The Reliability of Measurements of Foraging Behavior in Shorebirds: A Comparison of Real-Time and Slow-Motion Recordings

Abstract

Data reliability is the degree to which the data can be replicated by the same or different observers and should be one of the most basic aspects of scientific data assessment. Yet, for example, information on the reliability of feeding rate data in shorebirds is strikingly lacking in the literature. In this study, the reliability of foraging behavior metrics collected on Semipalmated Sandpipers (Calidris pusilla) and Short-billed Dowitchers (Limnodromus griseus) during their spring stopovers in Delaware Bay, Maryland, USA, was examined. Using video recordings of 30 foraging individuals scored by four observers, including one experienced observer, assessed: (1) to what extent the data on these two species' foraging rates can be replicated while viewing videos at real-time speed as opposed to videos viewed in slow-motion; and (2) whether intra- and inter-observer reliabilities vary between species and with the observer experience. The results demonstrate that scoring feeding rates in Short-billed Dowitchers does not require the analysis of the footage in slow motion, implying that data may be obtained in real time in the field, following an appropriate training of observers. By contrast, detailed scoring of different foraging techniques in small sandpipers requires either slow-motion reviews of the field-sourced footage, or the clumping of foraging techniques into tactile and visual types to provide sufficient reliability of the data collected in the field if filming of birds is not feasible.