Management of wild canids and trophic cascades: How is vegetation influenced by top-order predators?

Abstract

Classic trophic cascades are state changes in ecosystems initiated by top trophic-level organisms (predators) and transferred via herbivores to a third or lower trophic level of organisms (plants). This research aimed to determine the potential influence of the dingo, Canis familiaris, on changes in vegetation via predation on large macropods. The objectives were to review trophic cascade theory and assess its relevance in temperate and arid Australia, develop new efficient field-based methods for estimating macropod grazing density and herbaceous biomass and determine the influence of macropod grazing and other environmental variables on changes in herbaceous vegetation in a field-based experiment to assess the potential for dingo-driven trophic cascades to occur in temperate Australia.<br/>The wolf–elk–willow model of a trophic cascade in North America provided a case study for a comparison of environmental influences between arid Yellowstone and arid and semiarid south-eastern Australia. It revealed that climate stability and a predictable resource supply sustain strong trophic interactions that are critical to Yellowstone's trophic cascade. In contrast, the renowned variability of the arid Australian climate means that resource availability is unpredictable and unlikely to produce trophic interactions of similar strength to those of Yellowstone. This critical difference means that a dependence on classic trophic cascade theory may risk limiting our understanding of predator–prey–plant interactions in Australia.<br/>A camera-trapping method estimated macropod grazing density indices that initially correlated well with pellet counts (R2 = 0.86) but were less reliable between years, likely due to the variability in pellet deposition rates between plots and kangaroo densities, which are influenced by interannual variability in seasonal conditions. Reliable above-ground herbaceous biomass estimates gained through the Photographic Estimation Technique yielded regression coefficients (R2) of 0.80–0.98 and 0.81–0.97 between estimated and validated biomass samples in temperate-zone and arid-zone sites, respectively.<br/>A grazing exclusion experiment utilised five vegetation surveys of vascular plant taxa to measure composition, cover and biomass of the herbaceous ground layer vegetation in 20 fenced (ungrazed) and unfenced (grazed) plots in two different grassland communities over three years. Environmental variables were recorded and the macropod grazing density quantified. Macropod grazing suppressed biomass increase but did not change species composition, cover, biomass or Shannon–Wiener diversity in either community. Dominant and subdominant species persisted despite seasonal fluctuations and an overall increase in biomass and cover in both treatments and communities. Vegetation responses were more influenced by environmental processes than herbivory in the absence or simulated presence of dingo predation. The cover and biomass changes that occurred due to macropod grazing did not produce compositional changes concordant with a trophic cascade.<br/>Key results from this research show that (1) predation by dingoes can indirectly increase cover and biomass but this does not necessarily lead to a change in state of the vegetation; (2) environmental influences are likely to predominate in south-east Australian ecosystems as the variable climate governs irregular nutrient availability, which potentially limits trophic interactions; (3) effects of predation and herbivory on vegetation are likely to be temporally and spatially constrained due to inconsistent nutrient flow through trophic levels, which inevitably causes insufficient energy for sustained top-down influence; and (4) there is a need to apply precise field experiment methodologies to identify the consistency and effect of trophic interactions and accommodate climate instability in trophic cascade research in Australia.