Predicting growth in mixed-species tree plantations: An Australian Wet Tropics case study

Daniel Manson (2012). Predicting growth in mixed-species tree plantations: An Australian Wet Tropics case study PhD Thesis, School of Biological Sciences, The University of Queensland.

Attached Files (Some files may be inaccessible until you login with your UQ eSpace credentials)
Name Description MIMEType Size Downloads
s348594_phd_finalthesis.pdf Final Thesis application/pdf 2.67MB 19
Author Daniel Manson
Thesis Title Predicting growth in mixed-species tree plantations: An Australian Wet Tropics case study
School, Centre or Institute School of Biological Sciences
Institution The University of Queensland
Publication date 2012-02
Thesis type PhD Thesis
Supervisor Susanne Schmidt
Jim Hanan
Jerry Vanclay
Total pages 137
Total colour pages 8
Total black and white pages 129
Language eng
Subjects 06 Biological Sciences
Abstract/Summary Large-scale deforestation in the tropics and subtropics has led to a rapid decline in the ecosystem services formerly provided by natural forests. Reforestation occurs predominantly as monocultures of a small selection of mostly exotic tree species with commercial significance. Concerns over the sustainability of large-scale monocultures and their inability to restore ecosystem services formally provided by forests have raised suggestions that increasing species diversity in plantations would be beneficial in many situations. Mixed-species plantations consisting of native trees have potential to fulfil numerous ecological services and provide a sustainable source of forest products. The absence of operational-scale demonstrations of mixed-species systems and suitable methods for predicting growth and yield in mixtures currently limits the commercial adoption of these systems. The objective of the research presented in this thesis was to develop methods to predict the growth of mixed-species plantations at new sites through analysis of existing data from mixed-species trials. I used two process-based forest modelling systems, stand productivity model ‘3-PG’, and spatially explicit mixed-species competition model ‘SExI-FS’ to analyse data from 52 mixed-species plantings of native trees in the wet tropics of Australia. These models were chosen as most suitable for simulating the growth of component species at new sites and for predicting the interactions between species within stands. The plantations were established in the 1990’s to create new timber production systems composed of native Australian tree species as logging of native rainforests ceased with their World Heritage listing. Although several studies have investigated the plantations, a stringent analysis and detailed growth modelling had not been performed. I used a database of tree growth measurements and collated additional data to provide an empirical basis to develop parameterisations for the predictive process-based modelling approaches. The 3-PG forest model was calibrated to predict species’ growth potential at contrasting sites. Parameterisations were developed for the SExI-FS modelling system to predict the effect of inter-tree interactions on the growth of component species in mixtures. Together, these methods constitute a novel approach for advancing the results of plantation trials to future industrial-scale applications of mixed-species tree systems. I identified matrices of growth responses for 10 priority tree species to a range of environmental and soil variables. The Site Index (SI) of Eucalyptus and rainforest tree species exhibited unique responses to texture and drainage characteristics of the soil profile, with soil type being the strongest predictor of SI for rainforest species including Elaeocarpus grandis and Flindersia brayleyana. I parameterised the 3-PG forest model to predict SI of Eucalyptus pellita, Elaeocarpus grandis and Flindersia brayleyana and successfully calibrated the Fertility Rating parameter to model the relative suitability of soil types for growth of species. Final 3-PG parameterisations were effective in predicting 58, 55 and 71% of the variation in SI across sites for Elae. grandis, E. pellita and F. brayleyana respectively. Species were parameterised effectively in the SExI-FS modelling system using estimated SI models and shade tolerance variables obtained through parameter estimation techniques. Models successfully predicted growth in a range of mixed-species plantations, accounting for 54, 64, and 66% of the variation in growth reduction of individual trees (from maximum potential growth) in validation plantations for Elae. grandis, E. pellita and F. brayleyana respectively. Despite the complex dynamics of tree mixtures, the research presented here demonstrates that a systematic approach for simplifying representative variables of the processes determining the growth of trees in plantations allows reliable quantification of the growth of mixed-species plantations. Together, the modelling investigations constitute a logical framework that provides a basis for developing growth models of mixed-species tree plantations elsewhere. This study focuses on the above-ground interactions in tropical plantations and could easily be tested with mixtures in other regions. Investigation of the below-ground interactions between species and the long-term competition in mixed stands may further aid future model development and optimisation of tree species assemblages.
Keyword Mixed-species plantations
forest growth models
site index
carbon forestry
biodiversity conservation
Rainforest plantation
tree competition
Additional Notes Pages that should be printed in colour: 28, 31, 32, 36, 38, 93, 127, 129

Citation counts: Google Scholar Search Google Scholar
Created: Tue, 21 Feb 2012, 08:42:25 EST by Daniel Manson on behalf of Library - Information Access Service