Background/Question/Methods Efficiency of pollination services by social insects strongly depends on the pollinators' nest locations. The prediction of pollinators' nest locations is important in aspects of pollination ecology and conservation ecology. For the prediction of pollinators' nest locations, we must know the factors of deciding their nest locations. Here, we focused on the nest locations of bumble bees in the field because they are major pollinators of numerous plants. We proposed Foundress-Max hypothesis where nest site is chosen to maximize the foraging rate of a bumble bee foundress before or during colony establishment. Based on the Foundress-Max hypotheses, we constructed an optimization model using the net rates of energy intake. We compared the predictions of optimization model with the actual locations of six nests of Bombus ardens found in the field.

Results/Conclusions The Foundress-Max hypothesis was supported by nest locations in the field. We concluded that nest site is chosen to maximize the foraging rate of a bumble bee foundress before or during colony establishment. From results, we discuss the application of optimization model to predict nest locations of bumble bees.