Comparison of Bruce's formula and other Methods for Log Volume Estimation

A database consisting of measurements of 167 logs collected from the sample trees of E. hybrid and A. nilotica from Gujarat State was used to compare the performance of Bruce's formula with that of the Huber, Smalian, Newton, Sorenson, Subneiloid and two-end conic methods for estimating volumes of butt logs. Very low bias was (less than 1%) observed for all the methods, which waa indicative of the fact that the predictions of all the methods were bias free. The tolerance interval for Bruce's formula was small for short length logs while it was more for bigger logs compared to Huber, Smalian and Newton models. The MSE was generally lowest in case of Bruce formula for all size of logs thus Bruce formula performed better in comparison to other models as far as composite effect of accuracy and precision is concerned. Results of the correlation analysis indicate that there was a significant relationship between large end diameter and per cent error in estimating log volume. The error increasing large end diameter for all the formulas.