Estimating the community-level impact of the riparan alien species Mimulus guttatus by using a replicated BACI field experiment

Summary The community-level consequences of the invasion by Mimulus guttatus were studied. Twenty experimental plots were established in May 2006. Mimulus guttatus was removed from 10 plots, the remaining 10 served as control. Species composition was recorded repeatedly in the plots and the cover of all species estimated in order to identify changes following M. guttatus removal from the invaded community. The data were analyzed using split-plot ANOVA, DCA and RDA. The (a) square roots of species numbers (b) Shannon diversity H and (c) Shannon evenness E were used as importance values in the univariate analysis. The multivariate analysis was carried out on (a) species covers, to reveal changes in the proportions of species in the studied plots, and (b) binary data of species presence/absence to reveal the trends in species composition. Neither the univariate (p = 0.947, p = 0.16, p = 0.12141, DF = 3) nor multivariate (p = 0.832; p = 0.986) analysis revealed significant differrences in the community characteristics studied, following the removal of M. guttatus. It remains debatable whether this lack of response is due to the little or no real effect of the removal of M. guttatus or to the limited extent of the data. However, the control (invaded) plots had slightly higher mean number of species than treated plots, so it is very unlikely that the invasion of M. guttatus reduces species richness. The occurrence of M. guttatus in the Czech Republic is limited to sites frequently disturbed by flooding. Therefore, the occurrence of resident species seems to be constrained by the disturbance regime rather than competition with the invasive M. guttatus.