about us banner

WRC Home > About the WRC > CAIWQ

The Center for Agricultural Impacts on Water Quality: 2001-2005 Reserch Plan

PIs: Jim Anderson, Co-Director, Water Resources Center and Department of Soil, Water, and Climate; Patrick Brezonik, Department of Civil Engineering; William Koskinen, USDA-ARS and Department of Soil, Water, and Climate; Gary Malzer, Department of Soil, Water, and Climate; John Nieber, Department of Bioproducts and Biosystems Engineering; and Roger Becker, Department of Agronomy and Plant Genetics.

Contents

Importance

Objectives

Resume of Previous Investigations

Methods or Procedures

Literature Cited

Importance

All sectors of the US population, farm and non-farm alike, are demanding farming systems that protect and enhance the quality of our surface and ground waters, and at the same time assure a sustainable and high quality food supply. Pesticides and fertilizers used currently in crop production have been detected in both surface and ground waters associated with farming. Potential health and environmental risks associated with the presence of these agrichemicals in our water resource have not been defined satisfactorily, but detrimental effects have been observed. Future farming systems will be further constrained by environmental and social considerations (e.g., wetland preservation, wildlife sanctuaries, fewer farms) than are current systems.

From an agricultural perspective sustained use of fertilizers and pesticides are often necessary for a profitable enterprise. This combined with the application of animal wastes has affected both surface and subsurface transport of contaminants. Given the prohibitive cost involved in clean-up of water resources, there has been an increasing recognition that development and implementation of "best management practices" (BMPs) at the farmer scale is the preferred strategy for solving water quality problems.

The rate and extent of non-point pollution of water resources is spatially and temporally variable. Variations depend on soil distribution, climate, topography, subsurface geology, land use, land management strategies, and intensity of chemical applications. Agriculture contributes significantly to water quality problems. Major agricultural chemicals of concern in surface runoff are nutrients, principally phosphorus and nitrogen, and pesticides. Pesticides and nitrate-nitrogen are of concern in regard to potential contamination of ground water.

Entry of nutrients into surface water accelerates eutrophication of lakes and associated rapid growth of undesirable vegetation, algae blooms, loss of dissolved oxygen, alterations of fish populations, and loss of recreational status.

For more than 50% of lakes with impaired water quality and 60% of rivers with impaired water quality, agriculture has been identified as the source of the nutrient pollution. Nutrient losses from the Upper Mississippi Basin have been implicated as the major cause of hypoxia in the Gulf of Mexico. Locally, the Minnesota River accounts for 85% of the nutrient delivered to Lake Pepin on the Mississippi River and 1,500 tons/year of phosphorus.

As a result of these levels of impairment, states will be required to establish Total Maximum Daily Loads (TMDLs) of nutrients for watersheds. This will require sound review to evaluate the effectiveness of alternative management practices for water quality protection and economic viability.

Transport of soluble nitrate-nitrogen to ground water in water percolating beyond the root zone has been well documented. Adverse health effects for humans and livestock is the principal concern.

Pesticides are transported in solution phase to both surface and ground water. Sediment bound pesticide transport to surface water is another problem. Deterioration of drinking water quality and associated health hazards are the principal sources of concern related to pesticide contamination

Numerous classes of pesticides and their metabolites have been found by USGS surveys in almost every stream and fish sample and in about 60% of the shallow wells sampled in agricultural areas. Although most of the levels measured are far below health advisory levels, there are major concerns about their impact on human and ecosystem health. In addition, TMDLs of pesticides will be required for impaired watersheds.

Research is necessary to establish parameters that can be used to make site specific recommendations that provide for a profitable agriculture while minimizing impacts on surface and ground water. This research will provide strategies to protect surface and ground water. These strategies will involve a combination of practices to resolve apparent conflicts . between protecting surface and ground water quality.

Understanding both the water quality and economic implications of alternative practices is an important ingredient for making good investments and sound policy to reduce agricultural water pollution. A research effort to evaluate the effectiveness of alternative management practices and the associated costs and benefits of these practices for different types of agricultural producers needs to bee undertaken. Therefore, this project will have the following objectives:

Objectives

  1. Evaluate the impact of controlled drainage to mitigate nutrient loss from tile drainage systems in the Minnesota River Basin.
  2. Characterize, quantify and integrate the basic processes of sorption, desorption, degradation and transport of herbicides in soils to minimize the potential transport from point of application to surface and ground water.
  3. Develop and disseminate farmer-led and farmer-sanctioned water quality improvement initiatives and achieve measurable improvements in water quality.
  4. Estimate the costs and benefits of water quality improvements and assess the potential for differential economic impacts of specific nutrient BMPs.
  5. Continue public education programming to increase adoption of BMPs to improve water quality and maintain farm profitability.

