Annual ring density for lodgepole pine as derived from models for earlywood density, latewood density and latewood proportion

Abstract

Model-based prediction of annual ring density (RD) is necessary to manage forests for wood quality objectives. However, annual RD in lodgepole pine (Pinus contorta Dougl. ex Loud.) exhibits a high degree of variability making it a challenge to model. We compared two methods of predicting annual RD including (1) a ring component approach and (2) a direct approach. The former approach uses model-based estimates of earlywood density (EWD), latewood density (LWD) and latewood proportion (LWP) to calculate annual RD. The latter approach uses a single model with annual RD as the dependent variable. The two approaches were tested using a dataset which included sites on the western and eastern slopes of the Rocky Mountains, within the provinces of British Columbia and Alberta, Canada. The best models for EWD, LWD and LWP included ring number and ring width, while site-specific parameters indicated that sites on the western slopes differed from those on the eastern slopes. Component-based estimates of annual RD using only fixed effects explained 25 per cent of the variability, increasing to 63 per cent with random effects. The best model for a direct estimate of annual RD explained only 5 per cent of the variability using fixed effects, increasing to 55 per cent with random effects.