• Report

Optimising productivity of hardwood plantations: yield gap analysis for Eucalyptus globulus plantations in southern Australia

Go direct to:

Download report

Published Date

February 2024

Project number: VNC516-1920

This study aimed to improve the knowledge underpinning decisions on the location and management of hardwood plantations across southern Australia by (i) providing estimates of potential productivity (as wood yield), (ii) determining the main causes of differences between potential and actual yields (gaps), and (iii) determining the extent that management practices might reduce gaps. The study was based on (i) empirical analyses, (ii) a forest modelling and prediction framework suited to plot- and regional-scale applications, and (iii) case study demonstrations, focusing on age 10-years yields for planted Eucalyptus globulus.

Climate, and particularly water availability, was determined to be the primary limitation to yield across southern Australia. Variation in potential rooting depth as affected by soil properties also affected potential yield, including particularly through impact on survival on shallow soils. A soil-fertility effect, as land-use history, was evident for a subset of regions with average yield about 50 m3/ha less where there had not been a preceding Agriculture phase.

The APSIM Eucalyptus process-based growth model was used in the study because of the ability to include (i) complex soils (deep, with water, C and N), (ii) silvicultural flexibility (particularly the inclusion of N fertiliser, weeds, and slash management), (iii) science and software engineering credentials, and (iv) links to agricultural models and software support. Overall, the model simulations proved to be a useful way to examine the role of individual factors effecting growth and yield without the confounding effects of un-controlled factors that arose in some empirical analyses.

Industry training to use APSIM was provided, including through on-line material. It is envisaged that with on-going support industry will adopt this current modelling capability, and its future enhancements, either in-house or through consultants.


John McGrath, Thomas Baker, Philip Smethurst, Richard Harper and Neil Huth

Related Reports