The Canadian Wood Fibre Centre (CWFC) in collaboration with Canadian company Flash Forest is exploring a novel forest regeneration approach using multi-rotor Unmanned Aerial Vehicles (UAVs) to apply tree seed-loaded pods in evenly distributed patterns across wildfire sites and other sites identified as suitable for vertical direct application into natural or prepared microsites.

There are several reasons why development and successful deployment of this technology could benefit Canadian forestry.

First, it is considerably cheaper to regenerate a forest using high viability seed versus manual tree planting, and it is a cost-effective alternative to the use of nursery-grown stock which would quickly be exhausted if it was the only tree source available to restore the vast area that has most recently suffered catastrophic wildfire. Also, as a cost-effective alternative to manual tree planting, both government and forest companies can address their potential reforestation liabilities more quickly, although, historically, there have been poor results using direct seeding.

Secondly, this technology can be deployed on appropriate sites in a timely manner after a wildfire, and its use is particularly favorable on sites where intense fire has occurred so that a significant amount of the organic layer above the desired mineral soil on the forest floor has been severely burned. The organic layer also hosts potential competing vegetation stems, seed and roots, again, an advantage to tree seed artificially positioned on or embedded in the mineral soil using advancing technologies with seed-loaded pods.

Thirdly, it can be deployed where it would be unsafe to deploy tree planters because of dangerous deadfall, steep ground conditions, and other access limitations associated with natural disturbance; and even on these sites, it provides consistency of distribution with an application from above the standing physical impediments (25 to 30 m. height), using selective flight lines and targeted microsites.

And finally, a diversity of species can be applied on the same site and the resulting stand diversity can be manipulated to influence fuel characteristics, resulting in reduced future risk of wildfire occurrence and intensity.

“We have to look at this from a scaling-up perspective,” says Derek Sidders, Program Manager and Regional Liaison at CWFC. “How do we get to more areas faster and in a time sensitive manner, and how do we address the huge increase in the amount of reforestation that is necessary.”

CWFC has evaluated data and ground inspected sites where this UAV seeding technology has been deployed over the past three years and germination and survival has occurred.

As this Canadian-developed technology continues to evolve and positive evidence continues to mount, Sidders says that it gives forest managers and stakeholders another tool in their reforestation technology inventory.

“We need to use every technology available,” he says. Sidders added that if Flash Forest’s technology can be proven to be cost-effective, combined with scrutinizing and stratifying sites overhead and on the ground to identify appropriate bio geo-climatic conditions that are remote, have physical impediments, or where there is a need to treat large-scale sites in very short timeframes to avoid issues such as growth of competing vegetation, “it absolutely has a place if it proves to be successful on the ground as evidence demonstrates successful germination, establishment and growth of the pod-delivered crop to a sustainable size.”

Flash Forest collects a significant amount of its own seed in close proximity to where a disturbance has occurred. It has also developed in-house pod manufacturing technology capable of producing over two million pods per day with patent-pending moisture retention and transference capabilities, in what the company believes provides the seed in the pod, infused with 17 different microbial media and beneficial bacteria, with the best chance for germination and survival once it has been applied by the UAV. By using pods, the company can eliminate nurseries from the supply chain, allowing Flash Forest to respond within a matter of weeks or months, instead of years, to any forest disturbance.

To identify suitable sites, Flash Forest first conducts a desktop pre-assessment. Data such as burn classification and severity, solar isolation, mean precipitation, soil moisture content, slope degree, elevation and other indexes based on satellite imagery are analyzed and complemented by data collected from drone flyovers on each site. This data is then analyzed using artificial intelligence and a company-designed machine learning algorithm to optimize site selection and to remove naturally regenerating sites. Areas with abundant natural seed sources (serotinous cones or semi-serotinous seed), or sites not suitable for trees or better-suited for site prep and tree planting are excluded.

Based on this pre-assessment, company CEO Cameron Jones says that they can offer clients highly specialized seeding prescriptions according to the client’s prescribed species mix and the availability of microsites within well-defined polygons situated in disturbed areas. To date, virtually every site they have seeded has combined multiple coniferous species for diversity and seed establishment options and could add deciduous species.

“A more recent burn, such as in the last two years, is much easier for us to plant because we don’t have competing vegetation,” he says, adding that knowing the area’s mean summer precipitation is a critical parameter. “We like to plant sites that have 250 millimeters of growing season precipitation or above.”

Flash Forest uses heavy-lift drones adapted and modified from the agriculture industry. The drones can carry 70 kgs and are capable of firing 18 pods per second at 60 metres per second while flying at 7 metres per second. The company is able to apply about 10,000 pods per flight, which takes about 10 to 15 minutes.

According to Flash Forest Director of Silviculture Jeff Renton, the goal is to have the vast majority of the pods 60 to 80 per cent embedded into the soil surface for optimal germination potential. Test plots show that embedding to this level almost doubles the germination rate.

Historically, the company planted 20 pods for each established tree with the ultimate goal of establishing 1,500 evenly distributed trees per hectare at ‘Stocking Assessment.’ That is a gross total of 15,000 to 30,000 pods per hectare. Over the past three years, the average establishment rate after two full growing seasons has been 5 to 6 per cent across all sites they planted. This was calculated with individual pods being accompanied by flags and assessed over multiple years to avoid counting any natural regeneration.

In 2024, they reduced their pod use to 10 pods per tree and believe that with continued refinement of the pod recipe and improvements to microsite selection, they will be able to achieve a ratio of 5 pods per tree. Renton says that certain test plots manifesting specific characteristics related to rainfall, direction toward the sun, burn age and severity, and no vegetative competition delivered significantly better results.

CWFC and its partners will continue to work with Flash Forest to evaluate the potential of this Canadian technology for large scale use in favorable conditions and the management of naturally disturbed forest sites

For additional information, please contact Derek Sidders, Program Manager, Technology Development and Regional
Liaison at CWFC, NRCan at derek.sidders@nrcan-rncan.gc.ca or Jeff Renton, Director of Silviculture, Flash Forest, at
jeff.renton@flashforest.ca.