The mango is by nature a large tree, and for decades South African orchards were planted to suit it, with wide spacing that gave each tree room to fill its ground. A growing number of producers are now doing the opposite, planting trees closer together and keeping them small in pursuit of earlier crops and higher yields per hectare.
Few have tested the approach as long as Mohlatsi Farm near Hoedspruit, where some orchards were planted at high density in the early 1990s. Jaco Fivaz, farm manager at Mohlatsi, has three decades of data from these blocks and is convinced of the system’s value. “I strongly believe high-density plantings are the future for survival as a mango producer, but it is a mindset change,” he says.
Fivaz began his career as a researcher focused on mango pruning and high-density planting. His mentor, the late Dr Piet Stassen, introduced him to the high-density systems used in the apple industry. When he joined Mohlatsi, he found orchards already planted at 952 and 2 000 trees per hectare by earlier pioneers, possibly with the intention of later removing alternate rows and trees. Mohlatsi chose instead to keep the density and the challenge of keeping those orchards productive.
All new plantings on the farm now range from 1 667 to 2 000 trees per hectare. Most are spaced at 6 m x 1 m, against 6 m x 2 m or wider in conventional orchards, and the farm is experimenting with 5 m x 1 m and an ultra-high-density 3 m x 1 m.
Matching spacing to cultivar
Cultivar choice sets the limits of the system, and 16 years of data from Mohlatsi show clear differences. Tommy Atkins performs better at wider spacings of 6 m x 2 m or even 6 m x 4 m. Shelly and Sensation do much better at 6 m x 1 m, while Keitt gives its best yields at 5 m x 1 m. Fivaz points out that these results apply to Mohlatsi’s own orchards.
Across 30 years of records, Keitt has proved the best performer at high density. “It is still too early to make a call on the Agricultural Research Council’s new selections in high-density plantings,” Fivaz says. He expects Sun World’s selection Samango-3 (J82) to benefit from spacings of 6 m x 1 m or 5 m x 1 m.
Soil plays its part. Trees on sandy soils grow more slowly than on heavy red soils, a difference most pronounced in Tommy Atkins, and Fivaz prefers sandy soils for all cultivars, for slower growth and better fruit quality. Less is known about rootstocks, as there is no alternative to Sabre to compare, and results with plant growth regulators have varied. “Much more research on growth regulators is needed,” he says.
More trees, more fruit
The yield case rests on tree numbers rather than bigger trees. Keitt orchards at Mohlatsi planted at between 952 and 2 000 trees per hectare have produced 34 to 46 t/ha a year over the past 13 years, against 23 to 34 t/ha from conventional Keitt orchards over the same period. High-density blocks also reach higher yields sooner. Individual trees yield the same in both systems: “The higher number of trees just produces more fruit per hectare,” Fivaz explains.
Density comes with a strict condition: the trees must be kept small. Without pruning, closely planted trees crowd one another and shade out the lower canopy, where light drives flowering. Many producers have borrowed training systems from the deciduous fruit industry, such as central leader and closed vase forms. Fivaz regards central leader as the ideal, but warns against imposing it too soon, as pruning to a central leader within the first six or seven years reduces yield.
“We still have a lot to learn about pruning,” he admits. Timing matters as much as technique. Mango flowers on the tips of mature shoots, so the aim is to produce strong, healthy shoots as soon as possible after harvest, giving branches time to harden off before winter flower induction. Pruning just before induction should be avoided.
Where an orchard has flowered poorly, severe terminal bud pruning in June can encourage lateral flower buds to develop. “This is not the preferred practice, but it might be the only way to get flowering if the initial induction was not good enough,” Fivaz says. He advises producers to try it on a trial basis first.
On one point he is firm: pruning must stay a hand job. Mechanical pruning with orchard pruning vehicles should never be considered.
The cost of going dense
The main extra cost at establishment is the trees themselves. In the early years, fertilisation, pest and disease control and drip irrigation cost no more than in a conventional orchard, although differences appear once micro-irrigation is installed. Once trees are in full production, Fivaz sees no difference in normal running costs between densities. Mohlatsi works on a payback period of five to six years for high-density orchards, against seven to eight years for conventional ones.
Mechanisation is another matter. High-density orchards need orchard tractors and smaller, narrower spray carts, both more expensive than conventional equipment. The return comes at harvest. A tractor travels a shorter distance to fill bins, delivering more fruit to the packhouse each day and lowering harvesting and transport costs.
Labour demands are higher in the early years, with more individual trees to prune. Later, general orchard work needs much the same workforce as a conventional orchard, but pruning adds at least 10% to 15% to labour costs. Density has no effect on fruit quality, although smaller trees sometimes produce fruit with more blush.
Advice for a first block
Converting an existing orchard to high density is difficult, so Fivaz’s advice is aimed at new plantings. Producers should budget for higher establishment costs, expect pruning to demand more attention in the early years, choose tractors and spray carts carefully and keep pruning by hand. High density is not for producers who want to keep labour inputs to a minimum.
“Be well aware that pruning is the most important practice,” he says. For producers prepared to commit to that discipline, Mohlatsi’s data make a strong case: earlier returns, higher yields per hectare and an orchard that is cheaper to harvest.