Research Brief: Assessing Practices to Decrease Agricultural Phosphorus Losses in Wisconsin

Abstract

This study evaluates agricultural conservation practices for potential P loss reductions, considering the unique regions, soils, and cropping systems of Wisconsin. For reducing total P losses, four practices were particularly effective: decreasing tillage intensity, adding edge-of-field filter strips, incorporating perennial or overwintering crops, and early establishment of cereal rye cover crops. Shifting from intensive tillage to no-till consistently offers high P reduction, even on soils with high soil test P. Case studies emphasize that grassed waterways can significantly decrease total P losses in erosion-vulnerable fields, and avoiding winter manure application reduces the risk of P runoff. While most strategies target soil-bound P, long-term soil-test P drawdown through decreased P fertilizer application is uniquely effective at reducing dissolved P losses. Overall, this study highlights the potential benefits of controlling total P losses through a combination of in-field practices (tillage reduction and early planted cover crops), structural practices (filter strips and grass waterways), and cropping-systems changes (adding overwintering or perennial crops) in the most vulnerable fields.

Background

Agricultural lands are a major source of phosphorus (P) to surface waters. Numerous conservation and management practices have been proposed to address agricultural P losses, but their effectiveness can vary markedly, particularly when considering diverse soil types and agricultural systems of Wisconsin. This work is unique because it is the first known systematic examination of various cropland management combinations tailored specifically to 9 different regions of the state.

Goal

Evaluate potential benefits of agricultural practice changes in decreasing P losses in surface runoff from Wisconsin’s croplands. 

The results are valuable to farmers and farm advisors because the predicted reductions show the effectiveness of each practice for P loss reduction tailored by specific regions and soil types in Wisconsin. For the greatest benefit, locate fields with high baseline P losses, such as those with highly erodible soils or low residue cover, and use these results to identify where and when specific practices (or combinations) are most effective for P loss reduction. A range of combinations of practices can also be explored in an online tool under development.

The report shows that no single practice is universally “best” for every situation. Instead, practitioners are encouraged to weigh logistical, economic, and environmental trade-offs for each specific field and cropping system.

Methods

In this report, we used two main approaches to evaluate potential benefits of agricultural practice changes in decreasing P losses in surface runoff from Wisconsin’s croplands. 

First, researchers from UW-Madison used the Wisconsin Phosphorus Index (P Index), a model that has been calibrated specifically for Wisconsin’s crop and landscape characteristics with hundreds of site-years of on-farm measurements from commercial and research farms across Wisconsin. Using the SnapPlus software,  researchers  modeled over 300,000 unique field-scale scenarios, which strategically span a vast range of soil types, climate regions, and management conditions across the state. Science assessments in other states have typically reported broad geographic averages with high variability for different practices. A benefit of this report’s modeling approach is the ability to identify the specific contexts where different practices are effective. The results in this research brief highlight the percent reduction in P loss relative to baseline scenarios, but the full report also shares the raw estimated P losses in lb ac^-1 y^-1 (i.e. estimated average annual surface runoff P losses at the field edge).  

Second, surface P runoff data collected from the Wisconsin Discovery Farms program, and data from other Wisconsin runoff studies published in scientific literature, were used to provide context and targeted assessment of additional management practices. By integrating on-farm research from the Discovery Farms program, the results provide a “reality check” for the model and offer context on how practices perform on specific farms.

Results

Step 1 in reducing P loss: Reduce erosion. Four practices offer particularly high potential for reducing total P losses:

1. Decreasing tillage intensity. This creates less surface disturbance and more residue cover. Compared with chisel-disk tillage, P loss reduction was greatest for no-till, followed closely by strip till. Vertical till provided less benefit. 
2. Adding a perennial grass buffer (filter strip) or a grass waterway. These slow down water and trap sediment, reducing P losses.
3. Adding an overwintering or perennial crop (e.g., winter wheat or alfalfa) to an annual crop rotation , or converting annual rotations to perennial systems (e.g., pasture). Increasing perenniality creates more living roots and soil cover and protects soil from disturbance outside the period of  corn and soybean growth.
4. Adding an early-seeded winter rye cover crop. Getting the cover crop on early enough to establish enough biomass is key.

Of these practices, shifting from more intensive tillage (e.g., chisel-disk) to no-till often had large P loss reduction across regions and cropping systems, and shifting from chisel-disk to strip-till also had large benefits. For some cropping systems and soils, filter strips achieved even greater P loss reduction than a decrease in tillage intensity. When comparing among cropping systems, well-managed grazed perennial pastures generally had lowest P losses; aside from pasture, adding alfalfa in rotation with corn silage, or winter wheat in rotation with corn and soybean, could also decrease P losses from existing forage or grain cropping systems. Effectiveness of a cereal rye cover crop for P loss reduction varied markedly with planting date. Aerial seeding of rye into an existing crop canopy in August generally had high P loss reduction potential and often approached values for no-till or filter strips. In contrast, drill-seeding rye in October did not decrease P losses in most cropping systems and soils, and while drill-seeding rye in September often decreased P losses, this practice would be difficult to implement for many cropping systems.

Step 2 in reducing P loss: Once erosion is controlled, gradual P draw-down in fields with high or excessively high soil P content is the only consistent way to reduce dissolved P loss. 

