Evidence of increased hydrodynamic retention in the spawning grounds of large pelagic fishes in the western Mediterranean

Abstract

Hydrography shapes the reproductive and early life ecology of migratory large pelagics such as Atlantic bluefin tuna. While their spawning grounds have traditionally been linked to areas with suitable temperature, low productivity, and moderate surface mixing, other oceanographic processes are likely crucial to ensure that early life stages remain in favorable habitats. We hypothesize that retentive oceanographic patterns are a defining feature of these spawning areas, distinguishing them from surrounding regions. To test this hypothesis, we first evaluated the skill of a high-resolution, data-assimilative hydrodynamic model to represent the ocean surface circulation around the Balearic Islands, where the main spawning ground of the Western Mediterranean is located. We then used Lagrangian particle tracking to investigate retention and dispersion patterns at the regional scale during the reproductive season of Atlantic bluefin tuna, albacore tuna, and swordfish. Retention and dispersion analyses revealed that, during the reproductive season, surface circulation favors particle transport towards the spawning ground, where particles tend to remain. This shows that the Western Mediterranean spawning ground is governed by basin-scale hydrodynamic regimes that aggregate particles from neighboring regions, providing a mechanistic basis for their persistence over time despite occasional anomalous years.

1 Introduction

Migratory large pelagic species, perform long-distance migrations to specific areas where environmental conditions reproductive success and early life survival (Ciannelli et al., 2015). Understanding how environmental conditions in these areas influence fitness is essential for explaining the reproductive ecology of these species and for improving population management (Porch et al., 2019). In the western Mediterranean, the most important migratory large pelagic species include Atlantic bluefin tuna (Thunnus thynnus), albacore (Thunnus alalunga), and swordfish (Xiphias gladius). For these species, spawning areas are typically characterized by warm oligotrophic waters with low kinetic energy and complex mesoscale oceanography (Muhling et al., 2017; Reglero et al., 2014). Selecting spawning areas with favorable hydrodynamic features, such as eddies or fronts, can enhance larval retention and survival (Ottmann et al., 2021).

The Balearic Sea, located in the Western Mediterranean offers an ideal setting to investigate this hypothesis, as it hosts spawning grounds for Atlantic bluefin tuna, Mediterranean albacore, and swordfish (Torres et al., 2011). Previous research has linked spawning activity in this region to the Balearic Front, a salinity-driven density front formed by the confluence of resident and recent Atlantic waters (Alemany et al., 2010; Alvarez-Berastegui et al., 2014). Similar dynamics have been described in other spawning grounds of the Mediterranean (Russo et al., 2021) and the Gulf of Mexico (Lindo-Atichati et al., 2012). Moreover, recent work suggests that moderate surface mixing in this region may further shape tuna larval habitats (Díaz-Barroso et al., 2022).

However, characterizing the interannual variability of the Balearic Front and its role in retention and dispersal processes remains challenging. Satellite-derived salinity data lack the spatial resolution needed to precisely locate the front, while sea surface temperature alone cannot accurately track it during late spring and early summer (Reul et al., 2020; Balbín et al., 2014). Altimetry-based products also fail to resolve mesoscale variability at the scales relevant for larval ecology (Ballarotta et al., 2019). In this context, high-resolution, data-assimilative ocean models provide a valuable alternative for examining frontal variability and associated Lagrangian transport patterns.

The objective of this study is to characterize the hydrodynamic retention patterns in the main spawning grounds of migratory large pelagic species in the Western Mediterranean Sea by analyzing regional dispersion, retention, and residence times. In this framework, we test the hypothesis that hydrodynamic regimes around these spawning grounds (located around the Balearic archipelago) enhance transport into and retention within the main spawning ground during spring–summer, and we assess the associated interannual variability. We do so using a validated high-resolution data-assimilative ocean model to represent the surface circulation and we combine it with Lagrangian particle experiments to analyze decadal-scale retention and dispersion patterns of virtual particles during the spawning seasons of bluefin tuna, albacore, and swordfish (as they are the main representatives of large pelagic fishes in the area). For the first time, we provide a regional, multi-year assessment of the persistence of hydrodynamic retention in an ecologically relevant spawning area and explicitly examine how this persistence is modulated by interannual variability in the position of the salinity-driven Balearic Front.

2 Materials and methods

2.1 Study area

The Western Mediterranean Sea surface circulation is characterized by the inflow of Atlantic Water (AW) through the Strait of Gibraltar, which circulates cyclonically through the Mediterranean sub-basins (Millot, 1999) (Figure 1). The eastward Algerian Current flows along the North African coast, generating mesoscale anticyclonic eddies that sometimes detach from the coast and transport recent AW northward towards the Balearic archipelago (Escudier et al., 2016). The Algerian Current flows eastward through the Sardinia Channel and then splits near the Sicily Channel. One branch enters the Eastern basin, while the other turns northward into the Tyrrhenian Sea and later feeds the Northern Current, which flows along the Italian and French coasts towards the Balearic Sea (Salas et al., 2001).

Figure 1

Balbín et al., 2014). The green shaded area around the Balearic archipelago represents the area where the front was historically observed. (B) The red polygon indicates the maximum area of presence of tuna larvae in the surveys.

When approaching the Balearic Islands, the current splits into two branches (Lopez-Jurado et al., 1996): one crossing the Ibiza Channel southward, and the other deflecting northeastward to form the Balearic Current, which is also fed by northward inflows of recent AW through the Ibiza and Mallorca Channels (Vargas-Yáñez et al., 2025). This creates the Balearic Front at the confluence of the relatively resident and saltier AW with the more recent and relatively fresher AW (Barral et al., 2021).

This density front, primarily driven by salinity gradients, controls the surface circulation and mesoscale variability around the Balearic archipelago (Pinot et al., 2002; Ruiz et al., 2009). The early summer circulation of surface waters around the islands, strongly dependent on the front’s position, shows pronounced interannual variability (Balbín et al., 2014), which in turn influences the spawning grounds of bluefin tuna, albacore, and swordfish (Figure 1B) (Alemany et al., 2006, 2010; Balbín et al., 2014; Alvarez-Berastegui et al., 2014).

