ICES Journal of Marine Science: Journal du Conseil Advance Access originally published online on March 30, 2007
ICES Journal of Marine Science: Journal du Conseil 2007 64(3):519-530; doi:10.1093/icesjms/fsm022
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Published by Oxford University Press (2007)
Oceanic changes in the Sargasso Sea and declines in recruitment of the European eel
1 National Marine Fisheries Service, 28 Tarzwell Drive, Narragansett, RI 02882, USA
2 Ocean Research Institute, The University of Tokyo, 1-15-1 Minamidai, Nakano, Tokyo, 164-8639 Japan
3 University of Westminster, 115 New Cavendish St., London W1W 6UW, UK
Correspondence to K. D. Friedland: tel: +1 401 782 3236; fax: +1 401 782 3201; e-mail: kevin.friedland{at}noaa.gov
Friedland, K. D., Miller, M. J., and Knights, B. 2007. Oceanic changes in the Sargasso Sea and declines in recruitment of the European eel. – ICES Journal of Marine Science, 64: 519–530.Anguillid eel recruitment in the North Atlantic has declined in recent decades, raising concerns that climatic changes in the Sargasso Sea may be influencing oceanic reproduction and larval survival. There is a significant negative correlation between the North Atlantic Oscillation and long-term variations in catches of glass eel stages of the European eel Anguilla anguilla recorded by the fishery independent Den Oever recruitment index (DOI) in the Netherlands, lagged by one year. Ocean-atmospheric changes in the Sargasso Sea may affect the location of spawning areas by silver eels and the survival of leptocephali during the key period when they are transported towards the Gulf Stream. A northward shift in a key isotherm (22.5°C), defining the northern boundary of the spawning area, a declining trend in winds and transport conditions in larval transport areas, and a shallowing of the mixed layer depth could affect primary productivity in areas where leptocephali feed. The relationships between these ocean parameters and the DOI suggest that these changing ocean conditions could be contributing to declining recruitment of the European eel and probably also of the American eel (A. rostrata), but anthropogenic factors during their continental life stage must also be considered.
Keywords: American eel, European eel, North Atlantic Oscillation, ocean climate, recruitment
Received 7 September 2006; accepted 4 February 2007; advance access publication 30 March 2007.
 |
Introduction |
|---|
 Top Introduction Material and methodsResultsDiscussionReferences |
|---|
The declining recruitment of anguillid eels documented on both sides of the Atlantic Ocean in recent decades (Haro et al., 2000; Casselman, 2003; Dekker et al., 2003) has been hypothesized by Castonguay et al. (1994) and Knights (2003) to be related to changes in ocean conditions affecting the survival of leptocephali, their transparent leaf-like larvae that are born in the Sargasso Sea (Schmidt, 1923). These declines may also be related to reductions in the size of the spawning population of each species (Dekker, 2003, 2004), driven by anthropogenic factors such as habitat loss, pollution, parasite introductions, and overfishing (Haro et al., 2000; Feunteun, 2002). However, anguillid eels are facultatively catadromous fish that leave their freshwater and estuarine juvenile growth habitats to spawn in the open ocean, so most of the key life history events associated with reproduction and recruitment take place during the marine phase. It has been difficult to study the possible mechanistic links between changes in ocean conditions and reproduction and recruitment in eels because of an almost complete lack of information on key factors such as spawner escapement, egg production, and survival of larvae.
Maturing European, Anguilla anguilla, American, A. rostrata, and Japanese, A. japonica, silver eels make long migrations to spawn in the subtropical gyres of the North Atlantic or North Pacific oceans (McCleave et al., 1987; Tsukamoto, 1992). The distribution of the youngest larvae suggests that spawning takes place in specific areas that may facilitate mating and help place larvae in the appropriate west-flowing currents (Kleckner and McCleave, 1988; McCleave, 1993; Tsukamoto et al., 2002; Tsukamoto, 2006). In the case of the two Atlantic species, distinct temperature fronts present in winter and spring in the Sargasso Sea (Weller, 1991) form the northern boundary of the spawning area (Kleckner et al., 1983; Kleckner and McCleave, 1988; Tesch and Wegner, 1990); these fronts may be used as cues to help eels locate the spawning area (McCleave, 1987).
During the spawning season of Atlantic eels in winter and spring (Schoth and Tesch, 1982; Kleckner and McCleave, 1985, 1988; Wipplehauser et al., 1985; Tesch and Wegner, 1990) and in subsequent larval migration in summer and early autumn (Boëtius and Harding, 1985; Kleckner and McCleave, 1985; McCleave and Kleckner, 1987), the leptocephali of A. anguilla and A. rostrata are widely distributed in overlapping areas of the Sargasso Sea. After hatching, the leptocephali are present primarily in the upper 250 m of the ocean and are generally most abundant in the upper 100 m, although the degree of vertical migration is difficult to evaluate owing to net avoidance by day (Schoth and Tesch, 1984; Castonguay and McCleave, 1987). Leptocephali grow larger and have a longer larval duration than most fish larvae, the northern temperate species taking up to a year or longer before recruiting to estuarine and freshwater habitats as glass eels or elvers.
This long larval duration probably makes the larvae particularly vulnerable to changes in ocean currents and the availability of food. The dominant climate-forcing signal in the North Atlantic, the North Atlantic Oscillation (NAO), has been associated with a wide range of oceanographic and biological changes (Curry and McCartney, 2001; Parsons and Lear, 2001; Weijerman et al., 2005), including eel recruitment (Knights, 2003). However, it remains to be determined how alterations of the ocean-atmospheric system signalled by regime shifts in the NAO influence eel reproduction or larval survival. This is in part because many aspects of spawning and the early life histories of the species are poorly understood. Anguillid leptocephali appear to be adapted to grow in low-productivity open ocean environments as they drift west towards western boundary currents such as the Gulf Stream and Kuroshio, which transport them to their continental habitats (McCleave, 1993; Tsukamoto et al., 2002). They appear to feed at a very low trophic level, utilizing particulate organic matter (POM) such as marine snow, including discarded larvacean houses (Urochordata), faecal pellets, and possibly even dissolved organic matter (DOM) (Otake et al., 1993; Mochioka and Iwamizu, 1996; Pfeiler 1999). Availability of POM and DOM and the formation of marine snow are dependent on primary and secondary production, and their availability is strongly coupled to phytoplankton blooms in the Sargasso Sea and other regions of the North Atlantic (Turner, 2002).
