Intensive monitoring for bees in North America: indispensable or improvident?

Abstract

Continued reports of bee declines have prompted repeated calls for a national monitoring programme in the United States. Here, we argue that such a large‐scale surveillance monitoring programme would consume inordinate resources without providing the sought‐after data in a meaningful time period. Surveillance monitoring cannot provide reliable and timely measures of absolute or relative population size, species richness or diversity, host plant and nest site preferences or typical foraging ranges. In addition, the multitude of specimens captured by the passive traps commonly used to surveil overwhelm our identification capacity and, in some cases, may even accelerate species decline. Conversely, surveillance monitoring should be pursued when data can be collected non‐lethally for readily identifiable species or species groups (e.g., bumble bees) or when widely distributed sentinel plants can be observed (e.g., sunflowers). Lethal collections should continue at appropriate temporal intervals but only at a limited number of representative areas with accurate historic records. We advocate that emphasis be placed on a targeted, natural‐history approach aimed at addressing specific hypotheses and exploring solutions to conservation management problems. We already know enough about the drivers of bee decline to formulate and test potential solutions. While additional information is of value in understanding bee diversity, distributions and evolution, we do not need it to begin to take action; we should be focusing on the best ways to help bee species endure rather than attempting to accumulate even more evidence that they are in trouble.

Pursuing best practices for minimizing wild bee captures to support biological research

“…In fact, and though other methods including active ones have been traditionally used, passive sampling, and bowl traps in particular, have been promoted as a way to monitor changes in bee abundance and species occurrence (LeBuhn et al, 2013) and their use has increased in the last 15 years (Portman et al, 2020; Figure 1a). However, the suitability of passive methods to address these questions and to inform management and conservation measures has also been questioned lately (see Briggs et al, 2022; Portman et al, 2020; Tepedino et al, 2015; Tepedino & Portman, 2021; Woodard et al, 2020). These authors argue that passive and non‐targeted trapping cannot provide reliable and timely measures of changes in absolute or relative population size, species richness or diversity, and that these methods do not allow identification of key habitat and resource requirements to develop successful management actions.…”

“…First, passive trapping informs about the presence of an individual in a given place, but it is poorly informative about other aspects such as bee behavior, flower choice, or the delivery of pollination services. Therefore, for addressing certain questions, passive methods might not be the most suitable (Tepedino & Portman, 2021). For example, when assessing pollination services, important mismatches can occur between the insects captured by passive methods (pan traps, vane traps, Malaise traps, trap-nests, etc.)…”

“…Though the impact of lethal sampling on wild bee populations for scientific or conservation purposes dwarfs the impact of other anthropogenic stressors (Keilsohn et al, 2018), and it is certainly not among the main drivers of bee and insect declines (Miličić et al, 2021; Potts et al, 2010; Sánchez‐Bayo & Wyckhuys, 2019), reducing the numbers of captures is an ethical, logistical, and scientifically defensible long‐term objective (Drinkwater et al, 2019; Tepedino & Portman, 2021), especially for rare species or species of conservation concern (Minteer et al, 2014) and in intense spatially and temporally replicated sampling (Montgomery et al, 2021). The goal of minimizing harm to animal populations while simultaneously achieving scientific research objectives has been addressed with the “3Rs” approach since the 1950s (Russell & Burch, 1959).…”

“…The increasing availability of video and photographic devices, the growth of machine‐learning based image processing (Buschbacher et al, 2020; Høye et al, 2020) and the development of low‐ and noninvasive molecular methods such as DNA barcoding and environmental metabarcoding (Evans et al, 2016; Liu et al, 2020; Thomsen & Sigsgaard, 2019), call for a timely discussion on the potential alternatives to bee captures ( replacement ) and to the use of nonlethal methods ( refinement ) (Tepedino & Portman, 2021). The reduction in the number of captures can be pursued by estimating sample sizes needed prior to field sampling (LeBuhn et al, 2013) and by maximizing potential further uses of captured bees through improved sample preservation and deposition reporting methods (Packer et al, 2018; Stoeckle et al, 2010).…”

Estimating bee abundance: can mark-recapture methods validate common sampling protocols?

“…In recent years, evidence of declines in populations of bees and other pollinators has spurred strong interest in monitoring bee abundance over time and among habitat types (Packer & Darla-West 2021;Portman et al 2020;Potts et al 2010;Prendergast et al 2020;Westphal et al 2008). In the USA, nine bee species are now protected under the Endangered Species Act of 1973, and repeated calls for large-scale monitoring efforts led to the formation in 2020 of the US National Native Bee Monitoring Research Coordination Network (US Fish and Wildlife Service 2021; Tepedino & Portman 2021;Woodard et al 2020). Monitoring bee populations has both ecological and economic incentives, due to the important role of bees as pollinators in natural, agricultural, and urban environments (Kearns et al 1998;Potts et al 2016).…”