Airborne lidar provides reliable estimates of canopy base height and canopy bulk density in southwestern ponderosa pine forests

Abstract

Accurate estimates of canopy base height (CBH) and canopy bulk density (CBD) are critical inputs for many fire modeling and simulation programs. Commonly used LANDFIRE estimates of CBH and CBD are only available at 30 m resolution and have relatively low accuracy, and traditional field-based estimates can lead to inaccurate spatial estimates by assuming averages across spatial extents. Discrete-return airborne light detection and ranging (lidar) has become increasingly common in forestry applications, with demonstrated success in measuring canopy metrics and forest structure across broad landscapes and varied forest types. However, few studies have investigated the potential of airborne lidar for estimating these metrics in southwestern forests. This study developed an approach for estimating CBH and CBD using airborne lidar data and evaluated the accuracy of estimates at 20 m resolution. We also evaluated the predicted accuracy between managed and unmanaged forest stands. We employed a quantile-based method to derive CBH and utilized the Fire and Fuels Extension of the Forest Vegetation Simulator to estimate CBD from a lidar-derived tree list. Our results indicate that airborne lidar produced more accurate estimates of both CBH (R2 = 0.4; RMSE = 1.25 m) and CBD (R2 = 0.76; RMSE = 0.021 kg−3) as compared to LANDFIRE and other regional remote-sensing based estimates. We also report that airborne lidar is more accurate at estimating CBH in unmanaged stands versus managed stands, but CBD estimates maintain similar accuracy regardless of management history. Compared to methods developed in other regions, our approach resulted in relatively low R2 values, but our RMSE values were found to be similar or slightly improved. Our approach for deriving both metrics are conceptually and computationally simple making them especially valuable for use by land managers and ecologists alike. This study provides clear evidence that airborne lidar can be used to derive broad-scale, fine-resolution fuel data layers, which are commonly required in fire modeling and forest management activities and decision-making.