In this thesis the development of simulation models for important wood quality properties of Norway spruce (Picea abies) is shown. The models were integrated in the established single-tree forest growth simulator SILVA in combination with an also newly developed grading and bucking routine which is sensitive to wood quality. A main aspect of this work was the explicit spatial modeling of wood properties to cover correlations of wood traits with spatial tree structure and three-dimensional competition in the stand which result from silvicultural treatment. For this reason three new model components were developed and integrated in the forest growth simulator:

1. The new model for dynamic branch growth simulation, called “AMOK” is capable of simulating dimension, position and status of branches depending on spatial variability of competition in the crown layer of stands. The calculation of competition with a fisheye-approach is done for different crown segments separately. This makes a plausible modeling of competition and growth reaction of crown and branches possible.
2. The second simulation model, called “RAM”, covers the spatial extent of red rot in the stem caused by the fungus Heterobasidion annosum. The model predicts the spatial extent of the decay (height, diameter, form) as well as the degree of decay. Additionally the effects of red rot on tree growth are modeled.
3. The model “SILVSORT” picks up the information on branchiness, knottiness and decay and combines it with original output parameters from the forest growth simulator, like stem volume, stem form or ring width. A roundwood grading is simulated according to the spatial distribution of the wood properties in each stem following standard grading rules. The next step is the calculation of an optimal bucking solution for every stem. The optimizing model uses a genetic algorithm to find the economically best bucking solution.

The application range of the forest growth simulator is clearly extended by the integration of models simulating the spatial effect of silvicultural treatment on wood quality. This could be shown by scenario simulation. With the new models wood quality can be used additionally to timber volume as a predictor for the economic evaluation of forest stands and forest management actions. It could be shown that the integration of wood quality and quality sensitive grading and bucking in a single-tree forest growth simulator provides a valuable instrument to forest science and practice for the development and comparison of management strategies – an essential element for a consumer orientated timber production in the scope of multifunctional forestry.