What’s new and next in precision viticulture
by Courtney Llewellyn
Digital and precision viticulture is ever-evolving field, and while there’s not a lot of technology that’s ready to go into fields immediately, it’s good to know what will be available in the next five to 10 years. That’s what Hans Walter-Peterson, a viticulture Extension specialist with the Finger Lakes Grape Program, noted at the start of a recent virtual “tailgate” meeting.
Walter-Peterson introduced a few different projects being worked on. The first aims to determine bud mortality via thermal and multispectral imaging to guide pruning practices. The project was described by Cornell researcher Justine Vanden Heuvel.
“A lot of growers either don’t cut buds to see what their bud mortality is, or they just don’t want to cut buds,” Vanden Heuvel said. “Eventually, there will be good robotic pruners – but until then, how do we ID buds that are alive and buds that are dead?”
That’s where thermal imaging comes in. It can tell whether there’s water in a bud or not. Vanden Heuvel noted growers don’t necessarily need to know about every individual bud, but a detailed map of a vineyard can show the larger picture. Their research team is in the early stages of this thermal imaging project, hopeful it will grow in the coming years.
That team is also working on a low cost, vision-based yield prediction system. “We’re talking really low cost – using a smartphone to count clusters that are out there,” Vanden Heuvel said. Users would load their smartphone on a gimbal to stabilize it, put it on an ATV with lights and drive down a row at about 4-5 mph. Users would go out early in the season, before grape clusters were occluded by leaves, to have them counted.
“Our engineering team wrote software to ID and count clusters. It’s not super accurate, but it’s as accurate as the undergrad researchers we hired to count,” she said. “We can get really close by counting a few panels in a vineyard and multiplying out to cover the vineyard.”
The goal of this project is to lead to more efficiency while keeping things easy to use and inexpensive for growers. Vanden Heuvel said they’ve tried the technology on four different cultivars so far.
Also presenting during this “tailgate” meeting was Jim Meyers, Cornell Cooperative Extension regional grape specialist. Meyers spoke about a new satellite normalized difference vegetation index (NDVI)-based sampling protocol for grape maturation project in partnership with E&J Gallo Viticulture. The technology is already in use in California, because “a lot of technology is easier to use in large-scale models,” Meyers said.
“We wanted to go to one location in a vineyard in a 50-acre block and get a good sample – at least as good as samples from four different locations,” Meyers said. They can then look at the image taken from satellites and figure out the best location to take samples. “It takes significantly less time to figure out the best sample site, and you’ll get a better average sample,” he added.
Meyers also talked about a different kind of precision: the Cornell ENYCHP Vineyard Reports. These are sent out every morning, with the purpose of giving growers information very specific to their location. In New York State specifically, there is less of a climate difference across the Finger Lakes versus eastern New York (where vineyards are cultivated from New York City to the Canadian border). The reports include growing degree days, historical GDD, the weather forecast and historical precipitation information.
“How do these two things fit together?” Meyers asked. “We hope to start adding satellite imagery to the daily reports. What we did at Gallo was too big for what we’re doing in New York, but satellite imagery is getting higher resolution. Further down the line, we hope to leverage some research to use additional satellite imagery for disease diagnostics, scheduled sprays and more.”
Learn more about the Finger Lakes Grape Program, including any upcoming virtual meetings, at flgp.cce.cornell.edu.