Detecting the effects of introduced species:: a case study of competition between Apis and Bombus

Developing tools for rapid assessment of introduced species impacts is one of the most important challenges in invasion ecology. Most assessments of impact rely on correlational data or other indirect measures. Yet few studies have evaluated invasion effects using multiple, simultaneously applied monitoring and experimental approaches, in order to compare easily obtained metrics with more difficult but direct measures of reproductive success or population dynamics. In this study, I use data from an experimental test of introduced honey bee (Apis mellifera) impacts on native bumble bees (Bombus spp.) to address two major questions: 1) how well did observational data on niche overlap and spatial correlations between Apis and Bombus predict the results of experimental tests of competitive effects? and 2) how well did effects of the experimental Apis manipulations on Bombus foragers, which are easy to observe, predict changes in reproductive success of colonies, which are difficult to measure? Niche overlap between Apis and Bombus varied substantially, but increased to levels as high as 80-90% during periods of resource scarcity. Correlations between numbers of Apis foragers and numbers of Bombus foragers were also highly variable, but I detected a significant negative relationship in only one of the seven months observed. In contrast, the experimental results showed that mean numbers of Bombus foragers observed on a given transect increased significantly with greater distance from introduced Apis colonies. Of these three measures (niche overlap, correlations in abundances, and effects of experimental introductions), only the experimental data on forager abundances accurately estimated competitive effects on colony reproductive success previously reported for the same experiment, and the correlational data in particular completely failed to predict the effects observed in the experimental study. This work suggests that great caution is warranted in making assessments of invasion impact on the basis of spatial or temporal correlations between invasive and native species. Thus, investing in even small and limited experimental studies may be more valuable than extensive observational work in quantifying invasion impacts.