Ecological and Social Drivers of Tree Diversity in Coffee Agroforestry Systems

Abstract

In the face of biodiversity loss due to agricultural expansion and intensification, agroforestry has been proposed as an environmentally friendly form of agriculture capable of conserving biodiversity while supporting local livelihoods. However, how social drivers related to farmers’ decision-making and ecological processes affect the potential of agroforestry systems to serve as reservoirs for native species diversity and community composition is unclear. This dissertation aims to describe patterns of tree diversity and community composition in coffee (Coffea arabica) agroforestry systems as they compare to surrounding forests, and uncover the social drivers related to farmers’ decision-making and ecological processes giving rise to those patterns. Worldwide, there is an extensive overlap between coffee-growing areas and regions with high species richness and endemism considered biodiversity hotspots. This renders the issue of clarifying the sustainability and conservation value of shade coffee even more urgent. Otherwise, we risk losing important late-succession and conservation concern tree species, and simplifying the structural and floristic composition of mature forests. To uncover how the social factors related to farmers’ decision-making and ecological factors drive tree diversity and community composition in coffee agroforestry systems, a series of empirical studies were conducted based on surveys and field data collected in La Sepultura Biosphere Reserve in Chiapas, Mexico, between 2009-2013. Field research took place in two coffee farming communities within the reserve, where 50 farmers were interviewed, and 31 coffee agroforest sites and 10 forest sites were sampled. Forests were sampled to provide a reference for tree diversity and community composition in the absence of coffee management. Although a higher sample size would have permitted the inclusion of more covariates in statistical analyses without losing statistical power, there is no reason to believe that the results of this dissertation would change if a larger sample size were considered. This is because the magnitude of the effects detected were large and the p-values small, which underscore the robustness of the results of this dissertation. The three chapters in this dissertation correspond to individual studies. Chapter One investigates tree floristic patterns in coffee agroforestry systems and compares them to those in surrounding forests. Findings indicate that although at the landscape level coffee agroforestry systems hold similar tree species richness compared to forests, tree species community composition is significantly distinct. Coffee agroforests had a lower proportion of trees of conservation concern, a higher proportion of pioneer trees, were dominated by Inga spp., harbored lower tree species diversity at the plot level, and were composed of different tree species compared to native forests. This chapter raised questions with regards to whether these results were the result of farmers’ intentional tree selection criteria and preferences, or an unintended consequence of management practices. Chapter Two addressed these questions by examining the use of knowledge by farmers to manage coffee agroforests and the consequences on tree community composition relative to forests. In Chapter Two, results indicate that differences between agroforests and forests are primarily driven by farmers’ manipulation of tree community composition, which occurs according to their beliefs about the benefits and disservices of trees for coffee production. Tree community composition in coffee agroforest is dominated by the trees that farmers prefer and practically void of the trees they dislike as compared to the trees’ natural abundances in forests. These findings are novel and important because they clarify that the community composition changes observed in coffee agroforests are mostly an intentional consequence of management and not a byproduct. Finally, Chapter Three focuses on a subset of trees of particular conservation importance, trees of conservation concern (CC) and typical of old growth or late succession (LS) forests. This chapter investigates how management practices that affect shade tree density, basal area, and the proportion of Inga trees, mediated by land use legacies, affect the proportions of CC and LS trees in coffee agroforests. Findings indicate that management practices that sought to increase the proportion of Inga spp. trees had the largest negative impact on the proportions of trees of LS and CC, but the magnitude of the effects were dependent on land-use legacy. Among farms established on land previously used for pastureland or crop cultivation, the impact of farmers’ tree preferences and selection criteria on LS and CC trees were significantly higher than on farms established on forests without an agricultural history. These findings underscore that farmers’ sharp preference for Inga spp. trees undermines the potential of agroforests to conserve higher proportions of CC and LS trees. The results presented in each chapter of this dissertation allow for a more thorough understanding of the tree diversity patterns conserved in coffee agroforestry systems and the underlying social drivers related to farmers’ decision-making and ecological drivers that generate such patterns. The results of this dissertation seek to contribute new knowledge not only to the scientific community, but also to society so that better policies and strategies be devised that successfully conserve floristic diversity in the biodiverse areas of the world where coffee is cultivated.

Abstract from the paper’s OpenAlex record, shown for indexing and research. Copyright remains with the authors or the publisher. Read at the source ↗

The text the model read, with the source record’s inline markup (italics, sub- and superscripts) rendered and its character entities decoded; quote offsets refer to the raw text.