Assessing the regional-scale distribution of height growth of Cryptomeria japonica stands using airborne LiDAR, forest GIS database and machine learning

• Published in: Forest ecology and management Vol. 506; p. 119953
• Main Authors: Nakao, Katsuhiro, Kabeya, Daisuke, Awaya, Yoshio, Yamasaki, Shin, Tsuyama, Ikutaro, Yamagawa, Hiromi, Miyamoto, Kazuki, Araki, Masatake G.
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 15-02-2022
• Subjects: Airborne LiDAR; Cryptomeria japonica; Height growth map; Site productivity; SVF; IPCC; LiDAR; TWI; GPS; TMC; Site productivity; GIS; PRS; SRIA; Airborne LiDAR; WI; Cryptomeria japonica; PRW; RF; SI; Height growth map
• ISSN: 0378-1127; 1872-7042
• DOI: 10.1016/j.foreco.2021.119953

•Developed method for quantitative analysis of tree height growth at regional scale.•Assessed tree height growth patterns of Japanese cedar in two regions of Japan.•Forest stand age proved to be an important factor influencing tree height growth.•Environmental factors were also important, with their importance differing by region.•Growth patterns of the species differed among sites within a region.
The objective of this study was to develop a method for quantitative analysis of height growth in planted Japanese cedar (Cryptomeria japonica) stands at a regional scale. In two model regions (Gujo, Gifu and Kami, Kochi), overall tree height data were measured using airborne LiDAR, and target stand data were extracted by overlaying geographic information from a forest GIS database. A statistical model was used to quantitatively assess tree height data as a response variable, and forest age, climate and topographic factors as explanatory variables. Although there were errors in both regions due to data characteristics and unaccounted factors, tree height was estimated with high accuracy after bias correction (Gujo: rho = 0.98, Kami: rho = 0.90). The importance of climatic and topographic factors differed significantly between regions. In Gujo, which experiences relatively cool conditions, stand age (20.9%) and Warmth index (10.6%) were important determinants of tree height growth, whereas in Kami, with warm and rainy conditions, aspect (23.8%) and stand age (16.6%) were more important. The model developed in this study was used to predict the potential tree height of Japanese cedar stands aged 20–100 years. The potential growth pattern in Gujo was characterised by high growth in the warmer eastern and southern regions, but low growth in the higher-elevation and colder northern and western regions. In Kami, the pattern was more topographically complex, with lower height growth on southwestern slopes. Evaluating the potential growth at a site and predicting its future productivity will not only support suitable forestry management, but also allow sustainable and effective use of forest resources.