Point density and spatial resolution limit estimation of Leaf Area Index using discrete return LiDAR

View Researcher's Other Codes

Disclaimer: The provided code links for this paper are external links. Science Nest has no responsibility for the accuracy, legality or content of these links. Also, by downloading this code(s), you agree to comply with the terms of use as set out by the author(s) of the code(s).

Please contact us in case of a broken link from here

Authors Milodowski, David ; Swinfield, Thomas ; Coomes, David ; Williams, Mathew
Journal/Conference Name Methods in Ecology and Evolution
Paper Category , ,
Paper Abstract The total leaf area supported by a canopy, characterised by the Leaf Area Index (LAI), is one of the fundamental functional traits of forest ecosystems, underpinning biogeochemical cycles and surface energy balance, and mediating microclimate and habitat provision. Airborne LiDAR has emerged as a powerful method to measure LAI (or more specifically the Plant Area Index, PAI) and its variation across landscapes. However, survey characteristics such as point density exhibit high variance, even within the same survey dataset, due to factors such as flight line overlap and topography. Understanding how variations in point density influence estimation of PAI is therefore important in order to design and implement better surveys and correctly interpret associated uncertainties and biases. We derive analytical solutions describing the relationships between the point density of a discrete LiDAR survey and the maximum PAI that can be detected, and between the spatial resolution at which the point cloud is aggregated and the corresponding PAI estimates. We compare these predicted relationships against observations from a survey across a structurally diverse forested landscape in Malaysian Borneo. We show that the theoretical maximum observable PAI increases with point density in a predictable way following a logarithmic function. We also show that averaging the point cloud over a greater area imposes a negative bias in corresponding PAI estimates, and that this bias scales with canopy heterogeneity, but that it is possible to correct for this if the fine-scale spatial distribution of PAI is known, or can be estimated from representative regions of the forest canopy sampled at much higher resolution.
Date of publication 2018
Code Programming Language Python

Copyright Researcher 2022