Oil mallee plantings and arthropod biodiversity in the Western Australian wheatbelt : effects of host species, nutrition, and leaf chemistry

Abstract

Since European settlement, around 93% of the Western Australian wheatbelt has been cleared for agriculture, leading to a range of environmental problems, including erosion, salinity, and loss of biodiversity. Recently, oil mallees have been developed for use in the reduction and prevention of salinity, and in the production of oil, activated carbon, and electricity. While some work has been done on aspects of mallee ecology in order to maximize productivity, little is known about their usefulness as a source of biodiversity in natural and agricultural systems.This study concentrates on the canopy arthropod fauna of the mallees. While canopy research involving arthropods is common in tropical systems, there is limited information on temperate systems and still less relating to agro-forestry or conservation plantings in agricultural areas.Sampling was conducted in alleys of oil mallee vegetation and in remnant vegetation in the wheatbelt of Western Australia. Three mallee species: Eucalyptus polybractea, E. kochii subsp. borealis, and E. loxophleba subsp. lissophloia, and two native remnant species: Eucalyptus wandoo subsp. wandoo and E. astringens were used in the study. Trees were sampled for arthropods by canopy knockdown spraying in October 2005 and May 2006. Samples were sorted to the ordinal level in the laboratory. Coleoptera (beetle) specimens were identified to the species level. Leaf and soil samples were taken at each site and from each tree species in order to determine the levels of nutrition available to herbivorous arthropods. Leaves were also collected for terpenoid (essential oil) extraction and formylated phloroglucinol (sideroxylonal) analysis to determine the influence of leaf chemistry. Data were tested for homogeneity of variance and transformations were done where necessary. A range of statistical analyses including, analyses of variance, LSDs, coefficients of correlation, MDS ordinations, ANOSIM analysis, and the BEST procedure, were conducted on the data collected.It was determined that, for this study, chemical knockdown would be an ideal method for sampling such arthropods. A preliminary study examined the effect of repeated sampling of the same tree on canopy arthropod assemblages and found there was no effect of re-sampling on the ordinal level richness or total abundance of arthropods collected at the second sampling, six months later. As a consequence, we can be reasonably confident that the results in other sections of the thesis have not been confounded by the need to re-sample the same trees.The effect of oil mallee host species on canopy arthropod assemblages was examined. It was found that while there were minor differences in the presence or absence of some of the leaf blemishes recorded between species, with leaf folding being more prevalent on E. kochii and psyllids more common on E. loxophleba, there was no significant difference between the species in terms of ordinal richness or total abundance of arthropods. As there were very few differences observed between the three mallee species, it made it relatively simple to compare mallees generally with remnant vegetation. This was important, as mallees were treated as a single entity being compared to the two remnant species in other parts of the thesis.The ordinal richness and total abundance of canopy arthropods in two types of eucalypt vegetation, woodland eucalypts in remnant vegetation and mallee eucalypts in farm alley plantings was compared. Intuitively, we would expect native remnant vegetation to support a greater diversity of arthropods than any planted vegetation, simply by virtue of the native vegetation being in place for a longer period of time. The results of this study however, did not support this view. Arthropod richness and abundance were not significantly different between the tree species. Leaf blemishes and their associated sedentary arthropods also showed no significant differences between the species and overall very few differences between the mallee and remnant vegetation types were evident.The influence of major soil and leaf nutrients on arthropod assemblages, both in natural and planted eucalypt stands, were explored. It appears that arthropod abundance, in particular, is related to soil and leaf nutrient levels. There was, however, only limited evidence of increased arthropod ordinal richness in response to greater nutrient levels. Generally, high nutrient levels tended, instead, to reduce arthropod ordinal richness. Of the soil nutrients, phosphorus was the most influential, with high levels of phosphorus tending to relate to higher arthropod abundances. For leaf nutrients, phosphorus and nitrogen were important, with high levels of phosphorus being associated with lower ordinal richness, while high levels of leaf nitrogen were related to higher arthropod abundances. As high soil phosphorus and high leaf nitrogen tended to occur together, it is difficult to say whether one or the other is responsible for increases in arthropod abundance, though intuitively one would suggest that good soil nutrition led to enhanced plant quality, which in turn increased arthropod abundance.The role of leaf essential oils and other secondary plant compounds in determining arthropod abundance and ordinal richness was also examined. A number of secondary plant compounds were present in the host tree species tested, and there were wide variations in the levels of these compounds between the species tested. A range of relationships, including deterrent and attractant effects, were observed between arthropod assemblages and the various secondary plant compounds. Generally speaking, mallee species had high cineole and low pinene levels and remnant species had the opposite. Sideroxylonals showed no such pattern with vegetation type, being high in E. loxophleba and absent in the other mallee species. In terms of their influence on arthropods, the compounds varied in their effects. Pinene had a generally negative effect, while cineole had a generally positive one. Sideroxylonal, however, was more complicated in its effects as it had a negative effect on ordinal richness, but was positively correlated with Hemiptera numbers. This suggests that it is highly unlikely that any one of these compounds can explain the pattern of arthropod assemblages observed in isolation. It seems more probable that complex interactions between these chemicals cause changes in nutritional quality and palatability of foliage, influencing the feeding behaviour, development, distribution and abundance of herbivores, in turn affecting predator densities and feeding behaviour.The ways in which arthropod biodiversity may be influenced by the factors of host tree species, soil and plant nutrition and leaf chemistry were examined in more detail using the order Coleoptera as an example. Season of sampling was found to have an influence, with both beetle richness and abundance being higher at the first than the second sampling. Site, on the other hand, had very little influence. Of all the tree species, E. polybractea had both the highest beetle species richness, and high levels of beetle abundance. Strong similarities were apparent between beetle assemblages resident on the same tree species. No influence of soil or leaf nutrients on either beetle richness or abundance was observed. However, beetle abundance was found to have a negative relationship with leaf pinene. Cineole levels were lower in the remnant species (which tended to have similar assemblages), while pinene was high in E. loxophleba subsp. lissophloia and the remnant species. The other major leaf chemical examined, sideroxylonal had no significant impact on beetle richness or abundance. Statistical analysis selected cineole as the single factor best explaining the pattern of beetle assemblages observed, though this result should be treated with caution due to possible confounding of the results as a consequence of interaction between the factors.The broad aim of this research was to determine whether oil mallee plantings enhance arthropod biodiversity in agricultural landscapes. The results of this thesis show that oil mallees do support high levels of arthropod biodiversity. Overall, the mallees had a level of diversity, not dissimilar to that of high quality remnant vegetation. When planted in alleys across agricultural fields, they represent a significant change in the vegetative and architectural diversity of the landscape, and can have a positive influence on the environment by supporting beneficial arthropods and other native animals, reducing dryland salinity, and improving the aesthetics of the wheatbelt.Aside from their environmental credentials, oil mallees also provide the potential for farmers to make an income from something designed to benefit the environment. If oil mallee farming can be developed appropriately, it has the potential to benefit not only farmers and the environment, but the community of Western Australia as a whole.