purged until the temperature, pH, and specific conductance of each well had stabilized, generally after 10 to 15 minutes of pumping. The water-supply well for the facility was pumped until the parameters stabilized. The sample from the lagoon was collected using a 1" PVC screw-joint sampler with smaller diameter polyethylene tubing inserted inside. The sampler is approximately 10 feet long with a support on the leading edge to keep the sampler off the bottom of the lagoon. The internal tube was connected to a peristaltic pump that pumped the sample into the collecting bottle. KDHE laboratories analyzed the sample.

Samples were collected for both KDHE and KGS analyses. KGS samples for each well were collected in a 500-ml unacidified polyethylene bottle, a 200-ml polyethylene bottle acidified with 2-ml 6M HCl, and a 125-ml polyethylene bottle for the nitrogen-15 isotope sample. The samples were iced and cooled until analyzed at the water-chemistry laboratory.

Samples for KDHE were collected in 1000-ml containers that were either unacidified or pretreated with acid for samples used in the metals analyses.

Nitrogen-15 isotope samples were sent to the University of Virginia for analysis. Samples were frozen and sent on ice by Federal Express to the laboratory. Samples were thawed and a known aliquot of sample was dried and the residue burned to determine the nitrogen-15 ratio with a known standard (air). Results are reported in per mil units (parts per thousand) and are represented by ‰. Values are reported as enriched (+, greater than 0) or depleted (-, less than 0).

Geology

This area is approximately 5 miles north of the North Fork of the Ninnescah River and underlain by Permian-age bedrock. The depth to bedrock is approximately 20-40 feet. The overlying units consist of silts, clays, and fine to coarse sands of Quaternary age as shown by the cross section of wells at and near the site (figure 2).

Figure 2. Cross section of wells in T24S-R6W-Sec 22. Note variation in thickness of clay and clay silt zones in upper portions of section. Circles represent locations of wells.

Ground Water

Average precipitation in the area is approximately 28.5 inches per year. Most of the rainfall occurs during the growing season from April to September. Depth to water at the lagoon site was 10 ft below land surface (bls) at the time of facility construction in 1994. Because of the permeable soils throughout the study area ground-water recharge and water table response is rapid. Figure 3 shows the change in the water table throughout 1999 into 2000. A rain-gage and water-level recording site operated by GMD2 is located to the east of the swine facility near monitoring well 1N (figure 3). Rapid infiltration through the soil profile means that any contaminants also will move rapidly from the land surface to the water table.

Figure 3. Precipitation and ground-water level record for 1999 and 2000 at site in T24S-R6W-Sec 22. Data from GMD2 data collection station to the east of site.

Figure 4. Ground-water elevation and flow direction (shown by arrow in figure) in the vicinity of swine facility in Reno County.

Lagoon

A monitoring well is located at each corner of the lagoon (figure 4). The wells are 17 ft deep. Depth to water was approximately 10 ft in depth at the time of installation in 1994. A production well for use in the facility is located to the west of the lagoon and the swine-housing facilities. As indicated in the cross section of wells in the area (figure 2), the major portion of the aquifer is of fine to coarse sand. Horizontal hydraulic conductivity estimates from the county report (Bayne, 1956) indicate values from 12 to 72 ft/day.

Work by Kansas State University personnel (Ham and others, 2000) showed that the leakage rate from the lagoon was 0.03125 inches/day. This is equivalent to a maximum discharge of 11.4 inches/year or 0.95 ft/year. The leakage is not confined to a specific point below the lagoon so the maximum discharge rate must be evaluated over the area of the lagoon (96,000 ft²): 96,000 ft² x 0.95 ft/year = 91,200 ft³/year. Converting this to acre-ft: 91,200 (ft³/year)/43,560 (ft³/acre-ft) = 2.09 acre-ft/year (681,029 gallons) of potential leakage from the lagoon. The rate of leakage should be less than this amount because the lagoon is drained, and the effluent is used to irrigate cropland when a maximum depth is reached as specified in the swine facility operational permit issued by KDHE.

