Austral Ecology (2003) 28, 658–665 Effects of cover reduction on mulgara Dasycercus cristicauda (Marsupialia: Dasyuridae), rodent and invertebrate populations in central Australia: Implications for land management PIP MASTERS,* CHRIS R. DICKMAN AND MATHEW CROWTHER School of Biological Sciences and Institute of Wildlife Research, University of Sydney, New South Wales, Australia Abstract This study investigates the effect of cover reduction on the mulgara, Dasycercus cristicauda, a small marsupial classified as vulnerable to extinction, which occurs in areas of central Australia dominated by hummock grasslands. Loss or degradation of spinifex has been implicated in population declines of this species previously, but the importance of cover in maintaining quality habitat remains speculative. To determine the effect on D. cristicauda of cover reduction, caused by the harvesting of spinifex, we monitored population changes and changes in prey resources (rodents and invertebrates) before and after spinifex harvesting took place at a site near the Ayers Rock Resort, Northern Territory, Australia. Ten plots, each of 8.75 ha, were established and sampled from May 1994 to October 1995. Harvesting took place on five plots in August 1994, which reduced spinifex cover from 46 to 21% and the amount of spinifex >0.25 m high from 42 to 2%. Harvesting did not significantly affect the number of D. cristicauda known to be alive or captured despite other studies indicating that cover is an important habitat attribute. There was also no evidence that cover reduction impacted on the biomass of the invertebrate food resources. However, there was a significant reduction in the number of rodents captured. The lack of a response to cover reduction by D. cristicauda is possibly because the cover of Triodia remained high enough (above 15%) to sustain animals, and harvested areas were relatively small. This study therefore suggests that D. cristicauda can tolerate a moderate local reduction in cover of its preferred habitat. However, it remains possible that other land use practices that cause severe reduction of cover (including clearing for mining or fire prevention, grazing which may result in spinifex reduction through trampling, and fire management) will have more dramatic effects on D. cristicauda. Evaluation of such effects should be a priority for future research. Key words: habitat manipulation, mulgara, rodents, spinifex grasslands, threatened species. INTRODUCTION Maintaining high-quality habitat is one of the main objectives when managing species, especially if the target species are threatened and have small or declining populations. Although it is possible to corre- late the spatial variation in species abundance with habitat attributes, it is more difficult to determine which attribute, or combination of attributes, has the greatest influence on distribution and abundance. This can be approached using discriminant or other multi- variate statistical procedures (Seagle 1985; Haering & Fox 1995), but is best done by manipulative field experiments (Underwood 1990; Caughley & Gunn 1996). In several studies, habitat variables have been manipulated experimentally to determine their effect on mammalian distribution and abundance. For example, Kitchener (1972) manipulated cover and found that the quokka, Setonix brachyurus, increased in numbers on Rottnest Island, Western Australia. Price et al. (1994) reduced shrub cover and demonstrated an increase in Stephens’ kangaroo rat, Dipodomys stephensi. McMurry et al. (1994) induced successional changes in vegetation using fire and herbicides and showed that this resulted in increased population density and reproductive activity in cotton rats, Sigmodon hispidus. Finally, Cox et al. (2000) manipu- lated the cover of leaf litter and found that this influ- enced habitat use by black rats, Rattus rattus. The mulgara, Dasycercus cristicauda (Krefft), is a dasyurid marsupial that is classified as vulnerable to extinction, having declined over 50–90% of its histor- ical range (Maxwell et al. 1996). It is a sexually dimor- phic species, with females weighing approximately 60 g and males 100 g. It is an annual breeder, mating in autumn or winter with juveniles emerging in late winter or spring (Gibson & Cole 1992; Masters 1998; Dickman et al. 2001). The diet is broadly carnivorous, with invertebrates making up the major part of the *Corresponding author. Present address: Regional Conser- vation, Department for Environment and Heritage, South Australia, 37 Dauncey Street, Kingscote, South Australia 5223, Australia (Email: masters.pip@saugov.sa.gov.au). Accepted for publication May 2003.