Data-driven tree allometry of a planted forest in New Zealand, using machine learning and airborne lidar

This study presents a novel approach that uses machine learning and airborne lidar technology to establish data-driven tree allometry for a 20-year-old planted forest in New Zealand. The approach utilises airborne lidar measurements and a random forest classifier to derive a local allometric relationship between tree height and trunk diameter. This method offers an efficient alternative to labour-intensive traditional methods. The study concludes that the proposed approach effectively obtains precise local allometric correlations, with implications for large-scale forest inventories and ecosystem management in New Zealand.