The potential of aerial drones for thinning operations

The objective of this study, a desk study, was to evaluate the usefulness/potential of an airborne system in thinning operations in forests of Fennoscandia. The system is based on unmanned aerial vehicles (“drones”). The potential was evaluated from technical, environmental and silvicultural perspectives.

Thinning is a common silvicultural treatment in Sweden, with the first thinning typically done when a stand has a dominant height of 10-15 metres. There are different methods of thinning. Thinning from below where smaller trees are removed, uniform thinning where trees from all size classes are removed, and thinning from above where larger trees are removed. Some 300,000-350,000 ha of forest are thinned every year, and roundwood from thinnings makes up some 20-25 percent of the annual roundwood harvest.

Issues of concern in connection with thinning include:

• Thinnings tend to be done too late in the rotation, resulting in poorer diameter growth in the residual stand.
• Late first thinnings make stands more prone to wind and snow damage.
• Future stand development is too often impeded by neglect of pre-commercial thinning, making later thinning difficult.
• Spread of root rot in connection with thinning due to mechanical damage from tyres on roots and logging damage to stems in stands where stumps, the main ‘entry point’ for root rot, have not been treated against rot.
• Logging damage and ground damage by heavy machinery.

An obvious benefit from an aerial thinning system is that no strip roads are needed. Strip roads tend to cover around 18 percent of stands thinned, an area cleared with associated production losses. Studies indicate a production loss of 5–10 percent over 15 to 20 years. Many trees harvested in first thinning are strip road trees, suppressed and/or damaged trees, i.e. trees that must be removed. Thinning using an airborne system can be more selective, but broadleaves and large and deformed trees may pose problems due to inherent limitations of airborne systems with harvester heads.

Nurse crops, crops of trees or shrubs that foster the development of another tree species by protecting it, could be used more if an airborne system become an option. Nurse crops offer frost protection to seedlings and control of groundwater, increase growth and yield by 10 to 15 percent over a rotation, and increase seedling survival. Establishment of nurse crops, mainly used to protect seedlings from frost, and removal of these crops using normal systems is difficult and expensive. Logging damage levels and operational economics have limited the use of nurse crops. With airborne systems, damage could probably be held at acceptable levels, and the recurrent operations needed could be economically more viable. This would enable wider use of nurse crops, not only for frost protection, but also to increase production over the rotation.

Diameter distribution in drone-thinned stands will be wider than in stands thinned with normal systems. Broadleaves and large and deformed trees may be difficult to remove, as well as small trees not visible from above. This effect can be reduced by restricting the use of airborne systems to stands with few broadleaves and large and deformed trees.

An airborne system is an interesting option for removing spruce trees from sensitive sites like protected areas, buffer zones, and cultural heritage sites.

Alternatives to the system are worth considering, for example, using grapples instead of harvester heads on drones and using harvesters working from ghost trails. Problems with large and/or deformed trees and broadleaves may be eliminated with such a system. In addition, a grapple is not as heavy as a harvester head, thereby enabling larger payloads.

Technical perspectives

Thinning with an airborne machine that fells and delimbs selected trees and subsequently transports them to the landing is a new concept, so this evaluation is based on simulation techniques. Given the assumptions made, such a system with only one drone is not likely to be economically competitive against a conventional thinning system using a harvester and forwarder. This may change if one operator can manage multiple drones simultaneously. The results can be summarised as follows:

• Drone productivity is sensitive to tree weight, with an optimum load volume being close to the maximum payload.
• Drone productivity is sensitive to transport distance, especially for loads consisting of a small tree.
• Small trees result in low productivity. It will be difficult to position the harvester head to delimb and cut small trees positioned close to larger future crop trees. This was not considered in the simulations, and will cause further reductions in productivity.
• The desired extraction rates will be difficult to reach in thinning stands with a large variation in tree sizes, especially when the average stem weight is close to the safe payloads.

Pest control perspectives

During recent decades there has been a marked increase in disturbances caused by tree-killing bark beetles in all the world’s coniferous forests, leading to large losses of economic and ecosystem values. Following the drought event in 2018, there has been an unprecedented spruce bark beetle outbreak in southern Sweden. Drought lowers water availability, which reduces tree defences and increases tree susceptibility to spruce bark beetle attack, while higher temperatures increase flight activity and development of the beetles.

Timely removal of infested trees is considered the most effective way to reduce bark beetle populations, but this is difficult in practice. An emerging field of research is focusing on remote sensing techniques for detecting beetle-infested trees as early as possible. However, pest control efficiency is limited by availability of harvesting machinery before the new generation of beetles leave the infested trees. Properly adapted, the use of drone harvesters could contribute to pest control, probably not by preventing bark beetle outbreaks but possibly dampening the severity of future outbreaks.

Effects on stand production

We assumed that the technical nature of an airborne drone system enables reduced thinning intensities, and that forest management goals in production forests remain unchanged. This motivates increased thinning frequencies compared to the conventional harvester-forwarder system.

The Heureka analysis showed no differences in stand economics, stand production, carbon storage, or diameter growth between the two harvesting techniques, given that the forest management goals were maintained.

No inherent differences between the harvesting methods were found, since growth and stand density were approximately the same throughout the thinning phase, despite the difference in thinning intensities on individual occasions. The strip-road effect was not accounted for in the simulations, but it can be assumed to have an effect amounting to an extra 10 to 30 m3 per hectare over a rotation. This is not a massive volume admittedly, but on par with the extra volume gained through fertilisation some ten years before final felling, an investment often considered worthwhile.

Any future evaluations should include more detailed productivity estimates of the airborne drone system and experiences from practical thinning operations.

Biodiversity aspects

Thinning with drones is anticipated to yield positive environmental outcomes, primarily because of the elimination of strip roads. This will minimise the negative effects associated with their construction during traditional logging operations, for example wheel rutting and root damage. Thinning with drones means that there will be no understorey removal, promoting a denser vegetation layer. Drones reduce the risk of spreading invasive plants, as they do not drive over ground and carry seed or other matter from one location to another. Depending on the drone's range, there is potential for implementing smaller nature conservation measures outside the actively thinned stand.

To fully realise potential benefits, effective tools for planning and identification of structures, such as old broadleaves, from the air are necessary. Potential uncertainties or negative effects are dependent on the specifics of the drone harvesting method. For example, there is a potential risk of environmental disturbance to wildlife species from noise and the mere presence of drones that may differ from conventional logging disturbance. Generation of logging residues (GROT) and the potential covering of vegetation remains unknown until field studies have been conducted. Lastly, the possibility for the drone to reach areas currently inaccessible to machinery can cause disturbances to sensitive species in this type of refugia.

In summary, while thinning with drones presents several positive environmental outcomes, there are uncertainties and considerations that need further investigation and careful planning to maximise benefits and minimise potential negative impacts.

A final consideration

The transition to fossil-free vehicles in the forestry sector can reduce the sector's environmental footprint and be advantageous for the business's economy and future. The forestry sector can strengthen its market position by adopting fossil-free vehicles, which can be a differentiating factor and increase demand for forest-based products from environmentally conscious consumers and companies.