Austria's Agency for Health and Food Safety (AGES) announced on Monday, September 14, that it is developing new artificial intelligence tools designed to detect flavescence dorée and its insect carrier, the American grapevine leafhopper, before they can take hold in a vineyard.
The initiative combines drone-based aerial imaging with a tractor-mounted detection system, both aimed at giving plant protection services and winegrowers a faster, more precise way to respond to what AGES describes as the most serious threat to Austrian viticulture since the phylloxera crisis roughly 150 years ago.
A Disease That Moves Fast and Kills Faster
Flavescence dorée is caused by phytoplasmas — wall-less bacteria that cannot survive outside a host plant. In wild vegetation, these organisms tend to cause mild or unnoticeable symptoms. In cultivated grapevines, however, the effect is far more severe: an infected vine can die within just a few years.
What makes the disease especially dangerous is its mode of transmission. The American grapevine leafhopper, an invasive insect that lives almost exclusively in vineyards, picks up the phytoplasmas by feeding on an infected vine and then transmits them to healthy vines during subsequent feeding. Because a single insect can move repeatedly through a vineyard, the disease is capable of spreading across an entire plot within one growing season.
Diagnosis Is Harder Than It Looks
Identifying flavescence dorée in the field is not always straightforward. According to AGES, the agency is the only laboratory in Austria able to definitively confirm the disease using accredited methods. Part of the difficulty lies in symptom variability: signs differ depending on grape variety, and at first glance they can closely resemble those of stolbur, another phytoplasma-driven disease that is already widespread in Austrian vineyards. This overlap means visual inspection alone is not enough — laboratory confirmation is typically required to distinguish between the two.
A Spreading Footprint
Flavescence dorée was first identified in Austria in Styria in 2009. Since then, it has steadily moved northward. AGES points to the emergence of a new strain of the pathogen in 2018 as a turning point that accelerated the disease's spread. Today it is established throughout Burgenland, and this year it was confirmed for the first time in Lower Austria.
The insect vector's history in the country runs on a parallel — and earlier — timeline. The American grapevine leafhopper was first recorded in southern Styria in 2004, reached Burgenland by 2010, and arrived in Lower Austria the following year. Because the insect's spread has consistently preceded the disease's, Austrian authorities treat monitoring of the pathogen and monitoring of its carrier as two closely linked efforts rather than separate concerns.
When the Disease Is Confirmed, the Response Is Drastic
Any suspected case of flavescence dorée must be reported to the relevant state plant protection service, which then coordinates eradication or containment measures. AGES is clear that the disease cannot be managed through in-field treatment: infected vines have to be uprooted, and in cases where damage exceeds a certain threshold, entire vineyards may need to be removed to eliminate remaining sources of infection. That severity is precisely what makes early detection so valuable — catching an outbreak before it spreads can mean the difference between removing a handful of vines and losing a whole plot.
Three AI Tools, One Goal: Earlier Detection
The research project AGES has launched is structured around developing, testing, and eventually deploying AI systems on two main fronts.
The first uses aerial imagery captured by drones over high-risk vineyard areas. AGES is testing multiple image-analysis systems against one another to assess how well each performs — in terms of operational practicality, robustness, and accuracy — at locating and identifying signs of infection across terrain with different topographic features.
The second focuses on a tractor-mounted image-capture system paired with open-source, AI-assisted software. The goal here is a tool that can operate in real time during ordinary vineyard work — spraying, mowing, or other routine passes — at a cost low enough to make it practical for everyday use by growers, rather than a specialized instrument reserved for research institutions.
A third, complementary application is also in development: a system that automatically identifies adult leafhoppers caught in sticky traps. AGES says this tool would cut down the manual labor currently required to analyze trap catches and would make routine monitoring feasible directly at the farm level, without requiring specialized entomological expertise on-site.
Why It Matters for the Wine Industry
Earlier identification of infestation hotspots — whether at the scale of a region or a single vineyard — allows for more targeted and less destructive control measures. For an industry built around a crop that can be wiped out in a single season if the disease goes undetected, shaving weeks or months off the time between infection and confirmation could substantially reduce losses of grapes, one of winemaking's most fundamental raw materials, and give both plant protection authorities and growers a meaningfully faster window to act.
Source: Vinetur