2.2 Oceanographic in-situ and satellite observation data

2.2.1 In-situ data from hydrographic surveys

Twelve surveys were carried out during the spring/summer (June-July) from 2009 to 2019 (see Supplementary Table 1 in Supplementary Material for specific dates and sampling locations). Hydrographic stations were located in a standard sampling grid of 10 nautical miles (approximately 18.5 km). This spatial resolution resolves the position of the salinity front and the main regional mesoscale structures, which typically range from 50 to 100 km (Pinot et al., 2002). Conductivity, temperature, pressure (CTD) data from SeaBird911+ and SeaBird 25 were obtained from surface to 350 m. Hydrographic parameters (salinity, S; potential temperature, θ) were processed using the Sea-Bird Electronics Data Processing routines and calibrated and quality-controlled following Balbín et al. (2014).

2.2.2 Satellite data

Average daily means of sea surface temperature (SST) were obtained for the sampling region (red polygon in Figure 1) from the Copernicus Marine Service (the Mediterranean Sea High Resolution Sea Surface Temperature Analysis, product id: https://doi.org/10.48670/moi-00172, Buongiorno Nardelli et al., 2013).

2.3 Hydrodynamic modeling

The high-resolution data-assimilative Western Mediterranean Operational forecasting system (WMOP, Juzá et al., 2016; Mourre et al., 2018; Hernandez-Lasheras and Mourre, 2018), developed at the Balearic Islands Coastal Observing and Forecasting System (SOCIB), was used in this study. The WMOP model is a regional configuration of the ROMS model (Shchepetkin and McWilliams, 2005; version 3.4) with 2 km horizontal resolution and 32 sigma levels, providing a vertical resolution of 1–2 m near the surface. We used six-month-long yearly data-assimilative reanalysis simulations. Each year, the simulation is initialized on 15 March starting from the oceanic conditions provided by a ten-year free-run simulation at that date (Aguiar et al., 2020). These reanalysis simulations assimilate various datasets, including satellite SST and sea level anomalies, in-situ temperature and salinity profiles from Argo floats, and surface current measurements from the Ibiza Channel High-Frequency radar using a Local Multimodel Ensemble Optimal approach (Hernandez-Lasheras and Mourre, 2018). The assimilation cycle was performed every 3 days.

These simulations were nested within the 1/16° Copernicus Marine Service Mediterranean reanalysis (Simoncelli et al., 2019), provided through the E.U. Copernicus Marine Service Information system (https://doi.org/10.25423/MEDSEA_REANALYSIS_PHYS_006_004), using mixed active-passive conditions (Marchesiello et al., 2001) at the open boundaries. Daily river discharges were included from the six main rivers of the domain, based on data from the French HYDRO database and the Spanish water authorities. Atmospheric forcing came from high-resolution predictions by the Spanish Meteorological Agency (AEMET): the HIRLAM model (Undén et al., 2002; resolution of 3 h/5 km) for 2009–2018, and the HARMONIE-AROME model (Bengtsson et al., 2017; resolution of 1 h/2.5 km) in 2019. The generic length-scale method described in Umlauf and Burchard (2003) was used to represent vertical mixing in the model.

2.4 Model validation and identification of frontal interfaces

In order to identify the different oceanographic scenarios in the study area and to assess the adequacy of the hydrodynamic model to reproduce those scenarios, we conducted a comparison of the in situ information versus the model outputs. Then we explored the interannual variability of the frontal interfaces investigating the spatio-temporal variability of the Balearic Front.

2.4.1 Quantitative evaluation of hydrodynamic model simulations

We compared model results with satellite-derived sea surface temperature (SST) and CTD-derived temperature and salinity observations in the surface mixed layer to quantify model uncertainties in the study area. On the one hand, the accuracy of the model in capturing the interannual variability of the averaged SST fields over the whole spawning ground area during summer was assessed using satellite data, considering both the absolute SST values and anomalies relative to the 2009–2019 mean seasonal cycle. On the other hand, the capability of the model to reproduce the spatial variability of the surface temperature and salinity fields during the periods of the in-situ campaigns was evaluated by comparing CTD observations with the corresponding WMOP model outputs (extracted at 12:00 UTC on the sampling days).

Standard statistical metrics of model-observation differences were calculated, including mean error (BIAS), root-mean-square error (RMSE), centered RMSE (CRMSE), standard deviation difference (STDdiff), and correlation coefficient (r) for each survey dataset. The specific formulations of these metrics are provided in Appendix A.1, and Taylor diagrams summarizing these results are presented in Supplementary Figures 2, 3 in Supplementary Material.

2.4.2 Identification of frontal interfaces

The spatio-temporal variability of the Balearic Front was examined using CTD and numerical model data. This analysis had two main objectives: (i) to characterize the interannual variability of this key frontal interface in the large pelagic spawning ground and identify characteristic oceanographic scenarios, and (ii) to evaluate the capability of the hydrodynamic model to reproduce this variability and associated scenarios.

The analysis focused on the spatial variability of the salinity fields and their gradients. Surface salinity is a more reliable indicator of the location of the water masses and the Balearic Front in summer compared to surface temperature, due to the intense regional surface heating during early summer (Balbín et al., 2014). The spatial distribution of the salinity fields was derived from optimal interpolation of observed CTD salinity values at 10 m depth onto a 0.05°-resolution grid using the DIVA software (Data-Interpolating Variational Analysis, Troupin et al., 2012) with a correlation length of 0.3 °. The model maps were generated using the same method applied to the model estimates at the time and location of the CTD stations. Interpolated fields with an RMS error above 0.2 were excluded.

From these maps, two metrics were calculated using both CTD and model data to identify the location of the salinity front:

a. A front detection index based on the surface bidimensional salinity gradient isoline (Equation 1) following the criterion used in Barral et al. (2021). The 0.02 km⁻¹ salinity gradient isoline was used as the detection threshold. This value corresponds to a detectable change of salinity per kilometer, sufficient to resolve the boundaries between saltier Atlantic Water (AW) and more recent, fresher AW transported northward by the Algerian and Balearic Currents, marking the frontal interface. This approach allows us to identify physically meaningful frontal structures considering the small-scale variability in the salinity field.

b. The position of the 37.5 isohaline, which was used in previous studies to distinguish AW of recent Atlantic origin from resident AW (Balbín et al., 2014). This isoline corresponds to the boundary between the characteristic haline properties of these water masses.

