COS 83-1
Interspecific effects on establishment: How an introduced grass might exclude a native tree without altering the fire cycle
Invasive grasses are often associated with local declines in native plant richness and abundance, even without experiencing a fire. Without identifying a mechanism by which they impact native species, however, it is not clear whether grasses are simply moving in to areas that have become depauperate in native plants for other reasons. Mechanisms by which non-native plants actually have negative impact on native plants are rarely identified, and are especially difficult to address but important to predict for common but long-lived perennial species.
We tested for negative effects of an invasive perennial bunchgrass (Pennisetum ciliare, syn. Cenchrus ciliarus) on a native paloverde tree (Parkinsonia microphylla) in the Sonoran Desert. Because the most sensitive life stage is likely where competition will affect population dynamics, we hypothesized that germination or establishment of the delicate tree seedlings would be reduced by proximity to a high density of the invasive grass. We measured weekly emergence and establishment rates of paloverde seedlings in the field for 10 weeks. Seedlings were counted and marked on plots invaded by grass, where the grass was removed, and in native vegetation. We simultaneously conducted a parallel greenhouse experiment by planting paloverde seeds in tubs containing the invasive grass, a native shrub (Ambrosia deltoidea) that is a common ground cover in upland scrub, or no adult plant.
Results/Conclusions
Germination rates in the greenhouse were greater than 95% across all adult plant environments (p = 0.29). However, paloverde seedlings in the greenhouse had significantly lower seedling dry mass (p < 0.05) and survival (p < 0.05) in tubs with grass than in those with the native shrub or no plant. Fewer than 30 seedlings emerged on field plots, but across all plot types, and so is consistent with the greenhouse germination results. No seedlings on plots survived to establish, but of 75 observed off plot, only 1 established, well away from grass.
Proximity to the invasive grass did not affect germination of paloverde seeds, but did kill the sensitive seedlings. This generates a negative effect of the grass on recruitment by paloverde populations. Future research should determine the strength of negative effect the invasive grass exerts on its own seedlings, which observation suggests will be much smaller than its effect on paloverdes, as well as the negative density dependence of paloverdes on their own seedlings and on those of the invasive grass.