Movement ecology of gelatinous zooplankton: approaches, challenges and future directions

The open ocean is a vast and dynamic environment that contains a rich diversity of organisms ranging in size from microscopic plankton to baleen whales. The movements of these organisms within the epipelagic realm play a crucial role in ecological processes including nutrient cycling (Pitt et al., 2009; Gilbert et al., 2023) and export (Lebrato et al., 2013; Aksnes et al., 2023) and predator–prey interactions (Hansson et al., 2007; Gemmell et al., 2012), and ultimately influence the structure and function of marine ecosystems (Colin et al., 2015). Quantifying the detailed patterns of movement exhibited by marine organisms is essential for understanding the aforementioned processes as well as addressing conservation challenges, fisheries sustainability and understanding the impacts of a changing climate on the oceans.

The advent of satellite telemetry in the 1970s represented a critical step in both terrestrial and marine animal tracking, enabling researchers to remotely monitor the movements of large-bodied species (Buechner et al., 1971). In the ocean, these species were primarily limited to marine mammals (Stewart et al., 1989), sea turtles (Stoneburner, 1982) and sharks (Priede, 1984), where individuals were tracked across expansive oceanic territories. Subsequent developments in acoustic telemetry (McMichael et al., 2010), GPS tracking (Garrigue et al., 2015) and bio-logging (Kraska et al., 2015) opened new avenues for studying the fine-scale behaviors of these and other marine organisms, from foraging and breeding to social interactions and environmental responses (Crossin et al., 2014). Miniaturized GPS tags and archival data loggers allow researchers to track the movements of smaller marine taxa, such as seabirds (Montevecchi et al., 2012) and pelagic fish (Abascal et al., 2016), with unprecedented accuracy. These technologies enable scientists to unravel the complex interplay between individual behaviors and environmental conditions, shedding light on the mechanisms driving population dynamics and ecosystem function. Other technologies such as state–space models, machine learning techniques and network analysis approaches (Fig. 1) offer additional powerful tools for extracting meaningful insights from tracking data, revealing hidden patterns and processes underlying animal movements (Wang, 2019).

Although the integration of tracking technologies and sophisticated analytical methods has propelled the field of oceanic movement ecology into a new era, there remains a major group of marine organisms where the movement and life history patterns remain poorly understood. These are the oceanic zooplankton. Zooplankton make up the majority of animal biomass in the pelagic environment, yet our understanding of the movement patterns of the major zooplankton taxa is still poorly understood. The gap in our knowledge of oceanic zooplankton movement stems from the fact that although technological advances have miniaturized many of the ocean tracking tools, they remain far too large for most zooplankton species. For gelatinous zooplankton, a potential reason for the gap in knowledge of movement patterns stems from the traditional assumption that animals move in lock-step with oceanic surface currents and have often been modeled as passive particles (Hood et al., 1999; Johnson et al., 2005; Berline et al., 2013). Although behavior has been incorporated into spatial models for crustacean zooplankton and fish larvae since at least the 1980s (Boicourt, 1982; Wroblewski, 1982; Rothlisberg et al., 1983), behavioral considerations for gelatinous plankton has been more recent (El Rahi et al., 2020; Malul et al., 2024). This coincides with studies that have shown many species are capable of directed movements and can at least partially control large-scale movement patterns (Folt and Burns, 1999; McManus and Woodson, 2012; Michalec et al., 2017).

