Predicting wood and branch properties of Norway spruce as a part of a stand growth simulation system

The purpose of our study was to model how stem, branch and wood properties of Norway spruce (Picea abies (L.) Karst.) are linked to each other, and how they are distributed along the stem. The branchiness data were used to develop models for the crown ratio, the self-pruning ratio, i.e. height of the lowest dead whorl divided by the height of the crown base, number of living branches in a whorl, total number of branches in a whorl, diameter of the thickest branch in a whorl, diameters of smaller branches, and branch angle. The material on wood and fibre properties was used to develop models for wood density, early-latewood ratio, fibre length, fibre width, and cell wall thickness. Multilevel modelling approach was used to separate the variation of branch, wood and tracheid properties at region, stand and tree levels, as well as within a tree. Multivariate approach was used to simultaneously estimate the parameters of the equations. The independent variables were restricted to those collected in forest inventories or for forest management planning purposes. Such data sets are normally not very detailed. The models were connected to the empirical growth simulation system MOTTI and to a process-based growth model PipeQual. The dynamic growth model provides a description of tree growth down to the annual ring level. This structure will be used as a skeleton for calculating branch and wood properties. The growth model will update the stand and tree attributes from year to year and, consequently, change the distribution of the branch and wood properties within the stem. The variances of branch and wood properties and their covariances estimated during model development can be used to produce more realistic predictions compared to models which are fitted separately. When used in practice, the predictions of other models can be calibrated for a particular stand or tree if one dependent variable of the system is measured. The combined simulation system can be used for predicting the development of stem structure under environment controlled by different silvicultural management regimes. The results as such will also increase our understanding of the development of wood quality as a function of forest management, and the quality models are applicable independently of the growth models.