An app developed by the University of New England’s (UNE) Applied Agricultural Remote Sensing Centre (AARSC) can forecast yields in avocado orchards with a 93 per cent accuracy rate
 |
|
The new CropCount app is able to forecast avocado yields with 93 per cent accuracy
|
Using the world-first app, CropCount, avocado growers are able to more accurately predict yield levels with the technology boasting a 93 per cent accuracy rate.
Previously, research conducted in 2017 by UNE indicated avocado yield forecasting in a medium-sized orchard of 10,000 trees can be inaccurate by as much as 120 per cent.
The CropCount project was funded by HortInnovation and co-developed by creative agency Circul8.
The technology uses satellite imagery and targeted on-ground sampling to produce the accurate forecasts in a fraction of the time it would take growers to estimate yields using traditional methods.
An algorithm within CropCount then identifies which trees should be counted to provide growers with the more accurate snapshot of health and yield across the orchard.
AARSC director, Professor Andrew Robson, says the technology is far more time efficient for growers.
“It reduces the number of tree counts required from 70 to 1,000, to just 18. With CropCount, growers take nine hours to do their yield forecast, not the 500 hours they once did,” says Robson.
“With CropCount, growers record the fruit count for those 18 trees on the app, which sends the data to us. We do the processing and derive a yield map, fruit size map, tree disease map and a forecast for the block [or paddock], which goes back to the grower to inform their decision-making.
“The accuracy at the block level is better than 90 per cent. At the farm level, it’s generally better than that.”
Such a dramatic improvement in the accuracy is done through the consideration of yield variability which is commonplace in avocado production. Avocados suffer from irregular or alternate bearing, meaning trees which product 1,000 units one year may only produce a handful the next.
“It is very difficult to do a generic prediction for yield forecasting,” says Robson.
CropCount also accounts for other factors which may affect avocado yields such as the pruning of trees, diseases such as phytophthora, or weather events given each season is manually calibrated individually to provide growers with the most accurate estimate.
The app was built to work effectively when offline with the ability to sync when users were back online.
Currently, the CropCount app is its prototype phase with Robson and the team at UNE hoping to complete the full version of the app within the next six to 12 months. Its availability to the Australian market, including avocado growing regions in Western Australia and South Australia, remains a main priority.
The key markets of South America and South Africa, which account for over a quarter of the global avocado production, have also been identified as potential opportunities for the product.
Following the commercial rollout of the CropCount for avocados, it’s understood the technology could be translated to work across tree crops such as apples and stone fruit.