Terrestrial laser scan metrics predict surface vegetation biomass and consumption in a frequently burned southeastern U.S. ecosystem

Abstract

Fire-prone landscapes found throughout the world are increasingly managed with prescribed fire for a variety of objectives. These frequent low-intensity fires directly impact lower forest strata, and thus estimating surface fuels or understory vegetation is essential for planning, evaluating, and monitoring management strategies and studying fire behavior and effects. Traditional fuel estimation methods have applications for stand-level and canopy fuel loading; however, local-scale understory biomass remains challenging because of complex within-stand heterogeneity and fast recovery post-fire. Previous studies have demonstrated how single location terrestrial laser scanning (TLS) can be used to estimate plot-level vegetation characteristics and impacts from prescribed fire. To build upon this methodology, co-located single TLS scans and physical biomass measurements were used to generate linear models for predicting understory vegetation and fuel biomass as well as consumption by fire in a southeastern U.S. pineland. A variable selection method was used to select the six most important TLS-derived structural metrics for each linear model, where model fit ranged in R2 from 0.61 to 0.74. This study highlights a prospect for efficiently estimating vegetation and fuel characteristics relevant to prescribed burning via the integration of a single-scan TLS method adaptable by managers and relevant for coupled fire-atmosphere models.