Different colonization rates between exotic and native grasses as explanation for invasion success?

Background/Question/Methods

In California most native perennial grasslands have been replaced by exotic annual grasses and forbs. The success of invasive species might depend on their superior colonization abilities relative to native species.

We hypothesize that higher colonization rates of exotic annual grasses shift local interactions towards exclusion of native grasses. To test our hypothesis we established fifteen 80m² checkerboard grids of alternating patches of Bromus hordeaceus, a common annual invasive grass, and Nasella pulchra, a native perennial grass, and manipulated colonization by employing one of three different dispersal treatments to a checkerboard. First, we restricted natural dispersal by putting up seed fences around each patch, thus minimizing colonization effects. Second, we allowed natural colonization to take place, where we expected that B. hordeaceus has a larger colonization neighborhood and a higher colonization rate. Third, we equalized seed rain of both species within checkerboards, thus, equalizing colonization differences. This treatment simulated a possible restoration scenario. We established these plots in 2006 and let the vegetation establish (with no dispersal among patches) for the first two years. In 2008, we initiated the dispersal treatments and followed recruitment in 2009.

Results/Conclusions

We found large differences in seed production: B. hordeaceus produced 36000 - 95000 seeds/m² whereas N. pulchra produced only 1600 - 2400 seeds/m². As expected, in plots with natural dispersal recruitment of both species was higher in conspecific than in heterospecific patches: N. pulchra seedling density declined by 48% and B. hordeaceus density declined by 30% compared to their own patch types. In the equalized seed rain treatment, these differences were reduced only for N. pulchra (by 13%) but not for B. hordeaceus, indicating the stronger role of seed limitation for the native. Equalizing seed rain also increased recruitment of N. pulchra in its own patches relative to both the restricted and the natural dispersal treatments, indicating seed limitation even in its “own” patches. We did not find that B. hordeaceus was similarly limited in its own patches. However, even in the equalizing treatment, N. pulchra recruitment was still 35% lower in B. hordeaceus patches compared to its own, indicating the additional role of competitive exclusion.

These results point to an overwhelming propagule pressure of the exotic species that may reverse any local competitive advantage N. pulchra may experience. This might be the most important reason why annual exotic grasses are successful invaders and restoration of native grasslands is such a difficult task.