Satellite technology shows promise to tea tree oil industry

“It’s the first time I’ve done anything like this, and I have to say I’m pretty impressed with the results,” said the Coraki-based grower.

Ian’s mapping data was collated – along with yield records from the past five years – with the assistance of Extension Officer Alice Moore from Farmacist who manages the project, in conjunction with ATTIA and Dr Moshiur Rahman from the University of New England’s Applied Agricultural Remote Sensing Centre (AARSC). 

“Once Alice did what was needed to be done, she provided five years of our yield records to the University of New England where they assessed the maps using the remote sensing ‘Time Series’ platform to forecast crop yields,” said Mr Brown.

“I was expecting the forecast to come in within 20% of the actual percentage and couldn’t believe it when it came back within 1%. I have been growing and producing tea tree oil for 20 years and this sort of information is just excellent. I was very surprised and I’m glad I got on board.”

Mr Brown said the initial results have encouraged him to continue investigating the value of satellite technology. “Aside from yield forecasts, the images actually picked up parts of the crop that were performing well and those parts that weren’t,” added Mr Brown.

“So now, we’re ground truthing those areas to look at the soil conditions in parts that are underperforming and investigating what we can do to improve the soil and increase yield in seasons to come.”

Ms Moore said while the project is in its infancy, the positive results have certainly piqued grower interest.

“Ian was one of just two farmers involved in the first stage of the project so we need to be mindful that these results might not apply to every farm at this stage, however the first farm we did came back at 99% accuracy for plant biomass and the second farm was 96% accurate for oil yield and 95% for plant biomass,” said Ms Moore.

“We went in with a no promises approach because it was absolutely new for the [tea tree oil] industry. We were quietly confident, but it’s fair to say that the project team, as well as both growers were pleasantly surprised to be able to get such highly accurate yield predictions back.”

While satellite imagery is freely available, this is the first time it’s been specifically used and validated for the tea tree oil industry.

Forecasting models built at the University of New England have been applied to 14 different growing regions around Australia and learnings have been applied to multiple crops including avocadoes, citrus, mangoes, macadamias and sugar cane.

“I worked in the sugar cane industry for six years and that’s quite similar to the tea tree oil sector because you’re growing for biomass, not fruit or seed,” said Ms Moore.

“The sugar industry successfully used remote sensing for yield estimations and because of the similarities we were confident that it would work for tea tree oil growers as well,” added Ms Moore.

Ms Moore added while accessing free platforms is easy, the AgriFutures Australia project focused on an analysis of images using the Time Series approach – that requires additional assistance to get the yield predictions.

“There’s a range of benefits and it doesn’t matter whether you’re new to the industry or have been working in it for a long time; the benefits of knowing what your yield is going to be in the growing season rather than at harvest is significant,” said Ms Moore.

“This technology has the potential to help with labour planning by providing the yield estimation to the distillery. For example, it can indicate how many bins are required come harvest time. Yield predictions can guide marketing decisions around demand for the product.”

“Also, at the moment there’s no within field monitoring for tea tree oil so from an agronomic point of view we don’t know how much yield we’re losing due to soil constraints or other issues.”

“This makes the satellite imagery another data layer we can use to inform growers with as much information available to make decisions on farm that will ultimately increase their yield.”

The early validation of the benefits of satellite technology for tea tree oil growers is very encouraging said Professor Andrew Robson from the University of New England.

“Ten years ago, we were using single capture images to look at variations for tea tree plantations but there are limitations on single capture images as it doesn’t give you a lot of insight into historically how its performed,” said Andrew.

“Time Series allows us to go back in time and use Landsat or Sentinel Imagery which is freely available to look at how the crop has grown in five, 10, 15 years and match it to the yield or productivity achieved. The key benefit is you establish a benchmark and can better predict how the crop is going to perform.”

Professor Robson, who is the Director of the Applied Agricultural Remote Sensing Centre at the University of New England, said the use of the technology will be gauged by the needs of the grower.

“If they [growers] want better forecasts of production over time then public access remote images at 10m x 10m pixel resolution are sufficient. This resolution shows variability of growth over a farm block but if growers require more definition, say down to 30cm or even 2cm resolution, the possibility is also there,” added Professor Robson.

“It very much depends on how much growers want to spend and how practical and trustworthy the service provider is. So, there are lots of things to consider and what the grower actually needs is the most important one.”

For more information about this project please contact Extension Officer, Alice Moore via email alicew@farmacist.com.au.

For more information about the Producer Technology Update Program visit agrifutures.com.au/national-rural-issues/producer-technology-uptake-program/