As the dominant growth form of many Australian plant communities the continuing decline of Eucalyptus populations is of major concern. The severity of eucalypt dieback is greatest in rural areas where insect attack is most often implicated in the decline. Drought, salinity, chemicals and other agricultural practices are also believed to increase the susceptibility of trees to insect attack.

Eucalyptus rudis Endl. within the southwest of Western Australia has been identified as exhibiting extensive and progressive crown dieback and there is the need to understand the underlying causes. This is of crucial importance as the loss of E. rudis and fringing vegetation from southwest streamlines will increase nutrification and sedimentation of the water and result in a loss of flood control and habitat. The condition of E. rudis along the Preston River has generated substantial concern, with the leafminer Perthida sp thought to be responsible for canopy cover decline.

This project sought to identify variation in E. rudis canopy cover along the Preston River, how it may have changed with time and if there were any correlations with types of leaf damage, foliar quality, soil profile, soil nutrients and species richness. Canopy cover was found to increase significantly along the Preston River with distance upstream and was found to be significantly different between sites downstream of the Glen Mervyn Dam and those upstream. Leaf chew was the only type of leaf damage found to be significantly correlated with this change in canopy along the Preston River. Condensed tannins of foliage, the specific leaf weight, the proportion of total species that are weeds, and the mean number of psyllids (Creiis sp) on the leaves were significantly greater at sites below the dam than the sites above. Interspecific variations in foliar quality or resistance mechanisms, which may deter particular insect herbivores still need to be identified.

The limited examination of growth rings from trees at opposite ends of the study gradient indicates that tree growth has decreased over time. ¶¹³C measurements from rings representing the most recent 10-year period and a 10-year period 26-35 years ago indicate that water use efficiency has decreased at the severely defoliated sites. Along with anecdotal evidence for the defoliated sites, these results indicated that canopy cover has declined over the last decade.

The major distinguishing land use practice between less defoliated and severely defoliated sites was whether the sites were above or below the Glen Mervyn Dam, no other environmental variables (soil profile, nutrients) differed significantly between this division. However a multitude of other impacts (rainfall patterns, agricultural chemical, soil/water temperature, pathogens) would need to be investigated before any conclusions were reached.

The ability of E. rudis to undertake compensatory mechanisms in the glass house, with increasing total foliar N was confirmed. The ability of mature trees in the field to increase A_net would need to be confirmed, and how this may relate to increased soluble N and insect outbreaks should also be investigated.

This project has highlighted the complexity of interacting factors, which could contribute to the causes of the canopy cover decline along the study region of the Preston River, Indeed the original cause of the outbreak could not be determined within the scope of this project. From this project a number of recommendations have been formulated for further research which would contribute to an understanding of E. rudis dieback not just in the study region but address the widespread occurrence and severity of E. rudis dieback throughout southwest Western Australia.