Using LiDAR and remote sensing to model forest structure and assess wildfire hazard, improving understanding of fuel loads and landscape vulnerability.

Climate change is increasing wildfire severity and frequency, making it critical to better understand forest structure and fuel loads. This project focuses on using airborne LiDAR and remote sensing data to characterize forest structure and support the analysis of fire hazard across landscapes.

We use geospatial AI modelling to derive information on vegetation structure and distribution, supporting improved understanding of fire hazard at landscape scales.

This work supports environmental monitoring and decision-making by providing spatially explicit information on forest structure and potential fire hazard.

This work is part of the broader Wildland Foundations project: https://biodiversitypathways.ca/our-initiatives/wildland-foundations/

Collaborative geospatial work supporting regional land-use planning through detailed mapping and spatial analysis to inform decision-making across northern landscapes.

In early 2025, we began working with the Alberta Biodiversity Monitoring Institute (ABMI) to support the Ya’Thi Néné Lands and Resources Office in developing a detailed inventory of linear disturbances in northern Saskatchewan’s Athabasca Basin. There is currently a lack of detailed, comprehensive human disturbance maps in the region, particularly for linear features such as cutlines, which has limited the ability of regional land managers and Indigenous communities to assess landscape condition. 

The Athabasca Basin supports expansive wetlands, peatlands, mixedwood forests, and freshwater systems. The area forms part of the traditional territories of Indigenous communities who rely on the land for sustenance, cultural practices, and economic opportunities. Although the Athabasca Basin remains relatively undisturbed compared to southern regions, increasing exploration and development pressures—especially related to mining and mineral exploration—have led to the proliferation of linear disturbances.

This collaborative project is focused on digitizing human footprint features in six target areas using high-resolution satellite imagery. It also evaluates the feasibility of scaling up different mapping approaches—from manual digitization to semi-automated methods—across the wider Athabasca region

The pilot project is tied to a broader initiative to map linear features in the Athabasca Basin using lidar data and funding from the Weston Family Foundation.

Mapping forest structural complexity using remote sensing and modelling approaches to better understand ecosystem condition, disturbance, and recovery.

Forest structural complexity refers to the three-dimensional arrangement of trees and canopy layers and plays a key role in supporting biodiversity and overall forest health. This project maps forest structural complexity across Vancouver Island as an indicator of forest integrity using LiDAR, satellite imagery, and geospatial AI modelling.

We use remote sensing and modelling to quantify patterns of forest structure across landscapes, providing insights into how forests are shaped by natural processes and human disturbance. This work enables consistent, large-scale assessment of forest condition and recovery.

Forest structural complexity varies strongly with disturbance: recently logged and burned areas tend to exhibit lower complexity, while older forests show signs of structural recovery over time. Protected and conserved areas often maintain higher levels of structural complexity compared to surrounding landscapes.

This work supports conservation planning and environmental monitoring by providing spatially explicit information on forest condition and landscape change.

Collaborative work developing remote sensing approaches to monitor ecosystem condition and stress across large spatial scales, with a focus on forest ecosystems.

Understanding how ecosystems respond to human pressures and environmental change requires consistent, large-scale monitoring. This project focuses on collaborative work to develop remote sensing approaches for tracking ecosystem condition and stress across landscapes, with a focus on forest ecosystems.

Building on project-based analyses, we use satellite data and geospatial modelling to detect changes in vegetation, disturbance patterns, and environmental conditions over time. By integrating multiple data sources and analytical approaches, this work supports the identification of trends and emerging patterns across large spatial scales.

This work supports environmental monitoring, conservation planning, and decision-making by providing timely and spatially explicit information on ecosystem dynamics. Ongoing efforts are focused on advancing methods and scaling approaches to support broader applications across regions and ecosystems.

Developing a remote sensing and geospatial AI approach to detect early signals of forest ecosystem stress at global scales.

Current forest monitoring systems are largely reactive, detecting disturbance and degradation after it has already occurred. This project will focus on developing a remote sensing and geospatial AI approach to identify early warning signals of forest ecosystem stress at global scales.

By analyzing multi-sensor Earth observation data over time, this work will aim to detect subtle changes in forest condition driven by climate and human pressures. These approaches leverage advances in AI to identify emerging patterns that may indicate increased vulnerability before visible damage occurs.

This work will support global monitoring and decision-making by providing forward-looking information to help guide conservation and management efforts before ecosystems cross critical thresholds.