[Knot attributes of Korean pine plantation based on mixed effect model.]

Area of Science:

• Forestry Science
• Statistical Modeling

Background:

• Korean pine (Pinus koraiensis) is a key commercial timber species in Northeast China.
• Understanding knot attributes is crucial for improving timber quality.

Purpose of the Study:

• To establish mixed-effect models for predicting Korean pine knot attributes.
• To compare the performance of mixed-effect models against fixed-effect models.
• To identify optimal models for knot diameter, sound knot length, year of death, and knot angle.

Main Methods:

• Utilized 1534 knot data points from 60 Korean pine trees.
• Employed SAS software's NLMIXED and GLIMMIX procedures for mixed-effect model development.
• Evaluated model accuracy using Akaike information criterion (AIC), Bayesian information criterion (BIC), -2Log likelihood (-2LL), and likelihood ratio test (LRT).

Main Results:

• All mixed-effect models incorporating tree effects outperformed fixed-effect models.
• Specific random parameter combinations yielded optimal models for knot diameter and sound knot length.
• Optimal generalized linear mixed models were identified for year of knot death and knot angle, incorporating relevant variables.

Conclusions:

• Mixed-effect models are more effective than fixed-effect models for describing knot attributes in Korean pine.
• Integrating these models with pruning strategies can significantly enhance timber quality.
• This research provides valuable insights for managing Korean pine plantations.