Branch configuration in a narrow 2D fruiting wallResearch & Extension
Dr Gordon Brown reviews research conducted in Idaho, USA, on the impact of branch training in a fruiting wall on yield and fruit quality of young Fuji apple trees over five years.
With increasing labour costs and uncertainty of labour availability there is a need to improve labour use efficiency in all aspects of fruit production. Machine vision is being developed for a number of orcharding activities including robotic fruit harvesting. To allow maximum fruit identification and machine access a fruiting wall training system is desirable for optimal robotic harvester operation. An ideal fruiting wall will be tall with a thin canopy facing the inter row space with fruit evenly spaced within that canopy. The use of tall spindle trees in high density orchards has become popular for this purpose however this is a relative new training system and options for configuring the fruiting branches for machine harvesting have not been fully explored.
Fallahi, Fallahi and Mandavi have recently reported* on trials of three different branch configurations and the effect this had on Fuji fruit yield and fruit quality over five years in Idaho, USA (years 2 to 7 after planting). Trees in these trials were grown on Bud 9 rootstocks in upright single rows with 0.9m between trees. A drip irrigation system was used applying 100 per cent evaporative transpiration adjusted for ground shaded area for each training system with applications twice a week. They were trained onto a vertical 4 m tall trellis system with seven wires placed 45 cm apart from 60 cm above the ground level. The trees were trained into one of three different branch configurations.
- Standard tall spindle with 14 to 18 small feathers equally spaced around the central leader in all directions. These feathers were tied down to the trunk so that they were close to 110° from vertical.
- Overlapped arm trees had seven pairs of bilateral feathers approximately 45 cm apart which were tied horizontally along the trellis wires throughout the growing season. These feathers were allowed to grow along the trellis wire in each direction until they reached the trunk of the neighbouring tree such that each trellis wire had two branches growing on it in opposite directions between trees. Any vertical shoots growing from the horizontal arm were cut to about 15 cm at any time during the growing season to create a spur like structure.
- Tipped arm trees were trained in an identical manner to overlapped arm trees except the tip of each arm was pruned to terminate halfway between the trees such that each trellis wire only had one branch growing on it at any point along its length.
Figure 1. Tipped arm training system in third year of cropping. Photo: Journal of Agricultural Science.
Tree vigour and production pattern
Each year the trunk cross-sectional area was measured as an indicator of tree vigour and over the five years the trunks almost doubled in cross-sectional area. Each year the tall spindle trees were found to be more vigorous than the overlapped arm trees while the tipped arm trees had a vigour between the other two training systems. By 2016 the trunk cross sectional areas were 20.2cm2 for the tall spindle, 14.7cm2 for the overlapped arm trees and 17.5cm2 for the tipped arm trees.
Every tall spindle tree had more branches than the other two training systems and had a greater foliar density requiring additional irrigation due to the higher ground shaded area used to adjust quantity of water applied (data not provided). Trees trained to the tipped arm system had larger leaves than the other two training systems. These larger leaves had more than double the fresh weight of the other systems although their dry weight was similar indicating a higher degree of hydration and potentially water use efficiency.
Spurs and fruit in trees trained along the trellis wires cropped along the wires for the first four years of production. By the fifth year this regular cropping pattern became more random along the spurs that were developed. This suggests that the vertical shoots should have been pruned closer to the main arms rather than the 15 cm that was practised to maintain the regular 2-dimensional cropping pattern of fruit close to the trellis wire.
Fruit number per tree and biennial bearing
The tipped arm training system had fewer fruit than the other systems in 2012 although each year there was a steady increase in the number of fruit for this system displaying no tendency for biennial bearing. Both the overlapped arm and tall spindle training systems displayed biennial bearing habit such that in 2013 they had less than half the number of fruit than the tipped arm training system. This tendency for overlapped arm and tall spindle trees to display biennial bearing continued for the five years of the trial.
There was no statistical differences in the cumulative total yield over the five years between the growing systems indicating a similar yield efficiency. In any one year, however, the yield per tree reflected the fruit numbers such that the tipped arm system displayed a steady increase in yield year-on-year and having a higher yields than the other two systems in 2013 and 2015 when they were having an ‘off’ season.
Figure 2. Yield of fruit per tree with season and branch configuration
Fruit size and quality
The average fruit size for the overlapped arm system was smaller than the other two systems and ranged from 186 to 212 grams. The average fruit size for the tipped arm and tall spindle training systems was similar ranging from 212 (Tall spindle) to 255 (Tipped arm) grams.
No difference in fruit colour, sugar content or starch level at harvest was observed between the growing systems. The overlapped arm system however produced fruit that were firmer and with less water core than the other two systems. This response could be due to the smaller fruit encountered with this training system.
When training trees to a narrow 2D fruiting wall the branch configuration has a major impact on tree performance. The tipped arm trees did not display any tendency for biennial bearing and importantly appeared to have a higher water use efficiency. This training system produced similar yields of large quality fruit.
Take home messages
- This study revealed that standard tall spindle training systems produced trees that are vigorous and prone to biennial bearing.
- Fruit from trees with an overlapping arm training system also displayed biennial bearing and were smaller than those from the other systems.
- The tipped arm training system produced trees with large, well hydrated leaves suggesting greater water use efficiency with little tendency for biennial bearing producing a more reliable crop.
- Based on this study training trees to tipped arm configuration is recommended for upright wall fruit production to increase yield reliability, harvest labour use efficiency and preparing for robotic harvester operation.
*This research was published in the Journal of Agricultural Science, Vol 12 no 4 (2020) and can be read here: Branch configuration impacts on production, fruit quality, and leaf minerals of ‘Aztec Fuji’ apple trees in an upright single row high-density orchard system over five years.