Climate-sensitive growth modelling in Ontario
José Riofrio, PDF
Throughout the Hub sites, significant field inventory data is already available, which cover many growth characteristics, disturbance regimes and stand structures that could be considered for future silvicultural scenarios. The data, however, has been acquired over long periods using different methods across sites. Big data and machine learning approaches are well designed to be able to extract underlying trends from large datasets, where more traditional approaches such as regression may fail. In this project, we will invest in such approaches with the objective to link climatic data with the long-term forest mortality and growth data. To achieve this, Lara Climaco de Melo (PDF) will compile existing plot datasets across Hub sites and use machine learning techniques to predict mortality and growth from past climate. In conjunction with existing efforts, we will develop an open-source tool to facilitate exchange of inventory data between Hub sites, provinces, and companies. Outcome: A workflow to identify key climate variables that affect tree growth and mortality.
José Riofrio, PDF at University of British Columbia
Main Partner: Canadian Wood Fibre Centre
Professor: Bianca Eskelson
Riofrio, J., White, J.C., Tompalski, P., Coops, N., Wulder, M.A (2023) Modelling height growth of temperate mixedwood forests using an age-independent approach and multi-temporal airborne laser scanning data. Forest Ecology and Management 543;121137. https://doi.org/10.1016/j.foreco.2023.121137
Riofrio, J., White, J.C., Tompalski, P., Coops, N., Wulder, M.A (2022) Harmonizing multi-temporal airborne laser scanning point clouds to derive periodic annual height increments in temperate mixedwood forests. Canadian Journal of Forest Research, 52(10): 1334-1352. https://doi.org/10.1139/cjfr-2022-0055