Multi-scale analyses of exotic and native plant species diversity within Canada's endangered antelope-brush ecosystem

Native species richness is often significantly correlated with exotic species richness, but the direction and strength of this so-called native-exotic richness relationship (NERR) differs among studies. Furthermore, the potential influence of biotic and abiotic co-variates is less commonly considered. Our objectives were to determine (i) how two different components of scale, grain, and extent, influence the NERR, and (ii) if and how biotic and abiotic variables moderate the NERR. We addressed these objectives using data collected at 37 sites within the endangered Purshia tridentata Pursh (DC) (antelope-brush) shrub-steppe grasslands in southern British Columbia, Canada. We found a significantly positive NERR at our finest grain (1 m(2)), but no significant NERR at any of the three broader grains (10, 100, and 1,000 m(2)). Changing the extent of analysis did not affect the NERR. A multi-model inference approach revealed, at fine grains, (i) strong support for native species richness as a positive predictor of exotic species richness, (ii) Bromus tectorum L. (cheatgrass) as an important negative predictor of native species richness, (iii) foliar cover of P. tridentata as an important positive predictor of exotic species cover, and (iv) a limited predictive role for soil properties. We speculate that spatial patterning in the importance of facilitation relative to competition could influence the NERR within semi-arid ecosystems such as ours. Last, to facilitate comparative research and meta-analyses, we highlight the need for future research to be more explicit in its description of scale, and to more clearly define the grain and extent of each analysis.