Estimating Coarse Woody Debris Volume Using Image Analysis and Multispectral LiDAR

Abstract

Coarse woody debris (CWD, parts of dead trees) is an important factor in forest management, given its roles in promoting local biodiversity and unique microhabitats, as well as providing carbon storage and fire fuel. However, parties interested in monitoring CWD abundance lack accurate methods to measure CWD accurately and extensively. Here, we demonstrate a novel strategy for mapping CWD volume (m 3 ) across a 4300-hectare study area in the boreal forest of Alberta, Canada using optical imagery and an infra-canopy vegetation-index layer derived from multispectral aerial LiDAR. Our models predicted CWD volume with a coefficient of determination (R2) value of 0.62 compared to field data, and a root-mean square error (RMSE) of 0.224 m 3 /100 m 2 . Models using multispectral LiDAR data in addition to image-analysis data performed with up to 12% lower RMSE than models using exclusively image-analysis layers. Site managers and researchers requiring reliable and comprehensive maps of CWD volume may benefit from the presented workflow, which aims to streamline the process of CWD measurement. As multispectral LiDAR radiometric calibration routines are developed and standardized, we expect future studies to benefit increasingly more from such products for CWD detection underneath canopy cover. 3 of 28 quantify CWD volume based on field data and we believed that ML data would be beneficial to predictions on occluded areas. Variables we predicted would have a relationship with occurrence and quantities of CWD include the height of trees, active and passive vegetation indices, canopy cover, presence of visible CWD, shadows, water, and wetlands. We tested our models both in a calibration area, where the models were developed, and in a verification area 4 km from the calibration area, and compared the best models using ML variables with the best models without ML variables to assess the impact of infra-canopy information on volume estimation accuracy. We then generated a series of CWD-volume maps designed to illustrate the utility of our approach to aid forest restoration efforts and fire-hazard assessments.

The Forest Line Mapper: A Semi-Automated Tool for Mapping Linear Disturbances in Forests

“…Additionally, the FLM is capable of extracting and summarizing information from external raster images as attributes on forest-line segments. For example, the attributes in Figure 3 were summarized from a user-supplied external raster showing the spatial distribution of coarse-woody debris (CWD) across one of our areas of interest (see Queiroz et al [67] for details). CWD is an important element of linear-feature restoration in the boreal forest, providing microsites for regenerating seedlings and restricting the movement of humans and predators along lines.…”

“…The arbitrary option summarizes attributes along arbitrary (in this case, 40 m) segments. These three example maps show average coarse woody debris (CWD; fallen dead trees) volume on seismic lines and were derived from a CWD raster layer from a separate study(Queiroz et al, 2020).…”

Machine Learning Techniques for Fine Dead Fuel Load Estimation Using Multi-Source Remote Sensing Data

“…Skowronski et al [30] used aerial photography, airborne laser scanning, and cadastral datasets to test a reliable method for characterizing fuel loads at a local scale in New Jersey (USA). Queiroz et al [31] presented a methodology to map coarse woody debris in the boreal forest of Alberta (Canada) using optical imagery and an infra-canopy vegetation-index layer derived from LIDAR.…”

Lodgepole Pine Bole Wood Density and Decay Rate 1, 11, and 22 Years After Felling in Central Montana, United States

“…Sampling procedures for quantifying the abundance of DWM have been described in earlier work and include planar intersect (Warren and Olsen, 1964; van Wagner, 1968;Brown, 1971), area sampling (Bate et al, 2004), relascope sampling (Ståhl, 1998;Gove et al, 1999), photographic assessments (Fischer, 1981;Ottmar et al, 1998;Keane and Dickinson, 2007) and LiDAR (Marchi et al, 2018;Lopes Queiroz et al, 2020). Most methods measure volume initially, then biomass (mass per unit area) is calculated as the product of volume and wood density.…”