Naomi E. Davis

Interspecific and Geographic Variation in the Diets of Sympatric Carnivores: Dingoes/Wild Dogs and Red Foxes in South-Eastern Australia

Dingoes/wild dogs (Canis dingo/familiaris) and red foxes (Vulpes vulpes) are widespread carnivores in southern Australia and are controlled to reduce predation on domestic livestock and native fauna. We used the occurrence of food items in 5875 dingo/wild dog scats and 11,569 fox scats to evaluate interspecific and geographic differences in the diets of these species within nine regions of Victoria, south-eastern Australia. The nine regions encompass a wide variety of ecosystems. Diet overlap between dingoes/wild dogs and foxes varied among regions, from low to near complete overlap. The diet of foxes was broader than dingoes/wild dogs in all but three regions, with the former usually containing more insects, reptiles and plant material. By contrast, dingoes/wild dogs more regularly consumed larger mammals, supporting the hypothesis that niche partitioning occurs on the basis of mammalian prey size. The key mammalian food items for dingoes/wild dogs across all regions were black wallaby (Wallabia bicolor), brushtail possum species (Trichosurus spp.), common wombat (Vombatus ursinus), sambar deer (Rusa unicolor), cattle (Bos taurus) and European rabbit (Oryctolagus cuniculus). The key mammalian food items for foxes across all regions were European rabbit, sheep (Ovis aries) and house mouse (Mus musculus). Foxes consumed 6.1 times the number of individuals of threatened Critical Weight Range native mammal species than did dingoes/wild dogs. The occurrence of intraguild predation was asymmetrical; dingoes/wild dogs consumed greater biomass of the smaller fox. The substantial geographic variation in diet indicates that dingoes/wild dogs and foxes alter their diet in accordance with changing food availability. We provide checklists of taxa recorded in the diets of dingoes/wild dogs and foxes as a resource for managers and researchers wishing to understand the potential impacts of policy and management decisions on dingoes/wild dogs, foxes and the food resources they interact with.

Diets of native and introduced mammalian herbivores in shrub-encroached grassy woodland, south-eastern Australia.

Effective management of sympatric mammalian herbivore populations requires an understanding of interspecific interactions. At Wilsons Promontory National Park, Victoria, sympatric native and introduced mammalian herbivores are thought to be contributing to modification of shrub-encroached Coastal Grassy Woodland. We estimated the diets of the five terrestrial mammalian herbivore species present using microhistological techniques. The diets of introduced hog deer (Axis porcinus) and native swamp wallabies (Wallabia bicolor) consisted mainly of dicots. The diet of introduced European rabbits (Oryctolagus cuniculus) contained similar proportions of monocots and dicots. The diets of native eastern grey kangaroos (Macropus giganteus) and native common wombats (Vombatus ursinus) consisted mainly of monocots but kangaroos also consumed moderate amounts of dicots. Deer and wallabies consumed more native plants than did the other species and rabbits consumed more exotic plants than did all other species except kangaroos. Diet breadth was narrowest for kangaroos and broadest for swamp wallabies and hog deer. Overlap in food use by the five herbivores was high, particularly between deer and wallabies, and between kangaroos and both rabbits and wombats. Our results suggest that the potential impacts of native and introduced species on the vegetation of Coastal Grassy Woodland are similar, and that the entire herbivore assemblage will need to be managed to increase fine fuel loads if fire is used as a restoration tool.

Facilitative interactions between an exotic mammal and native and exotic plants: hog deer (Axis porcinus) as seed dispersers in south-eastern Australia

Endozoochory by exotic mammalian herbivores could modify vegetation composition by facilitating the dispersal and establishment of exotic and native plant species. We examined the potential for endozoochoric dispersal of native and exotic plants by exotic hog deer (Axis porcinus) in south-eastern Australia. We quantified the germinable seed content of hog deer faecal pellets collected in five vegetation types within a 10,500-ha study area that was representative of their Australian range. Twenty exotic and 22 native species germinated from hog deer faecal pellets and significantly more native species germinated compared to exotic species. Seedlings of the encroaching native shrub Acacia longifolia var. sophorae emerged, but no native trees emerged and the percentage of grasses that germinated was low (11%). The species composition of germinants was similar among the five vegetation types. We estimated that the hog deer population in our study area could potentially disperse >130,000 viable seeds daily. Our study shows how an exotic mammal can disperse seeds from both native and invasive plants and highlights the need for endozoochory to be considered more widely in studies assessing the impacts of exotic mammals on plant communities.

