Ecosystem processes are dependent on the interaction between biotic and abiotic elements of a system. Global biodiversity is currently under severe threat from habitat fragmentation and loss. While there is consensus that processes associated with fragmentation have direct impacts on biodiversity, we are just now beginning to understand how fragmentation affects ecosystem functions. Myremecochory is a crucial ecosystem function that limits intraspecific tree competition and locates seeds in habitat that helps facilitate germination. Since ants are a ubiquitous component in Australian ecosystems, we sought to determine the impacts of habitat fragmentation on seed removal. We asked: 1) Does habitat fragmentation alter seed removal in forest fragments versus continuous forest, in fragments of different size and at fragment edges? If yes: 2) is the change in seed removal linked to the number of ant species at a site, which species are present, or how many ant individuals are present? 3) Do changes to the environment, that are the result of experimental fragmentation, affect the rate of ant seed removal ?Our focus was the ant community at the Wog Wog Long-term Habitat Fragmentation Experiment in New South Wales, Australia. We conducted a seed removal experiment with Eucalyptus sideroxylon seeds and used linear mixed models and structural equation models for data analysis.
Results/Conclusions
Habitat fragmentation increased seed removal in large fragments and fragment cores. The number of seeds removed from a site was determined by the total number of ant individuals present at a site. However, fragmentation did not alter seed removal by directly altering the total number of ants at a site. Rather, fragmentation modified the environment so that more/fewer seeds were removed in different parts of the fragmented landscape, compared to continuous forest, when the same number of ants were present. Specifically, fragmentation reduced the number of seeds removed per ant in small fragments by reducing temperature there. Further, fragmentation increased the number of seeds removed per ant in medium and large fragments (and potentially fragment cores) by reducing grass cover there. More seeds were removed when grass cover was low. Our study helps to disentangle which aspects of ant biodiversity are impacted by fragmentation and are linked to seed removal. Our study also highlights that fragmentation driven changes to seed removal can be the result of direct effects of fragmentation on the environment that alter ant behavior.