Individual tree models for the crown biomass distribution of Scots pine, Norway spruce and birch in Finland

Abstract An increase in demand for wood energy is leading to more intensive harvesting of timber, including crown biomasses. In thinning operations, the harvesting method (full tree, part tree, and delimbed stem) and the minimum top diameter for industrial wood strongly affect the amount and quality of harvested biomass. When planning harvest operations, estimates should be based on typically available forest inventory data. In Scandinavia, the individual tree approach, with taper curves, is commonly used for the prediction of forest growth and timber assortments. Individual tree stem volume and biomass equations are commonly used to predict the total amount of crown biomass of a tree. To be able to predict the harvested biomass in integrated industrial and energy wood harvesting, the vertical distribution of crown biomass also needs to be known. This study presents individual tree models for the prediction of vertical distribution of living crown, needle and dead branch biomasses for Scots pine, Norway spruce and pubescent and silver birch. Mixed linear regression models were produced for the estimation of crown length and the heights to the lowest and highest dead branch. Tree height was the main predictor in the models, but predictors describing the competitive status of the tree and overall competition in the stand were also used. The Chapman–Richards model was used to estimate the distribution of the crown biomasses between the lower and upper limits of the crown. Data from different geographical regions and with different stand ages were pooled in order to obtain generally applicable models. The relative root mean square error of the estimate in the crown length models was 23.3% for pine, 19.7% for birch and 16.5% for spruce. In the models for the relative cumulative living crown masses the error was 11.7% for pine, 16.2% for birch and 14.3% for spruce. In the models for the relative needle masses the error was 15.1% for pine and 18.6% for spruce. Together with the biomass models for total living crown, whole crown, needles and dead branches, the models presented in this paper can be used for the estimation of the harvested volume and energy yield (MWh) using different harvesting methods and bucking options. The models can be used for the planning of harvesting operations, for the selection of feasible harvesting methods, and for the estimation of nutrient removals of different harvesting practises.