Month: December 2025

A New Webtool to Support On-Farm Decision Making with Single-Strip Evaluation Trials

Srinivasagan N. Subhashree1, Rahul Goel2, Manuel Marcaida III1, Juan Carlos Ramos-Tanchez1, and Quirine M. Ketterings1

1 Nutrient Management Spear Program (NMSP) and 2Department of Electrical and Computer Engineering, Cornell University

Simplifying On-Farm Research with Single-Strip Trials

On-farm research is a powerful tool to advance crop management, providing practical, field-specific insights. However, traditional designs such as the randomized complete block design, often referred to as replicated strip trials, are often difficult to implement on-farm due to time, equipment, and labor demands. With more farms collecting yield data with monitor systems, there is now an opportunity to conduct on-farm research using the Single-Strip Spatial Evaluation Approach (SSEA), which compares yield from one field-length treatment strip (at least two harvester widths) to two control strips using a spatial model. This new approach takes into account current year yield and pre-existing field variability, and reports results per yield stability zones: Q1 (high and stable), Q2 (high and variable), Q3 (low and variable), and Q4 (low and stable).  As one farmer noted, “Multiple years with the SSEA has helped us tune in where we could be improving nutrient placement… The practicality and ease of use makes it a welcome trial on a busy farm.”

Until recently, SSEA analyses required support from NMSP staff. To help farmers and advisors conduct these evaluations independently, we developed a new web-based tool (https://ssea-nmsp-tool.shinyapps.io/SSEA-tool-CornellNMSP/) that automates the analysis and provides result visuals and downloadable reports (Figure 1).

A screenshot of the SSEA tool showing the control and treatment strips.
Figure 1. Overview of the SSEA tool displaying the uploaded inputs such as the strip location and yield stability zone maps.

New Webtool to Support Spatial Evaluation of Single-Strip Trials

A free web-based tool was developed to automate the spatial analysis and provide interpretations of trial results without the need for statistical expertise. The tool was developed with feedback from a statewide advisory committee, who helped refine the visual outputs and ensure the tool met user’s needs. The interface includes four tabs: Inputs & Analysis, Results, Report, and About. When users first open the tool, only the Inputs & Analysis and About tabs are visible; the Results and Report tabs appear once the necessary files are uploaded, and the analysis is complete.

The Inputs & Analysis tab requires four inputs: (1) treatment and control strip locations, (2) a yield stability zone map, (3) temporal average yield layers, and (4) current-year yield data. The interface is designed to be simple and intuitive, with a satellite basemap, zoom tools, and checkboxes that allow users to view each layer individually. For farmers who share yield monitor data with the NMSP as part of the New York On-Farm Research Partnership, all tool inputs other than the strip location(s) are already prepared and shared in a ready-to-use format. Detailed instructions for creating strip shapefiles are available in the user guidelines found in the About tab.

Once the inputs are uploaded, the tool generates two key visuals: (1) a donut plot showing the distribution of yield stability zones in the field and in the strips, and (2) a confidence chart that summarizes the likelihood of yield benefit or loss as a result of the management change implemented in the treatment strip. These outputs appear in the Results tab. The tool then auto-interprets these results and compiles them into editable text boxes in the Report tab, allowing users to refine the language before downloading a polished, two-page PDF report. By delivering fast and easy-to-interpret results, the tool enables farmers to evaluate more trials and helps reduce key barriers to the adoption of on-farm research.

Case Study Results

A farm in central New York partnered with NMSP to test an agricultural product in a 23.5-acre corn silage field using the single strip approach. The treatment strip was two chopper widths wide, placed away from field edges, and positioned to allow equal-width control strips on either side. Strip locations, yield stability zone maps (derived from three years of historical yield data), temporal average yield, and current-year yield data were uploaded into the SSEA webtool.

A donut plot.
Figure 2: Single-strip spatial evaluation approach (SSEA) analysis results show a donut plot for zone distribution in the field and in the treatment strips (center strip and the control areas on both sides).

The zone distribution donut plot (Figure 2) confirmed that the treatment and control strips captured the major yield stability zones present in the field. In this field, the consistently low-yielding zone (Q4) represented the largest large portion (43%). The placement of the treatment strip was such that all four zones were represented but with more datapoints for zone Q2 (35%) and Q1 (34%) than for Q2 (20%) and Q3 (11%).