Because anguillid leptocephali live primarily in the surface layers of the oceans, their growth, transport, and eventual recruitment are likely to be affected by variability in the biological and physical processes in the upper few hundred metres. Changes in oceanographic conditions caused by ocean-atmospheric regime shifts signalled by the NAO in the Atlantic (Knights, 2003) or El Niño in the Pacific (Kimura et al., 2001; Kimura and Tsukamoto, 2006) have been hypothesized as possible factors contributing to the recent declines in northern temperate species of anguillid eels such as the European eel. Although the authors cited above found correlations between recruitment of glass eels and the NAO and ENSO indices, respectively, they were not able directly to determine the mechanisms controlling recruitment success. Developing and testing hypotheses about which specific oceanic factors have the greatest effect on recruitment success in anguillid eels requires exploratory and experimental research. Therefore, the objective of the present study was to determine whether decade-scale changes and regime shifts in key thermal, wind, oceanic currents, and productivity associated with the NAO are correlated with European eel recruitment and to hypothesize how these changes may have contributed to the observed declines in recruitment of glass eels.
|
Material and methods |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
The year-class strength or recruitment success of the European eel has been assessed in various ways, but the index of standardized catches of glass eels at Den Oever in the Netherlands [Den Oever recruitment index, DOI] is an important time-series because it is based on fishery-independent sampling, which makes it particularly useful in studies of multi-decade climate signals (Dekker, 1998). Glass eels have been collected each year since 1938 (except during World War II) at Den Oever, which is located at the outflow from Lake IJsselmeer (52°56'20'N 05°02'70''E) into the Wadden Sea. The eels are collected at night at 2-h intervals in spring with a 1-m2 dipnet with 1 mm mesh. The DOI data were log-transformed to approximate normality (ICES, 2005).
The temporal changes in the DOI were compared with the NAO using time-series techniques and a regime shift indicator. The NAO winter index (December through March) was used in the analysis (Hurrell et al., 2001); NAO is defined as the difference in normalized sea level pressure between Lisbon, Portugal, and Stykkisholmus/Reykjavík, Iceland, since 1864. Comparisons between NAO and DOI were made at a 1-y lag (NAOt vs. DOIt +1) to account for the duration of the leptocephalus stage. Trends in DOI and NAO indices were analysed with a smoothing filter. The smoothing filter was a Fast Fourier Transform filter that removes Fourier components with frequencies higher than a cut-off frequency of
|
|
t is the time (or more generally the abscissa) spacing between two adjacent data points. Larger values of n result in lower cut-off frequencies, and therefore a greater degree of smoothing. The function used to clip out the high-frequency components is a parabola with a maximum of 1 at zero frequency, falling to zero at the cut-off frequency defined above. We used a value of n = 5.
The regime shift indicator of Rodionov (2004) was used to assess whether there had been major changes in trends of the DOI and NAO in particular years. The method is based on sequential t-test analysis of regime shifts where, with each step of a time-series, it is determined whether the new values represent a significant deviation from the mean value of the current regime. If the test indicates a significant difference, the year is marked as one showing a potential regime change, and subsequent observations are used to confirm or reject the hypothesis. The hypothesis is tested using the regime shift index (RSI), which is calculated for each change point:
|
|
l is the average standard deviation for all one-year intervals in the time-series. RSI represents a cumulative sum of normalized deviations xi* from the hypothetical mean level for the new regime, for which the difference from the mean level for the current regime is statistically significant according to a Student's t-test. We also applied red noise estimation to account for time-series autocorrelation (Rodionov, 2006). The relationship between the DOI and NAO was tested using Pearson product-moment correlation at a 1-y time lag, because it takes leptocephali at least one year to migrate from the spawning area to Europe (McCleave, 1993). The temporal relationship between regime shifts was further examined by calculating a cross-correlation function between DOI and NAO over ±15 lags. The temporal variation in the DOI was then compared with ocean-atmospheric parameters in the region of the Sargasso Sea encompassing most of the likely spawning area of the European eel. These parameters were sea surface temperature (SST) distributions on the spawning grounds, wind in the larval transport area, currents in the larval transport area, and mixed layer depth (MLD) in the early larval development area. The longitudinal zone analysed is located on the eastern side of where the smallest sized leptocephali have been collected (McCleave et al., 1987), based on the assumption that spawning takes place slightly to the east of where larvae are actually collected, so allowing for some westward drift.
During the time period covered by the DOI, there have been changes in the distribution of SST in the North Atlantic and southern Sargasso Sea (Grey et al., 2000; Polyakov et al., 2005), where both European and American eels spawn. Temperature fronts in the Sargasso Sea may act as cues that help adult eels locate the spawning area (McCleave, 1987; Kleckner and McCleave, 1988). Therefore, we analysed the latitude of the surface expression of the 22.5°C isotherm, which is typically near the frontal zone at the northern limit of eel spawning (Kleckner and McCleave, 1988; Tesch and Wegner, 1990). The 22.5°C isotherm was selected because it has been located near the northernmost temperature front observed to form at the same water density in all previous studies of the region prior to 1994 (Kleckner and McCleave, 1988; Tesch and Wegner, 1990, Miller and McCleave, 1994). An "Isotherm Analysis" was conducted over a range of longitudes that encompass the spawning area (Figure 1) during the months of December–March to bracket the time of spawning (Kleckner and McCleave, 1988; Tesch and Wegner, 1990). Specifically, the analysis was carried out using the extended reconstructed SST data set of Smith and Reynolds (2003, 2004) for 22–32°N and 50–66°W. The latitude of the 22.5°C isotherm was determined for each of eight 2° longitude bins for each of the four months. The time-series was restricted to 1947–2004 to avoid problems with the DOI (missing values and the war years). Correlations were then calculated between the DOI and the latitude of the 22.5°C isotherm for each 2° longitude bin for each month from December to March. Correlations were made at a 1-y lag, as with the NAO data.
