Testing the effects of ant invasions on non-ant arthropods with high-resolution taxonomic data.

Abstract

Invasions give rise to a wide range of ecological effects. Many invasions proceed without noticeable impacts on the resident biota, whereas others shift species composition and even alter ecosystem function. Ant invasions generate a broad spectrum of ecological effects, but controversy surrounds the extent of these impacts, especially with regard to how other arthropods are affected. This uncertainty in part results from the widespread use of low-resolution taxonomic data, which can mask the presence of other introduced species and make it difficult to isolate the effects of ant invasions on native species. Here, we use high-resolution taxonomic data to examine the effects of Argentine ant invasions on arthropods on Santa Cruz Island, California. We sampled arthropods in eight pairs of invaded and uninvaded plots and then collaborated with taxonomic experts to identify taxa in four focal groups: spiders, bark lice, beetles, and ants. Spiders, bark lice, and beetles made up ~40% of the 9868 non-ant arthropod individuals sampled; the majority of focal group arthropods were putatively native taxa. Although our results indicate strong negative effects of the Argentine ant on native ants, as is well documented, invaded and uninvaded plots did not differ with respect to the richness, abundance, or species composition of spiders, bark lice, and beetles. One common, introduced species of bark louse was more common in uninvaded plots than in invaded plots, and including this species into our analyses changed the relationship between bark louse richness vs. L. humile abundance from no relationship to a significant negative relationship. This case illustrates how failure to differentiate native and introduced taxa can lead to erroneous conclusions about the effects of ant invasions. Our results caution against unqualified assertions about the effects of ant invasions on non-ant arthropods, and more generally demonstrate that accurate assessments of invasion impacts depend on adequate information about species identity.