Think Different

When evaluating research results, consider the parameters of the study.

While NutrientStar may have a way to go to become the fertilizer user's Consumer Reports, it is already making waves. An off shoot of the Environmental Defense Fund's (EDF) goal of improving nitrogen (N) utilization, NutrientStar has introduced a new research framework called the Technology Extrapolation Domain (TED). Initially put to use evaluating nitrogen management tools, it has the potential for widespread use with other products and services, as well.

"As a science-based organization, we took a look at tools to increase fertilizer use efficiency and felt there was a lack of data that was easily accessible and in the public domain," explains Karen Chapman, senior manager, Sustainable Agriculture, and administrator, NutrientStar. "NutrientStar wants to give farmers and CCAs data that helps them make good decisions leading to positive environmental outcomes on farms.”

Working with a blue-ribbon panel of agricultural scientists and a farmer advisory panel, EDF developed NutrientStar. The goal was an objective and transparent review process using hard data from field trials to evaluate nutrient management tools and put the information on performance into farmers' hands.

Greg Kneubuhler, G&K Concepts, Inc., an Indiana-based crop consultant, liked the idea of a better way to evaluate nitrogen management tools. He has run replicated N test plots for the past 15 years and uses those tools with his clients. He admits his experience using models for his growers has had mixed results. In 2016 and 2017, he became one of 20 consultants in Indiana, Illinois, Iowa, Michigan, Minnesota, Missouri, Ohio and Wisconsin working with NutrientStar to set up trials with growers.

While multiple modeling tools are available, only two companies agreed to make their systems available for comparison in the trials. They were Adapt-N and Climate FieldView. Reticence on the part of others didn't surprise Kneubuhler. "These tools are a work in progress," he says. "These companies took a big risk."

"The consultants laid out randomized, replicated, production scale, field trials run by farmers in their fields using their own equipment," says Chapman. "The trials compared tool-based recommendations with the grower's normal rate based on estimated yield goals."

To make the data meaningful beyond the trials, the TED framework was adopted from the Global Yield Gap Atlas project developed by the University of Nebraska. TEDs were defined by their crop growth influencers. These included long-term temperature, seasonality and aridity, as well as soil water holding capacity. The NutrientStar interactive website provides users with results by state and other contexts, including by TEDs so growers could compare field trial TEDs with their own TED.

Chapman notes that the value of the TEDs was that a grower in Ohio wanting to review data from a trial might find that one in Iowa was more similar to conditions in his fields than the Ohio trials were. She cautions that management factors, as well as seasonal weather variation, still need to be taken into account, but the long-term climate and soils would relate.

Results of the two-year study varied from TED to TED. Across all trials, Climate FieldView showed no statistically significant change in return over grower rates. On its face, Adapt-N results were more negative. Data from NutrientStar trials in 2016 and 2017, as well as data from earlier years, showed positive gains in N use efficiency, but a negative return to N compared to grower rates. With both tools, suggestions for improving results were included in the report. Results posted by state varied from positive to negative.

It was the broad context that created waves. Ohio State University researchers suggested that based on the results, neither tool had value for farmers in that state compared to an OSU developed regime. It was that kind of broad brush reaction that concerns Harold van Es, professor, Soil and Water Management, Cornell University. Van Es developed the model-based tool that evolved into Adapt-N. He feels the results, which were based on predicting N needs in May and meeting them with a single side dress application, did not reflect the dynamic, adaptive, N management nature of nitrogen management tools. He also questions other aspects of the study.

He fears the broad-brush interpretations such as made by Ohio State could turn off growers. "The TED is a useful framework, but if there are only one or two sets of trials in a region and they aren't executed correctly, it can look like a tool doesn't work well in that region," says van Es, noting that in many cases Adapt-N achieved both higher NUE and profits. "It can steer farmers away from a sustainability tool and away from trying to solve these problems."

Kneubuhler agrees that N modeling is a reactive tool and should be used that way. However, he notes that any comparison trial must have specific common parameters. He is confident the tools are here to stay. As he points out, N remains the most expensive fertilizer input and if there is a better way to manage it, people will use it.

"Models are tools, not the cure-all, end-all," says Kneubuhler. "Industry has made great strides, invested a lot of money and is starting to figure things out. At the end of the day, they will be of value, and people will use them."

Perhaps even more important over the long haul is the concept of an independent clearinghouse for farm-applied research and the TED framework for comparing it to a grower's own fields. "The fact that TEDs have been identified is the easy part," says Kneubuhler. "Their real value is to collect and aggregate data and extrapolate to other sites. By the same token, you need to do the research in a wide range of TEDs for the framework to have value."

Chapman has already received requests from growers and crop advisors for NutrientStar to carry out other product evaluations using the TED framework. " It is important for industry to adopt science-based field testing protocols to isolate the efficacy of any new products they wish to put on the market," says Chapman. "We are open to work with anyone willing to be transparent and share data. We walk this road as objectively as we can, presenting the data, but not sugar coating it."