Assessing the utility of LiDAR to differentiate among vegetation structural classes

Abstract

Representations of vegetation structure are critical for effective forest ecosystem management. Structure is conventionally characterized using aerial photographs and field measurements; however, such methods are time-consuming and subjective, yielding results that cannot be easily updated and lack the detail required for many management initiatives. In contrast, light detection and ranging data provide highly accurate and detailed height, cover and canopy structure estimates, offering an unparalleled information source for improving conventional methods. Although numerous metrics can be derived from light detection and ranging, three suites common to the literature include height percentiles, canopy height descriptors and canopy volume profiles. This study assessed these three metric types for differentiating among vegetation structural classes in the Southern Gulf Islands, Sidney, BC, Canada. Results indicate all metrics could significantly differentiate (i.e. p ≤ 0.01) between structural classes, but that the number of and types of metrics capable of differentiation decreased with increased structural age and complexity.