Iowa Initiative: making individualised decisions about fertilizer use

The Iowa Nitrogen Initiative, led by Castellano and Sotirios Archontoulis, is trying to figure out the best amount of nitrogen for each corn field by collecting data from farms all over Iowa. This should give growers the ability to strongly enhance the precision in fertilizer application rates.

Iowa leads the United States in corn production. Corn for grain production in Iowa for 2022 was estimated at 2.48 billion bushels, according to the USDA. “It’s an agricultural powerhouse”, Castellano points out.

For farmers to join the project, they need access to two increasingly common precision ag technologies: variable rate fertilizer application and GPS-based yield monitoring. Using historical yield data to identify spots expected to behave differently, a trial in a small area of a field is designed, typically about five acres.

Farmers who have the equipment for applying fertilizer at a variable rate often have insufficient evidence-based direction on how those rates should vary, says Castellano, William T. Frankenberger Professor of Soil Science and an Iowa State University professor of agronomy.

_{Text continues below picture}

Problems with satellite images and soil sampling

Currently, official recommendations are broad and ideal rates can vary widely from field to field and year to year. The research team aims to build models that offer far more granular guidance.

“On satellite images, the optimum greenness depends on the genetics, the hybrid variety of the crop. That causes all sorts of problems, because there are a million and one hybrids that get planted, and it’s impossible to calibrate tests for individual hybrids”, Castellano says.

“With soil sampling, there is a couple of problems. The nitrogen supply from the soil in our case is huge. It drips as much as the fertilizer. Predicting how much you are going to get in any given year is a challenge. The other more practical challenge is that, increasingly, fertilizer applications are banded in very narrow strips. And that makes sampling – to get a homogenous number – very challenging. And so, farmers seem to be not interested in that.”

Closing the gap

Castellano and his research team are now trying to close the gap between the precision ag advances made by engineers and the scientific understanding of agronomists. “The engineering is way ahead of the agronomists”, he points out.

The Iowa Nitrogen Initiative is currently running 270 on-farm trials this year across 72 different private farming operations. “Each trial has 5 different fertilizer rates”, Castellano explains. “On 5 replicated plots. These are scientifically robust experiments on farmer’s fields, what really differentiates this from previous on-farm research.”

_{Text continues below picture}

Moving beyond the average

According to Castellano, the trials have confirmed what has long been known. “And that is that optimum fertilizer rates can vary from 0 to over 300 kgs per hectare”, he says. “At the same time, however, the yields at the optimum fertilizer rate are much narrower. They really range from plus 1 ton per hectare to minus 1 ton per hectare. It’s a relative narrow range of yield that you get at the optimum rate, but the nitrogen that you need to get to the optimum rate is hugely variable. You never know what you are going to get from your soil.”

The nitrogen that you need to get to the optimum rate is hugely variable

The initiative should lead to the use of less nitrogen, Castellano points out. “We try to help farmers understand when they don’t need to apply as much fertilizer. And that’s about 30 percent of the time.”

Castellano says that farmers recognise that this is a real problem they face. There has been a 400% increase in trials from the project’s first year in 2022. “In the absence of information, they need to make good decisions, they now have to default to applying the average. Nobody wants to perform at the average, but right now they don’t have the information to move beyond that average.”

Tools to optimise nutrient management

The Iowa Nitrogen Initiative has the potential to provide farmers with information they need to make much more individualised decisions about fertilizer management in the future, Castellano says.

“We have the ultimate goal of 500 trials per year in the state of Iowa. We use biophysical process models, as tools to help farmers make better decisions. These APSIM models, developed by CSIRO in Australia, can be very accurate, but need a lot of data to help drive them. That’s the reason for so many trials. Farmers have thousands of management options, and we need to get a good representation of those.”

The initiative will lead to data and tools that farmers will utilise to optimise nutrient management, boost profitability and protect natural resources. The research team is developing three specific tools for growers.

Updated and more dynamic benchmark recommendations for nitrogen rates will account for differences in genetics, soil, management and weather. Farmers also will be able to see anonymised data from trials to see the real-world outcomes of various rates and practices.

_{Text continues below picture}

Potential outcomes

Forecasting will estimate ideal rates based on current and near-term predictions for soil and weather conditions. Weather has a disproportionate impact on nitrogen rates, Castellano says. “Weather is left out of the loop most of the time. This tool will follow the benchmarking tool, and will be super important.”

Hindcasting will help farmers look back at a prior growing year to explore how their crop’s nitrogen needs would have changed if they had done things differently, from planting a different hybrid to applying at a different time.

Google is an important partner of the Iowa Nitrogen Initiative

“The tools will be available to farmers through applications on their phones and computers”, Castellano explains. “They’ll be able to ask questions, and access the data in a dynamic way, that helps to understand potential outcomes that are related to the operations that they have in their fields.”

According to Castellano, Google is an important partner of the Iowa Nitrogen Initiative. A team of 10 Google employees, with design and engineering expertise, are volunteering their time to plan and facilitate the design sprint as part of a Google.org program that provides technical expertise to support nonprofits and civic entities.