Southern rust in corn can be found early with data from a drone

This figure shows the days each drone flight was conducted on the top half, and the bottom half shows examples of how the Texas A&M AgriLife team scored senescence. Credit: Texas A&M AgriLife photo by Aaron DeSalvio

Researchers at Texas A&M AgriLife found that they can use unmanned aerial systems, or drones, to predict southern rust epidemics in corn early enough to help growers avoid economic damage.

In Scientific Reports, a paper called “Phenomic Data-Facilitated Rust and Senescence Prediction in Maize Using Machine Learning Algorithms” explained how the work was done. Aaron DeSalvio is the main author. He is a graduate student at Texas A&M University in the Genetics and Genomics programme and works in the Department of Soil and Crop Sciences.

Seth Murray, Ph.D., a corn breeder at Texas A&M AgriLife Research and Eugene Butler Endowed Chair in the Department of Soil and Crop Sciences, and Tom Isakeit, Ph.D., a plant pathologist at Texas A&M AgriLife Extension Service in the Department of Plant Pathology and Microbiology, were in charge of the project. Alper Adak, Ph.D., a researcher with a post-doctoral degree, helped look at the data, and Scott Wilde, Ph.D., helped with the drone flights.

How to keep corn from getting southern rust

Isakeit said that Southern rust is the most important disease of the leaves of corn in the Upper Coast area of Texas. When susceptible hybrids are grown in years with a lot of rain, a fungicide treatment may be needed to keep yield loss to a minimum.

He said that southern rust is seen every year somewhere in the state, but that epidemics that hurt crop yields happen where it rains a lot. It will probably be seen first in the Lower Rio Grande Valley, where corn is planted first, but it will also affect the Coastal Bend, Upper Coast, Blacklands, and High Plains, which will be the last place to see it.

Isakeit said that severe outbreaks of this disease don’t happen every year. “Because the disease only happens sometimes, it is hard to get good information from variety trials about how susceptible hybrids are.”

Isakeit said that early detection lets people make smart decisions and use fungicides to stop damage. The fungicide should be used when the disease in the middle to upper canopy is not too bad.

Murray said that from a scientific point of view, the new ability to predict a southern rust outbreak in corn earlier is very exciting because it can help growers keep an eye on crops at key times when southern rust can do the most damage to their finances.

DeSalvio said that when they looked at the data, they found a link between the presence of southern rust at the time of grain filling and yield. This will have real-world implications for how precision agriculture is done.

“As a breeder, I never have time to write down anything about rust,” Murray said. “Now, I don’t have to because of UAS/drone tools, which are also more accurate. But what I find most exciting is that we can get a good idea of when grain fills up, which is a very good predictor of grain yield. Before, this was not really possible.”

The research

Orange, powdery spots on leaves are caused by southern rust. DeSalvio said that a severe infection kills leaves and causes plants to age too quickly, which stops full grain development.

“Spores of this disease will be blown from the tropical parts of North America into the southern U.S., and in wetter years, this can be a problem in Texas,” he said. “Last year was one of those kinds of years. We were able to predict the disease before we could see it with our own eyes because we found some visual signs that could be picked up by drones.”

Isakeit said that southern rust is first seen on the lower leaves. As the season goes on, the pathogen moves up the plant. Most of the crop’s yield comes from the middle and upper leaves, so rust on the lower leaves is not a big deal.

“The amount of rust and the stage of growth of the plant will tell you when to act,” he said. “How risk-averse the grower is is also a factor, as is what the weather is going to be like in the future. Aaron’s system can help get a better idea of how much disease there is.”

This study used UAS field-based high-throughput phenotyping to get high-resolution aerial images of elite corn hybrids planted in 2020 and 2021. Thirteen UAS flights were taken in 2020 and 17 in 2021.

The team was able to use UAS images to find signs of plant growth, such as changes in colour, that could be used to predict southern rust as measured in the field and senescence as measured using UAS mosaic images.

DeSalvio said that data collected by UAS allowed researchers to find early signs of southern rust and corn senescence before an expert could find them by scouting the field by hand. These early signs let researchers estimate how much southern rust and corn senescence there is.

“The size and detail of the images you can get from a drone also make it possible to cover more land and crops at a higher level of detail than you could if you just looked at the plants,” he said.

Rare chance for a student

DeSalvio said it is exciting that he has found and published research so early in his education.

“Before I started my doctoral programme, I was just a rotation student and a summer worker,” he said. “So, I’m very thankful to Dr. Murray and everyone else who helped me along the way for giving me this kind of head start.”

DeSalvio said that when he started working in the lab last year, the summer humidity was very high, which helped the rare pathogen grow. As part of projects with the U.S. Department of Agriculture-National Institute of Food and Agriculture and the Texas Corn Producers Board, Murray’s team was already using drones to collect data on their corn experiments. They flew as many as 25 times during the growth of the crop.

As the team used drones to measure the corn crop at different stages of growth, they started to notice something. With the help of old pictures and data, it was possible to see how plants looked earlier in the season and in previous years. Because the images and data were similar, they were able to write a paper for Scientific Reports.

“It’s an honour to have worked with such high-level researchers as Dr. Adak, who taught me how to do all the statistical analysis, Dr. Isakeit, who helped me analyse the effects of southern rust, and Dr. Wilde, who flew the drones for this particular data set,” DeSalvio said.

Today’s discoveries lead to solutions for tomorrow.

DeSalvio said that a lot of the background information he talks about in his paper came from researchers who looked at rust in wheat and other corn blights and pathogens with drones.

So, he said, “we definitely got ideas from other researchers, and we think our discovery will also help people who study other crops.”

“I think that, in the end, drones will help farmers, geneticists, and research programmes make better decisions about how to manage their land and fields and which lines are more resilient and better suited to the climates where they are grown.”

Further information: Aaron J. DeSalvio et al, Phenomic data-facilitated rust and senescence prediction in maize using machine learning algorithms, Scientific Reports (2022). DOI: 10.1038/s41598-022-11591-0

Journal information: Scientific Reports

Source: Texas A&M University