Accounting for observational uncertainty in plant-pollinator networks

As with all ecological systems, plant-pollinator networks are constructed from data subject to uncertainty. A missed observation could arise from true, underlying variation in the system or from simple bad luck during data collection. Despite the potential to arise through very different mechanisms, these missed observations impact downstream analyses in the same way, hobbling our ability to truly understand the dynamics shaping these networks. By borrowing statistical tools from the field of occupancy modeling, we demonstrate the potential gains in understanding possible by fully accounting for the role of observational uncertainty in plant-pollinator networks.