Long-term tracking and quantification of individual behavior in bumble bee colonies

Abstract

Social insects are ecologically dominant and provide vital ecosystem services. It is critical to understand collective responses of social insects such as bees to ecological perturbations. However, studying behavior of individual insects across entire colonies and across timescales relevant for colony performance (i.e., days or weeks) remains a central challenge. Here, we describe an approach for long-term monitoring of individuals within multiple bumble bee (Bombus spp.) colonies that combines the complementary strengths of multiple existing methods. Specifically, we combine (a) automated monitoring, (b) fiducial tag tracking, and (c) pose estimation to quantify behavior across multiple colonies over a 48 h period. Finally, we demonstrate the benefits of this approach by quantifying an important but subtle behavior (antennal activity) in bumble bee colonies, and how this behavior is impacted by a common environmental stressor (a neonicotinoid pesticide).

Behavioral variation across the days and lives of honey bees

“…These are supervised methods which require training and specified behavior to identify, and thus have focused on a few types of behavioral events which could be reliably identified. Alternatively, recent work has combined general methods of pose estimation with barcode tracking and applied this to bumblebees ( Smith et al., 2022 ); such pose estimation data could be used with unsupervised methods in order to identify complex behavioral patterns without training or a-priori specification ( Berman et al., 2014 ; Graving and Couzin, 2020 ). In contrast to these approaches, which use smaller colonies and shorter tracking periods of 2–7 days ( Gernat et al., 2020 ; Jones et al., 2020 ; Smith et al., 2022 ), in this study, we extract only trajectory data from barcode tracking, which enables the analysis of thousands of bees during their entire lifetimes in a timespan of several months.…”

“…Alternatively, recent work has combined general methods of pose estimation with barcode tracking and applied this to bumblebees ( Smith et al., 2022 ); such pose estimation data could be used with unsupervised methods in order to identify complex behavioral patterns without training or a-priori specification ( Berman et al., 2014 ; Graving and Couzin, 2020 ). In contrast to these approaches, which use smaller colonies and shorter tracking periods of 2–7 days ( Gernat et al., 2020 ; Jones et al., 2020 ; Smith et al., 2022 ), in this study, we extract only trajectory data from barcode tracking, which enables the analysis of thousands of bees during their entire lifetimes in a timespan of several months. Future work can merge these approaches or choose the methods most appropriate to specific biological questions, by combining aspects of supervised identification of behavioral events, unsupervised behavioral classification from pose estimation, and behavioral metrics calculated from trajectory data.…”

NAPS: Integrating pose estimation and tag‐based tracking

“…NAPS uses a hybrid tracking algorithm to provide high‐dimensional behavioural data compatible with recently developed computational ethology techniques. Similar hybrid tracking techniques have been introduced in the literature (Gal et al, 2020; Smith et al, 2022). The NAPS framework utilizes quantitative information about body part position and unique identifying markers to provide postural data with identity over long timescales and robust to complex interactions.…”

“…We imaged the arenas from above using a Basler acA5472-17um (Basler AG) camera recording 3664px × 3664px frames at 20 frames per second. Recordings were taken using a modified version of campy, a Python package developed for real-time video compression, to drastically reduce file sizes (Severson, 2021).…”

NAPS: Integrating pose estimation and tag-based tracking

“…NAPS uses a hybrid tracking algorithm to provide high-dimensional behavioral data compatible with recently developed computational ethology techniques. Similar hybrid tracking techniques have been introduced in the literature (Smith et al, 2022; Gal et al, 2020). The NAPS framework utilizes quantitative information about body part position and unique identifying markers to provide postural data with identity over long timescales and robust to complex interactions.…”