How have we studied seed rain in grasslands and what do we need to improve for better restoration?

Climate drivers of seed rain phenology of subtropical forest communities along an elevational gradient

Seed rain phenology (the start and end date of seed rain) is an essential component of plant phenology, critical for understanding population regeneration and community dynamics. However, intra- and inter-annual changes of seed rain phenology along environmental gradients have rarely been studied and the responses of seed rain phenology to climate variations are unclear. We monitored seed rain phenology of four forest communities in four years at different elevations (900 m, 1450 m, 1650 m, 1900 m a.s.l.) of a subtropical mountain in Central China. We analyzed the spatiotemporal patterns of seed rain phenology of 29 common woody plant species (total observed species in the seed rain), and related the phenological variations to seed number and climatic variables using mixed-effect models with the correlation matrix of phylogeny. We found that changes in the period length were mainly driven by the end rather than the start date. The end date and the period length of seed rain were significantly different between the mast and non-mast seeding years, while no significant elevation-related trend was detected in seed rain phenology variation. Seed number, mean temperature in spring (T_spr), and winter (T_win), summer precipitation (P_sum) had significant effects on seed rain phenology. When T_spr increased, the start date of seed rain advanced, while the end date was delayed and the seed rain period length was mainly prolonged by a higher seed number, T_win and P_sum. Forest canopy might have a buffering effect on understory climatic conditions, especially in precipitation that lead to difference in seed rain phenology between canopy and shrub species. Our novel evidence of seed rain phenology can improve prediction of community regeneration dynamics in responding to climate changes.

Limited seed dispersability in a megadiverse OCBIL grassland

Species from OCBILs (old, climatically buffered, infertile landscapes) are expected to exhibit reduced dispersability, linked to long-term stable environmental conditions. We tested the prediction of limited dispersability in plant communities in campo rupestre, a megadiverse OCBIL grassland. In a field experiment, we sampled sticky and funnel traps for 12 months along paired preserved and disturbed plots to estimate and compare annual seed density and richness. Our community-level seed rain assessment suggested widespread seed limitation in both preserved and disturbed areas, expressed by the low seed density in traps (average of 2.6 seeds/m2/day) and remarkably low seed density found for 90% of the morphospecies. Although seed density in campo rupestre was much lower than for other grasslands, species richness (92 seed morphospecies) was the highest reported in the literature. Higher seed density and species richness were captured in funnel traps, especially during the rainy season. Approximately 32% of morphospecies were registered exclusively in disturbed plots and 25% exclusively in preserved plots. This limited seed dispersability is of particular relevance for a better understanding of seed ecology and resilience in campo rupestre and may partially explain the poor recovery of these disturbed areas after soil removal.