The study area was divided into twelve subregions to describe the interannual variability of the position of the frontal zones and their intrusions in the Balearic Channels. This spatial categorization is shown in Figure 2, with boundaries defined along a line connecting the three main islands of the Balearic archipelago. From west to east, these areas include the Ibiza Channel, Ibiza Island, Mallorca Channel, Mallorca Island, Menorca Channel, and Menorca Island. The location of the front was determined based on the joint application of the two frontal detection criteria within each subregion (Figure 3), a positive detection with one of the two criteria being sufficient to identify the presence of the front.

Figure 2

Figure 3

2.5 Lagrangian analysis and retention-dispersion patterns

The analysis of the influence of oceanographic scenarios on retention and dispersion patterns during the reproduction and early life developmental stages of large pelagics was conducted at two scales: the Western Mediterranean basin and the region of large pelagic spawning grounds (Figures 1A, B). This approach allows us to compare the contribution of each scale. This analysis employed a Lagrangian approach to track virtual particle trajectories within the model simulations.

The virtual oceanic particle trajectories were generated using the TRACMASS Lagrangian code (https://www.tracmass.org/, Jönsson et al., 2015) driven by the two-dimensional surface velocity fields from the WMOP model simulations with an output frequency of 6 hours. The particles experienced advection by the simulated currents and a diffusive component through random turbulent velocities with a coefficient of 50 m²/s to represent the effect of unresolved processes. Vertical motion was not included, as the study was focused on surface drift, which influences the ecology of large pelagics during early life stages (Reglero et al., 2018). Particles are treated as passive surface tracers. The analytical method used in the TRACMASS trajectory model is the so-called stepwise stationary scheme (Döös et al., 2013), with a time step of 30 minutes.

Particles were released daily across the Western Mediterranean basin, between 35.10° and 44.35° N and between -5.60° and 9.00° E (Figure 1A), in a standard grid of 0.05° in longitude and latitude, resulting in a total of 24001 particles for each daily release and a total of 1433280 per year. Releases were performed daily at 12:00 UTC during the OEW (period covering the spawning and early life developmental periods of bluefin tuna, albacore tuna, and swordfish), which corresponds to June 1 to July 31 (Alemany et al., 2010; Saber et al., 2015; Millot et al., 2023), resulting in 60 simulations per year. A 15-day duration of the Lagrangian simulations was selected to account for the development period from egg to piscivorous stages in tuna (Blanco et al., 2019). The outputs were stored every 6 h, including the position (latitude, longitude) of each particle at each time step.

The final positions of all released particles in the Western Mediterranean basin were used to create density maps showing the retention and dispersion patterns at the scale of the whole study area. Overall density maps were derived from the cumulative number of particles across all simulated years, while interannual variability was assessed by comparing yearly cumulative maps.

In order to quantify these processes a time series of particle retention was developed. This series quantified the proportion of particles reaching the spawning grounds from remote versus local seeding areas. Additionally, the escape time of particles was examined from the same 15-day trajectories. Escape time measures how quickly particle trajectories escape from a domain. In our case, the escape domain is a 1° × 1° box centered on the release location; a particle is considered to have escaped when it crosses any of the four boundaries (i.e., when it moves beyond ±0.5° in longitude and/or latitude relative to its release position). The reported escape time is then obtained by averaging these individual escape times across all the 60 simulations. The domain represents the distance necessary to exit the spawning ground. The study flowchart combining observations, hydrodynamic modelling, Lagrangian trajectories and retention-dispersion analysis is represented in Figure 4.

Figure 4

3 Results

3.1 Quantitative evaluation of hydrodynamic model simulations

Modelled sea surface temperature (SST), averaged over the study area and sampling periods, closely matched satellite-derived SST, with a root mean square error (RMSE) of 0.19 °C for absolute values and 0.17 °C for anomalies relative to the mean seasonal cycle. As expected, absolute SST values varied depending on the timing of the sampling campaigns, with warmer conditions typically recorded in July compared to June. Anomaly analysis further revealed interannual variability, identifying relatively cold years (2010, 2013, 2016, and 2019) and warm years (2012, 2015, 2017, and 2018).

The model also reproduced the vertical temperature structure measured by CTD casts during the campaigns with good accuracy. At 10 m depth, the overall RMSE was 0.75 °C, with interannual variation ranging from approximately 0.3 °C to 1.3 °C. For salinity, the model yielded a global RMSE of 0.44 and a mean bias below 0.1. Correlation coefficients were higher for temperature (0.90) than for salinity (0.48), reflecting the high spatial heterogeneity of salinity in the transition zone between recent and resident Atlantic Water, and the limited availability of high-resolution satellite salinity data to constrain the model. A detailed summary of performance statistics is provided in Tables 1, 2, and Taylor diagrams illustrating model skill for both variables are shown in Supplementary Figures 2, 3. Overall, the model provides a realistic representation of the physical environment in the region and is considered suitable for the analysis of frontal dynamics and Lagrangian transport.

3.2 Interannual variability of the frontal interfaces

The distribution of sea surface salinity fields (Figure 5) showed the higher salinity values (>37.5) associated with the resident AW in northern latitudes (red to yellow) and lower salinity values (<37.5) in southern latitudes (light green to dark blue) associated with the more recent AW entering from the Strait of Gibraltar. The frontal interfaces were characterized by a complex eddy-driven circulation leading to the mixing of these two water masses.

Figure 5

Both metrics selected for detecting the front (i.e., the 37.5 isohaline and the 0.02 km⁻¹ salinity gradient isoline) revealed similar patterns of salinity intrusions in the channels, although with differences in the detailed shape of these patterns. Whereas the salinity isohaline provides a representation of the large-scale frontal displacement, the gradient-based metric highlights spatial heterogeneity along the frontal interface, revealing variations in its width and intensity associated with mixing and mesoscale activity. The analysis of the gradient metrics indicating the front location in the subregions defined in Figure 2 is summarized in Figure 3.

