Background: Seedling community assembly after disturbance sets the trajectory for future forest composition. Seedling recruitment depends on seeds reaching a given area and then successfully navigating the gauntlet of abiotic and biotic challenges in that spot. The importance of the dispersal step in this process is often under-appreciated, perhaps because the vast majority of seeds that reach an area are unable to establish.
Hypothesis: Seed dispersers are an important biotic filter affecting plant community assembly, therefore the composition of frugivores in a forest community influences seedling recruitment patterns and ultimately sets the trajectory for future forest composition.
Methods: To assess the influence of dispersers on forest regeneration, we compared seedling regeneration across three islands that vary in their frugivore communities. Guam has lost nearly all frugivores due to predation by a non-native snake, leaving only feral pigs for dispersal. The nearby island of Rota has a depauperate bird community but still has fruit bats, and the island of Saipan has a relatively intact bird community, but no bats or pigs. We performed belt transect censuses of the understory and adult trees across highly degraded, regenerating, and intact forest on all three islands. For each belt transect, we recorded the abundance of seedlings, saplings, and adults. To understand if seeds were likely dispersed from a nearby adult via gravity or from afar through animal-mediated seed dispersal, we determined whether each seedling and sapling had an adult conspecific within 2 meters.
Results: The disperser community on each island is reflected in the seedling community composition. First, there are few seedlings in the intact forest away from conspecific adults on Guam, compared to other islands. This is problematic for the trajectory of intact forest on Guam, as the future is likely to be less diverse. The opposite pattern is seen in the highly degraded forest, where there are more dispersed seedlings on Guam than on nearby islands, likely due to the pervasive impacts of pigs. However, many of these dispersed seedlings are from non-native species, foretelling a future forest dominated by non-natives. Rota, which is the only island with a significant bat population, shows a clear signal of bats with increased richness and seedling abundance of bat-dispersed species in intact forest.
Implications: This accidental experiment highlights the importance of dispersal as the first step of community assembly and demonstrates the critical importance of considering the frugivore community composition in passive forest restoration projects.
seed dispersal, community assembly, islands, seedling recruitment, frugivore, forest