Drones are useful tools for forestry management, but they may have unintended negative consequences that cannot be fully understood until their impact is assessed. As of 2022, there are 872,248 drones registered with the Federal Aviation Administration (FAA), with over 300,000 registered for commercial use. Drones are an optimal choice for forestry management because where humans struggle to traverse forest floors strewn with foliage and fallen limbs, drones can fly above the tangle. Not only are drones more maneuverable, but they can be equipped with advanced optics, LiDAR sensors, and onboard computers that can be programmed to operate the sensors and identify surroundings. Advancements in artificial intelligence and machine learning have led to the development of autonomous drones that pilot themselves and have the highest potential for forestry management.
Autonomous drones are being developed by Deep Forestry, a start-up from Sweden, alongside the U.S Department of Agriculture (USDA) and the European Commission. Deep Forestry claims that its drones can scan 1,000 trees every 20 minutes to accurately identify the species, height, diameter, trunk volume, and location of each individual tree. If these claims are true, this automated drone analysis will offer a bounty of new data for forest scientists and managers to work with. Current manual methods of assessing forests often only map 5 to 10 percent of trees and require assumptions and decision-making to be based on this limited data set.
Advancements in drone technology will likely be most impactful on fire management. Understanding the amount of fuel load, the total quantity of combustible contents in an area, is essential to effective fire management and the prevention of catastrophic wildfires such as the Bootleg fire that scorched over 400,000 acres. To calculate fuel load the U.S. Forestry Service advocates using the “photoloading” technique where field workers compare a section of forest or forest floor against a photo from a Forest Service Photo Guide, which provides estimates for the fuel load and how long a fire would last in that area. Photoloading is imprecise, varies depending on the impression of the individual inspecting the area, and is at best an estimate.
Drones can improve upon this system in two major ways. First, drones can more accurately measure forest biomass and calculate fuel load by relying on precise data rather than estimates. Second, drones equipped with near infrared light (NIR) cameras can measure the health of vegetation using the Normalized Difference Vegetation Index (NDVI), because healthy leaves reflect more infrared light than unhealthy or dead leaves. The farming industry is already using drones equipped with NIR and NDVI technology to measure crop health and variability. Combining more robust data on the amount of biomass with data on the biomass’ combustibility will create a more accurate and reliable estimate of fuel load.
Despite these on-paper benefits, widespread use of drones in forest management might bring myriad new issues. For example, access to previously inaccessible areas will provide more complete data, but it will also disturb what little habitat has managed to avoid human interference. A 2019 Cambridge meta-analysis of 30 published articles studying drones’ interaction with animals found that drones elicited a behavioral response from wildlife in two thirds of the studies. Of those studies, drones mostly impacted birds who attempted to escape 56 percent of the time and attacked the drone 11.1 percent of the time. The exact effect of drones on birds and other wildlife is unknown, but many birds are territorial and the smallest interactions can have lasting impacts. One inherent hindrance to drones’ invasion of the wilderness has been that drones can only fly as far as their human pilot can operate. Drones will be untethered once self-piloting drones become the norm and the human element is no longer necessary.
The increase of drone use in forests coincides with an existential crisis facing birds in North America. In just the past 50 years, the bird population of North America has decreased by almost three billion, a 25 percent decline. There is no one clear cause of this drastic decrease in population, but new disturbances in the forests and canopies will certainly not help. Furthermore, there is a lack of incentive to discover the extent of drones’ impact on wildlife. Similar to how the Forest Service was hesitant to find ecological value in the old growth because it would disrupt the agency’s business model, it is in the Forest Service’s interest to not find issue with drone use in forest management. Drones are more economically efficient at surveying forests because less labor is required, there is a decrease in human risk, and more data can be collected. With industry and agency aligned to increase drone use, negative impacts of drones will likely not be apparent until after their effect is felt.
Another possible issue drone use might bring is a false sense of security. Drones are excellent for collecting data, but only the data they are programmed to collect. While human surveying is costly and limited in scope, it allows for observation of unexpected phenomenon. Drones are not only incapable of making their own discoveries, but there is a risk data analysts will be satiated by the bounty of data collected and fail to consider what might be happening beyond the scope of the drone’s programming.
If the Forest Service does not self-impose limits to drone use, regulation via statute remains a viable tool to reign in drone usage. The FAA is responsible for regulating drone use at the federal level, but states and local municipalities are able to regulate drones as well. Federal regulations, FAA’s Part 107 and 44809 (a recreational exemption to Part 107), are broad and focus on national security. Drone regulations at the state and municipal level are more restrictive and show more promise to regulating drone use in the pursuit of environmental conservation.
In Oregon, one avenue to regulate drone use in forests can be through HB4066 section13, which makes flying a drone over, or making contact with, a “critical infrastructure facility” a Class A violation. Including certain federal land designations such as National Wildlife Refuges (16 USC § 668dd), critical habitats of endangered or threatened species (16 USC § 1533), or late-successional reserves (Northwest Forest Plan) to the list of “critical infrastructure facilities” would be an elegant way to regulate drone use without having to pass a new statute. One hurdle to this solution, however, is that section 13(3)(a) and (d) exempt the federal government and any person under contract with the federal government, respectively, from the “critical infrastructure facilities” limitations of section 13.
Regulation can still be implemented locally with city codes such as Portland City Code 20.12.180, which prohibits drone usage in any city parks. City codes similar to this are effective at achieving results, but are severely limited in scope. If drones do in fact negatively impact wildlife, nationwide collaboration at the local level would be necessary to prevent such impacts. Two structural issues facing future regulation are the inability of government bodies inability of government bodies to keep pace with technological advancements, and that drones are currently regulated by size and use more than they are by capabilities such as automation.
Drones will be helpful to forest management and environmental conservation generally, but more research is needed to understand their effects on wildlife before swarms of autonomous drones replace humans in monitoring forests. Lastly, while more data and more accurate data seems like a positive development, data will not solve the environmental issues facing the planet. Problem solving and difficult conversations between industry, politicians, and scientists are necessary to find solutions; the availability of more data might only lead to more analysis and hand wringing, further delaying the uncomfortable yet necessary conversations.
