Common, uncommon, and rare plant species survival: The consequences of post-dispersal seed predation in Missouri remnant and restored tallgrass prairies

Background/Question/Methods

Post-dispersal seed predation is a consequential determinant of seed availability for species establishment, and ultimately, community structure. For example, species with high intrinsic fitness and competitive ability may remain on a landscape if their seeds are preferentially consumed after dispersal. To determine how seed predation structures plant communities in tallgrass prairies, we quantified seed consumption and removal patterns of small mammals and insects, two groups of seeds predators that are highly prominent in tallgrass prairies. We investigated how seed consumption patterns are affected by plant species rarity, functional group (legume, grass, or non-legume forb), season, and whether they are affected by a site’s history. Our study sites include one remnant prairie, and a chronosequence of three restored prairies – seeded in either 2004, 2013, or 2017. For our study in the summer and fall of 2020, we selected six species within legumes, grasses, and forbs that are classified as common, uncommon, or rare in restored prairie communities. We set out 90 seed cafeterias at each of our 4 sites, with a total of 21,600 seeds per season. After leaving the cafeterias out for approximately one month for each season, we calculated what was left from the initial seeds we deployed.

Results/Conclusions

Our findings suggest that seed predation is a key aspect in plant community dynamics in tallgrass prairies. Predation rates varied substantially among species (highest rate of predation: 87% seeds consumed for Helianthus mollis, lowest rate: 0.04% for Danthonia spicata), and was especially high during the fall season (October-November). Preliminary results suggest small mammals are the largest contributor to seed loss in our study; insects had a negligible effect on the proportion of seeds consumed. Using a binomial mixed-effects model with site, functional group, rarity and treatment as fixed effects and plant species as a random effect, we found that site was a significant predictor for the proportion of seeds consumed (p < 0.01), with the lowest rates of seed predation occurring in the mid-age restoration site. However, our analyses suggest that seed predation rates do not vary substantially across species rarity and functional group. Thus, species that are uncommon and rare at our sites may have a disproportionately lower chance at seedling recruitment if their seeds are consumed at a comparable rate to common species. Therefore, highly consumed species that are uncommon and rare should be seeded more aggressively in restored prairies to promote plant biodiversity and uncommon/rare species establishment.