Jackson, L.E., Pulleman, M.M., Brussaard, L., Bawa, K.S., Brown, G.G., Cardoso, I.M., de Ruiter, P.C., García-Barrios, L., Hollander, A.D., Lavelle, P., Ouédraogo, E., Pascual, U., Setty, S., Smukler, S.M., Tscharntke, T., and Van Noordwijk, M.
Abstract: To examine management options for biodiversity in agricultural landscapes, eight research regions were classified into social-ecological domains, using a dataset of indicators of livelihood resources, i.e., capital assets. Potential interventions for biodiversity-based agriculture were then compared among landscapes and domains. The approach combined literature review with expert judgment by researchers working in each landscape. Each landscape was described for land use, rural livelihoods and attitudes of social actors toward biodiversity and intensification of agriculture. Principal components analysis of 40 indicators of natural, human, social, financial and physical capital for the eight landscapes showed a loss of biodiversity associated with high-input agricultural intensification. High levels of natural capital (e.g. indicators of wildland biodiversity conservation and agrobiodiversity for human needs) were positively associated with indicators of human capital, including knowledge of the flora and fauna and knowledge sharing among farmers. Three social-ecological domains were identified across the eight landscapes (Tropical Agriculture-Forest Matrix, Tropical Degrading Agroecosystem, and Temperate High-Input Commodity Agriculture) using hierarchical clustering of the indicator values. Each domain shared a set of interventions for biodiversity-based agriculture and ecological intensification that could also increase food security in the impoverished landscapes. Implementation of interventions differed greatly among the landscapes, e.g. financial capital for new farming practices in the Intensive Agriculture domain vs. developing market value chains in the other domains. This exploratory study suggests that indicators of knowledge systems should receive greater emphasis in the monitoring of biodiversity and ecosystem services, and that inventories of assets at the landscape level can inform adaptive management of agrobiodiversity-based interventions. [Copyright &y& Elsevier]