Lowering soil P content became increasingly important for achieving further P loss reduction for systems with lower baseline P losses, and was the only practice evaluated that consistently decreased dissolved P losses. 

The simulations indicated that relatively low P losses could potentially be achieved across most soils, regions, and cropping systems examined here by adopting combinations of multiple practices that decreased soil disturbance and increased plant biomass and residue cover.

For particularly vulnerable soils and cropping systems (e.g., high slope, low residue cover), multiple practices would often be needed to achieve strict water quality goals. 

Controlling disproportionate soil losses from vulnerable areas of a field is crucial to get the maximum impact from multiple practices. In a Discovery Farms study, a field that had long-term no-till and cover crops still had high P losses due to a small concentrated flow path. Installing a well-designed grassed waterway reduced runoff event total P losses by 75%, on average.

Table 1.1 summarizes the primary practices evaluated in this report and their relative effectiveness in reducing runoff P losses. Although the effects of different practices varied across regions, cropping systems, and baseline conditions, several practices consistently had high potential for reducing P losses. General qualitative conclusions are presented in this table. Numerical summaries are provided along with greater context in the full report. Notice that most practices are not effective in reducing dissolved P loss, except reducing soil test P over long periods of time. Learn more about caveats identified in each practice in the full report.

Practice	Category	Relative P Loss Reduction Potential	Caveats
Tillage reduction (P index model)	No-till, strip-till, or vertical-till vs. chisel-disk tillage	Higher (no-till, strip till), lower (vertical till)	Ineffective for dissolved P
Filter strip (vegetated buffer) (P index model)	Filter strip vs. none	Higher	Ineffective for dissolved P
Grassed waterway (Discovery Farms data)	Grassed waterway vs. none	Higher	Only applicable to fields with gully erosion; ineffective for dissolved P
Winter rye cover crop (P index model)	Oct.drill-seeded, Sept. drill-seeded, or Aug. aerial-seeded vs. none	Higher (Aug. aerial seeded), lower (Sept. drill planted), inconsistent (Oct. drill planted)	Ineffective for dissolved P; assumes consistent cover crop establishment, which is difficult to achieve
Winter rye cover crop (Discovery Farms data)	October drill-seeded vs. none	Inconsistent
Diversified rotation / perennial cover (P index model)	Corn-soybean wheat vs. corn-soybean, corn silage-alfalfa or grazed pasture vs. corn silage, grazed pasture vs. corn silage-alfalfa	Higher	Ineffective for dissolved P; wheat scenarios included straw retention
Soil test phosphorus reduction (P index model)	100, 50, 25 ppm	Higher to lower	More effective for soils with excessively high P. Effective for reducing dissolved P loss, but requires many years of low P inputs to achieve
Nutrient form (P index model)	Liquid dairy manure vs. commercial fertilizer	Inconsistent
Nutrient application timing (P index model)	Spring vs. fall	Lower to inconsistent	Effectiveness varied with tillage practices; spring application was more beneficial when fall-applied nutrients were not incorporated through tillage, such as in no-till and cover-cropped fields
Nutrient application timing (Discovery Farms data)	Non-frozen soil vs. frozen soil	Higher
Manure placement (Discovery Farms data)	Subsurface vs. surface	Inconsistent	Other practices (tillage, cover crop) co-varied with manure placement

Grounded in Measured Data

Analysis of P runoff data from Discovery Farms sites provided further insight on the P loss reduction from grassed waterways, changes in manure placement and timing, and cover crops.

• In one case study where a well-designed grassed waterway was installed in a field with long-term no till and cover crops, runoff event total P losses decreased by 75%, on average. This case study highlighted the significance of controlling disproportionate soil losses from vulnerable areas of a field. 
• A second case study analyzed the relationship between manure application and P loss across 14 farms and 671 runoff events. Manure applications on non-frozen soils were associated with 30% lower P losses in later runoff events, compared to manure applications on frozen ground, supporting the water quality benefits of avoiding winter manure spreading. Interestingly, there was no consistent difference in runoff event P loss following surface-applied manure and injected or incorporated manure, highlighting the potential trade-offs in dissolved and particulate P losses associated with manure placement. Runoff events occurring shortly after manure application tended to have higher P losses, but the overall relationship was relatively weak—that is, there was only a 20% reduction in runoff event P load for each year since P application. 
• A third case study evaluated the impact of an October drilled cereal rye cover crop in a corn-soybean rotation, and found no measurable benefit of this practice for P-loss reduction, consistent with predictions from the P-index modeling (late seeded cover crops can have other benefits, however, like decreasing nitrogen losses).

Conclusion

Overall, the results highlight the potential benefits of first controlling total P losses through a combination of in-field practices (e.g., tillage reduction, early planted cover crops), structural practices (filter strips and grass waterways), and cropping-systems changes (adding overwintering or perennial crops) in the most vulnerable fields. After addressing soil loss, gradual P draw-down in fields with high or excessively high P content can further decrease P losses, particularly the dissolved P component, along with optimization of form and timing of P application. In ongoing work, our team will continue to update this document in response to new data and stakeholder feedback, and we will begin to evaluate the effectiveness of agricultural practices in reducing nitrogen (N) losses.