The effect of seed dispersal limitations on the spatial distribution of a gap species, seaside goldenrod (Solidago sempervirens)

Abstract

This study examined the seed ecology of seaside goldenrod, Solidago sempervirens L., a gap species, on coastal dunes dominated by American beachgrass, Ammophila breviligulata. A test was done to determine whether lower densities of S. sempervirens on undisturbed primary dune grasslands were due to limited seed dispersal or barriers to seedling establishment. Seed rain was greater within blowouts and transition areas than in immature and mature grasslands. Blowouts and transition areas produced prominent seed shadows and served as primary seed sources. Both seed shadow measurements and inflorescence removal experiments suggested that the majority of seeds fell to the ground within 10 m outside blowouts. Experiments involving the hand broadcasting of seeds increased densities of seedlings in undisturbed areas. Plots with hand-drilled seeds showed no differences in percent emergence between transition and undisturbed areas; however, both were significantly greater than blowouts. Persistent seed storage in the soil was probably negligible. Seed bank densities at 5 m depth did not accurately reflect temporal or spatial patterns of seed rain. Often very low densities or no seeds were found in collected samples within an area. Furthermore, emergence percentages of field-buried seeds showed a significant drop-off after a single year. Thus, this study concluded that lower densities of plants in undisturbed areas were, in part, due to lack of seed dispersal from blowouts. Key words: seed dispersal, gap species, patchy disturbances, dune ecology, grassland ecology.