The acidic sap released when the mango ‘flower stalk’ is broken from the fruit will burn the fruit – and human skin – with some cultivars and some humans more susceptible than others.

The worldwide practice to avoid this problem is to harvest leaving some stalk still attached to the fruit. However, these stem ends can puncture the skin of other harvested fruit, and accidentally– broken stalks release sap, reducing quality of the fruit.

Fruit is then individually destalked then placed stem end down on racks to allow sap release. One Australian practice to avoid the careful stacking of stalked fruit into lugs (Fig. 1), with unloading and destalking onto racks in the packhouse, involved destalking fruit onto mesh benches under trees in the field. De-sapped fruit were then collected and taken to packhouses.

Figure 1. Mango fruit harvested with stems. Note care in packing fruit in the 2 kg lugs to avoid breakage of stems. 

In South America the de-stemming process has been partly mechanised (Fig. 2). Nonetheless the amount of care and double-handling makes this method of harvest relatively labour intensive, and thus expensive. With Australian labour rates, this harvest method costs around 40–50 c/kg mango.

With mango producers being low labour cost countries, the practice of harvesting fruit with stem continues. However, cost and labour availability drive alternatives in Australia – harvest labour is over AUD $25/hr in Australia but around AUD$25/day in Brazil. Only a few Australian growers, e.g., Carnarvon area growers and a few Mareeba area growers continue the practice of harvest with stem on.

Figure 2. Loading fruit to a mechanical destalker.

The Australian Mango Harvest Aid

The harvest of the Australian mango crop now involves destalking at harvest. Fruit picked directly by hand is twisted to break the stem and held inverted to allow release of sap. Fruit harvested using picking poles are manually de-stemmed soon after harvest. Pickers are given advice like ‘treat it like a hand grenade’, or ‘you have three seconds’ to get the fruit into an alkali solution (‘mango wash’), before sap permanently damages the mango skin.

The traditional mango plantings involved large trees on big row spacings. To avoid use of long poles and tree climbing, one innovation was the use of cherry pickers. Fruit was picked into crates and the picker had to move the platform to the ground to unload. The Le Feuvre brothers added long tubes from the picking platform down to on the ground bins. Mango wash was sprayed into the tubes, but fruit quality still suffered.

Today Hydralada offers a wheeled or tracked specialty cherry picker with two 20L mango wash tanks and an integrated forklift to carry a field bin. The operator picks into a sprayed bin mounted on the picking platform, with the platform lowered to empty to the field bin.

Another approach was the development of the Australian ‘harvest aid’. Ben Martin of Martos Mangoes in Bowen reports the first ‘harvest aid’ in the Bowen region was developed by Robert Vennard in 1995. A Holden ute chassis was used, with a small corrugated-iron tank holding alkali water used to irrigate an angled tarp that ended on 20kg field lugs. Fruit rolled from the tarp into the lugs. Martos built their first self-propelled harvest aid in 2002.

Haig Arthur, former owner of Acacia Hills Farm and a pioneer of the NT mango industry from the early 1990s, purchased one of the early picking machines from the Le Feuvre brother’s, modified it and EZY Harvester was borne in 1998. Ken Le Mesurier was an early picking aid builder in the NT with some units continuing in use, for example, by Skliros Produce.

Haughtons Engineering of Giru began producing one sided harvest aids with a picking platform and conveyor in 1999. Other regional engineering works also entered the field, including Davey Group in Toowoomba, Mackenzie Hydraulics & Engineering in Atherton, Total Ag&Fabrication in Acacia Hills, NT, and Greentech in Adelaide.

Early picking machines allowed picking from one row side only and usually had platforms for operators to work into the upper canopy of tall trees (Fig. 3). The systems often used an angled vertical angled tarp and a horizontal tarp sprayed with mango wash solution, with fruit thrown into the vertical tarp rolling across the tarps.

Figure 3. Harvest aid units with platforms for picking into the upper
canopy – for used on farms with tall canopy and narrow inter-rows.

Early self-propelled machines usually had one tarp on a horizontal platform and were used with lower height canopies. In the early units, fruit rolled into 20kg crates which needed to be shifted manually into 250–300kg wooden bins carried on the back or front of the machine. Later, this step was automated with conveyors and roller beds transporting the fruit directly to the bins. Plastic 400kg capacity bins also replaced the wooden bins, and hydraulics were added for bin lifting.

To generalise, most machines now allow for double sided picking. Now harvest aids come in ‘all colours’, driven by bespoke requirements of individual farms (Fig. 4). Perhaps the most fundamental design difference is in the choice of spray-to-waste or recirculation of mango wash. The required tank capacity is reduced in a recirculating system, but the potential for skin damage is increased, as sap builds up in the solution.

Figure 4. Mango harvest aids come in ‘all colors’, built to the bespoke needs of each farm.
Image provided by Total Ag&Fabrication.  

Design variations across the current fleet of Australian mango harvest aids include:

- Spray to waste or recirculation

- Length of unit – shorter for easier turning at end of row, longer for larger picking crews

- Tractor pulled or self-propelled, and if self-propelled 2- or 4-wheel drive or tracked

- Hydraulic or manual brakes

- Drive speed (slow for harvesting, faster for movement between orchards)

- Use of a final fresh water rinse to remove mango wash

- Use of hydraulic rams to level the picking platform when on sloping ground

- Use of picking platforms (with taller canopies)

- Whether fruit are thrown or placed onto the aid, which impacts the choice of the receiving surface, e.g, loose or tight tarp, stainless steel sheet or a conveyor

- Use of vertical over horizontal tarps where fruit are thrown into the aid

- System for carriage of empty field bins

- System for tilting of the field bin as it fills

- Automated lifting of conveyor belts to access hard to reach areas for cleaning

- Alarm systems for engines and machine functions, including safety switches, horns and cameras

- Flood lighting, allowing for night harvesting

- Seats and shade structures for operator comfort

- Storage areas for staff’s personal belongings, drinking water and picking sticks

- Location tracking systems to see position on farm.

Large farms now maintain fleets of a dozen or more harvest aids.

In a casual consideration, a harvest aid seems an inefficient investment – at over $100,000 for a self-propelled unit for an item used across a harvest season of approximately 8-weeks of the year. However, there is a strong driver in the reduction in harvest labour cost from 40–45c/kg to 10–25c/kg, depending on tree architectures, fruit load and harvest aid features.

With harvest of 12 tonnes of fruit per aid per day, payback within two seasons is expected. The instant tax write-off on assets of recent years is a bonus.

*Kerry Walsh is Professor – Plant Sciences, CQUniversity Rockhampton. Martina Matzner is an independent technology adoption advisor, who submitted a thesis on mango production as part of her university studies in Germany before coming to Australia, where she became involved in the development, and management of a large mango enterprise near Darwin. Somu Bhattacharya is a PhD student involved in a CQU mango research project on robot harvesting.