Agroforestry management, seasonal changes, biodiversity and multitrophic interactions of coffee arthropods

Abstract

Ongoing landscape anthropogenic modification in the tropics has caused biodiversity losses of many taxa, threatening key ecosystem processes. Traditional coffee agroforests, which are the only forest-like vegetation remaining in some regions of the tropics, have been emphasized as important refuges for biodiversity and may also contribute to pest supression due its complex vegetational structure. Agroforestry management largely varies among coffee farmers and may influence arthropod population patterns through changes in environmental conditions. Ecological interaction experiments between organisms are often used to predict and explain field patterns and processes. However, many population-driving factors such as environmental variables are determined by land-use management, making real world field patterns difficult to predict from small-scale experiments. The study was conducted on private farms located in the coffee growing region of Jipijapa, province of Manabi, Ecuador. The region is considered a hotspot for biodiversity conservation and its original vegetation has been converted to agriculture. We selected three coffee agroforestry types according to local management: simple-shade agroforests with low shade tree diversity, complex-shade agroforests with intermediate shade tree diversity, and abandoned coffee agroforests with high shade tree diversity. We assessed the role of agroforestry management and season on influencing the community of coffee-inhabiting mites (Chapter 2). We show that management intensification, from abandoned, complex-shade to simple-shade agroforests negatively affected species richness of mites on coffee. Further, more species and higher densities of mites were found in the dry season compared to the rainy season. We investigated the population dynamics of coffee pests (i.e. spider mites, leaf miners, and berry borers) in relation to agroforestry management. Furthermore, we assessed how specific developmental stages of coffee pests respond to agroforestry management (Chapter 3). We show that agroforestry management affects seasonal patterns of coffee pests with higher peaks being reached in more intensively managed agroforests. Further, there was an interacting effect of vegetational diversity and developmental stages of coffee pests. Overall, specific pest developmental stages built up higher densities in more intensively managed agroforests. We determined the relative importance of single environmental habitat variables on densities of coffee pests using hierarchical partitioning analyses (Chapter 4). We found that the environmental habitat variables temperature, humidity, and tree diversity explained most of the variation of coffee pest densities. Spider mite density was positively related to temperature, while leaf miner and berry borer densities were negatively correlated to humidity and tree diversity, respectively. Finally, we combined laboratory and field experiments to study the relative importance of biotic interactions and agroforestry management on densities of coffee pests in the field (Chapter 5). We show that preference matched fitness under laboratory conditions, leading to the prediction that coffee pests would avoid conspecific and heterospecific competitors in the field. However, coffee pest densites were positively correlated in the field, thereby contrasting with predictions made from the laboratory. Moreover, management intensity, which was defined by abiotic (temperature, humidity) and biotic (canopy cover, tree diversity, coffee density) habitat variables, proved to be more important than preference and fitness for influencing densities of coffee pests in the field. In conclusion, vegetationally diverse agroecosystems such as complex-shade and abandoned agroforests have the potential to conserve mite diversity, and therefore should be incorporated into landscape conservation programs. Seasonal and population structure need to be taken into account in arthropod population dynamic studies across different habitat types, because responses depended on season and developmental stages. Abiotic and biotic habitat variables determining a given habitat may play a key role for understanding the effects of land use on pest densities. Understanding such species density-environmental relationships provide insights on how to predict and manage pests in the field. Laboratory and field studies need to be combined in order to determine the relative importance of small and large spatial scale experiments influecing population densities in the field.