Files
Abstract
This thesis investigated wood and fiber quality from planted longleaf pine. Eight forest cutover and eight old agriculture field stands were sampled, with a total of 160 trees felled per site type. Disks were collected from each tree at multiple height levels. In chapter 1, “Models to predict whole-disk specific gravity and moisture content in planted longleaf pine”, non-linear mixed effects models were developed to predict the variation in wood and bark specific gravity (SG) with respect to relative height, age, and site type. Forest cutover sites had a higher whole-tree wood SG (0.504 vs 0.455) and bark SG (0.374 vs 0.347) than old agriculture field sites with the final wood and bark SG models explaining 50% and 37% of the variability, respectively. In chapter 2, “A computer vision approach to assess wood variability from whole-disk images of longleaf pine”, disks were surfaced and imaged. Estimates from images were in close agreement with reference measurements for wood volume (R2 > 0.99), bark volume (R2 = 0.96), outside bark diameter (R2 > 0.99) and inside bark diameter (R2 > 0.99). Additional measurements of disk shape and compression wood were estimated from the disk images.