Using TLS-measured Tree Attributes to Estimate the Above Ground Biomass of Individual Small Black Spruce Trees in Canadian Boreal Forest Peatlands

• Author / Creator

  Wagers, Steven M.

• Earth’s changing climate poses a number of potential issues for people around the globe. It has become increasingly important for researchers to have the ability to collect good data and build accurate, robust climate models that can help influence the direction of policy makers to mitigate the effects of climate change on as many people as possible. One important measurement pertaining to climate modelling is above ground biomass (AGB), for its ability to describe the amount of carbon stored within living vegetation, such as trees. Measuring AGB directly requires that trees be destructively sampled and weighed. An alternative to this method is to use allometric equations that are based on other, more easily measurable tree attributes as a way of estimating AGB. Advances in technology and the emergence of laser scanning technology has allowed for fast, highly detailed tree measurements. In the first chapter, this thesis explores the importance of developing accurate AGB models to measure the amount of carbon stored in forest ecosystems, and the ways laser scanning technology has changed the way researchers are able to measure the tree attributes needed for these equations. The research questions and hypotheses of this thesis are then posed based on the emergent uses of these technologies, and the needs for biomass mapping.
  In the second chapter, Terrestrial Laser Scanning (TLS) was used to develop point clouds of plots of small black spruce (Picea mariana L.) trees in the Taiga plains ecozone of the Northwest Territories. Tree attributes were measured to build allometric models to estimate individual tree AGB. The measurements used as predictor variables in the allometric equations created in this thesis were crown area, crown diameter, height, individual tree volume from a quantitative structure model (QSM), minimum bounding box volume of individual tree point clouds, diameter at breast height (DBH), and the products of crown area and height, crown diameter and height, and DBH and height. Multiple forms of these equations were created using both ordinary least squares (OLS) and weighted least squares (WLS) regressions, with either tree height or ground-measured biomass determining how much weight each tree received in the models. Model predictions were cross validated and ranked by average tree error, average RMSE and average adjusted R2. The best model was then tested against other established models to determine its viability. This thesis uses TLS point clouds to explore the applicability of tree attributes that can also be easily measured from airborne platforms such as airborne- and UAV laser scanners (ALS and ULS respectively) and provides the allometric equations that can be used for a common species found in a typical peatland environment in the Taiga Plains ecozone. A preliminary experiment to assess if the models using crown size and height could have the potential to be used with point clouds created by airborne methods was conducted. To simulate the decreasing point densities that are generally seen in ALS and ULS point clouds, rasters of varying cell size looking down on the trees were used to measure crown attributes such as crown area and crown diameter. A sensitivity analysis was then done to show how the AGB estimates given by equations provided by the best models using these tree attributes were affected as the raster cell size increased.
  The third chapter of this thesis examines the significance of the models created in Chapter 2, and comments on the implications of the results obtained. Potential future studies, including the possibility of flying scanning missions and using the AGB models from this thesis are also discussed.

• Subjects / Keywords

  • Terrestrial Laser Scanning
  • Biomass
  • Boreal Forest
  • Black Spruce

• Graduation date

  Fall 2021

• Type of Item

  Thesis

• Degree

  Master of Science

• DOI

  https://doi.org/10.7939/r3-ysdy-pm94

• License

  This thesis is made available by the University of Alberta Libraries with permission of the copyright owner solely for non-commercial purposes. This thesis, or any portion thereof, may not otherwise be copied or reproduced without the written consent of the copyright owner, except to the extent permitted by Canadian copyright law.