Therefore, gelatinous zooplankton is one group where current tracking technologies could begin to play an important role in elucidating plankton movement patterns (Table 1). These organisms were once assumed to be a trophic dead end (Verity and Smetacek, 1996). This perception has tempered some early interest in their ecology, but gelatinous zooplankton are now known to play significant roles in marine ecosystems. The advent and wide-scale deployment of video-based observational systems (Lucas et al., 2014; Lombard et al., 2024 preprint) has led to a growing recognition that gelatinous zooplankton are dominant components of oceanic ecosystems. This is primarily due to the fact that traditional net sampling techniques often destroy gelatinous zooplankton and has resulted in major underestimates of the abundance and ecological importance of gelatinous plankton. This recognition underpins the importance of accounting for gelatinous zooplankton communities for ecosystem and fisheries management (Aubert et al., 2018). In fact, some species of gelatinous zooplankton such as Mnemiopsis leidyi can exceed 50 animals per cubic meter over large areas (Kremer and Nixon, 1976) and affect zooplanktonic standing stocks (Kremer, 1979; Sullivan et al., 2001; Purcell and Decker, 2005; Finenko et al., 2006; Roohi et al., 2008), species composition (McNamara et al., 2013, 2014) and, by trophic cascading effects, even phytoplankton chlorophyll a concentrations (Sullivan et al., 2008; Tiselius and Møller, 2017) and fishery dynamics (Oguz et al., 2008; Roohi et al., 2010) of coastal communities. They have also been shown to have complex behavioral responses to environmental cues, which has been hypothesized to enable this delicate gelatinous predator to thrive in energetic, turbulent coastal ecosystems (Costello and Mianzan, 2003; Mianzan et al., 2010; Jaspers et al., 2018). However, traditional methods for studying gelatinous plankton movements, such as net sampling and visual observations, have inherent limitations, often providing only snapshots of their behavior. We cannot arrive at a mechanistic understanding of the population dynamics of these species using snapshots alone. These approaches limit our ability to understand and predict distributions and population biology of zooplankton.

Many species of gelatinous zooplankton have body sizes several orders of magnitude greater than other plankton taxa, which makes them more amenable targets for tracking devices, especially as devices continue to shrink in size owing to technological advances. Indeed, recent advancements in tracking technologies have begun to offer some insights into the movement ecology of gelatinous zooplankton, enabling researchers to start to unravel the mysteries surrounding their movements and behaviors. Understanding jellyfish movements is crucial for predicting their interactions with other organisms, their impacts on ecosystems and their responses to environmental changes. In this Review, we explore the recent advancements in ocean movement ecology and tracking of gelatinous zooplankton, and examine the challenges, opportunities and future directions of this rapidly evolving field.

To date, most behavioral information on gelatinous zooplankton has been gathered through observation in the laboratory (Costello and Colin, 1994; McHenry and Jed, 2003; Gemmell et al., 2019, 2021; Townsend et al., 2020; Costello et al., 2024). Although new tracking technologies are becoming more widespread, investigations into gelatinous zooplankton behavior and ecology in the natural environment has primarily utilized methodologies such as net sampling (Lynam et al., 2004), SCUBA diving (Colin et al., 2022; Potter et al., 2023) and submersibles (Daniels et al., 2021). These approaches have provided important measurements of jellyfish abundance, size and vertical distribution, as well as examinations of behavioral variations, predator–prey interactions (Cordeiro et al., 2022; Potter et al., 2023) and swimming speeds across different water depths. However, the vast majority of the data collected from these methods provides only a snapshot of their behaviors on timescales of seconds to minutes. We still lack essential information about long-term (hours, days, weeks) behavioral movement patterns of these oceanic animals. As technology has progressed, sensitivity of acoustic techniques has improved and tags and tracking technology has undergone a transformation, becoming increasingly compact and lightweight (Fig. 2). These advances have enabled researchers to extend tagging studies to much smaller organisms, including bumblebees (Mola and Williams, 2019) and sea scallops (Miyoshi et al., 2018). However, it is only recently that scientists have begun to recognize tagging as a feasible and promising approach for studying certain species of gelatinous zooplankton such as large jellyfish, specifically the medusal stages of scyphozoan and cubozoan species.

The introduction of echosounders with high resolution and sensitivity, employed through ship-based acoustic systems (Lynam, 2006; Kaartvedt et al., 2011, 2015), as well as remotely operated vehicles (ROVs) and video profilers (Graham et al., 2003; Klevjer et al., 2009) has facilitated measurements of jellyfish abundance, size and vertical distribution (Brierley et al., 2001, 2005; Lynam et al., 2006; Klevjer et al., 2009), as well as examinations of behavioral variations and swimming speeds across different water depths (Kaartvedt et al., 2007, 2011). Traditional acoustic methods employed in studies of jellyfish have focused on volumetric backscattering, but the low target resolution of water-based gelatinous animals has generally limited this method to measurements of abundance and distribution (Brierley et al., 2001, 2005; Båmstedt et al., 2003; Alvarez Colombo et al., 2009). Split-beam echosounders make it possible to accurately position an organism in the acoustic beam and thus make it possible to use acoustics to study the behavior of individual gelatinous organisms in situ (Kaartvedt et al., 2007) and provide insights into the activity patterns of jellyfish. This type of acoustic study in Lurefjorden, Norway, revealed diverse vertical migration patterns of the jellyfish Periphylla periphylla in relation to the diel cycle (Kaartvedt et al., 2011).

