Beyond floral visitation: Pollen use reveals selective resource allocation in a generalist solitary bee

A central challenge in characterizing species' dietary niches is ensuring that foraging data accurately capture the resources used, their relative importance, and how resource abundance shapes foraging patterns. Most studies infer diet breadth and resource-use patterns from observational records, yet such data may lack crucial contextual information documenting how a resource is used or which alternative options were available. For example, in plant-pollinator systems, floral visits involve collecting nectar, pollen, or both, which is allocated to either adult survival or offspring provisioning; visit data cannot distinguish the two, which is important because the resources allocated to offspring ultimately determine reproductive success and population persistence. Here, we experimentally quantified deviation between observed foraging behavior and resources allocated to reproduction by manipulating resource abundance, individually marking bees, and documenting resource use with direct observation, pollen-provision DNA metabarcoding, and pollen-provision microscopy. We show that observed floral visitation patterns did not fully capture the resources allocated to offspring provisioning; pollen from the most frequently visited plants was not the most abundant in provisions. Moreover, we found that resource abundance alone did not consistently explain the observed interaction patterns, challenging the assumption of resource abundance as a strong predictor of plant-pollinator interactions. Our results show that relying on visitation alone may portray a broader niche than is reflected in pollen use, and demonstrate how integrating behavioral foraging observations with offspring requirements can reveal resource-specific patterns that would otherwise remain hidden.