In most years (2009, 2010, 2012, 2013, 2014, 2015, and 2018), the salinity front remained predominantly south of the Balearic Islands (Supplementary Table 2). However, in 2016, 2017, and 2019, it shifted northward (Figure 5, Supplementary Table S2). Furthermore, in some years, recent AW displayed large intrusions through the channels, implying greater latitudinal variability of the front, particularly in 2015, 2016, 2017, and 2019.

The spatial variability of the front was also well reproduced by the simulations. A full set of annual salinity maps from the model is provided in Supplementary Figure 4 in Supplementary Material. Figure 6 compares salinity maps from the CTD and the model at 10 m for two characteristic scenarios: a southern front in 2018 and a northern front in 2019. While absolute salinity values were underestimated in the southern part of the sampled area in 2019, the relative position of the front in both years was consistent with CTD observations. These two years were selected as representative scenarios for comparing associated dispersion and retention patterns.

Figure 6

3.3 Retention–dispersion patterns (Lagrangian analysis)

Density maps of Lagrangian particles showed that the area around the Balearic archipelago functioned as a sink for particles released across the Western Mediterranean (Figure 7A). During June and July, the highest density of particles were observed in the vicinity of the Balearic Islands and to the east. The retention zone north of Mallorca and Menorca appeared consistently every year and was primarily associated with the retroflection of the southwestward Northern Current into the northeastward Balearic Current, which tends to trap particles flowing from the Gulf of Lion. When particles were released exclusively within the polygon marking the large pelagic spawning ground, the density map showed a strong retention pattern close to the islands (Figure 7B), with very little dispersion of particles beyond the polygon boundaries.

Figure 7

Although the main retention areas remained consistent across the years (all maps are provided in Supplementary Figure 5 in Supplementary Material), interannual variations in the spatial patterns were evident. To illustrate these differences, we present the density maps for the two characteristic scenarios in 2018 and 2019 (Figure 8). In 2018, the front was located south of the Balearic archipelago, resulting in higher particle densities both northward and southward of the Balearic Islands. In contrast, in 2019, with a front in a more northern position, the highest densities occurred north of Mallorca and Menorca, with relatively lower densities to the south (Figures 6A, B). The analysis of particles released only within the spawning ground (Figures 6C, D) showed that both scenarios retain most of the particles in the spawning ground area, but with more dispersion in 2019 (57.51% in 2018 vs 45.20% in 2019). This dispersion was also visible in the density map and currents of other years with a more northern front position (e.g., 2016 with 47.07% and 2017 with 41.24% particles retained). In summary, our results indicate that, for the years analyzed, a southern front position led to a larger accumulation of particles south of the islands, whereas a northern position resulted in greater dispersal of virtual particles (Supplementary Table 2).

Figure 8

The time series of particles retained within the spawning ground of the studied large pelagics, based on both particles released throughout the Western Mediterranean (Figure 9A) and those released exclusively within the spawning ground (Figure 9B), revealed significant variability among years (a 24% difference between years with maximum and minimum values in the first case and 30% in the second). The minimum local retention within the spawning grounds occurred in 2017 and 2019, which correspond to scenarios where the Balearic Front was located north of the archipelago. Years with a front located south (i.e., 2010, 2013, 2018) showed high particle transport from the Western Mediterranean into the spawning ground, but retention levels within the area varied considerably (Figure 9B).

Figure 9

As indicated by the previous analysis, the retention of particles around the Balearic archipelago and within the spawning ground was persistent across years, modulated by significant interannual variability (Supplementary Table 2). Extrapolating the intensity of retention and dispersion based solely on the frontal position was not straightforward. While the scenarios with a front located north generally exhibited higher levels of particle dispersion and lower retention within the spawning ground, the scenarios with the front located south of or within the archipelago displayed variable levels of retention and dispersion.

The maps of escape time (Figure 10) showed higher values in the central Western Mediterranean and around the Balearic archipelago, exhibiting strong spatial correlation with the particle density maps discussed in the previous section (Figure 8). This result confirmed that this area favors the retention of particles. However, yearly maps demonstrated that the spatial patterns in the Western Mediterranean are highly variable.

Figure 10

For the characteristic years of 2018 and 2019, two distinct escape time scenarios were identified (Figure 11). In 2018, the longest escape times were concentrated around the Balearic Islands, whereas in 2019 they shifted farther north of 40° N latitude. In both cases, higher escape times, occurred north of the frontal position. Moreover, the average escape time for the entire study period (Supplementary Figure 7 in Supplementary Material) revealed stable retention zones around the Balearic archipelago, highlighting the persistence of these hydrodynamic features over the decade.

Figure 11

4 Discussion

In this study, we analyzed the oceanographic surface retention and dispersion patterns in the Western Mediterranean affecting the early life ecology of large pelagic fish species such as bluefin tuna, albacore tuna, and swordfish. The results revealed that during the spawning and larval developmental period (spring–summer), the regional oceanography displayed Lagrangian patterns that favored transport towards the main spawning grounds located around the Balearic archipelago, where surface drifting particles tended to be retained. These findings support the ecological hypothesis that spawning grounds of large pelagic species are associated with geographic areas characterized by specific oceanographic features that differ from surrounding areas. The hypothesis that fish reproduction strategies are linked to specific hydrographic features has a long history (Bakun, 1996), and has been shown for different species and geographic areas (Suthers et al., 2023). Overall, our results reinforce that the Balearic Sea presents differential hydrodynamic settings that favors the persistence of suitable reproductive habitats for large pelagic fishes.

Our study show that the waters around the Balearic archipelago function as an accumulation area for surface-drifting particles in the Western Mediterranean basin. This oceanographic setting results from the interaction between relatively fresh new Atlantic waters and resident Atlantic Waters, which generate a frontal system characterized by strong salinity gradients. Located between the Balearic Sea to the north and the Algerian basin to the south, the archipelago lies in a transition region where these water masses interact, promoting persistent frontal structures and mesoscale variability. Regional topography and specific wind regimes further contribute to the formation of intense anticyclonic eddies (García-Ladona et al., 1996; Salat, 1995; Aguiar et al., 2022), while the Balearic Current acts as a dynamical boundary that tends to reduce cross-frontal exchange. Together, these features enhance residence times by limiting dispersion otherwise driven by frontal jet dynamics.

