Seed rain and seed bank dynamics in a remnant and chronosequence of tallgrass prairie restorations

Background/Question/Methods

Soil seed banks are natural reservoirs for ungerminated species post-dispersal (i.e., seed rain). Plant community dynamics are likely influenced by the composition and abundance of viable seeds stored in seed banks because seed banks are a major recruitment source. However, paired seed rain and seed bank data from tallgrass prairie ecosystems are limited, leaving the post-dispersal fates of seeds in these systems unclear. Because prairie restorations often do not reach comparable levels of biodiversity to remnant prairies, understanding the transition of seeds from seed rain to bank is critical for improving restoration efforts. We investigated the relationship between the seed rain and seed bank in a remnant tallgrass prairie and a chronosequence (2, 6-7, and 15-year-old) of prairie restorations in north-central Missouri. To assess seed rain, we deployed 50 artificial turf grass seed traps (0.1 x 0.1 m) per prairie in May 2019. We collected and replaced seed traps every two weeks from June to December 2019. In March 2020, we collected 40 soil samples (1500 cm3) to assess the seed bank. So far, we have collected over 88,153 seeds and germinated 4,707 seedlings representing at least 123 and 81 morphospecies, respectively.

Results/Conclusions

Preliminary results suggest clear differences in the number and composition of dispersed and germinable seeds between remnant and restored prairies. We found that only the youngest restoration had significantly more seeds falling (p < 0.001) and germinating (p < 0.001) compared to the remnant. In agreement with previous studies, we saw a sharp reduction in the number of germinable seeds in the seed bank compared to those falling in the seed rain for all prairies. There was a 76.9 - 85.9% loss in seeds when transitioning from seed rain to seed bank, with the oldest restoration and remnant prairie having the highest seed mortality rates. PERMANOVA analysis revealed that even 15+ years after initial seeding, seed rain (p < 0.01) and seed bank (p < 0.01) composition significantly differed between the remnant and restored prairies. Due to the close connection between the seed rain and bank to plant species' colonization, establishment, and persistence, our findings suggest that these restored prairies will remain compositionally divergent from remnants without outside intervention. Overall, our study provides additional evidence that restorations have not fully emulated the ecological processes and species assemblages seen at remnant systems.