Supervisor: Jim Kohen

Despite their ecological importance, the ants of the Kimberley region are poorly researched. Ants carry out important ecosystem functions such as nutrient cycling, seed dispersal and influence the physical and chemical structure of soil. The impact that any major disturbance, such as cattle grazing, may have on ants is therefore an important consideration. Kachana pastoral station, located 130km south-west of Kununurra in the East Kimberley, W.A., is an unusual station as the cattle are not used for export but rather for regeneration. Chris Henggeler manages the cattle with a 'pulse mosaic' grazing rotation. This is characterised by alternating high density cattle in small paddocks with periods of rest, to prevent overgrazing. These methods are based on the ideas of 'Holistic Management', a highly controversial decision making process that identifies the mulching, pruning and fertilising capabilities of cattle essential to maintaining the 'health' of the land.

This study aimed to determine the effect that different intensities of cattle grazing may have on the abundance and composition of ground-active ant assemblages at Kachana. Four treatments were sampled: high, medium, low and no (control) cattle impacts, with three (pseudo) replicates in each. Sampling was conducted with pitfall traps that were left open for a week. Vegetation, litter and canopy cover were estimated for each plot, and soil samples were taken. Pitfall trap catches were processed at Macquarie University, with ants sorted to morphospecies and abundance recorded. Soil samples were sent to the State Chemistry Laboratories in Melbourne for analysis of phosphorus and nitrogen levels.

Comparisons of the mean ant abundance as well as the average vegetation, litter and canopy cover in each treatment were tested with a one-way ANOVA. No significant treatment difference was found for the average vegetation, litter and canopy cover. A significant difference was found, however, between ant abundances in each treatment, with fewer ants located in the control sites. A graphical interpretation of substrate suggested some differences between treatments existed. The structure of ant communities was further analysed using the multivariate tests of multidimensional scaling, analysis of similarities and the similarity percentages procedure. While ant taxon richness did not differ between treatments, a difference between the species composition of each treatment was detected, with the cattle impacted sites dominated by opportunistic species associated with disturbance. The results of the soil analysis showed that while no differences between levels of nitrate and phosphorus were detected between sites, a difference was detected for ammonium levels.

In conclusion, the pulse-mosaic management of cattle at Kachana may be responsible for changes to the structure of ant communities in the area. Given the ecological importance of ants, alterations to their community structure may similarly affect the structure and function of the ecosystem in which they reside. Management implications are discussed.