External knot size and frequency in black spruce trees from an initial spacing trial in Thunder Bay, Ontario

Abstract

As growing space available for a given tree increases, crown size increases and branch size (and thus knot size) is generally greater. Increased tree spacing may also result in a higher knot frequency. Using a combination of nonlinear, multilevel mixed effects and generalized nonlinear modeling techniques, a series of equations were developed to predict size and number of knots with respect to vertical location in black spruce (Picea mariana [Mill.] BSP) trees from one of the oldest initial spacing trial in Thunder Bay, Ontario. The models developed in this paper focus only on live whorls and are intended to be linked with a growth and yield model. The maximum knot size model accounted for 74% of the total variation and had a root MSE of 2.03. Random effects terms accounted for an additional 3% of the total variation. The relative knot size model accounted for 32% of the total variation and no significant random effects were found. The knot frequency model accounted for 45% of the total variation and the root MSE was 2.11 and no significant autocorrelation or random effects were observed. The results of this study indicated that black spruce knot properties were relatively insensitive to tree spacing given that they were largely accounted for by bole and crown size covariates. Key words: maximum knot size, relative knot size, number of knots per whorl, plantation black spruce, nonlinear regression, nonlinear mixed effects