Figure 3: Single-strip spatial evaluation approach (SSEA) analysis results show a confidence chart that lists the probability of yield response of a certain size from the treatment that was implemented in the strip.

The confidence chart produced by the SSEA webtool showed a high confidence (dark purple, 81-100% confident) of a yield benefit in lower yielding zones with increases of 0.5 to 0.75 tons/acre for Q3 and 0.75 to 1 tons/acre for Q4; however, for high yielding zones, Q1 and Q2, this yield increase was not seen. Thus, the product that was tested by the farmer helped improve yields in the lower yielding zones only. The economic value of applying the product can be assessed by combining the information from the confidence chart, the distribution of yield stability zones in the farm, and the costs involved with applying the product versus the value of a yield increase. If the yield benefits outweigh the costs for Q3 and Q4, any field with a substantial area of these two zones could be targeted for use of the product while applications to fields with mostly Q1 and Q2 where a yield benefit is not expected. If targeted use in portions of the field is an option, the product would be used for Q3 and Q4 zones within different fields as well. For results to stand the test of time, it is highly recommended to test products or management changes across years and across multiple fields. The SSEA approach can combine information for multiple fields and years.

Conclusions

The single-strip approach provides a practical way to evaluate management practices on-farm while accounting for within-field variability and minimizing disturbance of field operations. The SSEA webtool provides a platform to evaluate single-strip trials using yield monitor data and yield stability zone maps.

Full Citation

This article is summarized from: Subhashree, S.N., R. Goel, M. Marcaida III, J.C. Ramos-Tanchez, and Q.M. Ketterings (2025). Enhancing on-farm research with a web-based single-strip spatial evaluation tool: Design, features, and applications. Agronomy Journal, 56(3): e70264. DOI: 10.1002/agj2.70264

Acknowledgments

This research was supported (in part) by Cornell Atkinson’s Center for Sustainability, Northern New York Agricultural Development Program (NNYADP), New York Farm Viability Institute (NYFVI), New York State Department of Agriculture and Markets (NYSAGM), New York State Department of Environmental Conservation (NYSDEC), and by intramural research program of the U.S. Department of Agriculture, National Institute of Food and Agriculture, Hatch NYC‐127459. The findings and conclusions in this publication have not been formally disseminated by the U.S. Department of Agriculture and should not be construed to represent agency determination or policy. For questions about these results contact Quirine M. Ketterings at 607-255-3061 or qmk2@cornell.edu, and/or visit the Cornell Nutrient Management Spear Program website at: http://nmsp.cals.cornell.edu/.

Keeping the Score on P-Index with Grid Soil Sampling

Manuel Marcaida III1, Kirsten Workman1,2, Karl Czymmek2, and Quirine M. Ketterings1

1Cornell University Nutrient Management Spear Program (NMSP) and 2PRO-DAIRY

Introduction
The New York Phosphorus Index 2.0 (NY P-Index) and the Northeast Region Phosphorus Index (NR P-Index) help farmers assess the relative risk of phosphorus (P) loss from their fields and make informed decisions about manure and fertilizer P applications. Both P-indices combine soil test phosphorus (STP) in four categories (<40, 40-100, 101-160, and >160 lbs Morgan P/acre) with a field P-Index score derived from field features such as soil type, flow distance to streams, and flooding frequency, and management practices such as incorporation or injection of manure, to derive a management implication (Figure 1).

Figure 1. With the New York and Northeast Region P-Indices, management implications are derived based on soil test P (STP) category and P-Index score, assessed using transport factors (Transport Score) and beneficial management practices (BMP Score). The management implication determines whether fields can receive manure to meet N-need of the crop (N-based), up to annual P removal (P-based), or no P allowed (Zero P). The example shown here is for the New York Phosphorus Index 2.0.

Farmers and advisors have increasingly looked at grid sampling to better management soil fertility for improved crop production. To address questions on how to use P-Index in grid sampling context where STP information are more spatially granular than traditional whole-field samples, we analyzed soil data from 20 corn fields across six New York farms. We compared P-Index results and management recommendations based on whole-field composite samples and grid-based STP data at three grid sizes (0.5, 1.0, and 2.5 acres).

