Lanterna Rally

Author: Leena E. Vihermaa

Publisher:

Published: 2010

Total Pages:

ISBN-13:

In plantation grown Sitka spruce, timber density is an important quality concern. Currently Sitka spruce timber meets the requirements for C16 strength grading which is the minimum requirement for construction uses. However, the margin is not exceeded by much and a small reduction in density could lower the log grading. Therefore it is essential to understand how timber density is impacted by site factors and climate in order to predict the potential effects of climate change on timber quality in Sitka spruce in the future. This has important economic implications for UK forestry. To assess the influence of site factors and climate on growth and resulting timber properties in Sitka spruce, three experiments were established; a large scale wood quality survey (?The Benchmarking experiment?), a detailed inter-site density study (?The Level II experiment?) and a continuous growth monitoring experiment (Dendrometer experiment). In the?Benchmarking? experiment, increment cores were collected from 68 sites over a geographically wide area in Scotland and Northern England in order to quantify the effect of selected site factors on density. These samples were not suited for climate analysis at annual ring level and therefore disks that allowed reliable dating of the tree rings were collected from three Level II sites. The Level II samples were also used in an acoustic velocity study and an assessment of the within-tree patterns in density. Material from both Benchmarking and Level II sites was used for modelling radial density. In addition to this the Level II data functioned as a test dataset for the different density models that were developed on the Benchmarking data. Density and other tree ring variables were measured with an ITRAX X-ray densitometer and WinDENDRO software. Some acoustic measurements were also carried out with a purpose built ultrasonic scanner to allow calculation of radial variation in stiffness. A dendrometer experiment was established to monitor short term growth variation in Sitka spruce in real time. The aim was also to use the growth data to date density profiles and hence identify causes for the density differences. Data analysis was carried out in R mainly using linear and non-linear mixed effects models, the dendrochronology software package dplR and methods of time series analysis. It was found that the largest part of the variation in density was between trees within each site. Both density and stiffness were mostly influenced by the growth rate or by another variable describing the tree vigour. Fast growth decreased density both by increasing earlywood proportion as well as decreasing the density of both earlywood and latewood. Models for the radial behaviour of density and stiffness could be fitted to the data, but random tree effects remained large. This limits the applicability of these models to new sites as reparameterisation would be required, which requires data from time consuming density and acoustic velocity measurements. The possibility of modelling the model coefficients from easily measurable stand and tree variables was investigated but the results were not promising. The Dendrometer experiment indicated that annual growth in Sitka spruce was initiated in late May and terminated by mid September. Ring widths detected by the dendrometers, micro core measurements and X-ray density based tree ring analysis differed, which caused difficulties in dating the density profile. To decrease these problems in the future the microcore sample storage protocol was revised. The dendrometer data indicated that the growing season was divided into several sections between which the growth rate differed. Latewood density and maximum density had the strongest correlations with the climatic variables. At one site the correlations included temperature and rainfall in April whereas at the other precipitation throughout the growing season or during the May-August period were important. Temperature was correlated positively with density variables and rainfall negatively.