Estimation of Individual Tree Stem Biomass in an Uneven-Aged Structured Coniferous Forest Using Multispectral LiDAR Data

Abstract

Stem biomass is a fundamental component of the global carbon cycle that is essential for forest productivity estimation. Over the last few decades, Light Detection and Ranging (LiDAR) has proven to be a useful tool for accurate carbon stock and biomass estimation in various biomes. The aim of this study was to investigate the potential of multispectral LiDAR data for the reliable estimation of single-tree total and barkless stem biomass (TSB and BSB) in an uneven-aged structured forest with complex topography. Destructive and non-destructive field measurements were collected for a total of 67 dominant and co-dominant Abies borisii-regis trees located in a mountainous area in Greece. Subsequently, two allometric equations were constructed to enrich the reference data with non-destructively sampled trees. Five different regression algorithms were tested for single-tree BSB and TSB estimation using height (height percentiles and bicentiles, max and average height) and intensity (skewness, standard deviation and average intensity) LiDAR-derived metrics: Generalized Linear Models (GLMs), Gaussian Process (GP), Random Forest (RF), Support Vector Regression (SVR) and Extreme Gradient Boosting (XGBoost). The results showcased that the RF algorithm provided the best overall predictive performance in both BSB (i.e., RMSE = 175.76 kg and R2 = 0.78) and TSB (i.e., RMSE = 211.16 kg and R2 = 0.65) cases. Our work demonstrates that BSB can be estimated with moderate to high accuracy using all the tested algorithms, contrary to the TSB, where only three algorithms (RF, SVR and GP) can adequately provide accurate TSB predictions due to bark irregularities along the stems. Overall, the multispectral LiDAR data provide accurate stem biomass estimates, the general applicability of which should be further tested in different biomes and ecosystems.

Above-Ground Forest Biomass Estimation using Multispectral LiDAR Data in a Multilayered Coniferous Forest

“…Overall, the results demonstrate that both SB and TB can be accurately estimated using LiDAR data and the RF algorithm in a structured forest with trees of different ages. In the case of the SB, the results are in accordance with most published studies that have analysed LiDAR data in different ecosystems (e.g., Allouis, Durrieu, Vega & Couteron, 2013;Cao, Coops, Innes, Dai & She, 2014;Cao et al, 2019;Georgopoulos et al, 2021). Thus, we can consider that SB is well described by the LiDAR-derived height-related metrics.…”

“…If we compare the results with those of studies carried out in the same forest using single-tree detection methods, it can be observed that the EABA approach resulted in slightly worse biomass estimation accuracy (in both SB and TB) (Georgopoulos et al, 2021). This is attributable to the various co-registration and position errors that can easily occur between the ground and Airborne Laser Scanning (ALS) statistics, which can shift the plot sides in different directions.…”

“…The reference SB and TB were calculated using species-specific allometric equations developed by Georgopoulos et al (2021). More specifically, these equations were developed using ordinary least square regression from 32 destructively sampled trees and were based primarily on DBH (Table 2).…”

“…The latest LiDAR sensors have the ability to gather vertical distribution information on different physiological processes across multiple spectral channels (Wallace, Nichol & Woodhouse, 2012). The additional, complementary, information derived from multispectral LiDAR data has been widely used in studies for improved land-cover and biomass estimation (Dalponte, Ene, Gobakken, Naesset & Gianelle, 2018;Georgopoulos, Gitas, Stefanidou, Korhonen & Stavrakoudis, 2021;Hopkinson, Chasmer, Gynan, Mahoney & Sitar, 2016). According to these studies, the AGB is highly correlated with the mean intensity values of the multispectral LiDAR data for the various forest layers.…”

“…The former usually aims to extract height-and/or intensity-derived metrics for each individual tree (Latella, Sola & Camporeale, 2021). However, the single-tree approach requires high pulse densities and its effectiveness is closely related to the accuracy of tree detection and of the crown delineation (Georgopoulos et al, 2021). Biomass estimates at the plot level, on the other hand, have been widely used in different biomes (Kelley, Trofymow, Metsaranta, Filipescu & Bone, 2021;Naesset, 2004;Silva Carlos Alberto et al, 2018) and include the integration of height-derived and other LiDAR-related metrics to predict forest characteristics.…”