Direct and indirect impacts of introduced species on community dynamics

Background/Question/Methods

The effects of herbivory by ungulates on individual plants or entire plant communities can be dramatic and might indirectly influence other associated species, such as pollinators. These effects may be especially strong when the affected species are keystone species in the ecosystem. The northern portion of the temperate forest of Patagonia harbors a unique triad of keystone mutualists comprised of a hummingbird (Sephanoides sephaniodes), a mistletoe (Tristerix corymbosus), and a marsupial (Dromiciops gliroides). The hummingbird is responsible for pollinating nearly 20% of the endemic woody plant species in this system. Aristotelia chilensis is most common species in the system and is the main host of the mistletoe. Thus, if herbivores reduce the density of A. chilensis, the number of mistletoes could also be affected, potentially disrupt a keystone mutualism between mistletoes and hummingbirds. Here, we examined the direct and indirect consequences of exotic ungulates on all the stages involved in this hostplant-parasite-pollinator system at 13 invaded sites and 13 intact sites where exotic ungulates were not present. Additionally, we experimentally excluded herbivores to assess their direct effects on an A. chilensis.

Results/Conclusions

The density of A. chinensis 16× greater in intact than invaded sites. An exclosure experiment shows that in just one month 99% of the A. chilensis leaves were damaged by deer outside the exclosure. By browsing on A. chilensis, exotic ungulates indirectly affect the population persistence of the keystone mistletoe. The mistletoe density was 83× greater in intact than in invaded sites. Because the mistletoe is the only resource of nectar for the hummingbird during the winter, hummingbird abundance decreased as the number of mistletoe plants decreased. Additionally, mistletoe host age distribution ranged between 6 and 42 years old in intact sites. In contrast, there were not infected host younger than 28 years at invaded sites. Finally, at all invaded site we failed to detect the presence of the marsupial D. gliroides, but the marsupial was present at the 13 intact sites. Consequently, we did not find mistletoe seeds dispersed at invaded sites and fruiting plants were three times less abundant at invaded than intact sites. Our results show that exotic ungulates, by browsing on A. chilensis, indirectly affected the keystone mutualisms, and had often dramatic cascading effects through the rest of the ecosystem, threatening a much larger set of interactions.