Acoustic telemetry involves tagging jellyfish with acoustic transmitters and deploying an array of receivers to detect their movements. This technique allows for continuous monitoring of jellyfish movements in three dimensions, providing detailed insights into their vertical and horizontal migrations, residence times in specific areas, and responses to environmental cues such as temperature and currents. Acoustic tags that relay position data to drifting hydrophones have demonstrated that medusae can move with many different trajectories relative to water currents (Diamant et al., 2023). This method was first used for the box jellyfish Chironex fleckeri in coastal Australia (Gordon and Seymour, 2009). Acoustic telemetry has proven particularly useful in studying the movements of large jellyfish species, such as the lion's mane jellyfish, Cyanea capillata, revealing complex migration patterns and seasonal fluctuations in distribution (Moriarty et al., 2012). The results showed that lion's mane jellyfish clearly exhibit active swimming behavior and are not passively planktonic. The animals swim faster during the night than during the day and the highest swimming speeds occurred during flood tides.

Satellite tracking using GPS satellites have been instrumental in unraveling complex, long-term movement patterns of terrestrial, avian and even some aquatic species. The method involves attaching satellite tags to animals, which transmit location data to orbiting satellites. This technique is of very limited use for gelatinous zooplankton owing to the fact that the vast majority of species never contact the atmosphere and GPS signals do not transmit through water. One species where GPS tracking has been used to help study the long-distance movements is the Portuguese man o' war, Physalia physalis (Ferrer et al., 2015). Because part of the animal is neustonic and partially exposed to the atmosphere at all times owing to the large gas-filled float, researchers used pop-up satellite tags for fish to elucidate the dispersal patterns and major oceanographic movement drivers for P. physalis. However, satellite tracking is still currently restricted by tag size constraints, which limits direct attachment and utility for smaller neustonic species such as Velella sp.

Biologging involves attaching data loggers or sensors to gelatinous zooplankton to record environmental variables, such as temperature, salinity and depth, alongside their movements. These devices may also incorporate accelerometers or gyroscopes to capture fine-scale behaviors, such as swimming speed and orientation. Time–depth recorders (TDRs) have provided insights into the vertical movements of individual jellyfish, uncovering more intricate swimming behaviors than previously recognized. Studies on a number of different species have demonstrated extensive vertical movements across the water column (Honda et al., 2009; Hays et al., 2012; Moriarty et al., 2012). For instance, vertical speeds up to 3.98 m min⁻¹ were recorded for Chrysaora capillata (Bastian et al., 2011). Additionally, Fossette et al. (2016) tracked the swimming patterns of the Pacific sea nettle, Chrysaora fuscescens, and showed that this species prefers to stay in shallow waters and does not vertically migrate. However, biologging tags showed that the jellyfish Nemopilema nomurai will often swim down to depths of up to 176 m, displaying a nocturnal pattern of deeper dives, while most individuals remained in the upper 40 m during the day (Honda et al., 2009).

Small TDRs have also been instrumental in demonstrating that jellyfish can actively reposition themselves in the water column over extended periods by maintaining constant depths for days to avoid crossing density gradients and thus remain in a given water mass (Hays et al., 2008; Bastian et al., 2011; Moriarty et al., 2012). This could allow for some directed control of movement patterns given that different water masses often travel at different rates or directions (Amos et al., 1971). Furthermore, tagging studies have compared jellyfish behavior to that of more traditionally active predators such as fish. For instance, Hays et al. (2012) found that the barrel jellyfish, Rhizostoma octopus, swam an average of 619.2 m day⁻¹ throughout the water column, displaying complex vertical search patterns, including Lévy flight movements previously unknown for gelatinous zooplankton (Fig. 2). These findings suggest that jellyfish may efficiently compete with fish for planktonic prey, with potential implications for ecosystem and fisheries management. Biologging approaches have provided novel insights into the behavior and habitat preferences of some gelatinous zooplankton, allowing researchers to correlate their movements with environmental conditions. However, the current size of biologging tags only makes them suitable for tracking large species of cnidarian medusae.

