How to manage orchard zones

Combining information from different imaging systems can help growers achieve more uniform trees and fruit. By Anna Mouton.

Variability is the bane of the pome- and stone-fruit grower’s existence. The different soil types of most orchards lead to non-uniform tree vigour that is exacerbated by applying the same fertiliser throughout.

Soil scientist Louis Reynolds of Fruitful Crop Advice advocates breaking orchards into smaller management areas or zones. Adapting each zone’s nutrition and other interventions allows growers to even out tree performance across the orchard over time.

“The trees in a zone have the same leaf area, so it’s much easier to give a pruning or thinning recipe,” said Reynolds. “And, at the end of the day, fruit quality is better because the balance between crop load and vigour improves.”

The good news is that new technology to enable zoning has moved out of laboratories into commercial orchards — Reynolds illustrated the underlying principles with real-world examples.

Drones and nitrogen

The first zoning approach Reynolds presented was the drone model. “We do one drone flight in January at the same time as we sample leaves,” he said. “We use the drone image when applying the first postharvest nitrogen in February or March and the spring nitrogen fertilisation in October or November.”

Reynolds bases the nitrogen applications on the canopy area generated from the drone’s 0.05 x 0.05-metre-resolution image of the orchard. He explained that the canopy area is a reliable indicator of vigour.

“Canopy area gives us the total growth for the season because it’s done in January,” he said. “Most pome-fruit trees stop growing by the end of December, so the drone image gives us an end picture.”

The drone maps the orchard into five vigour zones. Reynolds and the production manager confirm the drone results by visiting the orchard for a visual vigour assessment.

Pairing drone-generated GPS data with a GPS-enabled variable spreader allows Reynolds to tweak nitrogen application according to tree vigour. The aim is to achieve a 4 100 m² canopy area per hectare, which equates to 2.77 m² per tree in a 4.5 x 1.5-metre planting of 1 481 trees per hectare.

Reynolds shared results from a 4.5 x 1.5-metre Packham’s Triumph block in Ceres to show the drone model in action. The drone data from 2022 revealed an average canopy area of 3.23 m² with a 20% variation within the block. This had reduced to 2.94 m² with a 14% variation by 2024 due to variable nitrogen application.

“We apply nitrogen according to growth because nitrogen directly correlates with growth,” said Reynolds. “But keep in mind that nitrogen is not always the factor limiting growth. It could be wetness or irrigation or something else — you need to go to weaker zones in the block to see what the problem is.”

Identifying problem areas is in itself a benefit of drone imaging. Read more about how De Rust deploys this technology in The Drone Detective. Precise coordinates for weaker areas can also facilitate targeted compost or foliar-feed applications.

The drone model has allowed Dutoit Agri to save 7% on nitrogen fertiliser in their Eastern Cape orchards. They also use drone-image data and a GPS-enabled sprayer to apply plant growth regulators.

Satellites and quad bikes

Reynolds’s second method of reducing variation involves combining mainly normalised difference vegetation index (NDVI) satellite images with fruit counts performed using a quad-bike-mounted camera.

NDVI images are based on satellite measurements of reflected visible and near-infrared light. Healthy plants absorb mostly visible but not near-infrared light so the amount of reflected visible and near-infrared light correlates to the amount of vegetation.

Reynolds uses 0.8 x 0.8-metre resolution NDVI images taken in January at the same time as the leaf samples.

“With NDVI images, we need to do ground-truthing,” he advised. “We must validate each zone — that means feet in the orchard. This is a disadvantage of NDVI because it’s time-consuming and must be done annually for each block and each zone.”

Fruit counts performed at 35-mm fruit size provide an expected yield map. Reynolds pointed out that yield data gathered by combine harvesters has supported the revolution in precision farming of annual crops. “An expected yield map for an orchard is a step in the right direction,” he said.

The January satellite images inform the spring nitrogen applications. “For the postharvest nitrogen fertiliser, we integrate the yield map with the NDVI image,” said Reynolds. “We try to overlay the maps of crop load and vigour to give one map for the postharvest fertilisation.”

Optimising tree nutrition

Apple trees require about 1.8 kg of potassium per tonne of fruit. Reynolds reserves 20%–30% of the orchard’s allocation for a January application tailored to the yields of different zones.

He showed an example of a 2.9-hectare Rosy Glow orchard with a yield variation of 51–102 tonnes per hectare between the lowest and highest-yielding zones. “We apply the potassium to the areas that produce more fruit than we budgeted,” he said. “This helps with fruit size.”

Reynolds also presented the Rosy Glow orchard’s integrated NDVI and yield maps, which inform postharvest nitrogen application. The nitrogen requirement for each yield zone is calculated based on production, and this figure is adjusted based on the vigour zone.

The result is that the zone producing 51 tonnes per hectare received 100% of the 59 kg per hectare postharvest nitrogen calculated from its yield because its vigour is weak. But the zone producing 102 tonnes per hectare is too vigorous. Based on yields, it should receive 87 kg per hectare of nitrogen postharvest, but based on vigour, it won’t get any.

Reynolds’ message is that technology can improve orchard uniformity by allowing variable-rate fertiliser, lime, compost, and spray applications. Expected yield maps also improve labour allocation for hand thinning, yield prediction and harvest planning, and decisions about storing and marketing fruit.

Although he conceded that precision farming requires capital, Reynolds believes the benefits outweigh the costs. “If your orchards are more uniform, your fruit will store better and longer,” he concluded. “And it will be easier to farm.”

This article is based on a presentation at the 2024 Hortgro Technical Symposium. Go to the Hortgro YouTube channel to watch Reynolds and other speakers at this event.