MOE001 - Quantifying Effectiveness of Nutrient Land Application Options within Regional Aquifer Systems and Municipal Wellhead Protection Areas

This project was directly funded by the Ministry of the Environment through the Nutrient Management Joint Research Program. The program supported the development of environmentally effective and scientifically robust management practice options for land application of agricultural and non-agricultural source materials in Ontario.

Please contact Clara Tucker, Water Resources Scientist (Ministry of the Environment) at 416-314-0583 for more information.

Lead researcher

Dr. David Rudolph, Department of Earth Sciences, University of Waterloo

Objectives

The goal of this project is to assess the effectiveness, performance and efficiency of regional-scale reductions in land application of nutrients as an optional practice designed to mitigate risk of nitrate groundwater contamination in Ontario's critical glacial aquifer systems.

The four project objectives to accomplish this goal are:

1. Estimate, using innovative field techniques, the transient nitrate mass loading rate below the root zone at multiple locations within an established agricultural land parcel operating under reduced nutrient application practices.
2. Evaluate the utility of commercially available and validated one-dimensional numerical models for unsaturated flow and transport to effectively estimate nitrate mass loadings under various geologic and hydrogeologic conditions.
3. Establish scientifically-based methods to spatially scale-up the nitrate mass load estimates determined through Objectives 1 and 2 to the field scale, multiple-field scale, and capture zone scale within in a complex glacial environment with a focus on quantifying related uncertainty.
4. Use a sophisticated capture-zone scale groundwater flow and transport model to evaluate the spatial and temporal effectiveness of the implemented reductions in land applied nutrients to mitigate nitrate groundwater contamination in the vicinity of a municipal well field.
   

Expected benefits

The combined use of field data, appropriate data scaling and advanced modeling techniques will be used to permit regional assessment of potential risk to a water supply and provide an approach to evaluate the utility of any recommended land application options.

Results

In agricultural areas in Ontario excessive application of nutrients may negatively impact underlying groundwater quality and associated municipal water supplies. This research focused on quantifying the effectiveness of regional scale reductions in nutrient applications in lowering nitrate concentrations in groundwater systems where Beneficial Management Practices (BMPs) were implemented.

Work was performed at the Thornton Well Field, near Woodstock, ON where groundwater from several of the production wells has chronically high levels of nitrate. One hundred and eleven (111) hectares of agricultural land was purchased in 2002 within this municipal drinking water supply's well head protection area and a nutrient management plan (BMP) was implemented in 2003 that involved a change in crop rotation and significant reductions in the rate of fertilizer application (40 to 50% reduction). Innovative field techniques and laboratory analyses established scientifically-based methods to spatially scale-up point scale nitrate mass load estimates at 8 monitoring stations to the field scale, multiple-field scale, and capture zone scale within a complex glacial environment.

Detailed coring of the unsaturated zone (3->20 m beneath the root zone) after three years of implementing BMPs showed that average nitrate concentrations had decreased 60% (from 22 to 8 mg NO3-N/L), total stored mass of nitrate reduced by 20% to 60% , and mass flux through the unsaturated zone reduced an average of 45% . Monitoring changes in the unsaturated zone was a more sensitive and effective way of detecting early time improvements caused by the BMPs than the sampling of the underlying groundwater. The total reductions in nitrate mass flux of 45% compare well to the estimated 40% to 50% reduction in nitrogen application.

An evaluation of 50 commercially available 1-D models indicated the SHAW, HELP, RZWQM and HYDRUS models were the most applicable and capable for estimating nitrate mass flux. Overall, the RZWQM model performed the best and realistically represented field-measured recharge rates.

Predictive upscaling of the point data was undertaken by instrumenting an additional 7 monitoring stations in a parcel of land adjacent to the original 8 stations. Accurate upscaling depended not only on topography, surface soils, and land use, but also on surface runoff, ephemeral flows, and shallow geology. Significant improvement in upscaling prediction can be attained through the collection of soil cores to a depth of approximately 1 m or just below the root zone. Point scale data were extended to the field scale using various extrapolation methods (arithmetic averaging, contouring, Thiessen polygons, and land type classification) and all methods agreed closely with each other at larger scales.

Overall the analysis suggests that even in complex hydrogeologic settings, useful information can be obtained by extrapolating point data over larger regions and that simple averaging of point data can provide a reasonable estimation of regional scale characteristics. Regional monitoring of groundwater nitrate concentrations throughout the study area has not suggested significant improvements in water quality as a result of the implemented BMP. This is likely due to the long time lag required for these impacts to be seen in the deep subsurface. Where shallow multilevel monitoring wells have been installed, however, clear evidence of reductions in groundwater nitrate concentrations have been documented.

A 3-D capture zone scale, variably-saturated, flow and transport model (FeFlow) was then used to evaluate the spatial and temporal effectiveness of the implemented BMPs. A 22% reduction in average municipal well nitrate concentrations was predicted and was similar to the field data estimate of 15% to 20%, however, the full benefit of the BMPs was not predicted to be seen at the wells for at least 5 years.

Alternative scenarios were modeled where minimum loads were simulated and additional areas subjected to nutrient reductions, and the estimated final average well concentrations reduced by 37% and 30%, respectively. The overall time required for the full benefits of the implemented BMP strategies to be realized at the municipal wellheads ranged from 5 years to 20 years depending on the specific well and the scenario considered.

This approach can be applied to other well fields to provide insight with respect to the magnitude and timing of water quality changes and assist in water resources management decisions. It also provides a physically based tool to estimate the effects of different management alternatives which aid in the decision making process regarding the management of groundwater supplies.

In addition this approach can be generalized and applied at other well fields affected by non-point sources of contaminants. This information can also provide valuable input to assist with further data collection efforts that can be used to reduce model uncertainty.

Parameters studied

• nitrate
• subsurface flow

Recommendations relevant to nutrient management

Overall, these field and modeling results indicate that the implementation of BMPs in selected areas within the capture zone is a promising strategy to improve the excess nitrate conditions at supply wells.

Related information

Other projects funded through the Nutrient Management Joint Research Program:

• in 2006
• in 2007