Thu, Aug 05, 2021:On Demand
Background/Question/Methods
Agricultural intensification poses a threat to wildlife conservation, pollinators in particular. In this study, I examine the distribution of pollination services in Costa Rican (CR) coffee agroecosystems using spatial models. Bee pollinators can significantly increase the yield of coffee, a crop that is critical to CR’s culture and economy, providing an incentive to understand how to support pollinator-friendly coffee farms. Coffee is farmed along a wide spectrum of management strategies in CR, which impact the provisioning of pollination services. Spatial models are a useful tool, providing an opportunity to quantify ecosystem service (ES) benefits of pollination. In this study, I use the InVEST Crop Pollination Model, an ES modeling tool, to address the following questions: Where are predicted pollinator-friendly coffee farms? How does landscape context and farm management impact pollinator-friendliness of a coffee farm? How does isolation from natural areas impact pollination services on coffee farms? I use the original InVEST model to address these questions, and I also adapt its parameters to account for the diversity in coffee farm management common in CR.
Results/Conclusions These spatial models suggest that the predicted pollinator-friendliness of coffee farms varies greatly depending on both landscape context (i.e. farm surroundings) and farm management strategies (i.e. shady agroforests vs. sunny monocultures). The most pollinator-friendly farms predicted with these models are located furthest from urban areas, near forest, which aligns with previous empirical work. By simulating different farm management strategies within the coffee land-use class, a novel extension of this InVEST model, we learn that the model is very sensitive to changes in coffee-dominant landscapes. More specifically, when parameters were manipulated to simulate “sun” and “shade” coffee scenarios, the pollinator-friendliness of some farms varied greatly whereas others did not see great change. Shade trees can provide critical nesting and floral resources for bees within coffee farms that help buffer the detrimental effects of isolation from native forest pollinator habitat, which my adapted model was able to characterize for the first time. This application of the InVEST Crop Pollination model provides insight into how to account for hidden ecological heterogeneity that may be traditionally obscured with spatial models. Spatial models can help us prioritize areas of conservation concern in agricultural landscapes to plan to preserve biodiversity when farming. A priority should be gathering more empirical data to ground-truth spatial models and understand their applicability in different farming landscapes.
Results/Conclusions These spatial models suggest that the predicted pollinator-friendliness of coffee farms varies greatly depending on both landscape context (i.e. farm surroundings) and farm management strategies (i.e. shady agroforests vs. sunny monocultures). The most pollinator-friendly farms predicted with these models are located furthest from urban areas, near forest, which aligns with previous empirical work. By simulating different farm management strategies within the coffee land-use class, a novel extension of this InVEST model, we learn that the model is very sensitive to changes in coffee-dominant landscapes. More specifically, when parameters were manipulated to simulate “sun” and “shade” coffee scenarios, the pollinator-friendliness of some farms varied greatly whereas others did not see great change. Shade trees can provide critical nesting and floral resources for bees within coffee farms that help buffer the detrimental effects of isolation from native forest pollinator habitat, which my adapted model was able to characterize for the first time. This application of the InVEST Crop Pollination model provides insight into how to account for hidden ecological heterogeneity that may be traditionally obscured with spatial models. Spatial models can help us prioritize areas of conservation concern in agricultural landscapes to plan to preserve biodiversity when farming. A priority should be gathering more empirical data to ground-truth spatial models and understand their applicability in different farming landscapes.