If you fertilize your orchard at a flat rate, you could be missing out on fruit yields and leaving potential profits on the table. Because of the variability of orchard ground, a flat rate might be too high for some parts (resulting in wasted money spent on fertilizer) or not enough for others (resulting in less than optimum fruit yields).

The same could apply for many other orchard inputs. For example, not all trees in a block need the same thinning treatment, the same pruning strategy, or the same amount of water.

Dave Allan, with Yakima Valley Orchards in central Washington, figures variability in the orchard is probably costing growers a 15 percent shortfall in target yields.

Precision agriculture, a farming method commonly used in field crops, allows growers to measure and map variable factors that might be impacting their bottom line so they can manage the farm in zones.

It involves gathering as much information as possible, using vision, sensor, and monitoring technologies, in order to make site-specific decisions.

David Granatstein, sustainable agriculture specialist with Washington State University, said precision agriculture began in field crops, such as soybeans, because profit margins were slim and farmers wanted to improve their production efficiency. He thinks it could have application in tree fruits to help growers fine tune their management practices and improve uniformity of production.

Allan, who farms with his son Travis, sees several potential benefits of precision agriculture technologies. He would like to have aerial imagery—from an aircraft or drone—to help decide how much fertilizer is needed in various parts of the orchard. Then, they’d need a machine to apply variable rates of fertilizer. Currently, their tractor driver tries to adjust the fertilizer rate according to the vigor of the trees as he drives along.

“The bottom line is making everything in that orchard look exactly the same,” Allan said, noting that when orchards are planted on hillsides, trees in the dips grow strongly while those in shallow soil on the ridges often don’t have adequate growth.

“It’s about optimum tree vigor,” Allan said. “You want to make it so that everything in that orchard is optimum. If you’re farming some ground where the soil’s uniform, this doesn’t have application, but if the soil is not uniform—which is probably 80 percent of what we farm—then it has application.”

He’d like imagery to help him fine tune irrigation, with emitters putting out variable rates of water, according to trees’ needs. Similarly, he’d like to know where to target applications of the growth regulator Apogee (prohexadione calcium).

At bloom time, he wants to distinguish heavy-blooming parts of the orchard from those without much bloom and have a computerized sprayer that can apply appropriate amounts of thinning material.

Allan would also like to know how much light is hitting the orchard floor at a given time of day so he could identify areas of the canopy that aren’t getting enough light.

One of the problems in terms of mapping variation in an orchard is that it can be difficult to pick up GPS signals in the interior because of the trees and the trellises, he said, so there needs to be technology to address that.

Allan said the technology is not at a point where it is easy for growers to implement. “But I think if someone was willing to go to work on this, there’d be a pretty good market. The basic problem with precision agriculture is we do not have out-of-the-box equipment that will work inside the canopy.”

Craig O’Brien, a grower at Prosser, Washington, said that about a decade ago, university researchers and representatives from John Deere brought an autonomous vehicle to test at his orchard. One of the challenges was that it was difficult to pick up satellite signals inside the orchard canopy.

Though O’Brien finds the technology interesting, he thinks it will be difficult to integrate it into orchards because of all the unpredictable factors, such as sprinklers, insects, and the wind, that scientists don’t have to deal with in a lab.

But research on new technologies for tree fruits is under way.

Dr. Jingjin Zhang developed a technique for measuring how much sunlight comes through the tree canopy as part of her doctoral project at WSU’s Center for Precision Agriculture and Automated Systems. Her device uses Lidar remote sensing technology (see “Precision Ag Tools”) mounted on a bar that can be driven up and down rows on a four-wheeler fitted with GPS.

WSU horticulturist Dr. Stefano Musacchi is doing research with the machine in orchards with the goal of identifying the optimum pruning intensity for trees based on the amount of light reaching the orchard floor. He envisions that an orchard supply company might provide light mapping as a service to growers. However, it could be five years before the technology is ready to use.

Bruce Allen, president of Columbia Reach Pack in Yakima, is already doing site-specific farming the old-fashioned way. He has employees go through the orchard with paint to identify trees and weak spots that need attention.

Allen expects to be using drones with sensors, once the licensing and operational issues are resolved, but said the first thing that’s needed is a way to collect yield data for specific parts of the orchard, he said. That would help them identify low-yielding areas as well as allow them to measure the impacts of practices they use.

“If you have a weak spot, maybe you can change the fertilizer or the water and get the trees in more of a balanced state, but you don’t really have the yield data that’s very specific on a block,” he said. “With yield data, you can measure what the economic benefit is versus what the economic cost was. It’s the missing link.”

Gathering detailed yield data would be more complicated in tree fruits than in row crops because, in addition to tonnage, growers need to know the size and color of the fruit, which affect their returns.

Data overload

Karen Lewis, Washington State University extension specialist, said there are plenty of monitoring technologies available, but the risk is getting overwhelmed with too much data. It has to be processed and analyzed to make it useful. Ideally, the data would be accessible wirelessly via a smartphone or tablet.

Dr. Jim McFerson, manager of the Washington Tree Fruit Research Commission, also stressed that information must be actionable. Along with the hardware, there must be software that translates the data into something that horticulturists can use.

“Drones are the superfood of technology right now,” he said. “That’s fine, but I think it’s overselling the platform. The harder part is: What are we measuring and how are we developing decision models based on that information? It’s easy to imagine a drone out there buzzing through the orchard. It’s a lot less dramatic when you say, ‘I have 12,000 images of apples. What am I going to do with them?’ ” •