An assessment of pollinator abundances and environmental relationships at cultivar, field, and landscape scales in cultivated pumpkins on the Texas High Plains

In Texas, pumpkins (Cucurbita pepo) are alternative crops with 5,000 to 8,000 acres grown annually. Common practices for pumpkin production include hive rentals for supplemental pollination services, usually from managed honey bees (Apis mellifera) and bumble bees (Bombus impatiens). For production of cucurbits, native squash bees can supplement managed bees or be the sole facilitator of pollination. During production, pumpkin flowers are a pulse resources of approximately four-five weeks in duration, providing nectar, pollen, and oils to squash bees and other insects. Native squash bees in the genera Peponapis and Xenoglossa have co-evolved with the large tubular flowers of cucurbits, while honey bees have been associated with cucurbits in North America only since the 17th century.
Although western Texas produces most of the pumpkins in the state, and require sufficient pollination by insects to sustain yields, little is known of the status of native squash bee populations in the region. Pumpkin producers could benefit from knowledge of native, wild resources that could be used for crop production. Moreover, improving our understanding of the dispersion of squash bees across agricultural landscapes could guide strategies for the conservation of ground nesting resources for native bees. Understanding how individuals survive during most of the non-cropped year, underground, remains a key question, and is supported by this first study of native squash bees in regional farming systems. In this study, twenty-three pumpkin fields on the Southern High Plains of Texas were visited and pollinator communities were sampled to estimate densities and describe relative abundances and variation across different bee taxa among growing-season weeks, fields and cultivars. In addition to comparisons of bee densities, a binary logistic regression was used to determine relationships of flower occupancy by squash and honey bees to field, cultivar and land cover factors. In 2016, we collected 1952 bees from 12 fields, with 73.4 percent of the total number of bees collected native squash bees, 25.8 percent honey bees, and 0.1 percent bumble bees. Squash bees were more abundant than managed honey bees in all fields including those with managed honey bee and/or bumble bees at the field edge. Year 2 (2017) mirrored the 2016 numbers with 82 percent squash bees and 15.2 percent honey bees (out of 3511 total number of individuals). Forty percent (2016) and 60 percent (2017) of the total number of bees were recorded in the first week of both years. Declining numbers of bees across weeks could indicate natural seasonal mortality, environmental stressors causing mortality, or sampling bias associated with enclosed pumpkin canopies and edge effects (i.e. higher squash bee densities at margins as pumpkin canopies enlarge over the 4-5-week flowering period). In the 2017 analysis, honey bees showed preference for the Turban pumpkin cultivar, and this was the only cultivar where honey bees were more abundant than squash bees across both seasons. The results of binomial logistic regression showed that mean bee-occupancy of flowers varied by week and by field in 2016, and in 2017 varied significantly by these and additionally by cultivar, mainly attributed to the Turban-honey bee relationship. Percent wildland within 250 meters of pumpkin fields, the presence of field-level managed hives, pumpkin field area, nor cultivar richness were significant predictors of flower-occupancy across fields by squash bees or honey bees, emphasizing a need to further investigate the frequency of soil nesting behaviors within pumpkin fields and among field rotations.
This study provides baseline information to develop hypotheses and to support further studies on native squash bees and their contributions to pumpkin production on the Texas High Plains. Overall, across annual crop rotations benefits from squash bee foraging could be offset by environmental stressors resulting from production practices that affect soil nesting or pollinator health, yet the effects of field management practices on native squash bee populations are unknown. Further information on foraging and nesting resource requirements and densities or number of bees required for pollination would support a better understanding of the biology and value of native bees in pumpkins, to ultimately guide conservation actions for native squash bees on farms.