Resume of Previous Investigations

The Agricultural Experiment Station established this interdisciplinary research effort in 1986. Principal objectives were to 1) develop the laboratory capability to evaluate the presence of pesticides in the environment; 2) establish field research sites to determine best management practices to minimize the movement of nitrate and pesticides to ground water; and 3) determine possible mechanisms for movement of agrichemical through soils to ground water. Research has focused on the basic behavior of chemicals and nutrients in the environment, practices for responsible management of chemicals, and modeling chemical behavior in the environment.

In order to determine the potential impact of agricultural practices on water quality requires an understanding of how water and chemicals move in soils. Field methods for measuring water and chemical movement have been evaluated. Methods include suction lysimeters, zero-tension pan samplers, and wick samplers. This combined with other USDA funded research has indicated the importance of macropores on water and agrichemical movement to ground water.

Research has shown that fanning systems in sand plain settings using reduced herbicide inputs through banded applications, reduced nitrogen application based on proper credits and soil tests, using residue management for erosion control can be effective in reducing movement of agrichemicals to ground water. Similar efforts need to be conducted in areas where surface runoff is also of concern. The Minnesota River Basin is a major agricultural production area, and the river is listed as one of the 20 most polluted in the country. There is a need for systems research to establish the relationships between surface water quality and the agricultural management systems.

The results of the Princeton, Minnesota, MSEA farming systems study showed that the corn/soybean ridge conservation tillage system reduced the entry of atrazine metabolites into groundwater compared to both the conventional tillage continuous corn and sweet corn/potato systems (Lamb et al 1998). Results suggest that the corn/soybean system reduced the potential risk of herbicide contamination relative to the other two systems. Yield measurements indicated grain yields for corn were greater for the corn/soybean rotation than under the continuous corn system, resulting in larger gross returns for rotation corn. Nitrate-N concentrations in the upper one-meter saturated zone were also least below an area cropped with a corn/soybean rotation and greatest beneath the two cropped areas with the sweet corn/potato rotation. These results demonstrate that the ridge tillage corn/soybean system provides a financially viable and more environmentally benign alternative to a conventional full-width tillage continuous corn production.

To better understand the economic and environmental implications of alternative production practices, this study examined the linkages between agricultural production decisions, profitability, and water quality. In the Midwest, agricultural production on approximately 7 - million hectares of sandy soils threatens to impair, or impairs, ground water quality caused by leaching of agrichemicals. Results from this study demonstrate the vulnerability of groundwater to damage from agrichemicals in the sandy outwash plains. They also indicate that with "careful management, the soils can be cropped intensively without undue damage to water resources," (Lamb et al 1998). The conservation ridge-tillage system for corn and soybeans provides an excellent example of how improvements in farm management practices can reduce the potential for leaching of dangerous agrichemicals into the groundwater. In contrast, the conventional tillage sweet corn/potato system indicates the potential threat to groundwater resources from current management practices.

Combining the economic analysis with the findings from the fanning systems and water quality study yields mixed results: The economic analysis revealed that for grain production, the ridge tillage com/soybean system performed better than the conventional continuous corn system in terms of relative profits and risk, as measured by net revenue stability. These results suggest that improvements in groundwater quality in outwash sand plains agricultural areas in the Midwest could be achieved through adoption of the ridge-tillage corn/soybean system in lieu of the conventional full-width tillage continuous corn system. A movement away from continuous corn and toward the corn/soybean system represents a win-win outcome for agricultural producers and the environment.

In contrast, the economic analysis suggests that the sweet corn/potato production system yields relatively high, although significantly variable, net profits. Thus, the cropping system that poses the most significant threat to groundwater quality on the sandy plains yields the largest potential profits of the three studied. This is a disturbing finding. It suggests that producers will likely be reluctant to adopt more environmentally benign agricultural practices unless they have economic incentives or mandates to do so. Substantial financial risk associated with the volatility in year-to-year net profits on sweet corn /potato rotations has had a dampening effect on the expansion of production of these crops.