Water Chemistry

Figure 5 shows a nitrate-N contour of the various sites sampled by KDHE in September 1999. In general the entire area surrounding the swine facility is high in nitrate-N probably due to the presence of many sources in the area including fertilizer application and animal waste from both barnyard and septic-tank sources.

Figure 5. Contour map of nitrate-N concentration from KDHE sampling in September 1999 and KGS/KDHE sampling in December 1999. Note the general high nitrate-N throughout the area. Lower nitrate-N at monitoring well #7 (4.8 mg/L) indicated by hachure marks.

Figure 6. Graph of variation of nitrate-N concentration with time. Arrows indicate probable periods of fertilizer application. Note general increase in nitrate concentration during spring as indicated by arrows.

Chloride

The chloride concentration of most of the observation wells around the facility (wells 6, 8, and 9) and the supply well have similar chemistry at any given sampling event (1.9 to 22 mg/L, Appendix A, figure 7). Monitoring well #7 has a higher concentration of chloride than the regional ground-water chemistry. The chloride concentration in well #7 was 159 mg/L in December 1999 and the concentration in the lagoon was 547 mg/L.

Figure 7. Chloride concentration for wells near swine facility plus waste lagoon. Note similarity of chloride concentration from all monitoring wells except well #7. Well #7 has concentrations between that of the lagoon and the regional water chemistry, indicating probable mixing of waters.

The ratio of nitrate/chloride is another tool for determining sources of nitrate (Williams et al., 1998). Figure 8 shows the similarity of the nitrate/chloride ratio for the majority of wells near the swine facility sampled in December 1999. The domestic wells sampled in December 1999 show different ratios than the swine facility wells and monitoring well #7 shows an even larger difference from all of the other wells. The lagoon sample represents ammonium/chloride ratio because the nitrate-N concentration is negligible. The difference in the ratios suggests that different sources are responsible for the chemical values. As discussed previously the wells around the facility represent water that has been impacted by dry-land farming practices for the most part. The domestic wells represent the impact of human and/or other confined animal waste facilities near the wells. Monitoring well #7 appears to represent the mixing of the regional ground water with leakage from the lagoon.

Figure 8. Nitrate/chloride and ammonium/chloride ratios for December 1999 sampling at swine facility. Monitoring well #7 and the domestic well samples show very different ratios than the other monitoring wells and the supply well.

Figure 9. Range of nitrogen-15 values and probable sources from multiple studies (adapted from Heaton, 1986).

Animal-waste sources have a signature that is generally greater than +10‰. Nitrate-nitrogen concentrations often are above the drinking-water limit of 10 mg/L. Another possibility for an enriched value (greater than +10‰) is called denitrification. This occurs when bacteria break down the nitrate-N to nitrogen gas. This process usually results in nitrate-N concentration of below 1 mg/L as well as a high d ¹⁵N value (greater than +10‰).

The effects of various processes of the nitrogen cycle on the enrichment of d ¹⁵N are shown in figure 10. The figure illustrates the various sources of nitrogen to the environment and gives an indication of processes that result in changes in the d ¹⁵N signature of products from the various reactions. In general if biological processes occur such as bacterial use of nitrogen or animal use of plant material with the end product of manure or urea, the process preferentially uses the available ¹⁴N isotope. The net result is that the ¹⁵N isotope is concentrated in the remaining form of nitrogen. In the case of the study at Reno County, the dominant form of nitrogen in the ground water is nitrate.

As can be seen from the diagram (figure 10), animal waste starts as a +5‰ value. Any volatilization process of ammonia from the urea or manure will result in an additional increase of up to +3‰. Any biological intervention such as nitrification of ammonium in the barnyard will also increase the d ¹⁵N value of the resulting nitrate. Any biological use of the organic nitrogen in the system will also result in an increase in d ¹⁵N. The resulting nitrate will be high in concentration (usually if coming from a point source such as a confined feeding operation or septic system) with an enriched d ¹⁵N signature.

