Seed dispersal to revegetated isolated rainforest patches in North Queensland

Abstract

Vegetation structure, floristics and distance from a seed source influence dispersal of seeds to revegetated sites. This study investigates the effects of distance from a seed source on dispersal of seeds to three restored rainforest sites (all approximately 10 years old) on the Atherton Tablelands, North Queensland. The three sites were located at varying distances from the nearest seed source, one being adjacent to a rainforest reserve and two being surrounded by pasture and isolated from the nearest rainforest by distances of 600 m and 2 km. Floristics and vegetation structure (LAI, lateral cover at ground level, stem density and groundcover) were similar between sites, suggesting that the three sites would be equally suitable for use by seed-dispersing fauna. Differences in propagule dispersal to the three sites were therefore likely attributable to distance from seed source rather than vegetation structure. Over the decade since establishment a far higher number of native species—dispersed by flying fauna, ground-dwelling mammals, wind and varied dispersal agents—had been recruited (in higher densities) to the adjacent site than to either of the isolated sites. The two isolated sites had far higher levels of exotic recruits (characteristic of early successional stages) than did the adjacent site, owing to the proximity of these sites to pasture containing exotic grasses and forbs. A very small number of recruited species were extremely abundant at each site, while the majority of new species were rare, this pattern being particularly pronounced at the distant sites. Owing to isolation from a seed source, recruitment of species typical of intermediate and late-successional stages is occurring far more slowly at both isolated sites compared with the adjacent site. Recovery of these distant sites to self-sustaining mature rainforest may not occur without human mediated seed dispersal. Mammal-dispersed species in particular are unlikely to reach these sites unless contributed by humans. Recovery of tropical rainforest systems is likely to occur more quickly and be more successful if restoration is concentrated around the perimeter of existing rainforest rather than in isolated patches.