Surface ocean dynamics favoring transport toward the spawning grounds, as well as the retentive patterns around the Balearic archipelago during summer, were persistent throughout the decade analyzed (2009–2019). The frontal system was characterized using two complementary approaches: an isohaline that defined the water−mass boundary and a horizontal salinity−gradient metric that identified the dynamically active frontal interface. Both diagnostics consistently indicate that the position of the sea−surface salinity front is a key control on retention and dispersion within the spawning grounds.

At basin scale, the regional circulation and the Balearic Sea configuration reduce surface dispersal, favoring retention processes, which can enhance survival by modulating predator–prey encounters and food availability, consistent with mechanisms described for other systems that supports fish spawning grounds (Werner et al., 1993, 2001; Paris and Cowen, 2004; Bakun, 2006). In the case of bluefin tuna, the Balearic Current and associated frontal structures promote spatial segregation of spawning habitats that reduces predation pressure on early life stages (Ottmann et al., 2021), and eddy−driven aggregation processes can locally enhance prey availability once larvae become piscivorous (Reglero et al., 2011; Uriarte et al., 2019). Previous studies also highlighted that moderate surface mixing around the Balearic archipelago and the specific location of the salinity front favor bluefin tuna larval habitats (Alvarez-Berastegui et al., 2014, 2018; Díaz-Barroso et al., 2022) and that the area, located in the center of the Western Mediterranean, favors larval survival and growth due to its thermal regimes and prey distribution (Fiksen and Reglero, 2022).

The role of mesoscale structures and Lagrangian retentive patterns in shaping bluefin spawning habitats is also evident in other regions. In the Gulf of Mexico, spawning habitat has been associated with circulation patterns that limit advection (Teo et al., 2007; Muhling et al., 2011). Similarly, in the Slope Sea, high−resolution modelling experiments demonstrated that mesoscale circulation features and retention mechanisms strongly constrain the suitability and persistence of spawning habitats (Rypina et al., 2019; 2021). Hydrography and thermal regimes are also main elements of bluefin tuna larval habitats in the south of Sicily in the central Mediterranean (Russo et al., 2021). In the Balearic Sea, the hydrodynamic conditions not only favor bluefin tuna but other large pelagics; for example, the persistent accumulation zone identified south of Menorca (Figure 6b, 9), observed consistently over several years in this study (see Supplementary Figure 6), corresponds to a well−known hotspot for albacore tuna larvae (Álvarez-Berastegui et al., 2018), and the specific hydrodynamic conditions of the Balearic Sea also provide adequate habitat for swordfish larvae (Tugores et al., 2026).

Considering the relevance of circulation−driven retention processes in shaping larval habitats of temperate tunas, changes in dispersion–retention patterns characterizing spawning grounds emerge as a critical factor for survival success during early life stages (e.g. Russo et al., 2022). In the Balearic Sea, and over the study period, we identified three main spatial configurations of the position of the front relative to the archipelago: north, south, or central within the archipelago. Changes in frontal position and circulation regimes between years can therefore influence larval survival through their effects on retention, prey availability, and exposure to advective losses. Accordingly, years with stronger advection away from the area could yield lower larval survival. These scenarios typically arise in years when the Balearic Current flows north of the archipelago. The surface dynamics and location of the front are a complex process set by multiple factors such as the wind regime (García et al., 1994) and temperatures in the north−western Mediterranean that control the development of cold intermediate waters, which determine the strength of density gradients (Balbín et al., 2014). Understanding these processes will be crucial for modelling larval survival and for developing early-warning systems for recruitment scenarios of large pelagic species in the region.

It is important to emphasize the value of in situ data and reliable numerical models which are fundamental for advancing the integration of environmental variability into fisheries assessment (FAO, 2024). The valuable hydrographic dataset available in the study area enabled validation of the data−assimilative modelling tools used in this work. These models provide high−resolution representations of the mesoscale variability of surface currents, which are presently not accessible by satellite observations only. Future studies coupling Lagrangian dispersal with species−specific transport and survival modules will help anticipate biological responses to shifts in retention–dispersion dynamics that, under climate change, may deviate from historical baselines (Schroeder et al., 2016).

5 Conclusion

This study reveals that the main spawning grounds of migratory large pelagic species in the Western Mediterranean are characterized by persistent basin-scale driven hydrodynamic retention patterns. We show that surface circulation during the reproductive season consistently favors transport toward, and retention within, the Balearic spawning area. Although the position of the salinity-driven Balearic Front exhibits marked interannual variability, the region acts as a long-term accumulation and retention zone for surface-drifting particles, with variability in retention intensity modulated by frontal configuration. These findings provide a mechanistic explanation for the long-term persistence of this spawning ground and highlight the role of regional circulation in shaping early life habitats of large pelagic fishes. By explicitly linking frontal dynamics, transport pathways, and retention processes, this work advances the understanding of spawning habitat functioning and highlights the importance of high-resolution physical modeling for improving larval ecology studies and fisheries assessment in the Mediterranean.