|
Variations in the winds in the Sargasso Sea were also compared with the DOI at a 1-y lag. Winds can affect larval transport or other conditions in the Sargasso Sea, so influencing recruitment. Wind data for the months February–May for the years 1949–2003 over an area that includes the spawning and early stage larval transport area were used. This area includes part of the Gulf Stream through which many older leptocephali probably pass (Figure 1). Wind data were obtained from the NCEP/NCAR (National Centers for Environmental Prediction/National Center for Atmospheric Research) re-analysis (Kalnay et al., 1996). The vector wind velocities (u and v vectors) were compared with the DOI. To evaluate the directional effect of wind on larvae, Ekman volume transports were calculated for selected time periods for both zonal and meridional transport vectors:
|
|
0 the density of seawater (1028 kg m–3), and 
w the wind stress. The absolute value of zonal and meridional wind stress was computed as
|
|
a is the density of air (1.29 kg m–3), Cd the drag coefficient, and U10 the zonal or meridional winds at 10 m. Drag coefficient was parameterized as a non-linear function of wind speed rather than a constant, based on Large and Pond (1981), and further modified for low wind speeds following Trenberth et al. (1990):
|
|
For the same time periods, Ekman depth was calculated as
|
|
is the latitude in radians. In addition to examining the time-series of Ekman transports for the Sargasso region, we also considered the climatological currents for the area using the US Coastguard's Mariano Global Surface Velocity Analysis (MGSVA) 1.0 (Mariano et al., 1995). The data have a monthly temporal resolution and a spatial resolution of 1° of longitude and latitude. The currents are derived from ship-drift estimates made by merchant and military ships. Our intention was to evaluate how variation in wind-driven transport may affect the movement of larvae into the major geostrophic transport currents.
MLD was also compared with the DOI because it is associated with nutrient circulation, which could have an affect on productivity in the Sargasso Sea and subsequently on growth and survival of larvae. We calculated the MLD at Hydrostation S located southeast of Bermuda (Figure 1), using temperature at depth data monitored at approximately bimonthly intervals by the Bermuda Biological Station for Research (http://www.bbsr.edu/). MLD was calculated for the two contrasting periods of eel recruitment, pre- and post-1985, as guided by the regime shift analysis (see above). Two different computational algorithms were used to account for deficiencies in the data. The -criterion method (Kara et al., 2003) is based on a change in temperature (T = 0.8°C) from the temperature at a reference depth (10 m). The Hmix-criterion (Lorbacher et al., 2005) method is based on an estimate of the shallowest extreme curvature of a near-surface temperature profile and has the added benefit of providing a goodness-of-fit criterion.
The time-series we compared in this study had varying degrees of autocorrelation. We sought to account for this autocorrelation in our correlation analyses (Pearson product-moment correlation) by adjusting the effective degrees of freedom of each test according to the procedure suggested by Pyper and Peterman (1998). The effective degree of freedom (N*) of a correlation between two time-series, in notation series X and Y, was estimated by
|
|
xx (j) and yy (j) are the autocorrelations of X and Y at lag j. Following Garrett and Petrie (1981), we took the autocorrelation at lag j of the cross-products of standardized time-series of X and Y. The probability associated with a correlation coefficient is designated as p, and as p* for a test with degrees of freedom based on N*. |
Results |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
Recruitment of the European eel, as represented by the DOI since 1938, has exhibited a multi-decadal pattern of change that is inversely related to the NAO winter index. The DOI was low during the 1940s and 1950s, followed by a sustained period of high index values that suggest favourable recruitment through the 1970s (Figure 2a). The DOI began to decline in the 1980s, reaching its lowest level in the last few years of the time-series. Although the decline in recruitment appeared to have begun several years earlier, the most prominent regime shift in the DOI to a low, but relatively stable, level appears to have begun around 1987. The NAO trend changed from positive to negative and back to a prolonged period of high positive values between 1938 and 2005 (Figure 2b). The most prominent positive regime shift in the NAO appears to have begun around 1981, coinciding with apparent initial decreases in the DOI. There was significant cross-correlation between the DOI and the NAO around lag 0 and interestingly around 8, which would support the lag between regime transitions in the two time-series (Figure 3). The cross-correlation at lag 0 would correspond to the correlation reported by Knights (2003) of a long-term negative relationship between the DOI and the NAO using a 1-y lag, and this relationship was reaffirmed in the present study (r = –0.352; p = 0.003, p* = 0.010). This correlation, we believe, may be related to a NAO effect on larvae of that year. The cross-correlation at the 8-y lag would suggest a different mechanism and would be consistent with the idea that the NAO effect on recruitment is delayed by the approximate time between larval recruitment and spawning.
|
|
The distribution of SST in the spawning area of eels in the Atlantic has changed over time, reflecting a general warming trend (Knights, 2003). The iso-correlate surface of correlation between the DOI and the latitude of the key 22.5°C isotherm shows that negative correlation peaked in January and February in an area centred at 60°W (Figure 4a). However, this region of the correlation surface was also associated with a high degree of autocorrelation in both the recruitment and latitude time-series. Although correlations in other regions of the response surface are significant in both uncorrected and autocorrelation-corrected significance tests, this region is not significant at the conventional 95% confidence level in the autocorrelation-corrected test. In looking at the correlation probabilities in closer detail, we are limited to stating that the correlations associated with January and February in the area centred at 60°W are significant at about a 93–94% confidence level (Table 1). The January and February isotherms moved from as far south as 27°N and 26°N during the 1950s to as far north as 29°N and 28°N in the most recent years of the time-series, respectively (Figure 4b). Therefore, the northward shift in the 22.5°C isotherm since about 1970 may have caused a northward shift in the spawning area, and this trend is correlated with decreases in recruitment.