ROVs and video profiler systems have been used for many years to provide information on difficult to access marine animals. Unlike tracking tag methods, these systems do not require attachment or direct contact to the animal(s) of interest. This provides the benefit of being able to track and follow animals of any size and without the worry of attached tracking devices affecting the animal's behavior. Video-based systems have the added benefit of being able to acquire subtle kinematic data such as during the pulsation pattern of siphonophore swimming or the ctene row beat pattern of ctenophores (Swift et al., 2009). It can also be used to make observations of predator–prey interactions and trophic patterns (Choy et al., 2017).

Early versions of ROVs used to quantify gelatinous plankton movement and behavior were constrained by light availability, limiting deep-sea observations under natural light conditions (Sørnes et al., 2008), and were often of short duration. More recently, ROVs have been able to track individual hydromedusae and ctenophores, taxa that are not currently suitable for tagging owing to their small size, for long durations exceeding 5 h (Katija et al., 2021). Although this provided a radical step forward in the ability to observe natural behaviors of taxonomic groups that were not previously attainable, the method was labor intensive, requiring an ROV pilot to manually follow each animal for the duration of the deployment. To help overcome this challenge, researchers have developed a class of tracking-by-detection algorithms called Deep Search and Tracking Autonomously with Robotics (DeepSTARia), which integrate machine learning models with imaging and vehicle controllers to enable autonomous underwater vehicles (AUVs) to make targeted visual observations of ocean life (Barnard et al., 2024).

One of the major challenges with tagging live animals is being able to attach the tags well enough to prevent dislodgement during the observation period, while minimizing the impact a tag will have on the animal's movement or behavior. This is a particular challenge for soft-bodied animals underwater. All tags to date have been deployed on nine species of relatively large cnidarian medusae ranging in size from 10 cm to 1.6 m in diameter. Fossette et al. (2016) outlined several best practices for attaching tags to larger gelatinous zooplankton. Three methods of attachment have been used successfully on cnidarian medusae. The methods are broadly classified as the ‘cable tie method’ (Fig. 2), the ‘suction cup method’ and the ‘glue method’ (Fossette et al., 2016). A new method type of soft hydrogel-based bioadhesive appears to provide reliable and rapid adhesion on soft-bodied marine animals (squid) and may prove to be a useful approach for gelatinous zooplankton (Duque Londono et al., 2024).

Animal-borne data loggers must also be retrieved in order to access the data. Short-term (i.e. hours) deployments have used an approach where a thin monofilament is tethered to a small fishing float to aid in visually tracking the jellyfish during deployment. At the conclusion of the deployment, the jellyfish and logger are retrieved by gradually pulling on the tether and/or by snorkeling to reach the jellyfish. To prevent impeding the jellyfish's movements, it is recommended that the tether length should exceed the water depth by at least 10 m (Fossette et al., 2016). This method would pose challenges in complex habitats such as kelp forests or where large amounts of floating Sargassum spp. is present. Another technique for short-term deployments in relatively shallow waters involves using an acoustic transmitter alongside the logger. This allows for active tracking of the jellyfish, with manual retrieval of the logger by a diver or snorkeler at the deployment's conclusion.

Retrieval options are more limited for long-term deployments lasting more than a few hours. A study on the movement patterns of the jellyfish Rhizostoma octopus relied on the logger tags to wash ashore and be found by beachgoers once they detached from the animals several weeks to months later (Hays et al., 2012). However, this approach's effectiveness depends heavily on local conditions and may not be universally applicable. Pop-up tags that use VHF radio-transmitters can aid in locating the tag at the surface, and pop-up satellite tags will detach at a specific time and transmit data directly through the ARGOS satellite system, eliminating the need for physical retrieval. However, owing to their size and weight, these tags have only been employed on the largest jellyfish of species, such as Nemopilema nomurai (Honda et al., 2009). Furthermore, the limited bandwidth of ARGOS restricts the transmission and retrieval of high-resolution behavioral data, such as accelerometry data.

