Retrieval of narrow-range LAI of at multiple lidar point densities: Application on Eucalyptus grandis plantation

Leaf area index (LAI) is an important forest structural parameter that can be used to characterize various biophysical processes such as photosynthesis, evapotranspiration, and carbon flux. Accurate monitoring of LAI therefore is crucial for efficient management of managed and natural vegetation ecosystems. Remote sensing techniques have proved useful in the quantification and monitoring of LAI in different vegetation types; however, most of the focus has been on vegetation with relatively large LAI ranges. This study aimed to investigate the utility of airborne light detection and ranging (lidar) data to estimate narrow-range LAI (min = 0.71, max = 1.56, mean = 1.08 ± 0.18) of intensively-managed Eucalyptus grandis plantations. The secondary aim of the study was to assess the effect of lidar point density on LAI retrieval. Reference LAI was quantified in 15 m radius sample plots (n = 46) using hemispherical photography. Akaike Information Criterion (AIC) regression was used to build candidate models that estimate LAI from lidar-derived height and density metrics. The correlations were investigated at different point densities, including the original (>6 points/m2) and reduced density levels (0.25-5 points/m2). Candidate models returned adjusted coefficient of determination (adj. R2) ranging between 0.65-0.83 (RMSE 7.0-10.0% of observed mean) depending on the number of predicting metrics included in the models. A model that had two non-collinear metrics was selected as a compromise model (adj. R2 = 0.67; RMSE = 9.7%); this model was comparable to the best model, which had many collinear metrics. Estimation accuracies were similar for lidar densities of the original, 2-5 points/m2 and less accurate for 0.25-1 point/m2. These findings demonstrated the capability of lidar in estimating observed LAI with low range and variation. The study also suggests the efficacy of moderate lidar point densities acquired at relatively low-cost surveys in attaining acceptable LAI estimation accuracy.