|
|
Wind speed and direction over the Sargasso Sea were also correlated with recruitment of the European eel. The u-vector velocities, especially those in March at latitudes 27–36°N, and to a lesser degree in February, suggest the importance the east–west aspect of the winds in the area (Figure 5). Uncorrected and autocorrelation-corrected correlation probabilities were virtually the same in these data, largely because of the lack of autocorrelation in the wind-speed data. The v-vector velocities were generally weakly correlated with the DOI (Figure 6). Moreover, autocorrelation-corrected probabilities associated with the March v-vector velocities were mostly non-significant at the 95% confidence level. The wind speed of u-vector winds decreased from 6 m s–1 to <3 m s–1, whereas v-vector wind speed increased slightly over the period. These correlations reflect a weakening of southwest winds over the northern Sargasso Sea (the latitudinal extent of the Sargasso Sea is taken to be 25–35°N) in recent decades. Ekman transports of near-surface waters of the northern Sargasso during March declined to negligible levels in a comparison of mean transports by decade (Figure 7a–f). Ekman transports during the 1950s and continuing through the 1970s were to the south and southeast, whereas transports during the 1980s through to the period 2000–2005 became more to the southeast and eventually declined to negligible transports. Ekman mixing depths also decreased, as shown by a similar analysis of mean depth during March by decade (Figure 8a–f). Waters of the northern Sargasso had Ekman depths on the order of 30–50 m during the first three decades of the time-series, whereas the values shallowed to 10–20 m in recent years.
|
|
|
|
Climatological current data for the Sargasso Sea and surrounding ocean area during March show that the weak currents of the Sargasso Sea are bracketed by the west-flowing Antilles Current to the south and the east-flowing Gulf Stream to the north (Figure 9). Although most current vectors of the Sargasso Sea are very small, it is apparent that the currents of the southern Sargasso Sea (25–30°N) tend to the west whereas the vectors of the northern Sargasso (30–35°N) tend to the east and southeast.
|
Spring MLD has shallowed in the Sargasso Sea, suggesting that related changes in transport and the nutrient dynamics of leptocephalus habitats have also occurred. Owing to the elimination of some MLD estimates because of poor model fits, we compared MLD before and after the DOI regime shift demarked by the year 1985 (Figure 2). These comparisons indicate that the MLD during the period February–May has decreased since 1985 (Figure 10) and, for the spring months of March, April, and May, the decrease has been on the order of 50–100 m. Therefore, the regime of low DOI and high NAO is associated with decreased MLD.
|
|
Discussion |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
Our findings provide evidence of linkages between declines in recruitment of the European eel and specific environmental changes within the spawning and early larval development areas of eels in the Sargasso Sea. European eels may be responding to NAO-forced changes in fronts and currents affecting spawning and the survival of leptocephali in a similar manner to that proposed for the Japanese eel A. japonica, which are linked to El Niño forcing (Kimura et al., 2001; Kimura and Tsukamoto, 2006). Japanese eel migration in the North Pacific is believed to be affected by the salinity front associated with the North Equatorial Current, which has moved north in recent years, resulting in poor transport conditions for leptocephali. The northward displacement of thermal fronts in the southern Sargasso Sea would be expected to have similar ramifications for the migration and transport of European eels. However, our analysis suggests that a number of ocean condition parameters have changed in the Sargasso Sea, which in combination form a more complex set of factors that may influence recruitment. Such changes could also have affected recruitment of the American eel because of the proximity of its spawning area to that of the European eel and similarities in the early migration pathways of both species.
Because of the many factors that could be involved, the finding of a correlation between the DOI and the NAO does not resolve the issue of what is causing the declines of the northern temperate anguillids. However, there is mounting evidence that the NAO is related to significant changes in the physical and biological structure of the North Atlantic, affecting a wide range of marine organisms, from zooplankton to fish (Parsons and Lear, 2001; Weijerman et al., 2005). In the northern Sargasso Sea, specifically the Bermuda Atlantic Time-Series site, SST, MLD, salinity, primary productivity, new production, and TCO2 appear to be related to the NAO in recent years (Bates, 2001; Nelson et al., 2004). In our analysis, we detect associations between eel recruitment and thermal, wind, and mixing parameters in areas of the Sargasso that correspond spatially to eel spawning and larval ecology. These parameters co-vary with the NAO and with each other, suggesting that they may be acting synergistically on eel recruitment.
Changes in the temperature structure of the Sargasso Sea could alter thermal and density features that affect the oceanic stages of the eels of the Atlantic. Our analysis of the location of the 22.5°C isotherm during the early spawning season indicates that there has been a gradual warming trend of surface waters of the region that began around 1970. Warming trends in the global oceans have been reported (Levitus et al., 2000), and both warming and cooling periods have been noted in the upper layer of the North Atlantic region, including the southern Sargasso Sea (Molinari et al., 1997; Grey et al., 2000; Polyakov et al., 2005). The latitude of the 22.5°C isotherm reaches its southernmost position sometime during the period January–April (Halliwell et al., 1991), which corresponds to the spawning season, before moving north and eventually going subsurface as the SST rises across the whole basin in summer (Steinberg et al., 2001). That isotherm fluctuated to the north and to the south several times between 1940 and 1970, but in the last few years of the analysis, the latitudes during each of the four months have been farther north than in any year since 1940 (Figure 4b). Knights (2003) also showed positive correlations between Sargasso Sea temperature anomalies at 100–250 m and the DOI between 1952 and 1995, which may be related to the same trends of warming in the region.
The 22.5°C isotherm is located within the most distinct portion of the fronts that consistently form at the 25.6 density isopycnal in the Sargasso Sea (Kleckner and McCleave, 1988; Eriksen et al., 1991; Pollard and Regier, 1992; Miller and McCleave 1994). It is a useful indicator of the northern limit of spawning by both species of eels in the Atlantic, because studies on the distributions of the leptocephali of both species indicate that they spawn south of the isotherm (Kleckner and McCleave, 1988; Tesch and Wegner, 1990; McCleave, 1993). Frontal jets also form in association with these fronts (Eriksen et al., 1991), which appear to transport a variety of leptocephali species eastwards (Miller and McCleave, 1994). Therefore, changes in the latitude or intensity of these fronts may affect both the spawning location and the subsequent transport of the leptocephali to continental habitats, if transport out of the Sargasso Sea is more efficient with more southerly spawning. If this is true, a northward movement of the fronts that form the northern boundary of spawning may have resulted in increased retention of leptocephali within the Sargasso Sea gyre, which could reduce the number of successful recruits each year.
