Avian Persistence in Fragmented Rainforest

Loss and deterioraton of indigenous habitat increasingly affect natural populations worldwide. As a result of these processes, new selection pressures are imposed upon organisms, increasing local extinction rates. Simultaneously, reduced movement among remnant patches lowers colonisation rates and affects demographic and genetic population parameters. Yet, organisms with comparable life histories often respond to habitat disturbance in various ways. Why so is a matter of great importance to evolutionists and conservationists alike.

To address the question what factors determine the persistence of species in fragmented habitats, an international team led by Belgian ecologist Luc Lens studied the relative impacts of forest deterioration and fragmentation on the persistence of eight forest-restricted bird species within 430 ha of rainforest remnants in south-east Kenya. Three species are endemic to the Taita Hills, which is part of the Eastern Arc biodiversity hotspot. Over the past decades, the indigenous forest has been reduced to 12 patches, of which only the three largest ones (94-179 ha) are inhabited by all study species. The nine other remnants are tiny (1-8 ha) and heavily disturbed, and host breeding populations of a subset of species only.

The researchers used data collected during six years of trapping, marking, and recapturing more than 3,000 birds to estimate species-specific ability to move among the forest remnants. To estimate stress tolerance, the team relied on earlier studies showing that when birds are under stress, bones in the hind limbs grow longer on one side than on the other. It was determined which species suffered the most stress by comparing measurements of modern birds to those of museum specimens captured when the forest was relatively undisturbed. Based on these estimates, it was shown that more mobile species occupied a higher proportion of patches than expected from their estimated stress sensitivity. Likewise, less sensitive species occupied a higher proportion of patches than predicted from their estimated level of mobility. Together, dispersal rate and change in asymmetry explained an astonishing 88% of the observed variation in patch occupancy between the eight study species.

Results from this study are in agreement with the metapopulation concept in biology, which states that at a landscape level, the presence of species in habitat patches reflects a balance between chances of local extinction and chances of colonization of empty patches. The important lesson to learn for conservation biologists is that long-term conservation tactics may fail unless they include action both within sites, to minimize habitat deterioration, and across landscapes, to maximize dispersal.