Figure 10. Effects of processes in the nitrogen cycle on d ¹⁵N values for different forms of nitrogen, particularly nitrate.

Nitrogen-15 isotope values at the Reno County site

Figure 11 shows the range of values of samples collected in December 1999 from monitoring wells around the lagoon, the lagoon, the supply well, and several homes in the area. The values from monitoring wells 6, 8, and 9 indicate that a fertilizer source is responsible for the high nitrate in these wells. The values from the lagoon and monitoring well #7 (downgradient along the flow path from the lagoon) both show d ¹⁵N values in the animal-waste range.

Figure 11 also shows the nitrate-n concentration and d ¹⁵N values for samples collected in March 2000. The nitrate-N values are higher in all wells and the d ¹⁵N values are in the fertilizer range for all wells except monitoring well #7. The lagoon was not sampled in March 2000, and the d ¹⁵N value was not determined for this period.

Figure 6 showed the variation in the nitrate-N concentration in the vicinity of the swine facility since 1995. Nitrate concentration has definitely increased in the wells in the spring. The fertilizer signature for the d ¹⁵N associated with these values strongly indicates rapid ground-water recharge in the area.

Figure 11. Nitrogen-15 and nitrate-N values for wells at swine facility and surrounding domestic wells. Samples collected in December, 1999 and March 2000.

The cross section of well logs from the area (figure 2) showed the presence of mostly fine to coarse sands within the aquifer with overlying sandy to clayey loam soil. The distance between well logs is a minimum of 300 feet (monitoring wells at site) up to 1 mile. The presence of seasonal high nitrate-N and variation in d ¹⁵N values suggests that the overlying thin silty clays to clays may not be continuous and do not prevent nitrification of ammonium sources to nitrate.

The rainfall data and depth to ground water (figure 4) show that there is a rapid response and subsequent rise in the ground-water table when certain types of rainfall events occur. This rapid response indicates that any dissolvable material applied to the surface will quickly reach the ground water if sufficient water is available. The time lag is probably minimal, on the order of weeks to months at the most.

Conceptual Model

The swine facility has been in operation since 1994. The bottom and sides of the lagoon are sealed with bentonite to a depth of 2 feet and compacted. In spite of this barrier zone, a seepage investigation by KSU researchers found that the lagoon leaks at a rate of 0.03125 inches/day. This is equivalent to a maximum discharge of 11.4 inches/year. The rate of leakage will be less than this amount because the lagoon is drained and the effluent used to irrigate cropland when a maximum depth is reached as specified in the swine-facility operational permit issued by KDHE.

The bentonite used in the lagoon has a specified cation-exchange capacity that is in the range of 78 to 100 centimoles of negative charge per kg of bentonite. This means the bentonite has a finite ability to absorb the ammonium that is produced by the swine and is discharged with the wastewater to the lagoon. If the ammonium reaches the unsaturated zone, it is available either to be nitrified by bacterial processes to nitrate or if a reducing zone exists beneath the lagoon then the ammonium will most likely be stored in the limited clay and silt available in the upper unsaturated zone beneath the lagoon.

The lagoon has a measured rate of leakage. The chloride concentration of both the lagoon and well #7 (downgradient) is much higher than any of the other monitoring wells, the supply well, or any of the other wells sampled in the area. The only exception is one domestic well which is located downgradient from a small feedlot located in T24S-R6W-Sec 21 (Appendix A). The high chloride and very low ammonium-N (0.85 mg/L) at this well probably indicates an impact from the underlying Permian bedrock aquifer. A nitrogen-15 sample was not collected at this site.

