Wood quality predicting systems as a tool for differentiation and sorting of wood raw material

The project will emphasize the description of bending strength, modulus of elasticity when bending and shrinkage and swelling in radial, tangential and longitudinal direction. These properties are mainly dependent on density, fibre angle, fibre length and micro fibril angle. The relationship of these will be investigated in the project. The variation of the properties will be related to the differences within trees, between trees in a stand and between stands. Tree level and stand level variables will be used to model the variation in wood properties. The relationship between small clear wood samples and commercial timber dimensions will be investigated. In the project "Norwegian timber as raw material, added value and industrial possibilities", a databas e including a large number of wood and fibre properties, and stand and single tree variables, has been created. This data will be used in the proposed project to reach the main objective. In the mathematical modelling, statistical analyses such as varianc e component analysis and random coefficient models will be used. Simulation models will be used as a tool for predicting variation in wood properties. Wood property maps and profiles will be an educational way of presenting variation within trees and betw een trees. Modelling wood properties will help us understand which mechanisms that determine wood quality parameters on both single tree and stand level. If the models for wood quality parameters prove to have an acceptable level of prediction, they could be used to guide the wood raw material with the desired properties to the most suitable finished wood product. This would perhaps result in a finished product with both the desired level of quality and less variation in quality.