Spruced up: U of T biologist uses drones to help build a better tree

Ingo Ensminger, a biologist at University of Toronto Mississauga, is using a drone to help breed better spruce trees.

Ensminger and his collaborators are working on a suite of modern genomic resources to improve tree breeding and selection, using large white spruce progeny trials established in Quebec and Ontario by the Canadian Forest Service (CFS). This progeny data represents 20 years of tree-breeding research, where scientists have selected 2,000 different genetic lines of spruce seedlings; they expect to find spruce seedlings among those lines that will efficiently adapt to environmental stresses.

Along with his CFS collaborator Nathalie Isabel, and in association with industry partner PrecisionHawk, Ensminger will use a drone carrying optical sensors for leaf-spectral measurements that will ultimately help distinguish the best genetic lines based on their “optical fingerprint.” The project is funded by Ontario Genomics.

“The main idea behind the optical fingerprint is the fact that plants constantly interact with their ever-changing environment,” says Ensminger (left). “This involves adjustments in pigments that are responsible for the colour of the leaf. These adjustments take place within minutes and over the course of the season in order to mitigate stress and maximize the yield of photosynthesis. Adjustments in leaf pigments accurately reflect the fitness of a plant.”

His team will use samples of seedlings from the ground to assess fitness, using existing tests. After measuring the optical properties of these leaves, they will compare findings with data collected by the drone – data that is captured in a fraction of the time and on a much larger scale. The ground data, he notes, will help the team better understand the drone data, and calibrate growth models accordingly.

“The resolution of the drone sensor will be good enough to distinguish individual seedlings,” Ensminger says. “We think we can assess performance of individual plants and quickly identify the most robust genetic lines during a short drone flight from fields with several thousand individuals.”

The investment by Ontario Genomics’ Pre-Commercialization Business Development Fund will support deployment of a software application created by PrecisionHawk that will eventually become available to breeders and forest managers through the Algorithm Marketplace.

“This will be a software tool made for users in the breeding and forest management industry,” says Ensminger.

“While the users are not specialists in remote sensing and plant physiology, this software will help them analyze drone data and determine the fitness of individual trees or entire forest stands. They’ll be able to quickly assess when trees are water stressed or when the growing season ends, so they can adjust their management practices accordingly.”