Individual-tree distance-dependent growth models for uneven-sized Norway spruce

Abstract

Abstract Individual tree selection (ITS) is one option to manage uneven-sized forest ecosystems. However, scientifically based field guidelines adapted to ITS and economic profitability are rare, often because there is a lack of suitable tree models to use in growth and treatment simulations. The objective of this study is to develop individual-tree distance-dependent growth models focusing on Norway spruce dominated uneven-sized stands. Three models of different complexity, but with the same structural basis, are presented, followed by some examples of growth patterns for the subject trees. The data include 1456 trees (307 sample trees) collected from five sites in southern Sweden. The basic model (S) depends on subject tree size as the predictor, the second model (SD) adds distance to competitors as a predictor, and the third model (SDC) adds crown ratio as a predictor to the structure. R2Adj increases with number of predictors from 0.48 to 0.58 to 0.62. The levels of RMSE improve accordingly from 5.02 cm2 year−1 (S) to 4.43 cm2 year−1 (SD) and 4.26 cm2 year−1 (SDC). The present calibration range and model structures primarily make the models suitable for management simulation of individual-tree selection of Norway spruce in southern Sweden. The format of the models allows for further extension with additional predictors and calibration data with greater coverage.