INTRODUCTION

The Bay of La Paz (BoP) is the largest basin in the Gulf of California (Fig. 1a). The region is highly dynamic, with extensive seasonal and interannual variability due to atmosphere-ocean interactions. In winter, the prevailing winds are predominantly northwesterly, with high and persistent speeds exceeding 10 m·s^-1. In summer, southeasterly winds blow at approximately 5 m·s^-1, with frequent calm periods (^{Monreal-Gómez et al. 2001}).

Figure 1 Location of the Gulf of California and Bay of La Paz (Mexico). Isolines indicate bathymetry (m). Symbols are CTD cast (+), zooplankton samples in winter (∆), and zooplankton samples in summer (☐). 

The BoP is also recognized for its high biodiversity, which has been linked to the hydrodynamics of the region (^{Monreal-Gómez et al. 2001}, ^{Durán-Campos et al. 2015}, ^{Coria-Monter et al. 2017}), involving the presence of a quasi-permanent cyclonic eddy that exerts a high impact on the planktonic ecosystem by supporting production at high trophic levels. Indeed, this cyclonic eddy promotes a nutrient-rich Ekman pump that fertilizes the euphotic layer (^{Coria-Monter et al. 2017}). This induces a differential distribution in phytoplankton between diatoms and dinoflagellates (^{Coria-Monter et al. 2014}) and a differential aggregation of zooplankton inside the eddy field (^{Durán-Campos et al. 2015}), which then impacts the whole pelagic ecosystem, particularly the planktonic ecosystem.

Copepods are probably the most abundant multicellular organisms on Earth (^{Mauchline et al. 1998}, ^{Kiørboe 2011}) and in marine ecosystems, where they are of prime importance due to the position they occupy in the trophic web (^{Brierley 2017}). Copepods are indiscriminate or suspension particle feeders (^{Boltovskoy 1999}). Therefore, they can be predatory feeders (carnivores), mixed-mode feeders (omnivores), or strictly herbivorous (^{Ohtsuka et al. 1996}); however, under certain circumstances, some organisms can switch from one feeding mode to another (^{Tartarotti et al. 2014}). As copepods represent a direct link between the lower and higher trophic levels, including those species that support important fisheries around the globe (^{Richardson 2008}), copepods can also be indirectly considered as organisms with high commercial value. Additionally, copepods contribute to the removal of carbon dioxide from the atmosphere through the sedimentation of inorganic and organic carbon compounds included in their fecal pellets and thus affect the functioning of the biological carbon pump (^{Brierley 2017}).

Typically, zooplankton biomass is indicative of zooplankton production, and the estimation of this parameter is essential for evaluating the trophic structure and function in any aquatic ecosystem (^{Melão and Rocha 2004}). Changes in zooplankton biomass are closely related to several factors, including variations in the salinity field (^{Vera-Mendoza and Salas-de-León 2014}), the temperature regime (^{Webster and Lucas 2012}), the availability of food (^{Durán-Campos et al. 2019}), and the mortality rate due to predation.

Another source of variability in zooplankton biomass is the presence of hydrodynamic processes, which modify the hydrographic structure of the water column and exert a remarkable effect on productivity by introducing nutrients into the euphotic zone with a consequent enhancement in phytoplankton biomass; in turn, this leads to an increased availability of food for the zooplankton community (^{Estrada et al. 2012}). These hydrodynamic processes are present throughout the water column at different scales, including internal waves, fronts, and eddies (^{Mann and Lazier 2006}).

Mesoscale eddies (radii of 10-100 km) are high-energy structures of prime importance in any marine ecosystem (^{Gaube et al. 2018}). These structures are recognized as cyclonic, anticyclonic, and mode-water eddies with a notable impact on the planktonic ecosystem (^{McGillicuddy 2016}). It has been demonstrated that the presence of cyclonic eddies modulates the structure and biomass of the zooplankton community in diverse parts of the world, including southwestern Gran Canaria, the Canary Islands (^{Hernández-León et al. 2001}), off the coast of northwestern Africa (^{Yebra et al. 2004}), the Mediterranean Sea (^{Molinero et al. 2008}), the Sargasso Sea (^{Eden et al. 2009}), the Pacific Ocean off central-southern Chile (^{Morales et al. 2010}), Hudson Bay in Canada (^{Estrada et al. 2012}), and the Madagascar Channel (^{Noyon et al. 2019}).

