The Challenges of Spraying by Drone

Posted on

About Tom Wolf (Nozzle_Guy)

Tom Wolf is based in Saskatoon, SK and has 33 years research experience in the spraying business. He obtained his BSA (1987) and M.Sc. (1991) in Plant Science at the University of Manitoba, and his Ph.D. (1996) in Agronomy from the Ohio State University. Tom focuses on practical advice that is research-based to improve the efficiency of producers.

See all posts by Tom Wolf (Nozzle_Guy).

Spray application by drone is here. It’s common practice in South East Asia, with a very significant proportion of ag areas now treated that way. Estimates from South Korea, for example, suggest about 30% of their ag area being sprayed by drone. It’s in the US, too. The Yamaha RMax and Fazer helicopters, which pioneered drone spraying in Japan dating back to the mid 1990s, have been approved for use in California since 2015.  DJI, the world’s largest drone manufacturer, introduced their ag model, the Agras MG-1, to North America in 2016. Many other spray drones are available or in development.

As William Gibson, the author of Johnny Mnemonic, once said, “The future’s here, it’s just not widely distributed yet.”

DJI Agras MG-1 spray drone (Source:

Proponents of drone spraying cite a drone’s ability to access areas where topography is a problem, such as steep slopes, where productivity of manual application is much lower, or low areas where soil moisture prevents ground vehicles. Operator exposure is reduced compared to handheld application.

Opponents talk about productivity and cost factors compared to manned aerial application, spray drift, and rogue use.

Before drone spraying becomes commonplace, two important things need to happen.

  1. Federal laws need to be updated to accommodate the unique features of remotely piloted aircraft systems (RPAS), as they’re now called. Current laws make many assumptions unique to manned ships, and the process to correct that will require some patience. A thorough review for US laws, and their shortcomings, can be found here.
  2. Federal pesticide labels need to permit the use of drones for application. As of August, 2021, Canadian labels have no such registered use.

There is no doubt that we need to prepare for a future that includes spraying by drones. Features such as topography adjustment for height consistency and autonomous swath control are already essentially standard, and the capabilities that improve control and safety will continue to develop.

And yet I’ve been nervous about the prospect of pesticide application with drones. My primary concern is around – you guessed it – spray drift. Because a drone payload is relatively small (about 5 to 25 L, depending on the model), application volumes will need to be low to have any sort of productivity. How low? For manned aircraft with a 200 to 600 gallon hopper, 2 to 4 US gpa (18 to 36 L/ha) are the lowest commonplace volumes. The lower volumes require a Medium spray quality (among the finer sprays in modern boom spray practice) to achieve the required coverage.

It’s a simple concept: the less water is used, the smaller the droplets need to be to provide the necessary droplet density on the target. Drift control with coarser sprays requires higher volumes, and true droplet-size-based low-drift spraying can’t really happen at volumes less then, say 5 to 7 US gpa.

At 2 to 4 US gpa, a drone would be able to do perhaps 1 acre per load. While OK for spot spraying, it represents a serious productivity constraint for anything larger.  There will be a push toward lower volumes, perhaps 0.5 to 1 gpa (5 to 10 L/ha). The only way these will provide sufficient coverage is with finer sprays, ASABE Fine to Very Fine, with expected problematic effects on off-target movement and evaporation. These fine droplets are also more prone to the aerodynamic eccentricities of aircraft.

Vortices from the rotor can create unpredictable droplet movement (Source:

The current regulatory models for aerial drift assessment in North America, AgDISP and AgDRIFT, are not yet able to simulate drone application. But by entering finer sprays into these models for their conventional manned rotary wing aircraft, we can see that buffer zones will be higher. Much higher. And that outcome will give pause to regulators. Failure to control the movement of a spray is, and should be, a problem.

Estimated Buffer Zones (calculated by AgDISP) for a reference rotary wing spray aircraft, using three pesticide toxicologies and two spray qualities.

Furthermore, ultra-low volume (ULV) sprays can change the efficacy of some products, and these will require new performance studies. At this time, regulators are seeking information not just on spray drift, but on product efficacy, operator and bystander exposure, and crop residues.

Regulators are currently collecting spray drift and efficacy data from drones. Since the drones available in today’s market do not conform to a common design standard like fixed or rotary winged manned aircraft, each model may have its own characteristics and need its own study. Some will have rotary atomizers, others will use hollow cone hydraulic sprays. Some will have electrostatic charging, others may propose special adjuvants.

Once data are assessed, there will likely be restrictions in flight height, flight speed, wind speed, spray quality, water volume, perhaps air temperature and relative humidity (or Delta T). This is not new to spraying, as current labels already constrain use for both ground and aerial spray application, more so for aerial.

The obvious question is how these proper application practices can possibly be assured. Operators will need more than just regulatory approval to use a drone, they will require proper training, similar to what a commercial aerial applicator now receives prior to operating a business.

Recall that our aerial applicators are governed by national organizations, the NAAA in the US and the CAAA in Canada. These organizations are in regular contact with federal regulators to assure compliance. They also help fund research into application efficacy and safety. They organize conferences in the off-season and calibration clinics in the growing season. At these, flow rates are confirmed and deposited droplet size is measured. Spray pattern uniformity is assessed and corrected as necessary.

Should drone applications be exempt from these controls? I don’t think that would be wise. Are we ready to implement them? Absolutely not.

These requirements would change the drones’ economic model. And despite these precautions, a drone may still leave the control of a pilot due to unforeseen technical or human events.

In the US, Yamaha does not sell their drone helicopters. Instead, they deploy their own teams to make the applications. This way, they have assurance that only trained and experienced pilots use the technology.

As the industry gears up for the first registrations, we see drone service companies take a leading role in testing. Much is being learned via legal applications of liquid micronutrients, for example, or limited use of pesticides under approved research permits. And I’m pleased to see the recognition of drift management in these efforts through the use of low-drift nozzles. We are off to a promising start.

Requests for drone use are in progress at our regulatory agencies. The outcomes of their risk assessments will provide important initial guidance, and food for thought and discussion. In the meantime, the drone development continues at a rapid pace, with new features and greater capacity at each iteration.