Machine learning predicts forest soil fungal diversity from drone images
Combining drone data and machine learning can help cover more ground in monitoring forest soil health, University of Alberta research shows. The findings are published in the journal Forest Ecology and Management. Using both tools to map and monitor soil fungal diversity—a key indicator of a healthy forest ecosystem—proved highly effective and could help reduce the need for boots-on-the-ground…
New research from the University of Alberta demonstrates that machine learning can predict forest soil fungal diversity by analyzing drone images. This technique may offer a cost-effective and scalable alternative to traditional soil sampling methods. Dr. Cameron Carlyle, a professor involved in the study, explains that soil fungi play a crucial role in various forest processes, and monitoring their diversity is essential for maintaining ecosystem health.
The study focused on the alpha diversity (number of different fungal species in a specific spot) and beta diversity (changes in fungal species types across the forest). Researchers collected 538 soil samples from a 26-hectare area in a Chinese nature reserve and used drones to gather high-resolution images, tree height measurements, and leaf light reflection data.
By employing a random forest model, they found that machine learning could predict around 53% of the beta diversity and moderate levels of alpha diversity (28-45%). While the technique cannot entirely replace on-site sampling, it can enhance the coverage and efficiency of monitoring fungal diversity across large and diverse forest areas.
This tool could aid in forest restoration, conservation, and long-term monitoring efforts, allowing managers to track underground soil health more efficiently.
Written by urgent.news from Phys.org's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
