Delayed ant response to forest fragmentation: Evidence from the Wog Wog long-term habitat fragmentation study in Australia

Background/Question/Methods

Native habitats are crucial for the preservation of endemic ant populations, yet these habitats are increasingly threatened by fragmentation as a result of habitat loss. In Southeast Australia, the introduction of pine plantations during the past 30 years has resulted in the fragmentation of native Eucalyptus forests. Current theory predicts that populations remaining in remnant patches of Eucalyptus will demonstrate lower probability of occurrence and decreased abundance as a result of isolation and smaller habitat area. Additionally, there is the expectation that edge habitats will favor opportunistic generalist species because of the interaction of two different habitats. We tested these predictions with two species of ants, Leptomyrmex erythrocephalus and Aphaenogaster longiceps, collected over 24 years at the Wog Wog Long-term Habitat Fragmentation Experiment in New South Wales, Australia. Sampling began in 1985 and in 1987 clear-cutting took place followed by the establishment of the pine plantation. Twelve remnant patches of Eucalyptus are maintained in the pine matrix. We analyzed the response of these two ant populations immediately following fragmentation (1-4 years), at an intermediate post-fragmentation period (5-8 years) and late post-fragmentation (23 years).  

Results/Conclusions

Abundance of A. longiceps in the continuous forest and remnant patches was similar while abundance in the pine plantation showed a consistent decreasing trend with the lowest abundance values in year 23. The probability of occurrence for both A. longiceps and L. erythrocephalus in pine plantations was consistently lower than control sites and remnant patches. Interestingly, the probability of occurrence in control sites and fragments was similar. This suggests that these two species of ants are becoming increasingly isolated on the remnant patches and that the pine matrix, with increased age, is less habitable. It also indicates that habitat patches can support ant populations and that isolation does not always increase the likelihood of extinction. No consistent effects of edge habitat were detected in the immediate and intermediate time periods, however by year 23 there were significant decreases in occurrence probability for both species. This indicates that edge effects on remnant patches may not be detected until late pine plantation growth. Our study, unique in its long-term perspective, lends credence to the idea that long-term studies are necessary in order to understand the response of ground dwelling invertebrates to the fragmentation of their habitat.