With already thin profit margins and increasingly uncertain farm labor and other input costs, precision agriculture technology could improve New England's small and medium-sized farms' efficiency, productivity, and resilience. Unfortunately, factors such as up-front costs and validation of the technology's accuracy in the region remain a barrier to adoption. A research team at UNH led by Benjamin Fraser, visiting assistant professor and director of the Basic and Applied Spatial Analysis Lab, has shown that unmanned aerial vehicles (UAVs), commonly used in precision agriculture, are able to provide effective surveillance of fields planted with corn, including brown-midrib (BMR) corn, an important variety for silage production.
BMR corn provides key silage advantages to dairy farmers, but it is more expensive to grow than many other varieties and is susceptible to disease late in the growing season. Monitoring BMR corn is therefore critical for the New Hampshire dairy industry, but it is also time- and labor-intensive, and field-level inspections often miss early signs of disease. A recent paper presents findings from eight weeks of UAV surveillance of New Hampshire corn fields that assessed its ability to analyze corn characteristics at field- and plot-scale levels. The paper shows that the UAV imagery can differentiate between varieties of corn and estimate crop yields with high accuracy.
"The findings demonstrate that low-cost, consumer available (or 'off-the-shelf') UAV sensors with limited spectral range are highly likely to produce accurate results and that the imagery can be used in several ways to inform future corn farming practices," says Fraser.
Precision monitoring of corn
The applications for precision agriculture tools such as UAVs are varied, from monitoring for weeds and diseases to calculating yields to optimizing harvest timing and site selection, and they are used extensively on large farms in Midwest and Western states. Yet, at this time, usage of precision agriculture methods remains low, about 25%, on small Northeastern farms, largely because of the up-front investment required.
The paper adds to a growing body of research indicating that precision agriculture does provide important advantages in the long term. Overall, it promises to lower costs, particularly for labor, and deliver better outcomes for farmers, bolstering the sustainability of commercial agriculture on small farms in New Hampshire and throughout New England.
The paper, published in Agricultural Research , provides a case study for the use of precision monitoring of corn to collect field- and plot-specific data. The experiment was conducted on UNH agricultural fields planted with brown-midrib (BMR) and non-brown-midrib (non-BMR) varieties. BMR corn has been in use and studied for a century, is easily digested by dairy cows, and can improve milk production. However, BMR corn is susceptible to disease risks and grows and develops quickly, requiring frequent monitoring.
The UAV imagery data was multispectral, meaning that it was acquired across multiple color bands. Using red edge and near infrared wavelengths and a machine learning classification of corn varieties, the researchers were able to distinguish the subtle differences between BMR and non-BMR corn by field with accuracies of up to 98.7%. Narrow-band red edge image data showed high potential for estimating corn yields.
"The team explored ways that UAV imagery could inform field-specific management practices to reduce crop damage and costs," says Fraser. "It brought many areas of expertise, including Tom Beaudry, a certified crop advisor for dairy producers in New Hampshire, Vermont, and Massachusetts, Carl Majewski, a UNH extension specialist, and Peter Davis and Aaron Palmer, UNH farm managers."
The team's research mitigates risks for farmers looking to work with new remote crop monitoring technologies by demonstrating the accuracy and utility of UAV observations. UAVs provide farmers with an affordable, flexible tool for proactively monitoring plant pests and diseases and assessing leaf area and yield. Using the data for consistent, reliable modeling of crop health and yield also provides vital insight for food management and for improving production methods.
"Our team is planning to work with additional private farms in the upcoming field seasons," concludes Fraser. "We'll look to quantify direct causes and amounts of loss within corn fields using the lessons learned from this research."