Changes in the winds and wind-driven Ekman transports in the Sargasso Sea could also have an effect on larval transport out of the Sargasso Sea. We found a reduction of u-vector winds during late winter, in particular during March, which was highly correlated with the pattern of decline of the DOI. Any reductions in easterly winds in the southern Sargasso might slow the rate of westward transport of leptocephali toward the Florida Current or into the Gulf Stream, or increase the likelihood of some being retained in the Sargasso Sea gyre by eddies if these winds are important in maintaining the westward Antilles Current in the southern Sargasso Sea. Transport in the Antilles Current would help leptocephali reach the Gulf Stream, which transports European eel leptocephali into the North Atlantic Drift that carries them towards continental waters (McCleave, 1993). American eel leptocephali must also reach the Florida Current or Gulf Stream, but they later have to detrain from that current system to recruit to North America.
Although the Antilles Current probably transports recently spawned leptocephali westwards, leptocephali become widely distributed in summer as the surface layer warms in the Sargasso Sea (Boëtius and Harding, 1985; Kleckner and McCleave, 1985; McCleave and Kleckner, 1987). This suggests that the transport of many leptocephali is not direct. The flow structure in the southwestern Sargasso, through which anguillid leptocephali must pass to enter the Florida Current, is complex, but the Antilles Current northeast of the northern Bahamas appears to be present throughout most of the year (Amos et al., 1971; Ingham, 1975; Gunn and Watts, 1982; Fiadeiro and Veronis 1983; Olson et al., 1984; Reverdin et al., 2003; Meinen et al., 2004). Anguillid leptocephali are abundant in this region in October (Kleckner and McCleave, 1982), so it may be important in transporting leptocephali out of the Sargasso Sea.
Changes in the winds in the northern Sargasso Sea, which have reduced southward Ekman transports, could also be contributing to increased larval retention in the Sargasso Sea gyre. Southward Ekman transports could tend to move leptocephali that detrain from the southern edge of the Gulf Stream in eddies towards the westward flow farther south in the Sargasso Sea. A reduction in this transport in summer and autumn when leptocephali leave the Sargasso Sea could affect the transport of some, resulting in increased retention in the northern Sargasso Sea. These possible factors, coupled with the potential for changes in current velocities, directions, or recirculation patterns (Curry and McCartney, 2001; Bersch, 2002; Molinari, 2004) influencing the transport of leptocephali across the North Atlantic basin to Europe, could contribute to the declines in European eel recruitment. Modelling exercises of the transport of European eel leptocephali suggest that after spring spawning, leptocephalus migration trajectories may ensure a degree of synchronization with later spring/summer plankton blooms (Kettle and Haines, 2006). Hence, changes in the speed or the pathway of transport, or reductions in primary productivity in the northern and eastern regions of the extensive larval migration could reduce the growth and survival of leptocephali.
Interannual changes in the productivity and temperature structure of the surface layer in the larval development area of eels in the Atlantic could be influenced by the recent reductions in the MLD of the northern Sargasso Sea observed here and previously (Palter et al., 2005). Decreases in the MLD could be associated with reductions in upward mixing of deeper nutrient-rich water, resulting in less primary production in the surface layer, especially during spring thermocline overturns (Kimura et al., 2001; Knights, 2003). Although the factors influencing the upward flux of nutrients into the euphotic zone are complex and poorly understood, the strength of winds and the actions of eddies have been proposed as important mechanisms in the mixing of nutrients into the surface layer (McGillicuddy and Robinson, 1997; McGillicuddy et al., 1998; Stammer et al., 2001; Palter et al., 2005). These types of factors have been associated with increased primary productivity after hurricanes have traversed the Sargasso Sea (Babin et al., 2004).
If changes in the MLD significantly influence the primary productivity or community structure of the Sargasso Sea, then the growth or survival of anguillid leptocephali may have been affected. The feeding ecology of leptocephali is poorly understood, but they appear to feed on POM such as marine snow, including discarded larvacean houses (Otake et al., 1993; Mochioka and Iwamizu, 1996), and gain their nutrition at the microbial trophic level (Otake et al., 1993; Pfeiler, 1999). Therefore, changes in water column structure that influence primary productivity and the production of marine snow in the Sargasso Sea could affect the availability of food for leptocephali during their critical first months of feeding and for growth during their larval migration.
There is also some evidence that larval growth or the duration of larval migration may influence the timing and size of glass eels upon arrival, suggesting that oceanic conditions may be affecting recruitment on a long-term or seasonal basis. Dekker (2004) suggested that leptocephalus size has changed on a multi-decadal scale, and except for the last part of the time-series, glass eel size may have declined along with recruitment. Desaunay and Guerault (1997) analysed these patterns and also examined the otolith microstructure of glass eels to estimate their hatching dates and migration histories. These findings are intriguing in relation to the findings of the present study, but many questions about the early life history of the European eel complicate interpretations of the causal mechanisms of these types of observations (McCleave et al., 1998). Reductions in growth rates or lengthening of the duration of larval migration could result from the oceanic changes described here, but more basic life history information is needed to help resolve the questions about larval duration, migratory behaviour, and routes of migration of eels in the Atlantic before links can be made with changes in oceanic conditions.
The correlations we have found between key environmental factors acting in the spawning and larval development areas of eels in the Atlantic provide preliminary evidence that oceanic factors may have contributed to the recent declines in recruitment of European and American eels. These correlations with specific factors in the ocean provide the first conceptual basis for linking recruitment to ocean-atmosphere changes in the Sargasso Sea, but the lack of knowledge about the marine phase of the eels makes it currently impossible to determine the importance of the correlations. Our general lack of knowledge about eels is not limited to the early larval stages however, and oceanic changes may act on later larval stages as they approach or cross continental shelves. For example, Knights (2003) suggested that unfavourable changes in wind-driven coastal currents could be important in the transport and survival of glass eels and juveniles in the North and Baltic seas.
