The pressing issue of biodiversity loss has reached a pivotal moment. Achieving the targets of the 2030 Global Biodiversity Framework and the European Biodiversity Strategy demands more than a commitment (see Guidetti et al., 2008; Rinaldi, 2021; Yates et al., 2019); it necessitates an in-depth comprehension of biodiversity patterns and ecological interdependencies, as well as vigilant monitoring of the biodiversity changes and effects of anthropogenic activities on our planet's varied habitats to develop effective conservation and mitigation strategies (e.g., Knight et al., 2020; Perino et al., 2022). The intensification of human activities within our marine environments including the expansion of offshore infrastructure for energy production are reshaping ecosystems, making effective monitoring crucial. Through regular monitoring of biodiversity, we can gain insights into the state, trends and subtle shifts within ecosystems, and pinpoint areas where conservation efforts are most urgently needed (Proença et al., 2017).
Today, our ability to collect vast amounts of data on biodiversity has reached unprecedented levels. Many different technological solutions are currently used to monitor biodiversity in our oceans (e.g., acoustics Powell and Ohman, 2012, remote sensing Basedow et al., 2019, molecular methods Langlois et al., 2021, in situ imagingPicheral et al., 2022) with the increasing popularity of (semi)-autonomous platforms (such as data buoys), which allow for the collection of data from several sensors simultaneously (e.g., SINTEF OceanLab Observatory, 2023). These new techniques, such as automated data collection from in-situ sensors and DNA-based techniques with high quality standard reference libraries, have improved and overcome laborious and costly traditional sampling and data acquisition techniques (see Capurso et al., 2023; Fu et al., 2021). As biodiversity varies across different spatial and temporal scales, selecting representative sites and time points that capture this variation is crucial to obtaining a comprehensive understanding of a given ecosystem. Biodiversity monitoring activities are also often resource limited and face practical constraints such as budget, time, and available resources. Balancing these limitations with the need for representative data is a constant challenge.
Zooplankton in the ocean are incredibly diverse and abundant, and maintaining this diversity is crucial for maintaining the health of marine ecosystems (Charron, 2012). Zooplanktonic species have an intermediary role in the aquatic food webs, linking primary producers with higher trophic levels. Thus, they play an important role in biogeochemical cycling (Roman et al., 2002; Steinberg et al., 2012), contribute to energy transfer in pelagic food webs, and impact prey-predators dynamics and distributions (McQuatters-Gollop et al., 2019). Furthermore, zooplankton, and specifically copepods, are sensitive to changing environmental conditions, which makes them a suitable bioindicator (Batten et al., 2019; Thackeray et al., 2016). However, the abundance and distribution of zooplankton is known to be highly variable across spatial and temporal scales such as seasonal changes, diel vertical migration, and small-scale aggregations. This variability makes it difficult for monitoring programs to assess where and when to sample in a time- and cost-effective manner. Here, we use calanoid copepods as model taxon for testing a data-driven, adaptive sampling approach. This approach relies on the use of real-time data from an acoustic sensor to inform the sampling timing and depth. The sampling involves collecting water for environmental DNA (eDNA) metabarcoding to assess the presence of different marine Calanoida, and plankton net samples for traditional visual taxonomic classification to assess the age structure of sampled calanoid copepods. The adaptive sampling approach presented in this study aims to improve the quality of environmental monitoring while also reducing the time and cost expended for sampling.