Field research in Minnesota has shown that atrazine desorption, and hence its leaching potential, is lower than is predicted from laboratory studies and explains why numerous transport models overestimate pesticide leaching potentials. Our research also demonstrated that degradation products of atrazine are more strongly adsorbed onto soils than is the parent compound and do not desorb appreciably. Hence, it appears that these products will not leach to ground water or be released into surface waters. Further research showed that complexation of atrazine with soluble soil organic materials (humic substances) doubled as pH increased from 4.5 to 7.5. Such complexes can dominate the fate and transport of herbicides in surface soils. As we move to new farming systems with less reliance on herbicides, future research with the more environmentally benign "newer generation" herbicides must include characterization of basic herbicide/soil chemistry processes so successfully carried out for atrazine within the MSEA research program.

The ultimate goal of this research is the development of sustainable farming systems, with the supporting knowledge base, for less reliance on agrichemical inputs. These farming systems incorporate a tactical management concept of responding to management needs (post-priori decision making) versus the traditional a priori strategy (preplant inputs) in which inputs are based on anticipated needs. Such management concepts include, use of cover crops, substitution of previously fixed nitrogen (N) for fertilizer-N application, plant tissue and soil testing, variable-rate agrichemical application, seed bank assessment models.

Methods or Procedures

To meet Objective 1, the following activities will be conducted at the Southwest Research and Outreach Center-at Lamberton.

The field scale water quality simulation model ADAPT (agricultural Drainage and Pesticide Transport) will be used to identify agricultural areas that may be prone to contribute to nutrient loading to nearby streams and rivers. The ADAPT model includes routines for soil hydrology (soil freezing, snowmelt, runoff, infiltration, macropore flow, drainage, evapotranspiration, and seepage); erosion; crop growth, uptake, and yield; and nutrient (nitrogen and phosphorus) and pesticide transport. Recently, the ADAPT model was calibrated and evaluated for continuous corn with conventional tillage on a poorly drained Webster clay loam soil in southern Minnesota.

A paired ditch design will be used to evaluate the impact of drainage ditch modification on water quality. The concept of using a paired ditch design comes from research at the sub-watershed experimental areas, a control and a treatment area, and two periods of study, a calibration and a treatment period. Discharge and water quality from both ditches will be monitored during the calibration period. For the treatment period, one ditch will remain unchanged while the other ditch will be modified. ditch modification options include 1) small retention structure, 2) vegetation or 3) ditch design (terracing, increased sinuosity). Regression analysis of discharge and N03-N will be used to evaluate differences in water quality form the calibrations and treatment periods for the treated and untreated ditch. The aim of ditch modification is to reduce N03-N and increase residence time of tile outflow in the ditch to promote denitrification and/or nitrogen uptake by growing vegetation.

The following activities will be conducted at the Southern Research and Outreach Center at Waseca.

Nine tile depth and spacing plots at the Agro-Ecological Research Farm have been equipped with monitoring instrumentation. The effect of tile spacing and depth on transport of nutrients and profitability will be evaluated.

Four controlled-drainage plots have been developed with water table control structures and surface runoff flumes. Optimum water table management strategies to reduce the loss of nitrate (N) from the fields while maintaining crop yield will be investigated.

To meet Objective 2, the following activities will be conducted in cooperation with USDA-ARS scientists at the Rosemount Research and Outreach Center.

The field dissipation of herbicides varies spatially at the landscape scale and is controlled by the spatial variation in the soil properties that affect herbicide-soil interaction, giving rise to the possibility of reducing filed losses of herbicides through site-specific management. The degree to which observed spatial and temporal variations in pesticide concentrations may be explained by coinciding variations in soil properties will be tested. A 6.0-ha field has been identified as part of a watershed where four man-made terraces exist, such that each is about 2.5 ha and each has its own system for collecting runoff. Each terrace is composed of similar soils and there is no water movement between them. The site has previously been used as twenty-year biosolids study where municipal biosolids were applied to the terraces (Linden et al., 1995). Biosolids have not been applied to the area in more than 10 years allowing for natural erosion process to take place on the rolling landscape. This gives rise to high variability in soil properties like organic carbon, pH, and clay content on a relatively small scale. Making a case for site-specific management of herbicides on this extreme end will ensure the practicality of pursuing such investigations on fields-with less variability.

Soil samples will be taken on each terrace at georeferenced locations that approximate a 25-m grid sampling scheme that optimizes the accuracy of the variability in OC, pH, and soil texture (Mulla and McBratney, 2000; McBratney and Pringle, 1999). GIS and mapping software packages will be used to accurately depict the spatial variability of soil properties through semivariogram and terrain analyses. The four terraces will be cultivated in corn with acetochlor and isoxaflutole applied preemergence at uniform recommended rates. A subset of the georeferenced locations that cover a range in observed soil properties will be chosen. Using sorption methods developed in our laboratory (Berglof et al., 2000; Celis and Koskinen, 1999) and traditional batch equilibrium techniques, sorption (Kd values) of acetochlor and isoxaflutole will be assessed to determine the distribution and variability of the sorption coefficient in the field and to determine the relationship between Kd and soil characteristics. Two-foot soil cores will be taken at the same locations throughout the field season in order to determine herbicide concentration in the surface soil and also at different intervals in the soil profile over time. Acetoclor extraction will be performed on an automated robotic system and analyzed on GC/MS (Marek and Koskinen, 1996; Marek et al., 2000). Runoff samples for each terrace will also be analyzed for the two chemicals.