In addition, complex ocean-atmosphere changes and anthropogenic factors may be acting simultaneously and interactively to control eel recruitment. Many anthropogenic factors have been listed as likely to reduce eel population sizes or to reduce the number of spawners each year (Haro et al., 2000; Feunteun, 2002; Casselman, 2003; Dekker et al., 2003; Dekker, 2003, 2004), so the causes of the declines of northern temperate eel species cannot be attributed simply to any single cause. Determining cause-effect relationships is complicated by a lack of basic information on silver eel escapement and on the spawning and survival of eggs and larvae. More long-term research on specific changes in the Sargasso Sea and North Atlantic, and on the biology, reproductive ecology, and recruitment patterns of eels in the Atlantic is needed. The potential causal mechanisms proposed here, however, provide a framework for future studies to examine the relationships between ocean-atmosphere changes and recruitment of anguillid eels.
|
Acknowledgements |
|---|
We thank J. A. Hare and W. Dekker for productive reviews of the manuscript.
|
Footnotes |
|---|
* Present Address: B. Knights, King's College London, 150 Stamford Street, London SE1 8WA, UK
|
References |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
-
Amos A. F., Gordon A. L., Schneider E. D. (1971) Water masses and circulation patterns in the region of the Blake-Bahama Outer Ridge. Deep-Sea Research 18:145–165.
Babin S. M., Carton J. A., Dickey T. D., Wiggert J. D. (2004) Satellite evidence of hurricane-induced phytoplankton blooms in an oceanic desert. Journal of Geophysical Research 109:C03043 doi:10.1029/2003JC001938.[CrossRef]
Bates N. R. (2001) Interannual variability of oceanic CO2 and biogeochemical properties in the western North Atlantic subtropical gyre. Deep-Sea Research II 48:1507–1528.[CrossRef]
Bersch M. (2002) North Atlantic Oscillation-induced changes of the upper layer circulation in the northern North Atlantic Ocean. Journal of Geophysical Research 107:3156 doi:10.1029/2001JC000901.[CrossRef]
Boëtius J. and Harding E. F. (1985) A re-examination of Johannes Schmidt's Atlantic eel investigations. Dana 4:129–162.
Casselman J. M. (2003) Dynamics of resources of the American eel, Anguilla rostrata: declining abundance in the 1990s. In Aida K., Tsukamoto K., Yamauchi K. (Eds.). Eel Biology(Springer, Tokyo) pp. 255–274.
Castonguay M., Hodson P. V., Moriarty C., Drinkwater K. F., Jessop B. M. (1994) Is there a role of ocean environment in American and European eel decline? Fisheries Oceanography 3:197–203.
Castonguay M. and McCleave J. D. (1987) Vertical distributions, diel and ontogenetic vertical migrations and net avoidance of leptocephali of Anguilla and other common species in the Sargasso Sea. Journal of Plankton Research 9:195–214.
Curry R. and McCartney M. (2001) Ocean gyre circulation changes associated with the North Atlantic Oscillation. Journal of Physical Oceanography 31:3374–3400.[CrossRef][Web of Science]
Dekker W. (1998) Long-term trends in the glass eels immigrating at Den Oever, the Netherlands. Bulletin Francais de la Pêche et de la Pisciculture 349:199–214.
Dekker W. (2003) Did lack of spawners cause the collapse of the European eel, Anguilla anguilla? Fisheries Management and Ecology 10:365–376.[CrossRef][Web of Science]
Dekker W. (2004) Synthesis and analyses. Slipping through our Hands — Population Dynamics of the European Eel PhD thesis,(Faculteit der Natutuurweten Schappen, Wiskunde en Informatica, The Netherlands)127–145.
Dekker W., Casselman J. M., Cairns D. K., Tsukamoto K., Jellyman D., Lickers H. (2003) Worldwide decline of eel resources necessitates immediate action: Québec declaration of concern. Fisheries 28:28–30.
Desaunay Y. and Guerault D. (1997) Seasonal and long-term changes in biometrics of eel larvae: a possible relationship between recruitment variation and North Atlantic ecosystem productivity. Journal of Fish Biology 51:317–339.[CrossRef][Web of Science]
Eriksen C. C., Weller R. A., Rudnick D. L., Pollard R. T., Regier L. A. (1991) Ocean frontal variability in the frontal air–sea interaction experiment. Journal of Geophysical Research 96:8569–8591.
Feunteun E. (2002) Management and restoration of European eel population (Anguilla anguilla): an impossible bargain. Ecological Engineering 18:575–591.[CrossRef][Web of Science]
Fiadeiro M. E. and Veronis G. (1983) Circulation and heat flux in the Bermuda Triangle. Journal of Physical Oceanography 13:1158–1169.[CrossRef][Web of Science]
Garrett C. and Petrie B. (1981) Dynamical aspects of the flow through the Strait of Belle Isle. Journal of Physical Oceanography 11:376–393.[CrossRef][Web of Science]
Grey M., Haines H., Troccoli A. (2000) A study of temperature changes in the upper North Atlantic: 1950–94. Journal of Climate Change 13:2697–2711.
Gunn J. T. and Watts D. R. (1982) On the currents and water masses north of the Antilles/Bahamas arc. Journal of Marine Research 40:1–18.[Medline]
Halliwell G. R., Cornillon P., Brink K. H., Pollard R. T., Evans D. L., Regier L. A., Toole J. M., et al. (1991) Descriptive oceanography during the frontal air–sea interaction experiment: medium- to large-scale variability. Journal of Geophysical Research 96:8553–8567.
Haro A., Richkus W., Whalen K., Hoar A., Busch W-D., Lary S., Brush T., et al. (2000) Population decline of the American eel: implications for research and management. Fisheries 25:7–16.
Hurrell J. W., Kushnir Y., Visbeck M. (2001) The North Atlantic Oscillation. Science 291:603–605.[CrossRef][Web of Science]
ICES. (2005) ICES Document CM 2005/I: 01, Ref. G, ACFM.
