Larval Drift Research Articles

Temperature selection and the final thermal preferendum of snow crab (Chionoecetes opilio, Decapoda) from the Barents Sea

The snow crab (Chionoecetes opilio) is an invasive species new to the Barents Sea that expands its geographic range by larval drift and adult migration. To evaluate the potential spreading of the species in the Barents Sea, we investigated temperature selection and the final thermal preferendum (FTP) of nine adult males in a free choice horizontal temperature gradient (~ 1.0–5.5 °C) for 24 h. The crabs displayed clear behavioural thermoregulation – at test start they explored the entire temperature range but eventually gravitated towards a FTP zone of 1.0–1.6 °C (mean 1.4 °C) after 6 h in the gradient. Our tests show that adult male snow crab is limited to cold waters, and suggest a spreading further into the Euro-Arctic shelf seas.

A multidisciplinary analytical framework to delineate spawning areas and quantify larval dispersal in coastal fish

Assessing larval dispersal is essential to understand the structure and dynamics of marine populations. However, knowledge about early-life dispersal is sparse, and so is our understanding of the spawning process, perhaps the most obscure component of biphasic life cycles. Indeed, poorly known species-specific spawning modality and species-specific early-life traits, as well as the high spatio-temporal variability of the oceanic circulation experienced during larval drift, hamper our ability to appraise the realized connectivity of coastal fishes. Here, we propose an analytical framework which combines Lagrangian modelling, network theory, otolith analyses and biogeographical information to pinpoint and characterize larval sources which are then grouped into discrete spawning areas. Such well-delineated larval sources allow improving the quantitative evaluations of both dispersal scales and connectivity patterns. To illustrate its added value, our approach is applied to two case-studies focusing on Diplodus sargus and Diplodus vulgaris in the Adriatic sea. We evidence robust correlations between otolith geochemistry and modelled spawning areas to assess their relative importance for the larval replenishment of the Apulian coast. Our results show that, contrary to D. sargus, D. vulgaris larvae originate from both eastern and western Adriatic shorelines. Our findings also suggest that dispersal distances and dispersal surfaces scale differently with the pelagic larval duration. Furthermore, 30.8% of D. sargus larvae and 23.6% of D. vulgaris larvae of the Apulian populations originate from Marine protected area (MPA), exemplifying larval export from MPAs to surrounding unprotected areas. This flexible multidisciplinary framework, which can be adjusted to any coastal fish and oceanic system, exploits the explanatory power of a dispersal model, fine-tuned and backed-up by observations, to provide more reliable scientific basis for the management and conservation of marine ecosystems.

Modelling of larval dispersal of Baltic flounder (Platichthys solemdali) revealed drifting depth as a major factor determining opportunities for local retention vs large-scale connectivity

For flatfishes, transportation of larvae to nursery areas is regarded a key mechanism for recruitment, with adaptations in larval behaviour to reach a suitable habitat. Here we model different possibilities of larval drifting of coastal spawning Baltic flounder Platichthys solemdali (recently identified as a species from European flounder P. flesus) to reveal opportunities for local retention vs large-scale dispersal to ensure settling in coastal nursery areas. Drifting depth, duration of drifting and effects of year and time during season were modelled using 1) a high-resolution local dispersal model, and 2) a large-scale connectivity database. The outcome revealed drifting depth as a major factor affecting larval dispersal. Drift at 10–22 m depth involved retention along the coast with the majority of larvae (≥94% or 69–93% according to 1 and 2, respectively) with end points ≤20 km from the coast enabling further successful migration to nursery habitats. Contrary, larval drift close to the surface resulted in advection with end points in the open sea (72–76%), i.e. loss of larvae, but with a small fraction (5–12%) displaying cross-basin connectivity. The results suggest, in agreement with depth distribution of spawning, a larval behaviour promoting drift in the lower part of the water mass, favouring retention close to coastal nursery areas. Obtained dispersal patterns may sustain both local recruitment but also connectivity with other areas, potentially explaining the low genetic diversity between areas for P. solemdali. Low inter-annual variability in dispersal patterns when drifting at 10–22 m depth suggests that larval drift is not a major bottleneck explaining recruitment variability in P. solemdali in the area. The study highlights the differences in life-history strategies of the species pair of flounder in the Baltic Sea; P. flesus spawning in the deep basins with extensive larval dispersal, and coastal spawning P. solemdali with, according to the model outcome, mainly local larval dispersal for sustaining a viable population, i.e. request for different management strategies.

Spawning by the European eel across 2000 km of the Sargasso Sea.

It has been known for about a century that European eels have a unique life history that includes offshore spawning in the Sargasso Sea about 5000-7000 km away from their juvenile and adult habitats in Europe and northern Africa. Recently hatched eel larvae were historically collected during Danish, German and American surveys in specific areas in the southern Sargasso Sea. During a 31 day period of March and April 2014, Danish and German research ships sampled for European eel larvae along 15 alternating transects of stations across the Sargasso Sea. The collection of recently hatched eel larvae (≤12 mm) from 70° W and eastward to 50° W showed that the European eel had been spawning across a 2000 km wide region of the North Atlantic Ocean. Historical collections made from 1921 to 2007 showed that small larvae had also previously been collected in this wide longitudinal zone, showing that the spatial extent of spawning has not diminished in recent decades, irrespective of the dramatic decline in recruitment. The use of such a wide spawning area may be related to variations in the onset of the silver eel spawning migration, individual differences in their long-term swimming ability, or aspects of larval drift.

