UC Berkeley

This dissertation combines two disparate subjects: bees and belonging. The first two chapters explore pesticide exposure for wild bees and honey bees visiting crop and non-crop plants in northern California agriculture. The final chapter utilizes surveys from a mentorship program as a case study to analyze sense of belonging among undergraduates.

The first chapter of this dissertation explores pesticide exposure for wild bees and honey bees visiting sunflower crops. Neonicotinoid seed treatments are commonly used in hybrid sunflower production, yet their impact on crop yield and pollinators, particularly wild bees, remains unclear. In this study, a matched pairs design was used to assess bee exposure to pesticides in treatment plots planted with thiamethoxam treated seed and control plots planted without thiamethoxam. Samples of field soil, sunflower pollen, sunflower nectar, pollen-foraging honey bees, nectar-foraging honey bees, and a sunflower specialist wild bee (Melissodes agilis) were analyzed for pesticide residues. The effects of thiamethoxam treated seed on bee diversity and crop yield was also quantified. Though thiamethoxam was only detected at low concentrations in one soil sample, the treatment still impacted bee diversity. M. agilis abundance and crop yield were both significantly lower in treatment plots. Conversely, wild bee richness was significantly higher in treatment plots, potentially due to resource partitioning. Furthermore, many pesticides not used in the sunflower fields were detected in the samples, some of which are known to negatively impact bees, highlighting the importance of landscape scale assessments in pesticide risk analysis. Taken together, these results suggest that thiamethoxam seed treatment in sunflower may negatively impact wild bee pollination services. Furthermore, withholding this prophylactic treatment could be economically advantageous to farmers and seed companies. Most importantly, this study exposes the necessity of including multiple metrics in pesticide risk analysis, since residue analysis for specific active ingredients may miss the broader impacts of pesticide exposure for bees.

The second chapter of this dissertation explores pesticide exposure for wild bees and honey bees visiting perimeter (i.e. non-crop) plants in an agricultural landscape. Perimeter plantings in the form of bee attractive hedgerows and forb strips are one conservation strategy to bolster bee abundance and richness in agricultural regions, but recent studies suggest that these plants may also harbor pesticides. As bee species richness is known to be higher in perimeter plantings than in the crop area itself, hedgerows and forb strips are uniquely suited for the investigation of pesticide exposure for a wide variety of bee species. In this study, a number of samples were collected within perimeter plantings for pesticide residue analysis, including: multiple species of wild bees, honey bees, flowers from four types of bee-attractive perimeter plants, and soil. This study also made use of silicone bands as passive samplers of aerial pesticide residues. A number of results arose from this study. The greatest number of pesticides were detected in aerial samples, followed by soil, however the greatest concentrations of pesticides were detected in flowers. Flowers in some cases harbored pesticides applied to crops in the greater landscape. Pesticide concentrations were also higher in bees than in soil and seemed to increase with increasing wild bee size. Honey bees and wild bees were found to be differentially exposed to pesticides and both contained pesticide mixtures of concern for bee health. These results highlight the necessity for a landscape scale approach to pesticide risk analysis, in addition to the inclusion of greater range of bee species.

The third chapter of this dissertation explores undergraduate sense of belonging as the result of a mentorship program in a large research university. Sense of belonging is a fundamental human need that precedes learning and innovation. In higher education, a strong sense of belonging is positively associated with a number of academically advantageous outcomes. Historically underrepresented student populations, such as transfer students, first-generation college students, and underrepresented minorities, are known to face greater barriers in feeling a strong sense of belonging in higher education. This study evaluated an effort to improve sense of belonging among undergraduate students through a novel academic mentoring program. The Berkeley Connect mentoring program at UC Berkeley began in 2014 and serves thousands of students each year. It is a tiered program in which faculty directors within each department mentor a small cohort of graduate students, each of whom then mentor approximately forty undergraduate students each semester. Three program elements are utilized to promote community and a sense of belonging on campus: one-on-one meetings, small-group discussions, and special events. Survey data indicated that sense of belonging increased over the course of the semester for students in the program, regardless of the demographics studied. One-on-one meetings were particularly effective for reaching program objectives. Since this mentoring program had positive impacts for large numbers of undergraduate students, including historically marginalized student populations, this may be a promising model to be applied more broadly across a range of higher education institutions.