Estimation of relative illuminance within forests using small-footprint airborne LiDAR

To evaluate canopy metrics at landscape or regional levels, remote sensing techniques have usually been employed. In particular, ratio metrics such as laser penetration index (LPI: %) calculated as the percentage of ground returns of laser pulses within the forest from light detection and ranging (LiDAR) data have proved to be closely related to canopy metrics obtained in the field. In our previous study, we successfully proposed a new methodology that automatically and easily separated the laser pulses into those of canopy and below-canopy returns. In this study, we assessed the effectiveness of LPI that was calculated using our methodology to estimate relative illuminance (RI: %) and also evaluated the optimal scale to calculate LPI for estimating RI. We found that LPI was closely related to RI (P < 0.01), and might not be affected by the increase of pulse density by overlapping surveys. We further found that a radius of less than 5.0 m in the cylindrical plot to calculate LPI might not be suitable for estimating canopy metrics, especially for LiDAR data with relatively low pulse density.