Tropical tree size–frequency distributions from airborne lidar

Abstract

In tropical rainforests, tree size and number density are influenced by disturbance history, soil, topography, climate, and biological factors that are difficult to predict without detailed and widespread forest inventory data. Here, we quantify tree size-frequency distributions over an old-growth wet tropical forest at the La Selva Biological Station in Costa Rica by using an individual tree crown (ITC) algorithm on airborne lidar measurements. The ITC provided tree height, crown area, the number of trees >10 m height and, predicted tree diameter, and aboveground biomass from field allometry. The number density showed strong agreement with field observations at the plot-(97.4%; 3% bias) and tree-height-classes level (97.4%; 3% bias). The lidar trees size spectra of tree diameter and height closely follow the distributions measured on the ground but showed less agreement with crown area observations. The model to convert lidar-derived tree height and crown area to tree diameter produced unbiased (0.8%) estimates of plot-level basal area and with low uncertainty (6%). Predictions on basal area for tree height classes were also unbiased (1.3%) but with larger uncertainties (22%). The biomass estimates had no significant bias at the plot-and tree-height-classes level (À5.2% and 2.1%). Our ITC method provides a powerful tool for tree-to landscape-level tropical forest inventory and biomass estimation by overcoming the limitations of lidar area-based approaches that require local calibration using a large number of inventory plots.

Deriving Tree Size Distributions of Tropical Forests from Lidar

“…Our approach complements recent methods [23][24][25][26][27][28][29][30][31][32][33][34][35][36] as it (i) integrates already available information on tree geometry (captured in the leaf-tree matrix) and (ii) estimates the full range of size classes in tree size distributions. Tree size distributions can be used to derive different forest attributes (e.g., tree density or forest basal area) and allow us to estimate a forest's successional state or disturbances [50].…”

“…To improve the allometries, crown size relations can also be derived by tree crown detection algorithms based on lidar point clouds [32,33,52]. For example, Ferraz et al [33] reported tree and crown allometries based on detected single trees and their respective uncertainties. This approach requires high-resolution lidar point clouds, particularly for correctly detecting small-and mid-sized trees (with small crowns) in the understory where the lidar point density is normally lower.…”

“…Recent studies have followed the approaches of tree segmentation [32][33][34][35][36] by detecting single tree crowns from airborne lidar to estimate tree size attributes and frequencies. Those techniques interpret lidar point clouds and allow the detection of tree geometry relations (in particular for large trees).…”

A benchmark dataset for individual tree crown delineation in co-registered airborne RGB, LiDAR and hyperspectral imagery from the National Ecological Observation Network

“…The majority of crown delineation papers assess proposed methods using field-collected stem data in which a single GPS point, representing the position of the basal stem, is used to indicate the position of each tree [15]. There is high confidence that this type of evaluation data represents individual trees, because tree stems are often easy to identify in the field.…”

Impacts of Degradation on Water, Energy, and Carbon Cycling of the Amazon Tropical Forests

“…In addition, the increased availability of terrestrial laser scanning (TLS) and high-resolution, low-altitude unmanned aerial vehicle lidar could substantially increase the data availability and thus improve the overall quality of allometric equations and constrain the relative contribution of woody tissues to the total plant area (Calders et al, 2015;Stovall et al, 2018). Alternatively, techniques that extract individual tree crowns from lidar point clouds readily provide highly accurate local stem density and local size-frequency distributions (e.g., tree height or crown size; Ferraz et al, 2016Ferraz et al, , 2020. These distributions can be used to attribute DBH to individuals and generate initial conditions akin to forest inventory to the ED-2.2 model, and data-model fusion techniques that leverage the growing availability of data could reduce uncertainties on many model parameters, including allometry (F. J.…”