Analysis of a feller-buncher productivity: a case study of whole-tree harvesting from Marmara region, Turkey

Abstract

In fully mechanized forest harvesting systems, tree felling activities are mostly performed by using harvesters or feller-bunchers. In some regions of Turkey, where terrain conditions and stand characteristics are suitable, fully mechanized harvesting systems have been recently practiced by some of the logging contractors as private forest industry demands for large amounts of forest products throughout the year. Thus, performances of these newly practiced harvesting systems should be carefully analyzed in order to implement productive and cost-effective mechanized harvesting systems. In this study, productivity of whole-tree harvesting using a feller-buncher was investigated based on stand parameters including tree height, DBH, and volume. The DBH of the felled trees were divided into four classes (i.e. very small: 16-19 cm, small: 20-23 cm, medium: 24-27 cm and large: 28-31 cm) to investigate the effects of various DBH class on the time consumption of cutting stage and productivity of the feller-buncher. To estimate productivity of feller-buncher in harvesting operation, multiple linear and polynomial regressions were also developed and discussed after the interpretation of diagnostic plots. The results indicated that the average productivity of the feller-buncher was 74.96 m3 /h which was closely related with tree height (r = 0.63), DBH (r = 0.67), and volume (r = 0.67). The average moving time was the most time-consuming stage (60%), followed by cutting (29%) and bunching stages (11%). It was found that DBH classes caused statistically significant (p < 0.05) effects on the time spent on cutting stage and productivity of the feller-buncher. The cutting time and productivity increased from very small to large diameter classes, while bunching time increased from very small to small diameter and then medium diameter to large diameter classes. Polynomial regression had a positive impact on the performance of the estimation model of manually field-measured data based on the error parameters.