References

• 1

  AguiarE.MourreB.Alvera-AzcárateA.PascualA.MasonE.TintoréJ. (2022). Strong long‐lived anticyclonic mesoscale eddies in the Balearic Sea: Formation, intensification, and thermal impact. J. Geophysical Research: Oceans127, e2021JC017589. doi: 10.1029/2021jc017589. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 2

  AguiarE.MourreB.JuzaM.ReyesE.Hernández-LasherasJ.CutoloE.et al. (2020). Multi-platform model assessment in the Western Mediterranean Sea: Impact of downscaling on the surface circulation and mesoscale activity. Ocean Dyn.70, 273–288. doi: 10.1007/s10236-019-01317-8. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 3

  AlemanyF.DeuderoS.Morales-NinB.López-JuradoJ. L.PalmerM.PalomeraI.et al. (2006). Influence of physical environmental factors on the composition and horizontal distribution of summer larval fish assemblages off Mallorca Island (Balearic archipelago, Western Mediterranean). J. Plankton Res.28, 473–487. doi: 10.1093/plankt/fbi123. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 4

  AlemanyF.QuintanillaL.Vélez-BelchíP.GarcíaA.CortésD.RodríguezJ. M.et al. (2010). Characterization of the spawning habitat of Atlantic bluefin tuna and related species in the Balearic Sea (western Mediterranean). Prog. Oceanogr.86, 21–38. doi: 10.1016/j.pocean.2010.04.014. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 5

  Alvarez-BerasteguiD.CiannelliL.Aparicio-GonzalezA.RegleroP.HidalgoM.López-JuradoJ. L.et al. (2014). Spatial scale, means and gradients of hydrographic variables define pelagic seascapes of bluefin and bullet tuna spawning distribution. PloS One9, e109338. doi: 10.1371/journal.pone.0109338. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 6

  Álvarez-BerasteguiD.RegleroP.Aparicio-GonzálezA.BalbínR.HidalgoM.AlemanyF. (2018). Integrating reproductive ecology, early life dynamics and mesoscale oceanography to improve albacore tuna assessment in the Western Mediterranean. Fish. Res.208, 329–338. doi: 10.1016/j.fishres.2018.08.014. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 7

  BakunA. (1996). Patterns in the Ocean. Ocean processes and marine population dynamics (La Paz, Mexico: California Sea Grant College System/NOAA/Centro de Investigaciones Biologicas del Noroeste), 1–323. ISBN 1-888691-01-8.

  • Google Scholar

• 8

  BakunA. (2006). Fronts and eddies as key structures in the habitat of marine fish larvae: Opportunity, adaptive response and competitive advantage. Scientia Marina70, 105–122. doi: 10.3989/scimar.2006.70s2105. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 9

  BalbínR.López-JuradoJ. L.FlexasM. D. M.RegleroP.Vélez-VelchíP.González-PolaC.et al. (2014). Interannual variability of the early summer circulation around the Balearic Islands: Driving factors and potential effects on the marine ecosystem. J. Mar. Syst.138, 70–81. doi: 10.1016/j.jmarsys.2013.07.004

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 10

  BallarottaM.UbelmannC.PujolM.-I.TaburetG.FournierF.LegeaisJ.-F.et al. (2019). On the resolutions of ocean altimetry maps. Ocean Sci.15, 1091–1109. doi: 10.5194/os-15-1091-2019. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 11

  BarralQ. B.ZakardjianB.DumasF.GarreauP.TestorP.BeuvierJ. (2021). Characterization of fronts in the Western Mediterranean with a special focus on the North Balearic Front. Prog. Oceanogr.197, 102636. doi: 10.1016/j.pocean.2021.102636. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 12

  BengtssonL.AndraeU.AspelienT.BatrakY.CalvoJ.De RooyW.et al. (2017). The HARMONIE–AROME model configuration in the ALADIN–HIRLAM NWP system. Mon. Weather Rev.145, 1919–1935. doi: 10.1175/MWR-D-16-0417.1. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 13

  BlancoE.RegleroP.de RojasA. H.OrtegaA.de la GándaraF.FolkvordA. (2019). The effect of nutritional condition by two nucleic acid derived indices on the growth to post-flexion of Atlantic bluefin tuna and Atlantic bonito larvae. J. Exp. Mar. Biol. Ecol.519, 151182. doi: 10.1016/j.jembe.2019.151182. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 14

  Buongiorno NardelliB.TronconiC.PisanoA.SantoleriR. (2013). High and ultra-high resolution processing of satellite sea surface temperature data over Southern European Seas in the framework of MyOcean project. Remote Sens. Environ.129, 1–16. doi: 10.1016/j.rse.2012.10.012. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 15

  CiannelliL.BaileyK.OlsenE. M. (2015). Evolutionary and ecological constraints of fish spawning habitats. ICES J. Mar. Sci.72, 285–296. doi: 10.1093/icesjms/fsu145. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 16

  Díaz-BarrosoL.Hernández-CarrascoI.OrfilaA.RegleroP.BalbínR.HidalgoM.et al. (2022). Singularities of surface mixing activity in the Western Mediterranean influence bluefin tuna larval habitats. Mar. Ecol. Prog. Ser.685, 69–84. doi: 10.3354/meps13979. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 17

  DöösK.KjellssonJ.JönssonB. (2013). “ TRACMASS – A Lagrangian trajectory model,” in Preventive Methods for Coastal Pollution: Towards the Use of Ocean Dynamics for Pollution Control (Heidelberg: Springer International Publishing), 225–249. doi: 10.1007/978-3-319-00440-2_7

  • CrossRef
  • Google Scholar

• 18

  EscudierR.MourreB.JuzaM.TintoréJ. (2016). Subsurface circulation and mesoscale variability in the Algerian subbasin from altimeter-derived eddy trajectories. J. Geophysical Research: Oceans121, 6310–6322. doi: 10.1002/2016JC011760

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 19

  FiksenØ.RegleroP. (2022). Atlantic bluefin tuna spawn early to avoid metabolic meltdown in larvae. Ecology103, e03568. doi: 10.1002/ecy.3568. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 20

  Food and Agriculture Organization of the United Nations (2024). The state of world fisheries and aquaculture 2024: Blue transformation in action (Rome: FAO). doi: 10.4060/cd0683en

  • CrossRef
  • Google Scholar

• 21

  GarcíaM. J. L.MillotC.FontJ.García-LadonaE. (1994). Surface circulation variability in the Balearic Basin. J. Geophys. Res.99, 3285–3296. doi: 10.1029/93JC02114. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 22

  García-LadonaE.CastellónA.FontJ.TintoréJ. (1996). The Balearic current and volume transports in the Balearic basin. Oceanolog. Acta19, 489–497.

