The Role of Connectivity and Disturbance in Dynamic Landscapes: Squash Bees (Peponapis pruinosa) in Yolo County, California

Many species inhabit dynamic landscapes where habitat patches are destroyed and recreated. Despite the challenging conditions of these landscapes, where resources vary in space and time and species may go locally extinct as a result of disturbance, some species persist. Agricultural landscapes cover much of the earth's area and undergo regular disturbance, for example in the form of crop rotations and tilling. For my dissertation, I explored how these factors contribute to squash bee (Peponapis pruinosa) population persistence or extinction. Peponapis pruinosa is a univoltine, ground nesting bee. It specializes on and preferentially nests in soil beneath squash and pumpkin vines (Cucurbita spp.) that are grown as part of annual crop rotations. Chapter One investigates the role of within and between-year connectivity among neighboring squash fields in explaining P. pruinosa densities at 37 fields in 2010 and 2011. I found that between-year connectivity better explained P. pruinosa densities than within-year connectivity. This suggests that P. pruinosa is tracking its resources over time and that maintaining connectivity between years may contribute to P. pruinosa persistence within this landscape. In Chapter Two I predicted that sequential connectivity over multiple years (>2) may be a stronger driver than between-year connectivity in explaining P. pruinosa densities. In addition, I explored whether female and male P. pruinosa respond differently to connectivity. I calculated P. pruinosa density at 65 squash fields in 2012 and used generalized linear models to understand the role of sequential connectivity of squash fields across a three year period, as well as sampling date, sampling time, flower density, and field size in determining bee density. I found that connectivity was not a consistent driver of P. pruinosa densities across multiple years. This suggests that, within dynamic landscapes, the role of connectivity over time is, itself, variable. Connectivity may influence bee density in some years, but other factors likely outweigh the effect of connectivity in other years. In Chapter Three, I conducted an experiment to test the effect of tillage, a factor which may influence between year P. pruinosa persistence in agricultural landscapes. I established artificial P. pruinosa nesting aggregations in twenty mesh cages, each enclosing 9 square meters of squash field. I then tilled half of those nesting aggregations and, the following year, collected the offspring that emerged from each experimental cage. Using a Bayesian analysis, I found some evidence that end of season tilling results in reduced P. pruinosa offspring emergence the following year. In addition, I found that tillage did not affect the sex ratio of P. pruinosa offspring, but did cause females to emerge later than those emerging in control plots. Taken together, my dissertation shows that in agricultural landscapes, spatial and temporal connectivity among squash fields sometimes, but not always, benefits P. pruinosa and may favor persistence across the landscape. In contrast, standard tillage practices can be detrimental and may ultimately reduce P. pruinosa persistence.