Comparative Performances of Airborne LiDAR Height and Intensity Data for Leaf Area Index Estimation
Shezhou Luo,Jing M. Chen,Cheng Wang,Alemu Gonsamo,Xiaohuan Xi,Yi Lin,Mingjie Qian,Dailiang Peng,Sheng Nie,Haiming Qin
DOI: https://doi.org/10.1109/jstars.2017.2765890
IF: 4.715
2018-01-01
IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Abstract:Leaf area index (LAI) estimation based on remote sensing data has often relied on the use of spectral vegetation indices from optical data. However, it is difficult to accurately estimate LAI due to saturation of spectral signals. Light detection and ranging (LiDAR) systems have emerged as promising technologies for overcoming the saturation problem, and an increasing number of studies have been conducted on LAI estimation using LiDAR data. In this study, we compared the performance of LAI estimation using LiDAR height and intensity data, and explored the potential for estimating forest LAI using combined LiDAR height and intensity data. LAI estimation models were established using LiDAR height, intensity, and a combination of LiDAR height and intensity metrics based on a random forest regression algorithm. Our results show that the laser intercept index derived from LiDAR height or intensity data was the most important predictor for LAI. Field measurements of LAI at 64 sites were used to assess the power of various LiDAR metrics in predicting LAI. The results show that both LiDAR height and intensity metrics alone could reliably estimate forest LAI. However, compared to LiDAR intensity metrics [$R^{{\rm 2}}\,= \,0.610$ with root mean squared error (RMSE) of 0.664], LiDAR height metrics had a better predictive power ($R^{{\rm 2}}\,= \,0.765$ with RMSE of 0.562). Moreover, the combined LiDAR height and intensity metrics resulted in the highest LAI estimation accuracy ($R^{{\rm 2}}\,= \,0.809$ with RMSE of 0.501). Therefore, the combination of LiDAR height and intensity data has a great potential for improving the LAI estimation accuracy.