Impact of Reference Data Sampling Density for Estimating Plot-Level Average Shrub Heights Using Terrestrial Laser Scanning Data

Terrestrial laser scanning (TLS) data can offer a means to estimate subcanopy fuel characteristics support site characterization, quantification of treatment or fire effects, and inform modeling. Using field TLS within the New Jersey Pinelands National Reserve (PNR), this study explores impact forest phenology density shrub height (i.e., bed depth) measurements on estimating average heights at plot-level using multiple linear regression metrics derived from ground-classified normalized point clouds. The results highlight importance sampling when these are used train empirical models characterize characteristics. We document larger prediction intervals (PIs), higher root mean square error (RMSE), lower R-squared with reduction in number randomly selected reference samples available each plot. At least 10 random collected situ were needed produce accurate precise predictions, while 20 ideal. Additionally, leaf-on generally provided more predictions than those calculated leaf-off plots landscape. This highlights design will be for extrapolation new sites plots.