Although the impact of eddies on zooplankton biomass has been relatively well demonstrated worldwide, few studies on the effect of these structures on copepods have been conducted (e.g., ^{Yebra et al. 2004}, ^{Cruz-Hernández et al. 2018}, and ^{Jagadeesan et al. 2020}), mainly due to the uncertainties that remain about the impact of basin-confined eddies on the copepod community. As a contribution to this topic, the present study aimed to assess the biomass distribution of copepods in a cyclonic eddy system in the BoP, Gulf of California. Hydrographic data and zooplankton samples were collected during 2 oceanographic expeditions in 2 contrasting seasons: the winter of 2006 and the summer of 2009. We hypothesized that copepod biomass varies with respect to season and the hydrodynamics of the eddy system through its impact on the hydrographic properties of the water column and food availability. We intend for this study to contribute to a better understanding of the influences of eddies on particular and pivotal zooplankton groups, such as copepods, in one of the most productive marine ecosystems in the world (^{Álvarez-Borrego 2012}, ^{Arreguín-Sánchez et al. 2017}).

MATERIALS AND METHODS

In the BoP, high-resolution hydrographic data and zooplankton samples were obtained during 2 research cruises (Fig. 1a-b), onboard the R/V El Puma (Universidad Nacional Autónoma de México) in the winter (February 22-26) of 2006 and the summer (August 13-17) of 2009. Using a CTD probe (Neil Brown and SeaBird 19 plus instruments were used in 2006 and 2009, respectively), which was calibrated before each cruise by the manufacturer, we acquired data at 45 hydrographic stations across the bay (Fig. 1b). The CTD probe was lowered at a rate of 1 m·s^-1 and configured to store data at 24 Hz. To collect zooplanktonic organisms, the regions of the Alfonso Basin and the northern BoP near its connection to the Gulf of California were selected according to previous investigations that revealed the presence of a quasi-permanent cyclonic eddy (^{Monreal-Gómez et al. 2001}). Then, we defined the stations belonging to the center, edge, or outside of the cyclonic eddy system in both seasons. Thus, in these regions immediately following the CTD cast, oblique zooplankton sampling was carried out at 13 stations in winter and 9 stations in summer (Fig. 1b); in both cases, oblique zooplankton hauls were performed using a bongo system with a 60-cm diameter mouth equipped with a net mesh size of 333 µm. Zooplankton organisms were collected between a depth of 200 m and the surface with a haul time of 15 min at 1 m·s^-1. The stations were located in the Alfonso Basin and the northern reaches of the BoP near its connection to the Gulf of California. The water volume filtered during the haul was calculated using calibrated flowmeters (General Oceanics) placed in each net. Once onboard, the organisms were fixed for 24 h with a solution of 4% formalin plus sodium borate and then finally preserved in 70% ethanol. Ethanol represents one of the most common fluids currently used for general zooplankton preservation, as it preserves the colors of the organisms (^{Santhanam et al. 2019}). The samples were preserved in airtight bottles under dark conditions with special precautions during the preservation time; for example, ethanol was replaced every 3 months to avoid a yellow color in the samples, to avoid ethanol evaporation, and to prevent the oil solvent properties from being affected.

The CTD data were processed following the subroutines provided by the manufacturer and averaged over a 1 dbar interval. Then, the conservative temperature (Θ, °C), absolute salinity (S_A, g·kg^-1), and the density of seawater (ρ, kg·m^-3) were derived following the algorithms of the thermodynamic equation of seawater-2010 (TEOS-10) (^{IOC et al. 2010}). The geostrophic currents (u,v) were calculated from the CTD data according to the following (e.g., ^{Pond and Pickard 1995}): u=-1fρ∂p∂y;v=1fρ∂p∂x, where f (= 2Ω sinϕ) is the Coriolis parameter and p is the hydrostatic pressure derived from ρ. The geostrophic method for calculating the relative velocities between a pair of hydrographic stations, A and B, separated by a distance L follows the equation (v ₁ - v ₂) = 1Lf [∆Φ _B - ∆Φ _A ]. This is the practical form of the geostrophic equation used to obtain the relative velocities for 2 levels (v ₁ - v ₂), where ∆Φ _A and ∆Φ _B are the geopotential anomalies, which are null at the bottom (considered as the no-motion level). The geostrophic circulation patterns during the 2 research cruises were obtained using the geostrophic method for calculating the velocity relative to the bottom. These circulation patterns were analyzed at the base of the thermocline, which was obtained by the maximum vertical gradient method ∂T∂z .