  • Google Scholar

• 23

  Hernandez-LasherasJ.MourreB. (2018). Dense CTD survey versus glider fleet sampling: Comparing data assimilation performance in a regional ocean model west of Sardinia. Ocean Sci.14, 1069–1084. doi: 10.5194/os-14-1069-2018. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 24

  JönssonB.DöösK.KjellssonJ. (2015). TRACMASS: Lagrangian trajectory code (6.0). Zenodo. doi: 10.5281/zenodo.34157

  • CrossRef
  • Google Scholar

• 25

  JuzáM.MourreB.RenaultL.GomaraS.SebastianK.LoraS.et al. (2016). SOCIB operational ocean forecasting system and multi-platform validation in the Western Mediterranean Sea. J. Oper. Oceanogr.9, s155–s166. doi: 10.1080/1755876X.2015.1117764

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 26

  Lindo-AtichatiD.BringasF.GoniG.MuhlingB.Muller-KargerF. E.HabtesS. (2012). Varying mesoscale structures influence larval fish distribution in the northern Gulf of Mexico. Mar. Ecol. Prog. Ser.463, 245–257. doi: 10.3354/meps09860. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 27

  Lopez-JuradoJ. L.García LafuenteJ.PinotJ. M.AlvarezA. (1996). Water exchanges in the Balearic channels. Bulletin de lÍnstitut océanographique, Monaco, nº spécial 17, CIESM Science Series nº2, 41–64.

  • Google Scholar

• 28

  MarchesielloP.McWilliamsJ. C.ShchepetkinA. (2001). Open boundary conditions for long-term integration of regional oceanic models. Ocean Modell.3, 1–20. doi: 10.1016/s1463-5003(00)00013-5. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 29

  MillotC. (1999). Circulation in the western mediterranean sea. J. Mar. Syst.20, 423–442. doi: 10.1016/s0924-7963(98)00078-5. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 30

  MillotR.PoissonF.MacíasD.SaberS.AielloA.DominiqueE.et al. (2023). Reproductive traits and spawning activity of swordfish Xiphias gladius L. in the north-western Mediterranean Sea (Corsica). Fish. Res.267, 106811. doi: 10.1016/j.fishres.2023.106811. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 31

  MourreB.AguiarE.JuzaM.Hernandez-LasherasJ.ReyesE.HeslopE.et al. (2018). Assessment of high‐resolution regional ocean prediction systems using multi‐platform observations: Illustrations in the western Mediterranean Sea. New Front. Operational Oceanography, 663–694. doi: 10.17125/gov2018.ch24

  • CrossRef
  • Google Scholar

• 32

  MuhlingB. A.LamkinJ. T.AlemanyF.GarcíaA.FarleyJ.IngramG. W.et al. (2017). Reproduction and larval biology in tunas, and the importance of restricted area spawning grounds. Rev. Fish Biol. Fish.27, 697–732. doi: 10.1007/s11160-017-9471-4. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 33

  MuhlingB. A.LeeS.-K.LamkinJ. T.LiuY. (2011). Predicting the effects of climate change on bluefin tuna (Thunnus thynnus) spawning habitat in the Gulf of Mexico. ICES J. Mar. Sci.68, 1051–1062. doi: 10.1093/icesjms/fsr008. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 34

  OttmannD.FiksenØ.MartínM.AlemanyF.PrietoL.Álvarez‐BerasteguiD.et al. (2021). Spawning site distribution of a bluefin tuna reduces jellyfish predation on early life stages. Limnol. Oceanogr.66, 3669–3681. doi: 10.1002/lno.11908. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 35

  ParisC. B.CowenR. K. (2004). Direct evidence of a biophysical retention mechanism for coral reef fish larvae. Limnol. Oceanogr.49, 1964–1979. doi: 10.4319/lo.2004.49.6.1964

  • CrossRef
  • Google Scholar

• 36

  PinotJ. M.López-JuradoJ. L.RieraM. (2002). The CANALES experiment, (1996-1998). Interannual, seasonal, and mesoscale variability of the circulation in the Balearic Channels. Prog. Oceanogr.55, 335–370. doi: 10.1016/s0079-6611(02)00139-8. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 37

  PorchC. E.BonhommeauS.DiazG. A.ArrizabalagaH.MelvinG. (2019). “ The journey from overfishing to sustainability for Atlantic bluefin tuna, Thunnus thynnus,” in The future of bluefin tunas: Ecology, fisheries management, and conservation ( Johns Hopkins University Press, 2019. Project MUSE) 1, 3–44.

  • Google Scholar

• 38

  RegleroP.BlancoE.AlemanyF.FerráC.Alvarez-BerasteguiD.OrtegaA.et al. (2018). Vertical distribution of Atlantic bluefin tuna Thunnus thynnus and bonito Sarda larvae is related to temperature preference. Mar. Ecol. Prog. Ser.594, 231–243. doi: 10.3354/meps12516. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 39

  RegleroP.TittensorD. P.Álvarez-BerasteguiD.Aparicio-GonzálezA.WormB. (2014). Worldwide distributions of tuna larvae: Revisiting hypotheses on environmental requirements for spawning habitats. Mar. Ecol. Prog. Ser.501, 207–224. doi: 10.3354/meps10666. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 40

  RegleroP.UrtizbereaA.TorresA. P.AlemanyF.FiksenØ. (2011). Cannibalism among size classes of larvae may be a substantial mortality component in tuna. Mar. Ecol. Prog. Ser.433, 205–219. doi: 10.3354/meps09187. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 41

  ReulN.GrodskyS. A.AriasM.BoutinJ.CatanyR.ChapronB.et al. (2020). Sea surface salinity estimates from spaceborne L-band radiometers: An overview of the first decade of observation, (2010–2019). Remote Sens. Environ.242, 111769. doi: 10.1016/j.rse.2020.111769. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 42

  RuizS.PascualA.GarauB.FaugèreY.AlvarezA.TintoréJ. (2009). Mesoscale dynamics of the Balearic Front, integrating glider, ship and satellite data. J. Mar. Syst.78, S3–S16. doi: 10.1016/j.jmarsys.2009.01.007. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 43

  RussoS.TorriM.PattiB.MuscoM.MasulloT.Di NataleM. V.et al. (2022). Environmental conditions along tuna larval dispersion: Insights on the spawning habitat and impact on their development stages. Water14, 1568. doi: 10.3390/w14101568. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 44

