Not in wilderness: African vulture strongholds remain in areas with high human density

Abstract

Vultures constitute an important functional group in many ecosystems, providing crucial ecosystem services both in natural and humanized environments. These scavengers are facing massive declines worldwide, but in several African countries virtually nothing is known on populations' status and threats, hampering the development of adequate conservation strategies. In Guinea-Bissau, globally important populations of Hooded Necrosyrtes monachus and African white-backed vultures Gyps africanus were recently reported. Using the country as a study area, we aim to characterize human-vulture interactions in West Africa applying a multidisciplinary approach. We assessed the status and distribution of vulture populations using data from 1711 km of roadside transects, examined predictors of their distribution, and produced a nationwide population estimate for the Hooded Vulture, using an innovative method based on the relationship between the size of human population in settlements and vulture numbers. We conducted 47 stakeholder interviews to assess perceived roles played by vultures, and to investigate potential anthropogenic threats. Hooded vultures were strongly associated with high human population densities, whereas no relation was found between African white-backed and Rüppell's vultures and any of the tested predictors, which included cattle density, precipitation and Normalized Difference Vegetation Index, among others. We estimate a national population of 43347 Hooded vultures, the largest population reported in the species range. Respondents were generally aware of the services provided by vultures, especially waste and carcass removal, including in urban areas. Hunting for witchcraft and traditional medicine was the most frequently recognised threat, while poisoning was ranked as having the highest impact. We hypothesise that poisoning-related mortality may be affecting African white-backed and Rüppell's vultures' distribution and explain their scarcity in apparently highly suitable habitats. Our results suggest a mutualistic rather than a commensalistic relationship between vultures and humans, with important implications for designing and implementing conservation strategies.

Fig 1. Map of the study area.

Areas without red lines correspond to Sectors (an administrative division within Guinea-Bissau) that were not sampled for the study of vulture distribution and abundance (i.e. no road surveys were conducted). Orange areas are Sectors where interviews were carried out for the social surveying component of our study. The four major cities of the country are represented by their names in black writing.

Fig 2. Abundance of Hooded vultures.

Abundance of Hooded vultures Necrosyrtes monachus in sampled Sectors of Guinea-Bissau obtained from road survey transects. Grey areas represent Sectors that were not sampled (yet Hooded vultures are present and widespread in the Bijagós archipelago; authors pers. obs.).

Fig 3. Abundance of African white-backed and Rüppell’s vultures.

Abundance of Gyps spp. in sampled Sectors of Guinea-Bissau obtained from road survey transects. Gyps spp. consist mostly of African white-backed vultures. Grey areas represent Sectors that were not sampled (Gyps spp. were never observed in the Bijagós archipelago; authors pers. obs.).

Fig 4. Distribution over habitats of most abundant vulture species.

Distribution over habitats of most abundant vulture species during transect surveys in Guinea-Bissau mainland. (NI–Non-identified habitat). Gyps spp. consist mostly of African white-backed vultures.

Fig 5. Perceived roles played by vultures in Guinea Bissau according to stakeholders.

Perceived roles played by vultures in Guinea Bissau according to stakeholders, showing the frequency (in number of answers; n = 46) of interviews to respondents that regarded vultures as positive, negative, indifferent or irrelevant for each of the referred services.

Fig 6. Threat analysis exercise.

Threat analysis exercise, showing the number of mentions of each threat. In each interview, respondents could mention more than one threat, thus total exceeds sample size/number of interviews (n = 23).