Ingham M. C. (1975) Velocity and transport of the Antilles Current northeast of the Bahama Islands. Fishery Bulletin US 73:626–632.
Kalnay E., Kanamitsu M., Kistler R., Collins W., Deaven D., Gandin L., Iredell M., et al. (1996) The NCEP/NCAR 40-year reanalysis project. Bulletin of the American Meteorological Society 77:437–471.[CrossRef][Web of Science]
Kara A. B., Rochford P. A., Hurlburt H. E. (2003) Mixed layer depth variability over the global ocean. Journal of Geophysical Research 108:3079 doi:10.1029/2000C000736.[CrossRef]
Kettle A. and Haines K. (2006) How does the European eel (Anguilla anguilla) retain its population structure during its larval migration across the North Atlantic Ocean? Canadian Journal of Fisheries and Aquatic Sciences 63:90–106.
Kimura S., Inoue T., Sugimoto T. (2001) Fluctuation in the distribution of low-salinity water in the North Equatorial Current and its effect on the larval transport of the Japanese eel. Fisheries Oceanography 10:51–60.[CrossRef][Web of Science]
Kimura S. and Tsukamoto K. (2006) The salinity front in the North Equatorial Current: a landmark for the spawning migration of the Japanese eel (Anguilla japonica) related to the stock recruitment. Deep-Sea Research II 53:315–325.[CrossRef]
Kleckner R. C. and McCleave J. D. (1982) Entry of migrating American eel leptocephali into the Gulf Stream system. Helgoländer wissenschaftliche Meeresuntersuchungen 35:329–339.
Kleckner R. C. and McCleave J. D. (1985) Spatial and temporal distribution of American eel larvae in relation to North Atlantic Ocean current systems. Dana 4:67–92.
Kleckner R. C. and McCleave J. D. (1988) The northern limit of spawning by Atlantic eels (Anguilla spp.) in the Sargasso Sea in relation to thermal fronts and surface water masses. Journal of Marine Research 46:647–667.[Web of Science]
Kleckner R. C., McCleave J. D., Wippelhauser G. S. (1983) Spawning of American eel, Anguilla rostrata, relative to thermal fronts in the Sargasso Sea. Environmental Biology of Fishes 9:289–293.[CrossRef][Web of Science]
Knights B. (2003) A review of the possible impacts of long-term oceanic and climate changes and fishing mortality on recruitment of anguillid eels of the Northern Hemisphere. Science of the Total Environment 310:237–244.[CrossRef][Medline]
Large W. and Pond S. (1981) Open ocean momentum flux measurements in moderate to strong winds. Journal of Physical Oceanography 11:324–336.[CrossRef][Web of Science]
Levitus S., Antonov J. I., Boyer T. P. (2000) Warming of the world ocean. Science 287:2225–2229.
Lorbacher K., Dommenget D., Niiler P. P., Köhl A. (2005) Ocean Mixed Layer Depth: A Subsurface Proxy of Ocean-Atmosphere Variability. ECCO Report 38 http://www.ecco-group.org/.
Mariano A. J., Ryan E. H., Perkins B. D., Smithers S. (1995) The Mariano Global Surface Velocity Analysis 1.0. 55 USCG Report, CG-D-34-95.
McCleave J. D. (1987) Migration of Anguilla in the ocean: signposts for adults! Signposts for leptocephali? Proceedings of a Multidisciplinary Workshop on Marine Animal Orientation and MigrationSignposts in the Sea (Florida State UniversityIn Herrnkind W. F. and Thistle A. B. (Eds.). , Tallahassee) pp. 102–117.
McCleave J. D. (1993) Physical and behavioral controls on the oceanic distribution and migration of leptocephali. Journal of Fish Biology 43:(Suppl. A), 243–273.[Web of Science]
McCleave J. D., Brickley P. J., O'Brien K. M., Kistner-Morris D. A., Wong M. W., Gallagher M., Watson S. M. (1998) Do leptocephali of the European eel swim to reach continental waters? Status of the question. Journal of the Marine Biological Association of the UK 78:285–306.
McCleave J. D. and Kleckner R. C. (1987) Distribution of leptocephali of the catadromous Anguilla species in the western Sargasso Sea in relation to water circulation and migration. Bulletin of Marine Science 41:789–806.[Web of Science]
McCleave J. D., Kleckner R. C., Castonguay M. (1987) Reproductive sympatry of American and European eels and implications for migration and taxonomy. American Fisheries Society Symposium 1:286–297.
McGillicuddy D. J. and Robinson A. R. (1997) Eddy-induced nutrient supply and new production in the Sargasso Sea. Deep-Sea Research I 44:1427–1450.[CrossRef]
McGillicuddy D. J., Robinson A. R., Siegel D. A., Jannasch H. W., Johnson R., Dickey T. D., McNeil J., et al. (1998) Influence of mesoscale eddies on new production in the Sargasso Sea. Nature 394:263–266.[CrossRef][Web of Science]
Meinen C. S., Garzoli S. L., Johns W. E., Baringer M. O. (2004) Transport variability of the deep Western Boundary Current and the Antilles Current off Abaco Island, Bahamas. Deep-Sea Research I 51:1397–1415.[CrossRef]
Miller M. J. and McCleave J. D. (1994) Species assemblages of leptocephali in the subtropical convergence zone of the Sargasso Sea. Journal of Marine Research 52:743–772.[CrossRef][Web of Science]
Mochioka N. and Iwamizu M. (1996) Diet of anguillid larvae: leptocephali feed selectively on larvacean houses and fecal pellets. Marine Biology 125:447–452.