Satellite images of the Chla concentration as an indicator of phytoplankton biomass were obtained with a resolution of 1 km pixel from the NASA Ocean Biology Processing Group (OBPG) concurrent with the dates of both research cruises. The images were processed using levels 1 and 2 with SeaDAS software provided by the OBPG. Different masks/flags (STRAYLIGHT, CLDICE, LAND, and HILT) were applied to purge bad or low-quality data when the images were generated. To test the statistical significance of the satellite data and CTD data, correlation analyses were performed.

In the laboratory, the zooplankton samples were sequentially split with a Folsom mechanism, and individual copepods were identified following ^{Boltovskoy (1999)}; the copepods were counted and then separated using a Carl Zeiss stereo dissecting microscope. We concluded that the identification of the copepods at the group level nonetheless contributed to the knowledge of their ecology by visualizing patterns of distribution into the eddy field system and their relationship with the Chla concentration. Previous research has shown that the identification of marine zooplankton at the group level is enough for the assessment of trophic ecology and distribution patterns in different domains around the globe (e.g., ^{Ayón et al. 2008}), particularly in the BoP (e.g., ^{Durán-Campos et al. 2015}, ²⁰¹⁹; ^{Coria-Monter et al. 2020}). The copepods picked from the samples were pooled in a glass Petri dish and divided into 3 groups: calanoid copepods, cyclopoid copepods, and all copepodite stages. In this study, the identification and separation of each copepodite stage by group (calanoids or cyclopoids) were not considered; rather, all stages were combined. In this sense, we are aware that the third group may imply certain bias; however, it is known that copepodites (either calanoids or cyclopoids) tend to be omnivorous (^{Ohtsuka et al. 1996}, ^{Boltovskoy 1999}, ^{Tartarotti et al. 2014}). Therefore, we concluded that combining all copepodite stages was a valid approach considering that the aim of this study was to evaluate the role of physical forcing in the biomass of this group of organisms and its relationship with phytoplankton biomass, expressed as Chla.

To quantify the biomass for each group (wet weight, mg·100 m^-3) at each station, ethanol was removed using a Millipore system manually pumped through pre-weighed nitrocellulose membrane filters (0.45 µm, 47 mm diameter; Millipore Corp., USA). Then, using an analytical balance (Sartorius BP211D, resolution: 0.1 mg·210 g^-1), differences in weight were obtained. Finally, biomass (mg·100 m^-3) was obtained following the protocols described in ^{Durán-Campos et al. (2015)} and ^{Durán-Campos et al. (2019)}.

RESULTS

The geostrophic circulation patterns were analyzed at the base of the thermocline. In winter (2006), the mixing layer was 50 m deep; in summer (2009), it was 30 m deep. In both cases, the current patterns were characterized by the presence of a cyclonic eddy located in the Alfonso Basin; however, the current intensity and shape of the eddy differed between seasons. In winter (2006), the results showed an eddy with a diameter of 25 km that reached a velocity of approximately 50 cm·s^-1 at its periphery in the northern bay and occupied the deepest part of the bay and the Boca Grande region at its confluence with the Gulf of California, where exchanges between basins occur. In this region, the velocities increased, and the circulation pattern bifurcated into the bay and to the north (Fig. 2a-c). At 30-m depth, the geostrophic velocities (not shown) were slightly more intense than those at 50-m depth; however, the circulation pattern was similar at both depths. The circulation also showed a clockwise current in the southwestern region of the bay close to the coast. In summer (2009), at 30-m depth, the geostrophic velocities had a well-defined cyclonic eddy occupying the same area as in winter (2006) with a diameter of 30 km; however, at the base of the thermocline (30-m depth), the geostrophic velocities reached 70 cm·s^-1 at the periphery of the eddy (Fig. 2d-f).

Figure 2 Geostrophic currents (relative to the bottom) (cm·s^-1), surface chlorophyll a (mg·m^-3) distribution, and copepods biomass (diameter of the circles is proportional to the biomass values) (mg·100 m^-3). Top panel: geostrophic current pattern in winter 2006 at the base of the thermocline (50 m), surface chlorophyll a distribution, and biomass of calanoid copepods (a), cyclopoid copepods (b), and all copepodite stages (c). Bottom panel: geostrophic current pattern in summer 2009 at the base of the thermocline (30 m), surface chlorophyll a distribution, and biomass of calanoids (d), cyclopoids (e), and all copepodite stages (f). 