The d ¹⁵N of well #7 also is similar to that measured in the lagoon although it is lower in value. The chloride ratio of well #7 and the lagoon water suggests that approximately 30% of the water reaching well #7 is from the lagoon. This means that mixing of regional ground water and lagoon water has likely occurred. Because the nitrate-n value in well #7 is low (4 to 5 mg/L), the d ¹⁵N value is enriched relative to the other monitoring wells, and the chloride value is high. This suggests that leakage of the lagoon water may have caused a reducing water-chemistry zone in the subsurface that permits some denitrification of the nitrate-N in the regional ground water (13.6 mg/L nitrate-N in well 9 upgradient from the lagoon, fig. 4). The bicarbonate level in well #7 (447 mg/L) helps to support the idea of leakage from the lagoon but also may be an indicator that denitrification could be occurring in the subsurface, particularly if there is a reducing water chemistry below the lagoon.

Conclusions

Recommendations for Additional Work

1. Work is under way to obtain permission for collection of water samples for nitrogen isotope analysis from irrigation wells associated with the facility and down-gradient from the site.
2. Construction of a conceptual model of the flow and mass movement through the area underlying the lagoon will continue.
3. Additional fieldwork such as collection of a core under the lagoon for chemical and physical soil properties would provide additional confirmation of the occurrence of denitrification under the waste lagoon.
4. Collection of cores and installation of a monitoring well in the field where the waste is irrigated onto the field for disposal would permit evaluation of possible problems from movement of animal-waste nitrogen into the ground water in a non-reducing water-chemistry environment. Nitrogen-isotope work on these samples would permit definition of a limit to the north of the site in terms of the dimension of a possible plume of waste in the area.
5. Continue monitoring ground-water levels, water quality, and atmospheric conditions at or near the swine facility.

References

Davis, S. N., Campbell, D.J., Bentley, H.W., and Flynn, T.J., 1985, Ground Water Tracers: National Water Well Association, Worthington, Ohio, 200 p.

Gutentag, E.D.; and Weeks, J.B., 1980, Water table in the High Plains aquifer in 1978 in parts of Colorado, Kansas, Nebraska, New Mexico, Oklahoma, South Dakota, Texas, and Wyoming: U.S. Geological Survey, Hydrologic Investigations Atlas no. HA-642, 1 sheet, scale 1:2,500,000.

Ham, J.M., Reddi, L. N., and Rice, C. W., 2000, Animal waste lagoon water quality study: Research Report 99-123, Kansas Water Office, Topeka, KS, 138 p.

Heaton, T. H. E., 1986, Isotopic studies of nitrogen pollution in the hydrosphere and atmosphere: a review. Chemical Geology, v. 59, p.87-102.

Mueller, D. K. and Helsel, D. R., 1996, Nutrients in the nation’s waters — too much of a good thing?: U. S. Geol. Survey Circ. 1136, 24 p.

Townsend, M. A. 1999,Evaluation of sources of nitrate in ground water near Pretty Prairie, Kansas using nitrogen-15 isotope method: Kansas Geological Survey Open File Report 99-44, 14 p.

Townsend, M. A. 1997. Nitrate contamination of ground water in the vicinity of Haven, KS.: Kansas Geological Survey Open-file Report 97-79. 28 p.

Watts, K.R.; and Stullken, L.E., 1985, Generalized configuration of the base of the High Plains aquifer in Kansas: U.S. Geological Survey, Open-file Report no. 81-0344, 1 sheet, scale 1:500,000.

Weeks, J.B.; and Gutentag, E.D., 1981, Bedrock geology, altitude of base, and 1980 saturated thickness of the High Plains aquifer in parts of Colorado, Kansas, Nebraska, New Mexico, Oklahoma, South Dakota, Texas, and Wyoming: U.S. Geological Survey, Hydrologic Investigations Atlas no. HA-0648, 2 sheets, scale 1:2,500,000.

Williams, A. E., Lund, L. J., Johnson, J.A., and Kabala, Z. J., 1998, Natural and anthropogenic nitrate contamination of ground water in a rural community, California: Environmental Science and Technology, v. 32, no. 1, pp. 32-39.

Acknowledgments

Appendix A. Water-chemistry analyses from Reno County swine facility.

Samples collected by KDHE, KGS, and GMD2.