Technology developed at Melbourne University has increased Xylella detection accuracy in trials by up to 92%, while reducing uncertainty to less than 6%, across different plant hosts, including almonds and olives.
Reported in Nature Communications, the technology takes the research community a step closer to developing a rapid and more accurate large-scale screening process for at-risk crop species by enhancing the effectiveness of airborne scanning that uses hyperspectral imaging.
In our November 2018 E-news we shared an article on detecting pre-visual symptoms of Xylella in olive trees using spectral imagery.
While researchers have previously demonstrated the ability to detect Xylella in affected olive trees before symptoms appear, a common difficulty with the remote sensing algorithms that scan the hyperspectral images was attributing characteristics of a lack of water or nutrients to Xylella as opposed to other pathogens or environmental stresses.
“This is the same issue we have faced to date with identification of phylloxera infested vines using hyperspectral imagery before visual symptoms are evident,” said Suzanne McLoughlin, Vinehealth Australia Technical Manager.
Researchers at Melbourne University with their international partners, have been focusing on improving the ability of these remote sensing algorithms to ‘see’ where Xylella is and is not, to reduce the incidence of false positives.
After scanning 1 million infected and healthy trees in seven regions in Europe, they have reported that a novel algorithm has allowed them to distinguish the disease from water-induced stress and increase Xylella detection to up to 92% accuracy while reducing uncertainty to below 6% across different hosts, including almonds and olives, and across other vascular pathogens.
The technology available at Melbourne University, was trialed in Victoria last year as part of the National Xylella Preparedness Program. The project involved scanning several thousand hectares of healthy almond, citrus and olive trees with varying water and nutrient status levels as baselines to better adapt the Xylella detection models developed in Europe for the particular varieties and management practices in Australian agriculture. Read more on the research here.
“This technology is extremely exciting in the context of biosecurity in general. If a Xylella outbreak occurs in Australia, the methods developed out of this research could potentially be used to rapidly detect and limit the spread of the disease,” said Suzanne.
“We continue to see the applicability of this research for phylloxera given the similarities between Xylella and phylloxera in the delay of appearance of visual symptom expression and potential for spread in this time.
“We encourage this area of research to be explored for the early detection of phylloxera. This is vital given a lack of early detection for the pest.”
Xylella fastidiosa is the number one unwanted plant pest for Australia and a High Priority Pest for the wine industry. It is a bacterium which affects the xylem of plants where it blocks the movement of water, dehydrating the plant and causing death within a couple of years.
Known as Pierce’s Disease in grapevines and by other names in other host plants, the disease causes significant environmental and economic impacts across more than 550 commercial and ornamental plant species. Countries currently battling Xylella report that the key to limiting its spread involves early detection. This is difficult, however, given visual symptoms commonly aren’t apparent until 8-10 months after infection, during which time spread can occur from apparently asymptomatic plants.