Improved models of harvest-induced bark damage

Abstract

Key messageWe provided a precise quantitative analysis of the factors at the origin of bark damage during harvesting operations and developed a model able to predict them accurately. The major factors were the distance of trees to skid trails, the intensity of removals, the harvesting system as well as the interactions between the distance of trees to skid trails with harvesting systems, the average skidding distance, the tree species and tree height.• Context During timber harvesting, trees in the remaining stand may suffer bark damage resulting from tree-felling or log manipulation. Although a multitude of case studies and empirical observations provide qualitative and quantitative information with respect to the potential causal factors, the basic quantitative relationship between major factors of influence and the resulting degree of bark damage remains largely unclear.• Aims The objective was to provide a precise quantitative analysis of impact factors explaining the occurrence of bark damage during harvesting operations.• Methods Three different modelling approaches were tested: boosted regression tree (BRT), a generalised linear mixed effects model (GLMM) and Bayesian Markov chain Monte Carlo generalised linear mixed models (MCMCglmm).• Results The major factors with a significant impact on the occurrence of bark damage were the distance of trees to skid trails, the intensity of removals, the harvesting system and the interaction term between the distance of trees to skid trails with harvesting systems, average skidding distance, tree species and tree height.• Conclusion The final model includes the relevant major factors impacting on the infliction of bark damage during practical harvesting operations. Furthermore, it discriminates well with respect to the occurrence of bark damage, and it provides managers with a rational and conclusive tool for optimising harvesting operations.

Effect of Day or Night and Cumulative Shift Time on the Frequency of Tree Damage during CTL Harvesting in Various Stand Conditions

“…However, similar damage frequency levels were reported after the reduction of the distance between skid roads from 20 m to 10 m [34]. Furthermore, a clear trend towards the increasing probability of newly inflicted bark damage when the distance exceeded 20 m was observed [35].…”

“…Other factors that affect the probability of tree damage include the strip road's configuration [29], the amount and extent of road curvature [30], and the distance between strip roads [31][32][33][34]. Increasing the distances between strip roads leads to a higher probability of tree damage [35], as does the length of tree assortments, with longer processed logs causing more damage [7,23,24,[35][36][37][38].…”

Machinery-Induced Damage to Soil and Remaining Forest Stands—Case Study from Slovakia

“…Felling intensity is one of significant factors, which affects the damage intensity of the remaining stand. Some authors state that the percentage of stand damage increases with the increasing harvesting intensity [25][26][27]. The number of felled trees was determined by counting all fresh stumps.…”

High-resolution mapping of forest vulnerability to wind for disturbance-aware forestry

“…Machine learning approaches have also been successfully applied to wind disturbance modeling (see Hanewinkel et al 2004 for an early example) and recently especially tree-based ensemble models, such as random forests (RF) and boosted regression trees (BRT), have been popular and often shown to perform well in predicting wind damage (see Schindler et al, 2016;Kabir et al, 2018;Albrecht et al, 2019;Hart et al, 2019 for examples using RF and Díaz-Yáñez et al 2019 for BRT). While machine learning methods and additive models are able to more flexibly fit the data and account for non-linearities, GLMs have strengths in their straightforward interpretability and the robustness of predictions (Nakou et al, 2016;Albrecht et al, 2019).…”