A1 Vertaisarvioitu alkuperäisartikkeli tieteellisessä lehdessä
Exploring tree growth allometry using two-date terrestrial laser scanning
Tekijät: Yrttimaa, T.; Luoma, V; Saarinen, N.; Kankare, V; Junttila, S.; Holopainen, M.; Hyyppä, J.; Vastaranta, M.
Kustantaja: ELSEVIER
Kustannuspaikka: AMSTERDAM
Julkaisuvuosi: 2022
Journal: Forest Ecology and Management
Tietokannassa oleva lehden nimi: FOREST ECOLOGY AND MANAGEMENT
Lehden akronyymi: FOREST ECOL MANAG
Artikkelin numero: 120303
Vuosikerta: 518
Sivujen määrä: 13
ISSN: 0378-1127
eISSN: 1872-7042
DOI: https://doi.org/10.1016/j.foreco.2022.120303
Tiivistelmä
Tree growth is a physio-ecological phenomena of high interest among researchers across disciplines. Observing changes in tree characteristics has conventionally required either repeated measurements of the characteristics of living trees, retrospective measurements of destructively sampled trees, or modelling. The use of close-range sensing techniques such as terrestrial laser scanning (TLS) has enabled non-destructive approaches to reconstruct the three-dimensional (3D) structure of trees and tree communities in space and time. This study aims at improving the understanding of tree allometry in general and interactions between tree growth and its neighbourhood in particular by using two-date point clouds. We investigated how variation in the increments in basal area at the breast height (Delta g(1.3)), basal area at height corresponding to 60% of tree height (Delta g(0)(6h)), and volume of the stem section below 50% of tree height (Delta v(05h) ) can be explained with TLS point cloud-based attributes characterizing the spatiotemporal structure of a tree crown and crown neighbourhood, entailing the competitive status of a tree. The analyses were based on 218 trees on 16 sample plots whose 3D characteristics were obtained at the beginning (2014, T1) and at the end of the monitoring period (2019, T2) from multi-scan TLS point clouds using automatic point cloud processing methods. The results of this study showed that, within certain tree communities, strong relationships (vertical bar r vertical bar > 0.8) were observed between increments in the stem dimensions and the attributes characterizing crown structure and competition. Most often, attributes characterizing the competitive status of a tree, and the crown structure at T1, were the most important attributes to explain variation in the increments of stem dimensions. Linear mixed-effect modelling showed that single attributes could explain up to 35-60% of the observed variation in Delta g(1.3), Delta g(0)(6h) and Delta V-0(5h), depending on the tree species. This tree-level evidence of the allometric relationship between stem growth and crown dynamics can further be used to justify landscape-level analyses based on airborne remote sensing technologies to monitor stem growth through the structure and development of crown structure. This study contributes to the existing knowledge by showing that laser-based close-range sensing is a feasible technology to provide 3D characterization of stem and crown structure, enabling one to quantify structural changes and the competitive status of trees for improved understanding of the underlying growth processes.
Tree growth is a physio-ecological phenomena of high interest among researchers across disciplines. Observing changes in tree characteristics has conventionally required either repeated measurements of the characteristics of living trees, retrospective measurements of destructively sampled trees, or modelling. The use of close-range sensing techniques such as terrestrial laser scanning (TLS) has enabled non-destructive approaches to reconstruct the three-dimensional (3D) structure of trees and tree communities in space and time. This study aims at improving the understanding of tree allometry in general and interactions between tree growth and its neighbourhood in particular by using two-date point clouds. We investigated how variation in the increments in basal area at the breast height (Delta g(1.3)), basal area at height corresponding to 60% of tree height (Delta g(0)(6h)), and volume of the stem section below 50% of tree height (Delta v(05h) ) can be explained with TLS point cloud-based attributes characterizing the spatiotemporal structure of a tree crown and crown neighbourhood, entailing the competitive status of a tree. The analyses were based on 218 trees on 16 sample plots whose 3D characteristics were obtained at the beginning (2014, T1) and at the end of the monitoring period (2019, T2) from multi-scan TLS point clouds using automatic point cloud processing methods. The results of this study showed that, within certain tree communities, strong relationships (vertical bar r vertical bar > 0.8) were observed between increments in the stem dimensions and the attributes characterizing crown structure and competition. Most often, attributes characterizing the competitive status of a tree, and the crown structure at T1, were the most important attributes to explain variation in the increments of stem dimensions. Linear mixed-effect modelling showed that single attributes could explain up to 35-60% of the observed variation in Delta g(1.3), Delta g(0)(6h) and Delta V-0(5h), depending on the tree species. This tree-level evidence of the allometric relationship between stem growth and crown dynamics can further be used to justify landscape-level analyses based on airborne remote sensing technologies to monitor stem growth through the structure and development of crown structure. This study contributes to the existing knowledge by showing that laser-based close-range sensing is a feasible technology to provide 3D characterization of stem and crown structure, enabling one to quantify structural changes and the competitive status of trees for improved understanding of the underlying growth processes.
Turun yliopiston Tutkimustietojärjestelmän portaali