Livestock grazing impacts more land than any other use. Yet knowledge of grazing impacts on native fauna is scarce. This thesis takes a predictive approach to investigating the effects of livestock grazing on Australian woodland birds, employing some novel methodological approaches and experimental designs. These include methods of analysis to handle zero-inflated data and the application of Bayesian statistics to analyse predictions based on expert opinion. The experimental designs have enabled impacts of grazing to be separated from the frequently confounding effects of other disturbances, and to consider the effect of grazing on habitat condition in the context of different surrounding land uses. A distinguishing feature of many datasets is their tendency to contain a large proportion of zero values. It can be difficult to extract ecological relationships from these datasets if we do not consider how these zeros arose and how to model them. Recent developments in modelling zero-inflated data are tested with the aim of making such methods more accessible to mainstream ecology. Through practical examples, we demonstrate how not accounting for zero-inflation can reduce our ability to detect relationships in ecological data and at worst lead to incorrect inference. The impact of grazing on birds was first examined through the elicitation of a priori predictions from 20 Australian ecologists. This expert knowledge was then used to inform a statistical model using Bayesian methods. The addition of expert data through priors in our model strengthened results under at least one grazing level for all but one bird species examined. This study highlights that in fields where there is extensive expert knowledge, yet little published data, the use of expert information as priors for ecological models is a cost effective way of making more confident predictions about the effect of management on biodiversity. A second set of a priori predictions were formulated using a mechanistic approach. Habitat structure is a major determinant of bird species diversity and livestock grazing is one mechanism by which structure is altered. Using available information on the vegetation strata utilised by each species for foraging and the strata most affected by grazing, predictions of the impact of grazing on each bird species were formulated. We found that foraging height preference was a good predictor of speciesÂ’ susceptibility to grazing. This approach is a starting point for more complex predictive models, and avoids the circularity of post hoc interpretation of impact data. The confounding of grazing with tree clearing was addressed by examining the impact of pastoral management on birds in sub-tropical grassy eucalypt woodland in Southeast Queensland, where land management practices have made it possible to disentangle these effects. Changes in bird species indices were recorded across woodland and riparian habitats with and without trees across three levels of grazing, replicated over space and time. Tree removal had a dramatic influence on 78% of the bird fauna. 65% of species responded significantly to changes in grazing level and the abundance of 42% of species varied significantly with habitat, level of clearing and grazing. The impact of grazing on birds was most severe in riparian habitat. Finally, the extent to which landscape context and local habitat characteristics influence bird assemblages of riparian habitats in grazed landscapes is addressed. Over 80% of bird species responded significantly to changes in local riparian habitat characteristics regardless of context, while over 50% of species were significantly influenced by landscape context. The influence of landscape context increased as the surrounding landuse became more intensive. These results suggest that it is not enough to conserve riparian habitats alone but conservation and restoration plans must consider landscape context. The ability to predict which bird species will be most affected by grazing will facilitate the transformation of this industry into one that is both profitable and ecologically sustainable. Results from this thesis suggest that any level of commercial grazing is detrimental to some woodland birds. Habitats with high levels of grazing support a species-poor bird assemblage dominated by birds that are increasing nationally. However, provided trees are not cleared and landscape context is not intensively used, a rich and abundant bird fauna can coexist with moderate levels of grazing, including iconic woodland birds which are declining elsewhere in Australia.
| Identifer | oai:union.ndltd.org:ADTP/253649 |
| Creators | Martin, Tara Gentle |
| Source Sets | Australiasian Digital Theses Program |
| Detected Language | English |
Page generated in 0.002 seconds