Diets of Non-Native Deer in Australia Estimated by Macroscopic Versus Microhistological Rumen Analysis

ABSTRACT Estimating diet is often an important step in understanding and managing the impacts of ungulates, particularly for non-native species, but there is uncertainty about whether rumen contents should be assessed using macroscopic or microhistological methods or both. Introduced sambar deer (Cervus unicolor) have a large and increasing distribution in south-east Australia, and there is concern about their impacts on native and non-indigenous plant species. We estimated the diets of 102 sambar deer harvested during 2007–2009 using macroscopic and microhistological rumen analysis techniques. We identified 105 plant species in the diets of sambar deer, 61 identified with both techniques and 22 identified only macroscopically or only microhistologically. Rumen species richness was 70% greater using the microhistological technique. Estimates of taxonomic (i.e., monocot and dicot) and functional (i.e., fern, shrub—tree, forb, climber, grass, and grass-like) group composition by the 2 techniques were similar. Shrubs—trees dominated the diet (macroscopic, 49.7%; microhistological, 52.7%), followed by grasses (macroscopic, 22.7%; microhistological, 17.5%) and ferns (macroscopic, 20.6%; microhistological, 22.2%). We identified 9 non-indigenous plant species, 2 of which we identified using only the microhistological technique. We detected seeds of the weed blackberry (Rubus fruticosus aggregate), sometimes in large amounts, only with the macroscopic technique, whereas we detected foliage of that species with both techniques. Both techniques classified sambar deer as an intermediate mixed feeder closer to a concentrate selector—browser than a bulk and roughage feeder. However, both techniques detected seasonal differences in the percentages of taxonomic and functional groups in the diet; sambar deer were more grazers in autumn and more browsers in spring. Our results indicate that both macroscopic and microhistological techniques may need to be used when it is important to identify plant species in the diet, as is often the case for non-native ungulates. However, either technique can be used to estimate broader taxonomic and functional diet composition, including feeding type.

A systematic review of the impacts and management of introduced deer (family Cervidae) in Australia

Abstract. Deer are among the world’s most successful invasive mammals and can have substantial deleterious impacts on natural and agricultural ecosystems. Six species have established wild populations in Australia, and the distributions and abundances of some species are increasing. Approaches to managing wild deer in Australia are diverse and complex, with some populations managed as ‘game’ and others as ‘pests’. Implementation of cost-effective management strategies that account for this complexity is hindered by a lack of knowledge of the nature, extent and severity of deer impacts. To clarify the knowledge base and identify research needs, we conducted a systematic review of the impacts and management of wild deer in Australia. Most wild deer are in south-eastern Australia, but bioclimatic analysis suggested that four species are well suited to the tropical and subtropical climates of northern Australia. Deer could potentially occupy most of the continent, including parts of the arid interior. The most significant impacts are likely to occur through direct effects of herbivory, with potentially cascading indirect effects on fauna and ecosystem processes. However, evidence of impacts in Australia is largely observational, and few studies have experimentally partitioned the impacts of deer from those of sympatric native and other introduced herbivores. Furthermore, there has been little rigorous testing of the efficacy of deer management in Australia, and our understanding of the deer ecology required to guide deer management is limited. We identified the following six priority research areas: (i) identifying long-term changes in plant communities caused by deer; (ii) understanding interactions with other fauna; (iii) measuring impacts on water quality; (iv) assessing economic impacts on agriculture (including as disease vectors); (v) evaluating efficacy of management for mitigating deer impacts; and (vi) quantifying changes in distribution and abundance. Addressing these knowledge gaps will assist the development and prioritisation of cost-effective management strategies and help increase stakeholder support for managing the impacts of deer on Australian ecosystems.

Does diet constrain the occupation of high elevations by macropods? A comparison between Macropus rufogriseus and Wallabia bicolor

The ability to utilise a diet of shrubs or trees is key to the survival of herbivores in deep snow. However, reduction in snow depth with climate change may allow herbivores into higher elevations where herbfields are dominant. Wallabia bicolor occurs above the winter snowline of the Snowy Mountains in the subalpine zone, whereas Macropus rufogriseus, does not although it is present in alpine Tasmania. The winter diet of W. bicolor in the Snowy Mountains consisted of shrubs, trees, and herbs. With >60% of food sources (shrubs and trees) available above the snow, the change from occupation of habitat below the winter snowline to above requires little change in its diet. Consumption of shrubs, forbs and monocots by M. rufogriseus was similar between the Snowy Mountains and alpine Tasmania. M. rufogriseus includes a high proportion of shrubs in its diet; however, it may be excluded from snow-covered habitat due to a lesser ability to utilise poor-quality browse. Globally, migratory herbivores respond to deep snow with seasonal movements. However, W. bicolor and M. rufogriseus are not migratory and can only occupy higher elevations of the Snowy Mountains as snow depth and duration diminish. Because they do not currently occupy the alpine zone and the vegetation has not evolved to accommodate their presence, their impact on alpine vegetation is likely to be greater than migratory alpine grazers/browsers.

