Rating experiments in forestry: How much agreement is there in tree marking?

Area of Science:

• Forestry science
• Ecological management
• Quantitative ecology

Background:

• Tree selection in forest management is complex and influenced by numerous factors.
• Previous literature suggested high agreement among foresters in tree selection, a notion this study investigates.
• Marteloscope experiments provide a controlled environment to study human decision-making in forest management.

Purpose of the Study:

• To quantify the level of agreement among foresters when applying low and crown thinning methods.
• To explore factors influencing agreement in tree selection.
• To examine relationships between agreement measures and forest structure characteristics.

Main Methods:

• Analysis of data from 36 marteloscope experiments across Britain.
• Application of multivariate statistics, specifically Fleiss' kappa, to assess inter-rater reliability.
• Comparison of agreement levels between low and crown thinning methods.

Main Results:

• General agreement in tree selection among raters was found to be slight to fair, significantly lower than in comparable medical studies.
• Agreement was substantially higher in low-thinning experiments compared to crown-thinning experiments.
• Forest structure characteristics did not explain the observed levels of agreement, but conformity increased with stem diameter complexity.

Conclusions:

• Current agreement levels in tree selection for British forestry are lower than often assumed, necessitating enhanced training, particularly for crown thinning.
• Crown thinning, a key method for Continuous Cover Forestry, requires further standardization and training due to lower rater agreement.
• While forest structure complexity correlates with conformity, it doesn't fully explain agreement variability, highlighting the importance of human factors in forest management decisions.