Avocado Pollination Research

Avocado trees can have up to 1 million flowers but only 0.3% set fruit – this is three fruit in a thousand flowers! Further, flowering every second year can be substantially less, resulting in much smaller yields.

Plant and Food Research is working to increase pollination and fruit set success for better and more consistent yields from year to year. As pollination scientist David Pattemore says, “If we can increase fruit set by a small percentage, we can make a big difference for fruit production in New Zealand”.

Avocado trees, native to Central America, have an unusual system of flowering to prevent self-pollination, a system that has evolved for the insect diversity of a sub-tropical rainforest – not New Zealand.

Each tree can be almost covered by hundreds of thousands of tiny flowers. Look at the flowers one day and they may be female, but the next day the same flowers will be male. The timing of this change is different in different avocado cultivars. In some cultivars a flower opens in the morning as a female with a stigma, then closes at about midday. It reopens in the afternoon of the next day, but this time as a male with pollen. The flower closes in the evening and stays closed. In other cultivars, flowering is the other way round.

David Pattemore’s research has found that the colder the weather, the slower the female Hass flowers are to open; switching the entire system around. On many nights during changeable NZ springs, female flowers only open in the late afternoon or evening, and then can often stay open all night long.

Avocado trees are insect pollinated, so growers bring honey-bee hives into the orchard in the flowering season.

Most avocado orchards also have two different cultivars to aid insect pollinators, so when some trees have flowers at the female stage, other trees have male flowers.

Most commercial fruit is the Hass variety, predominantly because this cultivar changes colour to show it is ripe and this easy ‘ripe test’ is preferred by consumers. The secondary cultivar, usually only to aid pollination, is called the ‘type B’ and is inter-planted with the main Hass crop.

The poor fruit set and biannual crop yields are normal within the avocado and related to its evolutionary biology.

A heavy fruit set in one year inhibits the flowering in the following year. The fruit can also stay on the tree for up to a year, so fruit will still be present when the tree begins to flower again in the following spring, further inhibiting yields in the second year.

David’s team have manually pollinated avocado flowers and increased the fruit set to 5% so they know it is possible. It also tells them that there are other factors at play such as temperature/climate. David says that 5% would likely be too much for the tree, but they’re aiming for a natural system that will double or triple the set

PFR presently have 2 key projects that are looking to improve fruit set rates for avocado:

1. With the Avocado Industry Assn & MBIE, PFR is looking at what drives the 2-year cycle of flowering and fruiting, and whether understanding the floral biology and pollination requirements of avocado can improve fruit set in years that would otherwise produce few fruit.
2. In a broader study of new pollination systems for crops in NZ, PFR are working to deliver alternate pollination systems for orchardists in NZ, including bumble bees, native bees and flies.

PFR has found that in New Zealand orchards, only about three flowers in every 1000 produce fruit. Although honey bees are brought in for pollination, the bees are not necessarily the best avocado pollinators. Firstly the bees don’t find the avocado flowers as an attractive a food source as other plants which is not helped by the fact that avocado orchards will be in areas where preferred flowers are common (clover/citrus)

Individual bees tend to prefer visiting flowers that are same, such as either male or female flowers. That means they aren’t very good at transferring pollen from male flowers on one tree to female flowers on another.

Further the honey bee work day schedule does not always work well with the avocado flowering schedule. Bees tend to start work at 9/10am, peaking at mid-day and then tailing off in the afternoon. In cold, changeable New Zealand springs, on many days female Hass avocado flowers actually open in the afternoon or evening when the bee activity is low. Finally avocado pollen is very sticky and viscous and the honey bee struggles to collect it.

At present this situation can’t be helped by artificial pollination, which is used for kiwifruit, because avocado flowers produce very small amounts of pollen and the stickiness of it makes it difficult to work with – all which at present makes for a cost prohibitive proposition.

The research has been tracking the relationship between temperature and flowering. New Zealand can have cold, wet and very changeable spring weather, which differs a lot from the climate where avocados evolved.

The work involves tracking thousands of avocado flowers over the course of the flowering season and collating temperature/weather data.  David’s team would paint flowers on different days after different temperature conditions, and then go back to record fruit development. After about 2.5 weeks the team can tell if a fruit is developing. From here they tie on a tag and then come back weekly to monitor the fruit growth. In the first season, the team tracked 8650 flowers of which, 24 produced fruit.

Surprising results so far have NZ data showing that in cooler weather, female flowers can open late in the day and stay open through the night – this suggests the possibility of nocturnal insect pollination.

PFR is involved in research to look at alternative pollinators to bees – this research aims to benefit horticulture across NZ, including avocado growers.

Bumblebees are looking promising as they are not susceptible to the varroa mite. They’re tougher flying in wind and rain and their big, furry bodies help them do the work of up to 50 honeybees. However bumblebees have their own unique issues as presently there is no system for introducing bumblebee nests to orchards as there is with honey-bee hives. Honey-bees live in colonies of tens of thousands of individuals while bumble bees live in much smaller colonies of a 1000 or less individuals. Further, bumblebees do not live in movable hives but much smaller nests in the ground.

This related sideline of David Pattemore’s pollination work has seen his team looking at how bumblebees select a nest site and what makes for attractive nest real estate for a bumblebee.

Understanding bumblebee real estate choices meant finding the nests – enter Ollie the bumblebee dog. Ollie, an exceptionally handsome Hungarian Vizsla, was trained by David to find these nests.

Being able to understand the nests has enabled David’s team to begin trialing artificial nests in orchards to see how they can entice these insects to take up residence where they want them.

Tracking the relationship between the flower and pollinator for this work is labour intensive. To compare bees to flies to beetles, the team start by bagging un-pollinated flowers. Each flower is then unbagged by a technician who will offer it to a nearby insect. After the insect has visited the flower, they then remove the stigma from the flower and put it on a microscope slide. These are analysed back at the lab to record pollen grains deposited by different insects. At this stage, this system of observation will not work for nocturnal visitors.

Flies are looking like excellent pollinators of the avocado. In Australia research has shown some orchards to be almost entirely pollinated by flies. However flies do not have a structured colony like bees, so PFR is carrying out research at Lincoln into rearing flies en masse that might be supplied to orchardists during spring.