Can we extend the area of occupancy of the kipunji, a critically endangered African primate?

摘要

The recently discovered and critically endangered kipunji Rungwecebus kipunji is known from just two sites in southern Tanzania, with the bulk of its tiny population surviving as patchily distributed subpopulations within the Mt. Rungwe-Livingstone (RL) montane forests. We investigated the kipunji's habitat associations to determine its usage of different forest types, to predict occupancy across the study area and to identify unoccupied areas that might, with appropriate management, offer the best range of expansion possibilities. We developed kipunji-habitat-use models using 14 floristic, structural and geographical predictors for three different methods (non-spatial logistic regression, autoregressive and generalized estimating equations). The best models were selected based on Akaike Information Criterion minimization, and used to develop a habitat suitability model with interpolation techniques. Kipunji were associated with a closed canopy at mid to lower altitudes, and with specific tree communities, as highlighted by the interaction between canopy cover and a floristic axis, where fragmentation may be tolerated as long as resources are available. Predictions of habitat suitability closely matched those of current occurrence (42 km2), with little room for expansion within forests around their known range. Interpolation of occurrence probabilities suggested that the original habitat in areas located to the south of the current range, now deforested, would have been highly suitable for kipunji. Conservation management should concentrate on facilitating range expansion and connectivity between subpopulations to secure future kipunji numbers and genetic variation. Steps should be taken to improve forest quality and connectivity within the current protected areas, including (but not limited to) the Bujingijila corridor, which splits the kipunji subpopulations of RL. Kipunji's tolerance of some habitat disturbance suggests that restored forests do not necessarily need to be pristine to benefit the species. Additional management might consider the restoration of some of kipunji's former range by reforesting suitable non-forested habitat adjacent to their current range.Digital Object Identifier http://dx.doi.org/10.1111/j.1469-1795.2011.00474.x