The restricted range of the Ethiopian Bush-crow Zavattariornis stresemanni is a consequence of high reliance on modified habitats within narrow climatic limits

We attempt to describe and explain the peculiarly restricted distribution of the globally threatened Ethiopian Bush-crow Zavattariornis stresemanni. At a regional scale, models containing only correlates of land cover suggested a far wider distribution of suitable habitat in north-east Africa than the area actually occupied. However, models including only climate variables predicted the known distribution almost perfectly, and suggested that the species’ area of occupancy is delimited by a pocket of climate that is cooler, dryer and more seasonal than surrounding areas. The predicted probability of occurrence was low outside a narrow range of mean annual temperatures of 17.5–20°C. Within the area predicted to be climatically most suitable, records of Bush-crows were concentrated in 1-km cells of marginally but significantly lower normalised difference vegetation index (NDVI), indicating a preference for areas of lower photosynthetic activity. At a finer spatial scale within a 10-km2 intensive study site in the core of the species’ range, nests were located in 30-m cells of higher NDVI but always close to areas of lower NDVI. These areas of lower NDVI comprise open grassland, which standardised observations of individual birds showed to be the main foraging habitat. However, taller vegetation is also necessary for nesting and roosting; the average height of nests from the ground was nearly 5 m. Therefore, the species’ range appears to be defined primarily by a unique climate pocket within which it shows a preference for park-like habitats of grassland interspersed with taller vegetation, largely the result of clearance of vegetation by people and their associated grazers. The diet appeared unspecialised and a wide range of feeding methods was observed. Models estimate the species’ optimal climatic range to cover around 6,000 km2, of which perhaps 4,500 km2 has suitable land cover. We tentatively estimate the global population to be at least 9,000 breeding pairs, with a potentially larger additional population of non-breeding birds, particularly nest-helpers. Several climate models predict increases in both temperature and precipitation in the southern highlands of Ethiopia. The species’ narrow climatic range suggests that global climate change may therefore pose a serious threat to its long-term survival.