Catalog Home Page

Monitoring and modelling forest and woodland declines under a drying climate, from individual tree canopies to landscapes across the southwest of Western Australia

Evans, Bradley (2014) Monitoring and modelling forest and woodland declines under a drying climate, from individual tree canopies to landscapes across the southwest of Western Australia. PhD thesis, Murdoch University.

[img]
Preview
PDF - Front Pages
Download (146kB)
[img]
Preview
PDF - Whole Thesis
Download (2MB)

Abstract

Severe reduction in precipitation over the last three decades has resulted in forest and woodland decline and mortality across the southwest of Western Australia (SWWA). This precipitation-reduction driven drought is shown to have come with significant increases in diurnal thermal variability. These bioclimatic changes are shown to correspond to declining forest and woodland condition and in the case of the Northern Jarrah Forest, the mortality of the endemic keystone species Eucalyptus marginata (jarrah) and Corymbia calophylla (marri). Quantifying the bioclimatic link between these localized events is a key to understanding the dynamic drivers of change and to the development of ongoing adaptive strategies to reduce the impact and severity of forest and woodland declines. A novel measure of crown health, named the total crown health index (TCHI), is proposed and validated by combining the use of remotely sensed airborne digital multispectral imagery (DMSI) and in-situ assessments of crown condition. The approach provides a means to quantify changes in individual tree crown condition through space and time. Landscape scale remote sensing and climatologies are used to develop a methodology for mapping forest mortality and quantifying the bioclimatic envelopes of seven eucalypt dominated forest and woodland zones across the SWWA. Reported decline events across SWWA are investigated and together, this cross-scale range of methodologies, constituting a framework, can be applied at a range of scales from an individual tree up to landscapes. The extent of forest and woodland declines are spatially aggregated using remotely sensed imagery at 0.5m, 10m, 250m, 500m and climatologies of around 5000m. A set of bioclimatic thresholds are determined for each of the seven forest and woodland zones across SWWA. These thresholds could have been used to predict the declines across SWWA up to 12 months in advance. I conclude with a set of recommendations suitable for land managers and policy makers that suggests how to apply these methods and tools for the ongoing monitoring and modelling of forested and woodland areas.

Publication Type: Thesis (PhD)
Murdoch Affiliation: School of Veterinary and Life Sciences
Supervisor: Lyons, Tom
URI: http://researchrepository.murdoch.edu.au/id/eprint/20553
Item Control Page Item Control Page

Downloads

Downloads per month over past year