Spatial and temporal patterns in herbicide concentration will be mapped and analyzed against soil property data using multivariate and spatial statistical methods in an effort to elucidate any spatial correlation between herbicide concentration and soil properties and to what extent this variation takes place. The Kd values will also be mapped in order to visualize potential areas for herbicide sorption and loss and to see if these areas indeed correspond to what is seen in the field. We would expect to see herbicides detected lower in the profile for areas with a low Kd and vice versa for areas with a high Kd value. This information can lead to the creation of management zones based on soil properties that are fairly inexpensive and easy to measure (Oliveira et al., 1999). In pesticides, therefore, similar studies involving spatial variation of crop yield (Timlin et al., 1998), soil fertility (Mulla et al., 1996, and Mallawatantri and Mulla, 1992) have begun to evaluate site-specific management of pesticides. While Oliveira et al. (1999) determined that there is spatial. variability in Kd values of imazethapyr, no study currently exists in the literature that examines the spatial and temporal variability of the field dissipation of herbicides. Data generated from this project will provide a large and unique data set from which we can begin to evaluate the potential for site-specific management of pesticides.

To meet Objective 3, the following activities will be conducted for paired watersheds in the Minnesota River Basin.

The ADAPT model will be used to evaluate the long-term impacts of changes in farm nutrient management for the paired watersheds studies. We have had extensive experience with this model in Minnesota at a variety of scales, ranging from the field plot (Davis et al., 2000), the commercial farm (Mulla et al., 2001 b), and the major watershed (Dalzell et al., 2001). In Bevens Creek watershed (83,000 acres), a Minnesota River Basin watershed very similar to Huelskamp Creek and Seven Mile Creek, Dalzell (2000) used the ADAPT model to simulate nitrate loads for three years. Results from the ADAPT model agreed very closely with measured nitrate loads from an ISCO 6700 located at the mouth of the watershed. A similar approach will be used to calibrate the ADAPT model against baseline monitoring data in each of the four watersheds in this study. The model will then be validated using the additional years of monitoring after BMPs have been established in the treated watersheds of Huelskamp Creek and Seven Mile Creek. Validation will also be performed for non-baseline water quality monitoring in the control watersheds.

Once the ADAPT model has been calibrated and validated, it can be used with long-term climatic records (SO years) from existing nearby climatic stations (Mankato, Minnesota) to investigate the long-term water quality impacts of the control versus treated watersheds. several scenarios will be investigated for long-term modeling efforts to provide information for an economic evaluation of BMPs. These include increased adoption of conservation tillage, increased adoption of phosphorus application rate reductions and improved methods of application, increased adoption of N credits and reasonable yield goals, and increased adoption of late fall N applications. In addition, we will simulate water quality trends after increased tile drainage in the watersheds. Outputs from the model will give the mean loads of sediment, nitrogen, and phosphorus, as well a cumulative probability distributions for water quality impacts of each practice.

To meet the requirements of Objective 4 the following activities will be conducted.

While the ADAPT modeling results provide valuable information regarding the effectiveness of alternative management practices to reduce agricultural pollution, farmers will not adopt these practices unless they are economically viable. Understanding both the water quality and economic implications of alternative practices is an important ingredient for making good investments and sound policy to reduce agricultural water pollution. the problem facing decision makers in the Minnesota River Basin is a lack of information on the effectiveness of alternative management practices to reduce agricultural water pollution, and the associated costs and benefits of these practices for different types of agricultural producers. To better understand the economic implications of alternative practices, economic data from the annual on-farm surveys will be examined to analyze potential changes in profits associated with adoption of alternative BMPs. Data from the annual surveys will include production practices, input levels and costs, and yields. Since risk and uncertainty influence production practices as well as profitability, these factors will be incorporated quantitatively into the analysis to the extent possible.

The results of the above analysis will be used along with the ADAPT modeling results to determine abatement costs for various levels of reduction in phosphorus and nitrogen loads to surface waters. We have already used this approach in the Sand Creek watershed of the Minnesota River Basin very successfully (Johannsen et al., 2001).

