AI-Driven Determination of available Forest Biomass and Harvester Efficiency for GHG Reduction in New Brunswick

New Brunswick’s forestry sector currently relies on traditional harvesting methods that consume approximately over 50 million litres of fossil fuel annually to harvest 14 million cubic metres (m³) of timber. However, the careful selection of the optimal harvesting systems can contribute to the reduction of fuel consumption by a third for the same output. This presents a significant opportunity to save up to 15 million litres of fossil fuel each year, translating into an annual reduction of approximately 38,029 tonnes of carbon emissions. Additionally, the high cost of traditional harvesting methods often renders the extraction of net biomass economically unfeasible, limiting the province's ability to utilize biomass as a renewable energy source.

The objective of this project tackling the upstream portion of the forest biomass value chain are 1) to enhance LiDAR-derived Above Ground Biomass (AGB) models by integrating freely available, high-frequency Sentinel-2 satellite data and advanced AI tools for more accurate net biomass prediction, and 2) to refine and validate predictive tools for forest harvesting system choice productivity using advanced remote sensing technologies and artificial intelligence (AI). These improvements aim to optimize harvester operations and biomass estimation in New Brunswick’s mixed and hardwood forests, increasing efficiency and sustainability.

By addressing inefficiencies in current harvesting practices, the project will enable more precise and efficient resource extraction, significantly reducing fossil fuel consumption and operational costs. The expected outcomes include increased harvester productivity, cost-effective biomass extraction, and a measurable reduction in the province's carbon footprint, positioning New Brunswick as a leader in sustainable forest management and renewable energy production.