Modeling Approaches to Understanding Bumblebee Behavior and Population Decline

Heath, Kevin

Etd

Modeling Approaches to Understanding Bumblebee Behavior and Population Decline

Public Deposited

Understanding how anthropogenic disturbances affect plant–pollinator systems at the individual and population level has important implications for the conservation of biodiversity and ecosystem functioning. At the individual level, previous laboratory studies show that anthropogenic disturbances such as pesticides and pathogens, which have been implicated in the rapid global decline of pollinators, can impair behavioral processes needed for pollinators to adaptively exploit floral resources and effectively transfer pollen among plants. However, the potential for sublethal stressor effects on pollinator-plant interactions at the individual level to scale up into changes to the dynamics of plant and pollinator populations at the system level remains unclear. To address this question, we developed an empirically parameterized agent- based model of a bumblebee pollination system called SimBee to test for effects of stressor- induced decreases in the memory capacity and information processing speed of individual foragers on bee abundance, plant diversity, and bee–plant system stability over 20 virtual seasons. Modeling of a simple pollination network of a bumblebee and four co-flowering bee- pollinated plant species indicated that bee decline and plant species extinction events could occur when only 25% of the forager population showed cognitive impairment. Higher percentages of impairment caused 50% bee loss in just five virtual seasons and system-wide extinction events in less than 20 virtual seasons under some conditions. Plant species extinctions occurred regardless of bee population size, indicating that stressor-induced changes to pollinator behavior alone could drive species loss from plant communities. These findings indicate that sublethal stressor effects on pollinator behavioral mechanisms, although seemingly insignificant at the level of individuals, have the cumulative potential in principle to degrade plant–pollinator species interactions at the system level. Understanding the mechanisms behind bumblebee behavioral response to change is key to improving our ability to model and predict how pollinators respond to rapid human-induced change. Building on the work done with the SimBee model, we implement models of individual memory and decision-making to test behavioral response to simulated scenarios of rapid change. Characterizing the behavioral response to change variability of environments, probability of reward, and frequency of change provide insight into the role of memory and recency effects in adaptive decision-making. In environments with variation, memory provides an adaptive advantage to foraging bumblebees and models of decision-making that utilize memory outperform memory-less strategies. Our tests indicate that recency bias is a possible mechanism that allows bumblebees to adaptively respond to changing and variable environments when new information must be acted upon quickly. While we establish a foundation for exploring and modeling bumblebee behavior and decisions in plant-pollinator systems at the individual level, improving data collection on the dynamics of plant-pollinator interactions at the population level is critical for conservation efforts. Since long-term controlled experimental studies are difficult to execute, we utilize the citizen science Beecology project to lay foundational work for the automated classification of bumblebee behaviors in videos. Recent advances in deep learning have made rapid and accurate behavior classification of human behaviors possible, but these advances have not been applied to bumblebees. We address this by first creating a dataset of bumblebee action video clips using videos submitted by citizen scientists. The dataset was then used to train and test a two-stream convolutional network (TSN) to test the viability of using deep learning techniques for automated bumblebee behavioral classification. Our work highlights the need for a more robust dataset that can facilitate the use of deep learning architectures.

Creator