Search

Agricultural intensification is leading to conversion of cocoa agroforestry towards monocultures across the tropics. In the context of cocoa agriculture, arthropods provide a range of ecosystem services and dis‐services. Arthropod pests (e.g., mirids and mealybugs) can cause major damage to crops, whilst pollinators and natural enemies (e.g., predatory insects and parasitoids) have the potential to enhance agricultural yields. Understanding how intensification of cocoa farming affects different arthropod groups is therefore important in maximising the abundance of beneficial arthropod taxa and reducing pest burdens. However, little is known about the influences of agricultural intensification on tropical arthropod communities, especially in Africa, where ~70% of the world's cocoa is produced.Most research on arthropod communities considers data from different sampling methods separately, as proxies of abundance; whilst these proxies can be informative, estimating true abundance enables direct comparison between arthropod taxa, and therefore the study of community dynamics. Here, we develop a Bayesian hierarchical model that integrates data from three common arthropod survey techniques to estimate population size of arthropod orders and to investigate how arthropod community composition responds to farm shade cover (an indicator of management intensity).Our results show that eight of 11 arthropod taxa responded to farm shade cover; importantly, brown capsids (the primary pest of cocoa in Africa), Coleoptera pests and Hemiptera pests decreased with increasing farm shade cover, whilst Araneae (natural enemies) and Diptera (potential pollinators) were more abundant in shady farms.Synthesis and applications. To achieve lower pest burdens and higher abundances of potential pollinators and natural enemies, African cocoa farms should maintain a dense canopy of shade trees. The current shift towards high‐intensity cocoa farming in Africa could result in long‐term losses due to pest infestations and loss of arthropod‐mediated ecosystem services. [ABSTRACT FROM AUTHOR]

1. Like many wildlife populations across Africa, recent analyses indicate that African lions are declining rapidly outside of small fenced areas. Community conservancies – privately protected areas that engage community members in conservation – may potentially maintain wildlife populations in unfenced pastoralist regions, but their effectiveness in conserving large carnivores has been largely unknown until now. 2. We identify drivers of lion survival in community conservancies within the Masai Mara ecosystem, Kenya, applying mark–recapture analyses to continuous sight–resight surveys. We use the number of livestock and human settlements as proxies for potential human–lion conflict whilst controlling for environmental variables and lion socioecology. 3. Average lion densities within the Mara conservancies between 2008 and 2013 (11·87 lions 100 km-2) were 2·6 times higher than those previously reported in 2003. Survival rates varied amongst prides and were highest for lions utilizing central regions of conservancies. 4. The number of livestock settlements (bomas [corrals] and manyattas) that were not members of a conservancy, and that fell within a pride home-range, had a large negative effect on female survival and was the most influential external predictor in models. These nonconservancy settlements accounted for 37·2% of total observed variation in survival, whereas conservancy settlements, which benefit financially from wildlife through their membership, had no effect on lion survival. 5. Internal drivers of survival agreed with known ecology and social behaviour including age plus a negative effect of male takeovers on cub survival <6 months. Vegetation cover, prey availability and the density of grazing livestock within a pride’s range did not explain patterns in lion survival. 6. Synthesis and applications. We show that lion densities have increased substantially within the Mara conservancies over the last decade and suggest that the creation of community conservancies has benefitted their survival. This suggests that lions can survive outside of fenced areas within pastoral regions if communities gain benefits from wildlife. We highlight the importance of expanding existing conservancies beyond their current geographical and political scope and forming buffer zones if wildlife ranges outside them. We suggest that changing attitudes to predators should be a key goal of community-based conservancies. Further work is recommended to identify what specific aspects of conservancy membership promote lion survival. [ABSTRACT FROM AUTHOR]