The satellite-derived Chla concentration (as an indicator of phytoplankton biomass) showed remarkable coupling with the circulation pattern. During the winter of 2006, a distribution pattern was observed in the northern bay in connection with the Gulf of California, where high values (~3 mg·m^-3) formed a Chla-enhanced area with a belt shape around the cyclonic eddy that gradually decreased toward the center (Fig. 2a-c); during the summer of 2009, the Chla concentration was lower than that in winter, yielding the highest values in the southern coastal region with a secondary peak (0.8 mg·m^-3) forming a circular shape around the cyclonic eddy (Fig. 2d-f). The Chla and sea surface temperature results support these observations. As expected, a highly inverse and statistically significant correlation between Chla and sea surface temperature was found in both winter (R = -0.60, P = 0.0002) and summer (R = -0.69, P = 0.0001) (Fig. 3). To establish the coupling between the circulation patterns obtained at the times of our observations of Chla concentrations and the copepods studied (calanoids, cyclopoids, and all copepodite stages), the results of the biomass for each group were superimposed (Fig. 2). The results showed a distribution pattern with progressive changes (biomass decreases) from the connection with the Gulf of California into the bay, as well as from the periphery to the center of the cyclonic eddy. A peak of copepod biomass was observed on the western coast close to Punta Las Tarabillas (Fig. 1b), where an intermittent stream is located.

Figure 3 Inverse and statistically significant correlation between sea surface temperature (registered with CTD) and chlorophyll a. (a) Winter (R = -0.60, P = 0.0002) and (b) summer (R = -0.69, P = 0.0001). 

In winter, there were 2 regions with maximum biomass values for the 3 target groups: one located at the entrance of the bay, in the Boca Grande region, with values between 1.4 and 5.1 mg·100 m^-3, and the other one in the western region close to the western coast, with values >5.0 mg·100 m^-3 (Fig. 2a-c) associated with an intermittent stream. Calanoids and copepodites showed a change pattern within the eddy field, with high values associated with the periphery of the eddy and defining its circumference. During summer, the highest values, located in the western region close to the coast (>5.0 mg·100 m^-3), coincided with secondary high values around the periphery of the eddy (Fig. 2d-f). In winter, calanoid biomasses were higher (the differences were statistically significant, with P = 0.0200) than those observed in summer.

The cyclopoid biomass distribution in both seasons also showed a pattern of progressive change from the center to the edge of the eddy system, with the highest values observed at the stations on the edge and a marked peak in their biomass close to the western coast of the bay. To clarify the differences in copepod biomass for each group and between the hydrographic stations considered in this study, we defined the stations belonging to the center, edge, or outside of the cyclonic eddy system, and then the mean and standard deviation were calculated. The results are summarized in Table 1, revealing the differences among the 3 regions (statistically significant differences, with P < 0.0400) in winter and summer; the lowest values were found at the stations related to the center of the eddy, while except for the station close to Punta Las Tarabillas mentioned above, the highest values were observed at the stations on the edge of the eddy. According to the mean values of copepod biomass, there was a pattern of alternance of calanoids-cyclopoids between winter and summer; at the center and edge of the cyclonic eddy in winter, calanoid biomass was higher (the differences were statistically significant, with P = 0.0200) than cyclopoid biomass, and the opposite was found in summer (Table 1).

DISCUSSION

In the last decade, important efforts have been made to understand the causes that support the high biological productivity of the southern Gulf of California and adjacent regions, such as the BoP. To date, a systematic monitoring program has been carried out with multidisciplinary oceanographic expeditions in different climatic seasons. As a result, the presence of a quasi-permanent cyclonic eddy confined to the interior of the BoP has been documented in different seasons, and the origin of this eddy is attributed to a combined effect of the wind pattern and the bathymetry of the region (^{Monreal-Gómez et al. 2001}, ^{Coria-Monter et al. 2017}). This cyclonic eddy induces changes in the hydrographic structure of the water column, generating changes in the three-dimensional distribution of the hydrographic parameters of the BoP, and it promotes the elevation of the thermocline, pycnocline, and nutricline (^{Sánchez-Mejía et al. 2020}). Thus, the cyclonic eddy induces an uplift of nutrients (particularly nitrate), which fertilize the euphotic zone and induce high Chla values in the area of influence of the eddy (^{Coria-Monter et al. 2017}, ^{Sánchez-Mejía et al. 2020}). This eddy has also been reported to exert an important impact on planktonic communities, generating a differential distribution of phytoplankton from the center to the periphery, with a dominance of dinoflagellates in the center and a dominance of diatoms in its periphery (^{Coria-Monter et al. 2014}), while differential aggregations of the main functional groups of zooplankton have been documented (^{Durán-Campos et al. 2015}).