Key Findings

Homogeneous field conditions lessen the need for grid sampling
Most fields sampled were relatively homogeneous in terms of STP categories, with STP levels consistently classified as either A (<40 lbs Morgan P/acre) or D (>160 lbs Morgan P/acre) regardless of grid size (Figure 2). In such cases, whole-field composite sampling was sufficient for P-Index assessment and P management planning. Three fields showed noticeable variability across STP classes, with their whole-field averages in the STP range B (40-100 lbs Morgan P/acre). For these fields continued grid-based or zone sampling can help refine P management as well as address other fertility goals of the farm manager.

Figure 2. Distribution of soil test phosphorus (STP) levels within each field across different grid sampling resolutions. The bars show the fraction of each field that falls into four STP categories. Fields with more color variation indicate greater within-field differences in soil P.

Grid sampling is valuable for fields with variable STP levels
For the three fields whose STP levels were predominantly in class B (40-100 lbs Morgan P/acre), grid sampling revealed meaningful within-field differences (Figure 2) that would not be apparent from a whole-field composite sample. The finer scale information can help farmers and crop advisors designate P management zones, identifying whether portions of a field may warrant more conservative manure applications, or whether lower-P areas justify continuing N-based manure rates. The value of grid sampling is not in calculating grid-level P-Index scores, but in using the spatial pattern of STP to determine whether a single field-level P-Index assessment is adequate or whether zone-based management could enhance nutrient-use efficiency and advance environmental stewardship.

Coarser grid sizes provide comparable P management insights
Analysis on Fields E3, D1, and C1 showed that fields with STP levels in 40-100 lbs Morgan P/acre (class B) range gave similar management results whether sampled on 0.5-, 1.0-, or 2.5-acre grids at different P-loss risk scenarios (Figure 3). This finding suggests that coarser grids, such as 2.5 acres, can still capture the key spatial patterns needed to guide P management decisions. For farmers and crop advisors, this means sampling and testing can be streamlined without losing the detail needed to understand whether P-risk varies meaningfully across the field. While the P-Index is not intended for grid-level application, an initial grid sample can reveal where distinct P zones exist, allowing those zones to serve as the basis for future P-Index assessments.

Figure 3. Comparison of P-Index derived manure management implications using whole-field and grid-based soil sampling (0.5, 1.0, and 2.5 acres). Field maps show differences in soil test phosphorus (STP) levels, while the donut charts illustrate how the proportion of field area in each P-Index management category (N-based, P-based, or Zero P) changes across grid sizes and phosphorus loss risk scenarios.

Conclusions
Grid sampling for P management is most useful in fields with 40–100 lbs/acre STP levels, where nutrient variability within the field can change management recommendations. For fields with uniformly low or excessively high STP levels, whole-field composite sampling provides adequate information for nutrient management planning. When grid sampling is beneficial, a 2.5-acre resolution captures meaningful variability without the added cost of finer grid sampling. Using the New York P-Index 2.0 or Northeast Region P-Index together with grid-based data helps farmers make informed decisions that balance productivity, nutrient efficiency, and water quality protection.

Full Citation
This article is summarized from our peer-reviewed publication: Marcaida, M. III., K. Workman, K. J. Czymmek, and Q.M. Ketterings (2025). Grid-based soil sampling for Northeast Region phosphorus index assessment. Soil Science Society of America Journal, 89, e70156. https://doi.org/10.1002/saj2.70156

Acknowledgments
The authors would like to thank the staff of Champlain Valley Agronomics, Western New York Crop Management Association, and participating farmers. Funding came from the Northern New York Agricultural Development Program, the New York Corn and Soybean Growers Association via the New York Farm Viability Institute, the New York State Department of Environmental Conservation, the New York State Department of Agriculture, and the intramural research program of the U.S. Department of Agriculture, National Institute of Food and Agriculture, Hatch 2021-22-210. The findings and conclusions in this publication have not been formally disseminated by the U.S. Department of Agriculture and should not be construed to represent agency determination or policy. For questions about these results contact Quirine M. Ketterings at 607-255-3061 or qmk2@cornell.edu, and/or visit the Cornell Nutrient Management Spear Program website at: http://nmsp.cals.cornell.edu/.