Molinari R. L. (2004) Annual and decadal variability in the western subtropical North Atlantic: signal characteristics and sampling methodologies. Progress in Oceanography 62:33–66.[CrossRef][Web of Science]
Molinari R. L., Mayer D. A., Festa J. F., Bezdek H. F. (1997) Multiyear variability in the near-surface temperature structure of the midlateral western North Atlantic Ocean. Journal of Geophysical Research 102:3267–3278.[CrossRef]
Nelson N. B., Siegel D. A., Yoder J. A. (2004) The spring bloom in the northwestern Sargasso Sea: spatial extent and relationship with winter mixing. Deep-Sea Research II 51:987–1000.[CrossRef]
Olson D. B., Schott F. A., Zantopp R. J., Leaman K. D. (1984) The mean circulation east of the Bahamas as determined from a recent measurement program and historical XBT data. Journal of Physical Oceanography 14:470–1487.
Otake T., Nogami K., Maruyama K. (1993) Dissolved and particulate organic matter as possible food sources for eel leptocephali. Marine Ecology Progress Series 92:27–34.[Web of Science]
Palter J. B., Lozier M. S., Barber R. T. (2005) The effect of advection on the nutrient reservoir in the North Atlantic subtropical gyre. Nature 437:687–692.[CrossRef][Medline]
Parsons L. S. and and. Lear W. H. (2001) Climate variability and marine ecosystem impacts: a North Atlantic perspective. Progress in Oceanography 49:167–188.[CrossRef][Web of Science]
Pfeiler E. (1999) Developmental physiology of elopomorph leptocephali. Comparative Biochemistry and Physiology A – Molecular and Integrative Physiology 123:113–128.[CrossRef]
Pollard R. T. and Regier L. A. (1992) Vorticity and vertical circulation at an ocean front. Journal of Physical Oceanography 22:609–625.[CrossRef][Web of Science]
Polyakov I. V., Bhatt U. S., Simmons H. L., Walsh D., Walsh J. E., Zhang X. (2005) Multidecadal variability of North Atlantic temperature and salinity during the twentieth century. Journal of Climate 18:4562–4581.[CrossRef][Web of Science]
Pyper B. J. and Peterman R. M. (1998) Comparison of methods to account for autocorrelation in correlation analyses of fish data. Canadian Journal of Fisheries and Aquatic Sciences 55:2127–2140.
Reverdin G., Niiler P. P., Valdimarsson H. (2003) North Atlantic Ocean surface currents. Journal of Geophysical Research 108:3002 doi:10.1029/2001JC001020.[CrossRef]
Rodionov S. N. (2004) A sequential algorithm for testing climate regime shifts. Geophysical Research Letters 31: L09204, doi:10.1029/2004GL019448.
Rodionov S. N. (2006) The problem of red noise in climate regime shift detection. Geophysical Research Letters 33: L12707, doi:10.1029/2006GL025904.
Schmidt J. (1923) The breeding places of the eel. Philosophical Transactions of the Royal Society 211:179–208.
Schoth M. and Tesch F-W. (1982) Spatial distribution of 0-group eel larvae (Anguilla sp.) in the Sargasso Sea. Helgoländer wissenshaftliche Meeresuntersuchungen 35:309–320.
Schoth M. and Tesch F-W. (1984) The vertical distribution of small 0-group Anguilla larvae in the Sargasso Sea with reference to other anguilliform leptocephali. Meeresforschung 30:188–195.
Smith T. M. and Reynolds R. W. (2003) Extended reconstruction of global sea surface temperatures based on COADS data (1854–1997). Journal of Climate 16:1495–1510.[Web of Science]
Smith T. M. and Reynolds R. W. (2004) Improved extended reconstruction of SST (1854–1997). Journal of Climate 17:2466–2477.[CrossRef][Web of Science]
Stammer D., Böning C., Dieterich C. (2001) The role of variable wind forcing in generating eddy energy in the North Atlantic. Progress in Oceanography 48:289–311.[CrossRef][Web of Science]
Steinberg D. K., Carlson C. A., Bates N. R., Johnson R. J., Michaels A. F., Knap A. H. (2001) Overview of the US JGOFS Bermuda Atlantic Time-series Study (BATS): a decade-scale look at ocean biology and biogeochemistry. Deep-Sea Research II 48:1405–1447.[CrossRef]
Tesch F-W. and Wegner G. (1990) The distribution of small larvae of Anguilla sp. related to hydrographic conditions in 1981 between Bermuda and Puerto Rico. Internationale Revue der Gesamten Hydrobiologie 75:845–858.[CrossRef]
Trenberth K. E., Large W. G., Olson J. G. (1990) The mean annual cycle in global ocean wind stress. Journal of Physical Oceanography 20:1742–1760.[CrossRef][Web of Science]
Tsukamoto K. (1992) Discovery of the spawning area for Japanese eel. Nature 356:789–791.[CrossRef]
Tsukamoto K. (2006) Spawning of eels near a seamount. Nature 439:929.[CrossRef][Medline]
Tsukamoto K., Aoyama J., Miller M. J. (2002) Migration, speciation, and the evolution of diadromy in anguillid eels. Canadian Journal of Fisheries and Aquatic Sciences 59:1989–1998.
Turner J. T. (2002) Zooplankton fecal pellets, marine snow and sinking phytoplankton blooms. Aquatic Microbial Ecology 27:57–102.[Web of Science]
Weijerman M., Lindeboom H., Zuur A. F. (2005) Regime shifts in marine ecosystems of the North Sea and Wadden Sea. Marine Ecology Progress Series 298:21–39.[Web of Science]
Weller R. A. (1991) Overview of the frontal air-sea interaction experiment (FASINEX): a study of air-sea interaction in a region of strong oceanic gradients. Journal of Geophysical Research 96:8501–8516.
Wippelhauser G. S., McCleave J. D., Kleckner R. C. (1985) Anguilla rostrata leptocephali in the Sargasso Sea during February and March 1981. Dana 4:93–98.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  H. Kim, S. Kimura, A. Shinoda, T. Kitagawa, Y. Sasai, and H. Sasaki Effect of El Nino on migration and larval transport of the Japanese eel (Anguilla japonica) ICES J. Mar. Sci., October 1, 2007; 64(7): 1387 - 1395. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. E. Maes and F. A. M. Volckaert Challenges for genetic research in European eel management ICES J. Mar. Sci., October 1, 2007; 64(7): 1463 - 1471. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||