Tagging can also induce alterations in gelatinous zooplankton behavior (Hays et al., 2008). Although information is very limited, tagged jellyfish, regardless of the deployment method (glue or cable tie), seem to exhibit an initial dive away from the person deploying the tag, which likely stems from handling-induced stress. Fossette et al. (2016) found that, through a combination of accelerometer data and video observations, tagged jellyfish went through a ‘recovery’ phase. Tag-equipped jellyfish would first exhibit reduced activity compared with pre-deployment levels, with retracted tentacles and oral arms, and downward movement. After approximately 15 min, bell pulsation became more vigorous and regular, and activity levels stabilized. Longer-term impacts of tags on gelatinous zooplankton behavior are still unknown, but tags deployed on other species are known to impact drag and kinematics/buoyancy (Methling et al., 2011; Vandenabeele et al., 2015).

Movement ecology of gelatinous zooplankton is still in its infancy. Less than 5% of scyphozoan and cubozoan jellyfish species have been tagged thus far (Fossette et al., 2016), and we have virtually zero available data on planktonic ctenophores, salps, siphonophores and hydromedusae, which together represent over 1000 species. As data-loggers continue to shrink in size, we are hopeful this will increase the array of gelatinous zooplankton species that could become feasible candidates for tag deployment in the future. Moreover, the development of tags specialized for soft-bodied invertebrates (Mooney et al., 2015) holds promise for routinely gathering fine-scale and high-resolution behavioral and physiological data for gelatinous animals. One approach that can reduce the size of tagging equipment immediately involves attaching only a small acoustic transmitter to the organism, which is subsequently monitored by an AUV equipped with a broader array of environmental sensors. This approach has already been used with some fish species (Clark et al., 2013; Skomal et al., 2015). Because gelatinous zooplankton are less likely to be disturbed by a nearby foreign object than a fish with a highly developed visual system, AUVs could follow much closer to gelatinous plankton and, in AUVs with high-definition video capabilities, allow researchers to assess behavioral patterns as well. These types of systems also have the potential to provide insight into how movement of gelatinous zooplankton could influence the behavior of other species.

Significant progress is also being made towards fully autonomous systems (Fig. 1) capable of tracking individuals and aggregating planktonic animals using less intrusive methods such as stereo imaging. New classes of tracking-by-detection algorithms that utilize machine learning and artificial intelligence models coupled with high resolution video and vehicle controllers are being developed to allow AUVs to record extended, non-invasive measurements of natural behaviors in gelatinous zooplankton (Masmitja et al., 2022, 2023; Katija, 2023; Barnard et al., 2024) (Fig. 1). These methods offer the greatest potential for expanding our knowledge of many of the species for which tagging efforts are intractable owing to animal size, body composition or behavioral tag effects. This is because an AUV equipped with high resolution imaging and tracking algorithms would have the ability to track the movement and behavior of gelatinous zooplankton without the need for any tagging or physical contact with the target species (Yoerger et al., 2018). With the appropriate onboard power supply, these AUVs could potentially track animals continuously for days, weeks or even months at a time. This would allow for unprecedented access to the life history patterns of some of the most abundant, yet poorly understood groups of macroscopic marine animals on our planet. As technology continues to improve, this could be translated to even small zooplankton groups such as copepods.

The growing recognition of the importance of gelatinous zooplankton in oceanic ecosystems underscores how new tagging and tracking technologies should be embraced as important research tools in the toolkit of marine ecologists and oceanographers interested in the movement ecology, nutrient cycling processes, predator–prey interactions, life history and behavior of gelatinous zooplankton, as well as also fisheries sustainability and the impacts of a changing climate on the oceans. By complementing established techniques, these new methods will continue to enhance our knowledge of this important but understudied component of marine ecosystems.

The authors acknowledge Dr Andrew Biewener and participants at the Integrating Biomechanics, Energetics and Ecology in Locomotion Symposium for helpful discussions and feedback.