Interactions between plants and ants contribute to functional ecological communities. Plants provide food, shade and habitat and attract prey; while ants disperse pollen and seeds, defend against herbivores and cycle soil nutrients. Despite these close links, the congruence of plant and ant assemblages, and particularly how these patterns relate to environmental change, is poorly understood. Here we test the prediction that more highly structured and diverse plant communities will support more diverse ant assemblages. Secondly, we ask how altered environments might decouple patterns of diversity, across space or time. To answer these questions, we measured plant and ant species composition and abundance at 51 survey plots on a 550km transect spanning a temperature and rainfall gradient (13.3-17.3ºC mean annual temperature and 308-991mm mean annual rainfall) in South Australia. We calculated the congruence of current plant and ant communities across our gradient using symmetrical procrustes procedures that rotate, flip and scale two ordinations to achieve a maximum correspondence. To determine how future environments may change these relationships we conducted simulations under a range of projected climate scenarios, using the observed relationships to current climatic environments and compared the expected responses of each assemblage (plant or ant) and their congruence.
Results/Conclusions
Our surveys detected 363 plant and 220 ant species. We found ant and plant species richness were not correlated, but community compositions were correlated to each other and their corresponding abiotic environment. Environmental variables constrained 33% of observed variation in plant community structure and 45% of ant community structure. Temperature and rainfall were the strongest drivers of community structure, and outperformed edaphic and topographic variables, which had low explanatory power. The strongest environmental correlates to assemblage structure were maximum temperature (plant R2=0.87***, ant R2=0.18*) and mean annual precipitation (plant R2=0.86***, ant R2=0.28**). Symmetrical Procrustes rotation indicated a strong correlation between contemporary plant and ant communities (correlation=0.355, significance=0.004). Preliminary analysis of future congruence indicates ant and plant community similarity will decrease with future change, but further analysis (currently underway) is required to quantify the magnitude of this effect. Our results provide a template for biodiversity forecasting under climate change and justify additional research in to the degree to which mutualistic interactions can buffer the effect of a changing climate.