Habitat suitability modelling to improve conservation status of two critically endangered endemic Afromontane forest bird species in Taita Hills, Kenya

Abstract

Tropical montane forests are known to support many endemic species with restricted geographic ranges. Many of these species are however, faced with numerous threats, most notably from habitat loss and degradation, invasive alien species, and climate change. Examples include Taita Apalis and Taita Thrush. Taita Apalis (Apalis fuscigularis) and Taita Thrush (Turdus helleri) are species of birds listed as Critically Endangered by the Government of Kenya and the International Union for Conservation of Nature (IUCN). They are endemic to Taita Hills’ cloud forests in southeastern Kenya and protected under Wildlife Conservation and Management Act. As they face high risk of extinction, exploring their habitat suitability is imperative for their protection. To determine the current spatial distribution and the key ecogeographical explanatory factors and conditions affecting species distribution and indirect effects on species survival and reproduction, we employed Maximum Entropy (MaxEnt) modelling. This study was conducted in Ngangao and Vuria forests in June and July 2019 and 2020. Ngangao forest is gazetted as forest reserve and managed by the Kenya Forest Service whereas Vuria is non-gazetted and thus remains without official protection status. Ecogeographical explanatory variables; climatic, remote sensing-, LIDAR-, topography- and landscape-based variables were used in modelling and separate models were produced. 23 occurrence records of Taita Apalis and 30 of Taita Thrush from Ngangao and 21 of Taita Apalis from Vuria forests were used in the modelling. According to the models, less than 7% of the total area of Ngangao and Vuria forests was predicted as suitable habitat for Taita Apalis and Taita Thrush. This shows that these two species are more vulnerable to extinction from demographic stochasticity. Consequently, managing their habitats is critical for their long-term persistence. LIDAR-based canopy height range and elevation greatly influenced Taita Apalis distribution in Ngangao forest, with areas of high elevation (1620–1750 m a.s.l.) and having open middle-storey preferred. Elevation, slope and topographic wetness index (twi) were the major determinants of Taita Thrush distribution in Ngangao, where gentle sloping areas with moderately dry surfaces within high elevation (1620–1730 m a.s.l.) were favoured. Mean annual temperature, Euclidean distance to the forest edge, slope and land cover type greatly influenced the distribution of Taita Apalis in Vuria, with gentle sloping areas within forest interior made up of indigenous vegetation preferred. This study proposes reforesting open and degraded sites next to areas predicted as highly suitable for the two species; establishment of agroforestry belts based on indigenous trees on the boundaries of the two forests to reduce grazing and firewood collection pressure and enhance resilience to the edge effects; and enhancing forest protection through Participatory Forest Management.