To meet Objective 5, the following activities will be conducted.

Educational programming will be coordinated with existing agencies (UM Extension Service, MPCA, MDA, BWSR, USDA-NRCS, SWCD) and organizations (RCRCA, MNLICA) to develop an informational network for educating stakeholders and enhancing land use changes aimed at improving water quality. Educational programs will consist of workshops and forums for stakeholders on water quality issues, management strategies, and implementing BMPs. Workshops and forums will encourage stakeholder participation and stakeholder-researcher interactions. A moveable display will be developed to inform stakeholders and others about progress on the project and opportunities for participation.

Stakeholder participation is a critical component to implementing control measures and for attaining and maintaining water quality. field tours will be planned to show stakeholders management control measures used to improve water quality. Publications will describe key water quality issues, highlight accomplishments toward attaining water quality standards, and provide examples of water quality problems and solutions.

Literature Cited

Berglof, T., W.C. Koskinen, J. Brucher, and H. Kylin. 2000. Linuron sorption-desorption in field most soils. J. Agric. Food Chem. 48:3718-3721.

Celis, R. and W.C. Koskinen. 1999. An isotopic exchange method for the characterization of the irreversibility of pesticide sorption-desorption in soil. J. Agric. Food Chem. 47-782-790.

Dalzell, B.J., D.J. Mulla, and P.H. Gowda. 2001. Modeling and evaluation of alternative agricultural management practices in sand Creek Watershed. Soil Erosion Research for the 21" Century. J.D. Ascough II and D.C. Flanagan (eds.), Am. Soc. Ag. Eng., St. Joseph, MI, pp. 637640.

Davis, D.M., P.H. Gowda, D.J. Mulla, and G.W. Randall. 2000. Modeling nitrate nitrogen leaching in response to nitrogen fertilizer rate and tile drain depth or spacing for southern Minnesota, USA. J. Environ. Qual. 29:1568-1581.

Johansson, R.C., P.H. Gowda, D.J. Mulla, and B.J. Dalzell. 2001. Modeling economic effects of phosphorus BMPs in the Minnesota River Valley. J. Soil Water Conservation. (Submitted)

Lamb, J., R. Dowdy, J. Anderson, R. Allmaras. October 1998. Water quality in an irrigated sandy soil: ridge tillage in rotated corn and soybean compared with full-width tillage in continuous corn. Soil and Tillage Research 48(3).

Linden, D.R., W.E. Larson, R.H. Dowdy, and C.E. Clapp. 1995. Agricultural utilization of sewage sludge: A twenty-year study at the Rosemount Agricultural Experiment Station, University of Minnesota. Minn. Agric. Exp. Sta. Bull. #606-1995. p. 1-93.

Mallawatantri, A.P., D.J. Mulla. 1992. Herbicide adsorption and organic carbon contents on adjacent low-input versus conventional farms. J. Environ. Qual. 21:546-551.

Marek, L.J., and W.C. Koskinen. 1996. LC-MS analysis of 4-methoxy-6-methyl-1,3,5-triazin-2yl containing sulfonylurea herbicides in soil. J. Agric. Food Chem. 44:3878-3881.

Marek, L.J., W.C. Koskinen and G.A. Bresnahan. 2000. LC/MS analysis of cyclohexanedione oxime herbicides in water. J. Agric. Food Chem. 48:2797-2801.

McBratney, A.B. and M.J. Pringle. 1999. Estimating average and proportional variograms of soil properties and their potential use in precision agriculture. Precision Agric. 1:125-1 S2.

Mulla, D.J. and A. B. McBratney. 2000. Soil spatial variability. Handbook of Soil Science. Sumner (Ed.), pp. A321-352.

Mulla, D. J., P. Gowda, W.C. Koskinen, B. R. Khakural, G. Johsnon, and P.C. Robert. 2001b. Modeling the effect of precision agriculture: Pesticide losses to surface waters. Terrestrial field Dissipation Studies. E. Arthur and V. Clay (Eds.). ACS Symp. Ser., ACS, Washington, DC. (In press).

Oliveira, R.S., W.C. Koskinen, F.A. Ferreira, B.R. Khakural, D. J. Mulla and P.J. Robert. 1999. Spacial variability of imazethapyr sorption in soil. Weed Sci. 47:243-248.

Timlin, D. J., Y.A. Pachespsky, V.A. Snyder and R.B. Bryant. 1998. Spatial and temporal variability of corn grain yield on a hillslope. Soil Sci. Soc. Am. J. 64:764-773.