Skip to main content
GrowersHelp your business growResearch reports, publications, fact sheets and more Implementing precision agriculture solutions in Australian avocado production systems (AV18002)
Ongoing project

Implementing precision agriculture solutions in Australian avocado production systems (AV18002)

Key research provider: The University of New England

What’s it all about?

This investment is refinining and working towards commercialising technologies and innovations to help the avocado industry improve production and efficiency. There is a particular focus on delivering solutions to help growers predict yield, look at yield variability and map factors such as disease, to in turn support on-farm decision making.

The work builds on the Multi-scale monitoring tools for managing Australian tree crops initiative, supported by Hort Innovation under the Australian Government’s Rural R&D for Profit program.

Specific project activities include, but aren’t limited to…

  • Keeping the Australian Tree Crop Rapid Response map (available here) and its associated app (available here) updated with information on commercial avocado orchards. This mapping tool was a key output of the multi-scale monitoring tools program. At a top level it’s designed to assist with natural disaster recovery efforts and biosecurity work, but when combined with other innovations such as remote sensing and analytic technologies, it can be used to support on-farm decision making.

  • Developing a mobile app to provide avocado growers with up-to-date, high-resolution satellite imagery and other capabilities to support pre-harvest yield forecasts and mapping of tree health and vigour, yield parameters including fruit size, and disease with a focus on phytophthora.

Since the team’s last update, an independent review of the CropCount prototype app was delivered with recommendations for whether further development is warranted. While under assessment, validation of the methodology continued at orchards in the Bundaberg region. Derived yield (t/ha) and fruit number maps will be provided by ‘WebApp’, offering greater functionality for growers to interrogate the mapping products.

National mapping of commercial avocado orchards progressed, with all growing regions in Queensland, northern New South Wales and the Riverland in South Australia published to the Australian Tree Crop Map. Draft mapping for Sunraysia and Riverina regions is under review.

The Severe Weather app was updated to include historical events, such as thunderstorm cells and cyclones, as well as statistics presenting potential orchard impact area (hectares).

For ‘non-sampling’ yield forecasting, the team developed satellite based ‘time series’ models using imagery for large orchards in the Bundaberg, north Queensland, South Australian and Western Australian growing regions. Model accuracy in comparison to grower estimates is promising, with harvest yet to be completed.

Further evaluation of remote sensing-based crop climate models continued, with temperature and relative humidity recorded for 22 trees, representing high, media and low growth canopies. The data and associated tree yield will be analysed in the coming months.

The team reported that industry support, including provision of historical data, has been exceptional. Progress and results continue to be shared with growers and the sector via newsletters, meetings, websites, presentations and social media.


All the project’s industry applications are available via the National Tree Crop Mapping Project webpage on the University of New England website.

Read this article World-first yield forecasting technology offers avocado growers 93% accuracy, published on the GrowAG website

Since the team’s last update, the national mapping of commercial avocados orchards has progressed with new data published to the Australian Tree Crop Map and draft mapping available for review via the Industry Engagement Web App.

An Australian Tree Crop Severe Weather web app was also developed that overlays BOM information with commercial orchard locations, supporting the rapid identification of severe weather risks across the industry.

A prototype of the avocado yield forecasting app, CropCount, was successfully tested on orchard blocks in the Bundaberg and Mareeba region, allowing growers to assess fruit number and fruit size per tree in the orchard using high resolution satellite imagery and compare this to harvested yield.

Results confirmed the app’s potential benefit for the industry, with grower feedback from the prototype being used to further develop this technology as the team continues to investigate commercialisation pathways.

Work to further develop the remote sensing climate-based yield prediction model continued, with measurements taken from sensors on individual trees to be matched with final yield, to provide further insight into the influence of temperature and relative humidity on fruit number and size.


Check out the project’s Industry Web Apps and Maps online.

Watch this video about the CropCount yield forecasting app (4 mins)

This project team has commenced work to adapt technologies and innovations that will help the avocado industry improve production and efficiency. In this reporting period some field activities were disrupted due to Covid-19 travel restrictions, however progress continued in key areas.

National mapping of commercial avocado orchards was updated with the inclusion of the Bundaberg region in the Australian Tree Crop Rapid Response Map. The location and extent of avocado orchards for natural disaster response and recovery was successfully applied. Draft mapping for the Wet Tropics and Tablelands regions is currently underway.

During the 2019/ 20 growing season, sampling was conducted in the Bundaberg area and at Mareeba, with the measures used to derive yield forecasts and fruit number maps using satellite imagery. Accurate counts are essential for achieving strong calibrations to canopy reflectance and accurate predictions. Strong correlations between tree vigour and final yield (fruit number and size) were achieved for all orchards, following the final harvest of sample trees. 

To further develop the remote sensing climate-based yield prediction model, remote access, low cost temperature and relative humidity sensors were attached to individual trees representing high, medium and low growth canopies, with data logged from February 2020. The difference in temperature and relative humidity measured from the trees will be matched with final yield results, thus providing further insight into the influence of temperature and RH on fruit number and size.