Simulated Invasion Growth Rate
Simulated population dynamics, treating each species sequentially (within each of the eight assemblages for which population data are available) as the invader population, revealed invasion growth rates (IGRs) that were always positive (Fig. 4). This provides further support that the stabilizing effects of niche partitioning are sufficient for coexistence, even when considering the total overlap that occurs with (and potential competition from) multiple co-occurring species. The observed range of IGRs (0.007 to 0.356) was only marginally lower than those resulting from models assuming zero niche overlap between species (0.152 to 0.488, with populations achieving IGRs approachingR max). By contrast, observed IGRs were substantially larger than predicted in the absence of niche partitioning: in the latter case, only 10 of the 37 simulated populations had IGRs > 0.