Saving our forests

According to the IUCN, forests are one of the most important solutions to addressing the effects of climate change. Approximately 2.6 billion tonnes of carbon dioxide, one-third of the CO2 released from burning fossil fuels, is absorbed by forests every year. Increasing and maintaining forests is, therefore, an essential solution to climate change.

Today, the technology for identifying bark beetle attacks in the forest is largely based on manual work through visual checks of forest areas. Early attacks are almost impossible to see, while old attacks are easier to detect.

This technique, however, is time-consuming, expensive, and unable to spot early attacks so as to save the tree before it is too late. This is why innovative solutions must be created to swiftly and efficiently identify problem areas. A solution is to combine drones images with AI technology to quickly detect infected or stressed trees. This is fueling precision forestry in which we can rapidly gain insights and valuable information about large forest areas and identify problems where almost immediate intervention is key.

GLOBHE teamed up with  Arboair  to track the spruce bark beetle infestation in Swedish forests. By utilizing GLOBHE’s Crowddroning network of drone pilots in collaboration with Arboair’s image analysis platform, we were able to extrapolate valuable insights for forest owners, managers, and timber buyers. 

The reason why we chose to work with GLOBHE is the flexibility, speed, and extensive geographical coverage. With Crowddroning, our customers can quickly and flexibly get the data they need when they lack access to a drone for data collection,” Jacob Hjalmarsson, COO of Arboair.  

Drones take a large number of high-resolution photos over specific areas. According to the project, drones can be equipped with other sensors to gather other types of data, such as thermal imagery and LIDAR. In our case, only images are needed because we will identify infected trees from what they look like in images.

These captured images, however, need to be processed before starting to take information out of them. Below is a detailed explanation of the workflow.

An orthomosaic 2D map is like stitching all images together to create a large image that is geotagged and has both latitude and longitude information.

The resulting map is a high-resolution and recent representation of a geographic location, containing much more valuable information than an outdated low-resolution satellite image.

According to the Swedish forest authority, groups of trees having a rust-red top or with flaked bark, are likely infested by the spruce bark beetle. When the needles and crowns have a gray color or the needles have fallen off, the infestations are probably from previous years' infestations. Very recent infestation can only be seen by manual inspection as the bark beetles didn't cause any visible effect on the tree yet.

This means that we will look for both grey and rust-red topped trees when trying to identify trees from drone images. In addition, we will add a buffer area around all infested trees that require manual inspection so that forest owners know where to look in case some trees are suspected to be recently infected.

The final output is a GIS map showing the exact location of both infected, and stressed trees. This map is then delivered to the client, which could correspond to forest owners or municipalities. They will use this information to decide how to tackle this infestation. They will know exactly where each infested tree is and where all other trees might be attacked by bark beetles. Therefore, they can act much more quickly to save their forest from further infestation.

According to the  Swedish forest authority,  it is important that newly attacked trees are cut and transported out of the forest as soon as possible while they still have most of the bark left. Then the owner will get many spruce bark drills that are still in the bark. It usually does not make sense to remove occasional infested trees that have turned grey and where almost all the bark is gone because the spruce bark beetle had enough time to leave these trees.

If the owner has a lot of damage, it may be wise to harvest the entire stock. Such stocks are in such poor condition that they will probably be heavily infested next year as well. By harvesting these stands during the winter, owners can save the timber value and force the spruce bark beetle to give in to other stands next year, which have higher resistance, . 

AI-powered drones have proved to be very effective in quickly scanning and precisely detecting infected trees. It is possible to adopt this workflow with satellite images but using drones provides more flexibility, swift deployment, and higher resolution which minimizes mistakes and uncertainties.

As the bark beetles are quickly moving from tree to tree, it is essential to identify them early on. With traditional methods, infected trees are discovered too late, when the bark beetles have already migrated to nearby healthy trees. That's why an early discovery helps save both trees and money.

Scale-up is very possible with  Crowddroning by GLOBHE , because it connects to thousands of pilots that are capable to perform this work both in Europe and globally. GLOBHE's collaboration with  Arboair  allows this to be accomplished quickly and periodically to help forest owners and local authorities to minimize the effect of the spruce bark beetle. In addition, Crowddroning also directly creates jobs for local drone pilots and helps the local economy.