Training systems for trellised mangoes

Tree architecture is known to affect productivity and fruit quality, and can also influence tree management, improving the efficiency of activities such as spraying and harvesting. 

The Queensland Department of Agriculture and Fisheries (DAF) has been evaluating the benefits of three tree architectural training systems on trellis grown mangoes at Walkamin Research Station within an Australia / Philippines mango crop management project co-funded by the Australian Centre for International Agricultural Research (ACIAR). 

This ACIAR project is one of several complementary projects which collectively address the aims of DAF’s ground-breaking Small Tree High Productivity Initiative. 

The other projects are the current five-year AI13004 transforming subtropical / tropical tree crop productivity project co-funded by HIA, and an earlier project that commenced the initiative, funded only from DAF treasury allocation and royalty income. 

It is expected that over time, other projects will follow on from this work, as training system studies are a long-term endeavour.

The ACIAR project enables us to study trellising and training alternatives that were not possible to include in the concurrent AI13004 project. The aims of the ACIAR project are to evaluate espalier (or fence), fan and cordon training systems (picture 1) on high density, trellised mangoes—for their growth, flowering, and cropping efficiency. The training systems are being tested on four varieties including Keitt, Calypso, NMBP 1243 and NMBP 4069. 

Picture 1: Two-year-old Keitt tree trained with the fan training technique, with casual employee Chris Walsh at Walkamin Research Station.

Picture 1: Two-year-old Keitt tree trained with the fan training technique, with casual employee Chris Walsh at Walkamin Research Station.

Each of the training systems offer different scaffold configurations which may influence vegetative growth, flowering and cropping slightly differently. As mango is a terminal bearing tree, the number of terminal branches is important as it directly influences the numbers of flowers. Limiting trees to horizontal branching and the associated bending and training to reduce vigour is showing promise, however the technique can be time and labour consuming.

The training systems:

Figure 2: Three canopy training techniques (a) Fan, (b) Cordon and (c) Espalier (or fence) training techniques.

Figure 2: Three canopy training techniques (a) Fan, (b) Cordon and (c) Espalier (or fence) training techniques.

Fan – first order laterals are trained from low on the trunk to form a fan along the trellis. Branches are then tied to the trellis to fill the lower part of the trellis. As branches grow up they are spaced to radiate and fill the space on the wires, and secured to the trellis. 

Although this shape is more two-dimensional, one of the main benefits of this system is that it is similar to how the tree naturally grows, requiring little bending and tying of branches with the tree filling the trellis quickly (picture 1 and figure 2a). 

Cordon – first order laterals are encouraged to grow horizontal and then naturally bend up to vertical to fill the lower wire at the allocated space on the trellis, any main leader is also allowed to grow vertically. This method takes advantage of the upright growth habit of the mango (figure 2b). 

Espalier – a single leader is encouraged to go vertical while the first order laterals are trained horizontally. The leader is tipped at each point where it crosses the trellis wire, to encourage branching, and the laterals are tipped to encourage branching repeating this process until the tree fills the trellis. We found laterals need to be pruned to encourage sub-laterals (into the rows) to increase flowering positions (figure 2c). Pruning the laterals in spring tended to encourage heavy vegetative response, which competed with the single leader. Therefore, limiting pruning to the single leader in spring / winter to encourage branching is suggested. Pruning laterals is perhaps better left to summer when vegetative growth is more passive. 

An evaluation of the different training systems’ growth, flowering and yield, will start shortly. 

This research will assist the development of high density orchard systems that are consistently high yielding and suit a variety of growing conditions and cultivars. These training systems will need to be grown and evaluated over several years to fully understand the positives and negatives of each method. 


ACIAR funds international and domestic research to improve the productivity and profitability of agricultural systems in partner countries as well as providing benefits to Australian industries, such as the Australian mango industry. 

ACIAR have been a key funder of mango research in Australia for over 30 years including work on tree phenology, varieties and pests and disease management.

Article submitted by Dr Paula Ibell, and Dr Ian Bally, Queensland Department of Agriculture and Fisheries (DAF), Mareeba.