  RussoS.TorriM.PattiB.RegleroP.Álvarez-BerasteguiD.CuttittaA.et al. (2021). Unveiling the relationship between sea surface hydrographic patterns and tuna larval distribution in the central Mediterranean Sea. Front. Mar. Sci.8. doi: 10.3389/fmars.2021.708775. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 45

  RypinaI. I.ChenK.HernándezC. M.PrattL. J.LlopizJ. K. (2019). Investigating the suitability of the Slope Sea for Atlantic bluefin tuna spawning using a high-resolution ocean circulation model. ICES J. Mar. Sci.76, 1666–1677. doi: 10.1093/icesjms/fsz079. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 46

  RypinaI. I.DotzelM. M.PrattL. J.HernandezC. M.LlopizJ. K. (2021). Exploring interannual variability in potential spawning habitat for Atlantic bluefin tuna in the Slope Sea. Prog. Oceanogr.192, 102514. doi: 10.1016/j.pocean.2021.102514. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 47

  SaberS.de UrbinaJ. O.Gómez-VivesM. J.MacíasD. (2015). Some aspects of the reproductive biology of albacore Thunnus alalunga from the western Mediterranean Sea. J. Mar. Biol. Assoc. U. K.95, 1705–1715. doi: 10.1017/s002531541500020x. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 48

  SalasJ.García-LadonaE.FontJ. (2001). Statistical analysis of the surface circulation in the Algerian Current using Lagrangian buoys. J. Mar. Syst.29, 69–85. doi: 10.1016/s0924-7963(01)00010-0. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 49

  SalatJ. (1995). The interaction between the Catalan and Balearic currents in the southern Catalan Sea. Oceanolog. Acta18, 227–234.

  • Google Scholar

• 50

  SchroederK.ChiggiatoJ.BrydenH. L.BorghiniM.Ben IsmailS. (2016). Abrupt climate shift in the Western Mediterranean Sea. Sci. Rep.6, 23009. doi: 10.1038/srep23009. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 51

  ShchepetkinA. F.McWilliamsJ. C. (2005). The regional oceanic modeling system (ROMS): A split-explicit, free-surface, topography-following-coordinate oceanic model. Ocean Modell.9, 347–404. doi: 10.1016/j.ocemod.2004.08.002. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 52

  SimoncelliS.FratianniC.PinardiN.GrandiA.DrudiM.OddoP.et al. (2019). Mediterranean Sea physical reanalysis (MEDREA) (Version 1) [Data set]. Copernicus Monitoring Environment Marine Service. doi: 10.25423/MEDSEA_REANALYSIS_PHYS_006_004

  • CrossRef
  • Google Scholar

• 53

  SuthersI. M.SchaefferA.ArcherM.RoughanM.GriffinD. A.ChapmanC. C.et al. (2023). Frontal eddies provide an oceanographic triad for favorable larval fish habitat. Limnol. Oceanogr.68, 1019–1036. doi: 10.1002/lno.12326. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 54

  TeoS. L. H.BoustanyA. M.BlockB. A. (2007). Oceanographic preferences of Atlantic bluefin tuna, Thunnus thynnus, on their Gulf of Mexico breeding grounds. Mar. Biol.152, 1105–1119. doi: 10.1007/s00227-007-0758-1. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 55

  TorresA. P.RegleroP.BalbinR.UrtizbereaA.AlemanyF. (2011). Coexistence of larvae of tuna species and other fish in the surface mixed layer in the NW Mediterranean. J. Plankton Res.33, 1793–1812. doi: 10.1093/plankt/fbr078. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 56

  TroupinC.BarthA.SirjacobsD.OuberdousM.BrankartJ.-M.BrasseurP.et al. (2012). Generation of analysis and consistent error fields using the Data Interpolating Variational Analysis (DIVA). Ocean Modell.52–53, 90–101. doi: 10.1016/j.ocemod.2012.05.002. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 57

  TugoresM. P.TorresA. P.RegleroP.MartínM.BalbínR.TintoréJ.et al. (2026). Climate-induced shifts in swordfish spawning habitat in the Western Mediterranean. Prog. Oceanogr.240, 103610. doi: 10.1016/j.pocean.2025.103610. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 58

  UmlaufL.BurchardH. (2003). A generic length-scale equation for geophysical turbulence models. J. Mar. Res.61, 235–265. doi: 10.1357/002224003322005087

  • CrossRef
  • Google Scholar

• 59

  UndénP.RontuL.JärvinenH.LynchP.CalvoJ.CatsG.et al. (2002). “ HIRLAM-5 scientific documentation,” in Swedish Meteorological and Hydrological Institute (SMHI) (Norrköping, Sweden).

  • Google Scholar

• 60

  UriarteA.JohnstoneC.Laiz-CarriónR.GarcíaA.LlopizJ. K.ShirozaA.et al. (2019). Evidence of density-dependent cannibalism in the diet of wild Atlantic bluefin tuna larvae (Thunnus thynnus) of the Balearic Sea (NW-Mediterranean). Fish. Res.212, 63–71. doi: 10.1016/j.fishres.2018.12.013. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 61

  Vargas-YáñezM.OuradouM.MoyaF.BallesterosE.AlonsoC.SerraM.et al. (2025). A time series of water mass transports through the Balearic Channels using an ocean circulation inverse method: 1996–2022. Prog. Oceanogr.237, 103525. doi: 10.1016/j.pocean.2025.103525. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 62

  WernerF. E.MackenzieB. R.PerryR. I.LoughR. G.NaimieC. E.BlantonB. O.et al. (2001). Larval trophodynamics, turbulence, and drift on Georges Bank: A sensitivity analysis of cod and haddock. Scientia Marina65, 99–115. doi: 10.3989/scimar.2001.65s199. PMID:

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 63

  WernerF. E.PageF. H.LynchD. R.LoderJ. W.LawrenceS. R.LoughR. G. (1993). Influences of mean advection and eddy diffusion on cod (Gadus morhua) larval dispersion on Georges Bank. Fish. Oceanogr.2, 177–191.

  • Google Scholar