C‑DRONE
Agricultural drone flying over a farm field

C-DRONE GUIDE · 7 AUGUST 2026

Drone bird deterrence for vineyards and orchards: starlings, method, price

A flock of several thousand starlings can strip a vineyard plot or a cherry orchard bare within days, just weeks before harvest. Faced with that risk, classic deterrents — netting, gas cannons, predator kites — quickly lose their effect: birds are clever enough to habituate within days to a stimulus that never changes. This is precisely where a drone changes the equation: flown or programmed to intervene unpredictably, it plays on birds' instinctive fear of an unfamiliar flying object, one that never becomes a familiar fixture of the landscape. Here is what research shows about its real-world effectiveness, how a mission is organised at a vineyard or orchard, the regulatory framework that applies, and what this service costs in 2026.

Published on 7 August 2026, reviewed on 8 August 2026 — regulations in force as of August 2026.

Starlings and thrushes: a risk measured in lost harvest

The common starling, along with the thrush and the woodpigeon, is the bird French winegrowers and fruit growers worry about most as harvest approaches: moving in flocks of several hundred to several thousand birds, it can wipe out an entire season's work on an exposed plot within days, particularly on early-ripening grape varieties and cherry orchards, whose ripening window coincides with the autumn migration. In the United States, where the phenomenon has long been documented, bird damage to fruit crops is estimated in the millions of dollars every year; in France, an unprotected plot next to a hedge or a wood can lose several tens of percent of its harvest on the most exposed rows.

Traditional deterrents — anti-bird netting, detonating gas cannons, raptor-shaped kites, scare balloons — remain widely used, but all share the same weakness: a fixed stimulus, repeated at regular intervals or motionless in the landscape, quickly stops being frightening. A gas cannon that fires every ten minutes becomes, within days, mere background noise that starling flocks learn to ignore; a motionless predator kite eventually gets identified as harmless. This habituation explains the renewed interest in drones as a deterrence tool.

What research shows about drone deterrence effectiveness

The principle behind a bird-deterrent drone rests on a documented behavioural pattern: an unfamiliar flying object moving unpredictably triggers a far more lasting fear response in birds than a static or repetitive stimulus. A study published in 2019 in the journal Crop Protection by Zihao Wang and Andrea S. Griffin, of the University of Newcastle in Australia, tested this approach in the field: the researchers fitted a drone with a loudspeaker broadcasting bird distress calls and a taxidermied crow mounted underneath, to visually suggest a predator carrying captured prey. Trialled in vineyards across south-eastern Australia, the setup showed it could durably deter large birds such as ravens and cockatoos within a 50 m radius, and chase off smaller birds such as silvereyes, though the effect on the latter proved shorter-lived (see the study on Google Scholar).

A second study, published in 2022 in Computers and Electronics in Agriculture by Bhusal, Karkee, Bhattarai, Majeed and Zhang, of Washington State University, takes the idea further by automating the system: a fixed camera detects a flock of birds entering the vineyard, automatically triggers the launch of a compact drone tasked with dispersing them, then brings it back to base to wait for the next incursion. This system builds on two earlier studies by the same team showing that manually piloted flights, with a deliberately unpredictable flight path, were already enough to effectively keep birds out of a vineyard (see the study on Google Scholar). Common to both pieces of research: the unpredictability of the drone's movement matters at least as much as its mere presence in delaying habituation.

Manual flights or an automated system: two levels of investment

Two approaches coexist on the ground today. The first, and the most accessible, involves flights piloted manually by a drone operator, carried out several times a day at the hours when risk peaks — early morning and late afternoon, when starling flocks move between their roosts and feeding plots. This formula suits a targeted intervention over a few critical weeks well: véraison to harvest for vines, or the ripening period for cherry trees and other berry-bearing fruit trees, whose other seasonal uses are detailed in our guide on drones in orchards.

The second, closer to the research setups described above, relies on a fixed charging base paired with automatic detection: as soon as a camera or radar spots a flock of birds, a compact drone takes off on its own to intervene, then returns to recharge. The upfront investment is markedly higher, but monitoring becomes continuous with no dedicated operator — sound for a large estate, less so for a smallholding of a few hectares where occasional flights are enough. Between the two, a grower can also have the same provider handle grape ripeness mapping or orchard monitoring, pooling site visits over the same at-risk period.

What French regulations say

Unlike a spraying drone in vineyards, which may only fly under the strict derogation regime set by the law of 23 April 2025 — reserved for steep-slope plots or a serious sanitary hazard, as detailed in our guide on drone spraying in vineyards —, a bird-deterrent drone carries and releases no product at all: it falls under the standard open category, in sub-category A1 or A3 depending on its weight and how close nearby housing sits to the plot. The pilot needs the corresponding training — see our guide on A1/A3 training and the A2 certificate —, must be registered as an operator on AlphaTango and covered by professional liability insurance, as for any agricultural mission.

One point does call for particular care: a fully autonomous system, like the American research setup described above, falls outside the standard framework of European regulation, which in principle requires a remote pilot present for every flight. Such a system cannot, as things stand, be deployed in France without specific DGAC authorisation; current commercial offerings therefore remain on-demand piloted flights, rather than a continuous automated patrol. Since vineyards and orchards usually sit in open rural areas, checking the Géoportail restricted-zone map remains standard practice nonetheless, particularly near an aerodrome or a Natura 2000 zone, for which our guide on flying a drone in a nature reserve details the extra precautions needed when protected species are present.

The price of a drone bird-deterrence mission in 2026

The price depends on the formula chosen — targeted one-off flights or a package covering the whole at-risk period — and the surface to cover. Orders of magnitude observed in France in 2026 (excl. VAT):

ServiceObserved price (excl. VAT)
One-off deterrence flight (single pass, one plot)€100 to €200
Daily package over the critical period (several passes a day, 2 to 4 weeks)quoted, lower per-unit rate depending on frequency and surface
Prior scouting and flight plan (roost corridors, incursion times)€150 to €300, often included in a season package
Setting up a fixed-base system with detectionhigher investment, quoted based on the surface to monitor

The comparison that matters is not the price of the flight alone, but the harvest value preserved: a few well-timed targeted passes are often enough to limit losses that, left unchecked, run into several thousand euros. Request a quote stating the surface to protect, the species responsible for the damage, and the risk period targeted.

Does a drone replace gas cannons and netting? Not necessarily: it often comes as a complement, particularly during pressure peaks, precisely because it avoids the habituation that fixed stimuli suffer from.

Does it require a derogation like drone spraying in vineyards? No: a bird-deterrent drone carries no product, so it falls under the standard open-category framework.

Does the drone also disturb protected species? Non-lethal deterrence not targeted at any particular species is generally tolerated, but it is worth checking which species are present locally before any regular intervention, particularly in a Natura 2000 zone.

Request a free quote

Put it into practice

Also worth reading