|dc.description.abstract||Although theory underlying the invasion paradox, or the change in the relationship between the richness of alien and indigenous species from negative to positive with increasing spatial scale, is well developed and much empirical work on the subject has
been undertaken, most of the latter has concerned plants and to a lesser extent marine invertebrates. Here we therefore examine the extent to which the relationships between indigenous and alien species richness change from the local metacommunity to the interaction
neighborhood scales, and the influences of abundance, species identity, and environmental favorability thereon, in springtails, a significant component of the soil fauna. Using a suite of modeling techniques, including generalized least squares and geographically weighted
regressions to account for spatial autocorrelation or nonstationarity of the data, we show that the abundance and species richness of both indigenous and alien species at the metacommunity scale respond strongly to declining environmental favorability, represented here by altitude. Consequently, alien and indigenous diversity covary positively at this scale. By contrast, relationships are more complex at the interaction neighborhood scale, with the
relationship among alien species richness and/or density and the density of indigenous species varying between habitats, being negative in some, but positive in others. Additional analyses demonstrated a strong influence of species identity, with negative relationships identified at the interaction neighborhood scale involving alien hypogastrurid springtails, a group known from
elsewhere to have negative effects on indigenous species in areas where they have been introduced. By contrast, diversity relationships were positive with the other alien species. These results are consistent with both theory and previous empirical findings for other taxa, that interactions among indigenous and alien species change substantially with spatial scale and that environmental favorability may play a key role in explaining the larger scale patterns.
However, they also suggest that the interactions may be affected by the identity of the species concerned, especially at the interaction neighborhood scale.||en