In order to study the in situ effects of the agricultural landscape and exposure to pesticides on honey bee health, sixteen honey bee colonies were placed in four different agricultural landscapes. Those landscapes were three agricultural areas with varying levels of agricultural intensity (AG areas) and one non-agricultural area (NAG area). Colonies were monitored for different pathogen prevalence and pesticide residues over a period of one year. RT-qPCR was used to study the prevalence of seven different honey bee viruses as well as Nosema sp. in colonies located in different agricultural systems with various intensities of soybean, corn, sorghum, and cotton production. Populations of the parasitic mite Varroa destructor were also extensively monitored. Comprehensive MS-LC pesticide residue analyses were performed on samples of wax, honey, foragers, winter bees, dead bees, and crop flowers for each apiary and location. A significantly higher level of varroa loads were recorded in colonies of the AG areas, but this at least partly correlated with increased colony size and did not necessarily result from exposure to pesticides. Infections of two viruses (deformed wing virus genotype a (DWVa) and acute bee paralysis virus (ABPV)) and Nosema sp. varied among the four studied locations. The urban location significantly elevated colony pathogen loads, while AG locations significantly benefited and increased the colony weight gain. Cotton and sorghum flowers contained high concentrations of insecticide including neonicotinoids, while soybean and corn had less pesticide residues. Several events of pesticide toxicity were recorded in the AG areas, and high concentrations of neonicotinoid insecticides were detected in dead bees.
Keywords: agricultural landscape; crops; honey bees; pesticides; pollinator; varroa; virus.