Habitat-specific and season-specific faecal pellet decay rates for five mammalian herbivores in south-eastern Australia

The accuracy of population abundance estimates of mammalian herbivores from faecal pellet counts is potentially affected by pellet decay. We collected fresh pellet groups from hog deer (Axis porcinus), European rabbit (Oryctolagus cuniculus), eastern grey kangaroo (Macropus giganteus), swamp wallaby (Wallabia bicolor) and common wombat (Vombatus ursinus) (n = 300 per species) at Wilsons Promontory National Park, Victoria, Australia. We deposited five pellet groups per species per month within each of five vegetation types in the park, then monitored pellet group decay over 24 months. We demonstrate that age estimation of pellet groups was inaccurate and is unlikely to improve the efficiency of pellet counts. We present habitat- and species-specific estimates of pellet and pellet group decay using two measures: decay rate (the proportion of pellets surviving per unit of time); and mean time to decay. We explain how our data can be used to optimise faecal pellet count design, and to improve the accuracy of both indices and estimates of abundance from pellet counts. The variability observed in the decay of pellet groups among vegetation types, and for species among seasons, suggests that caution should be used if applying pellet decay rates over long time-frames or to locations with differing environmental conditions.

Vegetation management influences habitat use by mammalian herbivores in shrub-encroached grassy woodland

Abstract Context. Restoration of disturbed vegetation communities commonly involves altering vegetation composition and structure, attributes that can influence the suitability of habitat for fauna. Feedbacks may occur whereby changes to the vegetation affect mammalian herbivores, and unintended changes may prevent managers from achieving conservation goals. Aims. To understand how vegetation management affects habitat use by five mammalian herbivores, namely eastern grey kangaroo (Macropus giganteus), swamp wallaby (Wallabia bicolor), common wombat (Vombatus ursinus), European rabbit (Oryctolagus cuniculus) and hog deer (Axis porcinus). Methods. A management experiment (mechanical slashing of the encroaching shrub Leptospermum laevigatum) at Wilsons Promontory National Park, Australia, created slashed swales in addition to untreated dune and scrub woodland. In each vegetation stratum, we estimated the cover of L. laevigatum and quantified herbivore abundance by counting the standing crop of faecal pellets. Key results. Relative to untreated vegetation, mechanical slashing of L. laevigatum substantially reduced cover of this species above 200 cm, but increased its cover below 30 cm. On the basis of faecal-pellet counts, multispecies use of managed and unmanaged parts of the landscape differed substantially, with the differences principally driven by higher abundance of European rabbits and eastern grey kangaroos at slashed sites. Conclusions. The responses of three grazing species (kangaroo, rabbits and wombats) to vegetation management were predicted well by prior knowledge of diet and habitat preferences. This was not the case for the browser (swamp wallaby), nor for the grazer that consumes substantial amounts of browse in the study area (hog deer), and additional knowledge of the processes underlying their responses to vegetation change is required. Implications. Our findings highlighted that vegetation management can influence herbivore abundances in the managed system. An improved understanding of these associations will allow vegetation management plans to incorporate herbivore responses.

Interspecific variation in the diets of herbivores in an industrial environment: implications for exposure to fluoride emissions.

Atmospheric fluorides (gaseous and particulate) are deposited on, and absorbed by, vegetation. Ingested fluoride accumulates in calcified tissues of vertebrates, and if it is excessive, it may lead to dental and skeletal fluorosis. The prevalence, form and severity of the effects vary greatly between species. Foraging strategy can be an important determinant of fluoride exposure in herbivores, because foliar fluoride concentrations vary between plant species, for example, according to vertical and lateral position in the vegetation. We combined microhistological analysis of diet and analysis of foliar fluoride levels to examine interspecific variation in dietary fluoride exposure of macropodid marsupials (swamp wallaby Wallabia bicolor, red-necked wallaby Notamacropus rufogriseus and eastern grey kangaroo Macropus giganteus), in the buffer zone of an aluminium smelter in Victoria, Australia. Dietary niche differentiation between species was evident. The swamp wallaby and the red-necked wallaby were browsers or mixed feeders, depending on the classification system used. The eastern grey kangaroo was a grazer, consuming almost entirely grasses. However, foliar fluoride did not vary significantly between the main plant groups consumed. Our results indicate that interspecific variation in diet at this site is unlikely to explain variation in fluoride exposure.

The diet of the common wombat (Vombatus ursinus) above the winter snowline in the decade following a wildfire

The use of high elevations with deep snow cover presents a challenge to mammalian herbivores, which is exacerbated by subalpine vegetation dynamics such as slow regrowth following disturbance. We postulated that post-fire responses of common wombats (Vombatus ursinus) at high elevations would differ from those at low elevations. We examined the winter diet of common wombats in the Snowy Mountains in the decade after fire in burnt and unburnt areas and compared our results to published diet studies from low elevations. Optimal foraging theory predicts that as food resources become scarce herbivores respond by widening their choice of foods, yet we found that wombats have only marginally wider dietary breadth at higher than at lower elevations in terms of plant forms and diet breadth in terms of species was not greater. The use of shrubs and the tall herb Dianella tasmanica enables wombats to reduce the energetic costs of digging for food in snow. Able to survive fire in a burrow, the wombat is then capable of responding to reduced foraging opportunities following fire by broadening the range of species consumed and adopting foraging strategies that exploit temporally improved food quality, demonstrated by the greater proportion of grass consumed in burnt sites.