Detection and characterisation of the forest road network using LiDAR

点击翻译按钮获取中文摘要

Forest roads are an essential prerequisite for sustainable forest management providing access for silvicultural operations, such as planting, stand tending, and harvesting, as well as being critical for other activities such as fire suppression and access. Roads are also used to provide access for recreational activities including camping, hiking or hunting. Critical to the management of the roads network is maintaining up-to-date and accurate information on both the quantity and the quality of road surfaces within a forest estate. The aims of this project are first to develop and validate new approaches to accurately locate forest roads across Quebec, and second to predict the condition (or attributes) of forest roads. For location will take advantage of the high spatial accuracy and province-wide coverage of the most recent airborne laser scanning (ALS) data acquisition in Quebec. Road attribute predictions and temporal updates will be derived from a mixture of satellite imagery analysis, participatory GIS and deep learning. Our research will be conducted in four core areas, covering a range of environmental conditions affecting road development and condition across Quebec. The team is highly committed to advancing the use of ALS Canada. The updated road layers will be made available on an open-access platform (https://www.foretouverte.gouv.qc.ca/) supported by the Quebec Ministry of Forests, Wildlife and Parks, which will enable interested users to gain direct access to the research outcomes.