You are viewing an archived copy of this website captured Mon Nov 05 11:33:39 AEDT 2012

All Content © CRC for Forestry 2007

The Log - Issue Three - Study recommends strategies to improve estimates of bark thickness in radiata pine

Radiata pine logs - bark thicknessIntroduction

Harvesters estimate bark thickness from diameter measurements using inbuilt models in their onboard computers.  Bark thickness is used to estimate underbark diameter which is critical to log volume and value calculations and optimisation.  Errors in bark thickness estimation can reduce harvesting returns through rejection of undersize logs and poor optimisation.

Studies have shown that radiata pine bark is proportionally thicker near the base than higher on the trunk.  However, most harvesters don’t account for this thicker bark when estimating radiata pine bark thickness.


A study of radiata pine bark thickness using a dataset supplied by ForestrySA revealed three factors that determine errors in radiata pine bark thickness estimation by harvesters:
•    site differences in bark thickness
•    use of incorrect coefficients for the harvester inbuilt models
•    poor fit of the harvester inbuilt models to radiata pine bark.

Analysis of the data showed a trend for thicker bark on poorer sites.  Two models were derived: one for better sites and one for poorer sites.  Applying the incorrect model resulted in errors of up to 5% in estimated log volume and value.

Two approaches were identified to account for the thicker basal bark of radiata pine: a model developed by Gordon (1983) or using separate models for the bottom and the top of the trunk.  Both these approaches gave consistently better predictions of bark thickness and log volume and value estimation than the current approach.


Several strategies are recommended to improve bark thickness estimation.  In the short term, harvesters need to be calibrated regularly using the steps in the Best Practice Harvester Callibration guide because bark thickness is estimated from diameter measurements.  The correct model is also required for poorer and better sites.

In the long term, harvesters will need to use either Gordon’s (1983) model or the approach of using separate models to describe the bark thickness on the upper and lower trunk to account for the thicker basal bark of radiata pine.

Future directions

Program Three will engage with STANFOR-D (the organisation that manages the standards for programs in harvesting computers) to have the more appropriate bark thickness equations included in the harvester computers  as an option. This study will be reported in more detail in an industry bulletin.  For further information contact Martin Strangard.


Gordon A (1983) Estimating bark thickness of Pinus radiata. New Zealand Journal of Forest Science 13(3): 340–53.


Martin Strandgard

Research Fellow
Tel: 03 9250 6872