The results of this study showed that the cyclonic eddy influenced the copepod biomass distribution through its actions on the hydrographic conditions, inducing high biomass values around the eddy system, possibly as a result of several processes, such as ecological interactions, population dynamics, and feeding habits, particularly by the high Chla concentrations observed around the eddy, which potentially represent a source of food for the copepods of interest in this study. Our results agree with those obtained by ^{Hernández-León et al. (2001)}, who documented a high variability of zooplankton biomass in a cyclonic eddy around the Canary Islands (Spain), with low values in the core of the eddy due to the divergent effect induced by its physical structure.

The influence exerted by the presence of cyclonic eddies, particularly on copepod populations, has been analyzed in a few domains worldwide. The horizontal and vertical distributions of 2 calanoid copepod species off Northwest Africa were analyzed by ^{Yebra et al. (2004)}, who found that the presence of eddies, both cyclonic and anticyclonic, had an important influence on copepod populations through the effects of eddies on hydrography, but the eddies also generated an advection of cold-enriched waters in the form of Chla filaments, benefiting the populations of both species. In the southern Gulf of California, the vertical distribution of the calanoid copepod community within a cyclonic eddy was documented by ^{Cruz-Hernández et al. (2018)}, who found differences between the center and the periphery of the eddy, with marked changes in the vertical distribution of the species and the thermocline being the most propitious region for calanoid copepod survival.

The results obtained in this study could also be related to bottom-up mechanisms given the presence of the cyclonic eddy that impacted the phytoplankton communities; the eddy induced high Chla levels and then exerted an influence on the copepod biomass, which may benefit the high trophic levels (e.g., fish) that feed in this region, as was previously noted by ^{Durán-Campos et al. (2019)}. High Chla concentrations around the eddy are attributed to mechanisms of fertilization by nutrients documented in these structures along with the mixing processes that occur on the periphery, ensuring the availability of nutrients for the phytoplankton communities and thus for the zooplankton grazers, such as copepods (^{Mahadevan 2016}).

In the southern Gulf of California and inside the BoP, changes in phytoplankton biomass were observed with season, with high values (>3 mg·m^-3) of Chla during winter and low values during summer as a result of typical periods of heating and cooling of the surface layers in summer and winter; these changes induce mixing in winter, thereby increasing the concentration of nutrients available for phytoplankton and leading to high Chla values (^{Álvarez-Borrego 2012}, ^{Durán-Campos et al. 2020}). ^{Martínez-López et al. (2012)} also reported this pattern of high biological productivity during winter in the Alfonso Basin inside the BoP.

The progressive changes observed in copepod biomass around eddies could be induced by the advection generated by divergent movements produced by the cyclonic structure, as previously noted by ^{Hernández-León et al. (2001)} for the eddies of the region of Gran Canaria, Canary Islands. Along this line of thought, the high copepod biomass values found in the 3 groups analyzed in this study near the connection with the Gulf of California can be explained by the presence of a bathymetric sill where important processes (e.g., hydraulic jumps) fertilize the euphotic zone (^{Salas-Monreal et al. 2012}). However, the highest biomass was observed in the western portion of the bay close to the coast of Punta Las Tarabillas. This can be explained by the presence of a phosphate mining industry that fertilizes the region, causing phytoplankton blooms that enhance zooplankton production.

The low concentration of Chla observed at the center of the eddy in both seasons could be related to the predominance of certain heterotrophic phytoplankton groups (e.g., dinoflagellates) (^{Lee 2008}, ^{Coria-Monter et al. 2014}). In addition, the high copepod abundance associated with high Chla values in the form of a belt-shaped area is related to the predominance of diatoms, as previously reported by ^{Coria-Monter et al. (2014)}.

In summary, the results presented here highlight the influence of physical processes on Chla and copepod distribution, which offers evidence to increase our understanding of the physical-biological interface. A complete assessment of this coupling poses big challenges. In this sense, more complete in situ observations are needed to improve the evaluation of different aspects of eddies, including differences in hydrographic parameters and dynamics, which could affect zooplankton communities that support many commercially important pelagic fish species (which usually consume zooplankton, particularly copepods). Finally, an understanding of copepod ecology is key to understanding fisheries production and achieving better management of marine resources.