In this paper, we analyze databases [corrected] on birds and insects to assess patterns of functional diversity in human-dominated landscapes in the tropics. A perspective from developed landscapes is essential for understanding remnant natural ecosystems, because most species experience their surroundings at spatial scales beyond the plot level, and spillover between natural and managed ecosystems is common. Agricultural bird species have greater habitat and diet breadth than forest species. Based on a global data base, bird assemblages in tropical agroforest ecosystems were composed of disproportionately more frugivorous and nectarivorous, but fewer insectivorous bird species compared with forest. Similarly, insect predators of plant-feeding arthropods were more diverse in Ecuadorian agroforest and forest compared with rice and pasture, while, in Indonesia, bee diversity was also higher in forested habitats. Hence, diversity of insectivorous birds and insect predators as well as bee pollinators declined with agricultural transformation. In contrast, with increasing agricultural intensification, avian pollinators and seed dispersers initially increase then decrease in proportion. It is well established that the proximity of agricultural habitats to forests has a strong influence on the functional diversity of agroecosystems. Community similarity is higher among agricultural systems than in natural habitats and higher in simple than in complex landscapes for both birds and insects, so natural communities, low-intensity agriculture, and heterogeneous landscapes appear to be critical in the preservation of beta diversity. We require a better understanding of the relative role of landscape composition and the spatial configuration of landscape elements in affecting spillover of functionally important species across managed and natural habitats. This is important for data-based management of tropical human-dominated landscapes sustaining the capacity of communities to reorganize after disturbance and to ensure ecological functioning.

No abstract

Shade coffee certification programs have emerged over the past six years to verify that coffee marketed as ''shade grown'' is actually grown on farms that provide higher quality habitat for biodiversity. In spite of good intentions and an increasing market, little consensus exists on whether current criteria can successfully identify coffee farms of conservation significance. This paper provides the first ecological evaluation and comparison of shade-grown coffee criteria used by major certification programs. Using vegetative data, we evaluated criteria developed by the Rainforest Alliance, the Smithsonian Migratory Bird Center (SMBC), and the Specialty Coffee Association of America across a range of coffee agroecosystems in Chiapas, Mexico, to determine which management practices each program would certify. Fruit-feeding butterflies and forest bird species found in these coffee agroecosystems were compared with nearby forest reserves as indicators of biodiversity and conservation potential. These agroecosystems fall into three categories: rustic, commercial polyculture, and shaded monoculture. The rustic system contained significantly higher fruit-feeding butterfly diversity and an avifauna more similar to that found in forest reserves than the other systems. This was also the only agroecosystem that met the criteria for all certification programs, while the shaded monoculture fell short of all sets of criteria. This suggests that certification programs are succeeding in discriminating between the extremes of shade coffee production. Certification programs differed, however, in their treatment of the intermediate, commercial polyculture systems, reflecting different philosophies for conservation in managed ecosystems. Programs promoted by SMBC use high standards that would exclude all but the most diverse commercial polyculture or rustic systems to certify only those systems that support high levels of biodiversity. The program supported by the Rainforest Alliance only excludes the shaded monoculture while engaging the others in the move toward greater sustainability. The merits of each approach should be put to rigorous debate, and their ability to contribute to biodiversity conservation should be reflected in product marketing. This study suggests that further research can provide a stronger scientific basis and independent verification for the certification of green products that claim to enhance biodiversity conservation in tropical agroecosystems.