Risk of oil contamination of fish eggs and larvae under different oceanic and weather conditions

Abstract An oil drift model is applied to determine the spread of oil spills from different locations along ship lanes off southern Norway every month for 20 years. These results are combined with results from an egg- and larvae drift model for Atlantic cod (Gadus morhua) to determine their risk of being impacted by oil. The number of eggs and larvae exposed to oil contamination is connected to environmental conditions. The highest risk of overlap between an oil spill and cod in early life stages occurs during March and April when the eggs and larvae concentrations are highest. Spills off the west coast pose a greater risk because of the ship lanes’ proximity to the spawning grounds, but there is large interannual variability. For some spill locations the interannual variability can be explained by variability in wind and ocean currents. Simultaneously occurring onshore transports lead to a high-risk situation because both oil and larvae are concentrated towards the coast. This study demonstrates how results from oil drift and biological models can be combined to estimate the risks of oil contamination for marine organisms, based on the location and timing of the oil spill, weather/ocean conditions, and knowledge of the organisms’ life cycle.

Genetic diversity and population structure of a cyprinid fish ( Ancherythroculter nigrocauda ) in a highly fragmented river

In this study, samples of Ancherythroculter nigrocauda, an endemic fish in the upper Yangtze River, were collected above and below dams in the Longxi River, a tributary of the upper Yangtze River, China, to investigate the genetic impacts of dams. The mitochondrial cytochrome b gene (cyt b) and 13 microsatellite (SSR) loci were used to analyze whether dams have resulted in loss of genetic diversity of the two fragmented populations or caused genetic differentiation between them. The results showed that the haplotype diversity (0.488; 0.486), nucleotide diversity (0.084%; 0.082%) and average expected heterozygosity (0.652; 0.676) of the two populations were all at a low level, and recent bottlenecks were detected. However, there was no genetic differentiation detected by the low genetic differentiation index (F-st, cyt b: -0.1677, p = 0.99707; SSR: 0.00259, p = 0.81427). Besides, 11 pairs of half-sibling relationship were found between the two populations indicating that there were individual movements and gene flow between them. This could be the larvae moving from upstream to downstream when water spilled over dams in flooding season. Therefore, our analysis showed that the dams have caused a loss of genetic diversity of the populations of A. nigrocauda in the Longxi River, blocked the active upstream movement but allowing passive downstream drift of larvae.

Local egg production and larval losses to advection contribute to interannual and long-term variability of American lobster (Homarus americanus) settlement intensity

American lobster (Homarus americanus) egg production and settlement intensity were examined over a 19-year period (1995–2013) in the Gulf of St. Lawrence at the Magdalen Islands (MI), where the population is spatially isolated during the benthic phase. Settlement and hatch dates by year were back-calculated from observed young-of-the-year size structure and juvenile and larval growth models. Drift of locally released larvae, from stage I to the end of stage III, was simulated using an ocean circulation model. Settlement intensity was related positively to egg production and negatively to drift distance. There was a strong positive trend in settlement intensity explained largely by increasing egg production, as well as by declining larval duration and drift distance. In the last years of the study, settlement intensity may have been limited by nursery saturation. The results suggest that demographic connectivity through larval drift is highly dynamic in time and that it declined during our study period. The demographic dependence of the MI lobster population on other populations in the Gulf of St. Lawrence is probably low.

An Intermittent Stream Supports Extensive Spawning of Large‐River Native Fishes

AbstractIntermittent or ephemeral streams make up a large percentage of all stream habitats and may have significant roles in spawning, foraging, refugia, and early life history habitat for many fishes. From 2015 to 2017 we examined the use of Cottonwood Creek, an intermittent tributary in the Gunnison River basin, Colorado, by spawning Flannelmouth Sucker Catostomus latipinnis, Bluehead Sucker C. discobolus, and Roundtail Chub Gila robusta. We used combinations of a stream‐spanning picket weir and PIT‐tag antennae near the mouth of the creek to determine fish movement in and out of the stream, which was more effective for the suckers than for Roundtail Chub. Large numbers of suckers used the stream each year despite very different flow regimes. The timing of initial fish entry varied by 6 weeks across 3 years of study from March 17 to May 6. Immigration was only loosely and inconsistently correlated to abiotic factors (various stream temperature metrics and discharge) among years and was most heavily dependent on the presence of sufficient water in the stream to permit fish passage. Average residency times of suckers ranged from 26 to 37 d depending on year and species, and spawning fish lost 10–18% of their body weight depending on species and sex. Both sucker species displayed a high level of annual site fidelity—61–71% of fish returned the year after being tagged. Larval drift was sporadic and many larvae failed to evacuate the stream before flow ceased each year. While further research is needed to assess the overall recruitment contribution from Cottonwood Creek to the regional population, this study revealed use of an intermittent tributary by thousands of native Colorado River fishes, highlighting the importance of nonperennial waters for the completion of the life histories of some large‐river fish species.

Tolerable ranges of fluid shear for early life-stage fishes: implications for safe fish passage at hydropower and irrigation infrastructure

Egg and larval fish drifting downstream are likely to encounter river infrastructure leading to mortality. Elevated fluid shear is one likely cause. To confirm this and determine tolerable strain rates resulting from fluid shear, egg and larvae of three Australian species were exposed to a high-velocity, submerged jet in a laboratory flume. Mortality was modelled over a broad range of strain rates, allowing critical thresholds to be estimated. Eggs were very susceptible to mortality at low strain rates and 100% of golden and silver perch died once strain rate exceeded 629 and 148s–1 respectively. Larvae were less vulnerable than eggs, but mortality increased at higher strain rates and at younger ages. Most ages of larvae will be protected if strain rate does not exceed 600s–1, although a lower guideline of less than 400s–1 may be needed in areas where very early stage Murray cod larvae drift. Golden perch and silver perch were not susceptible to shear once maturity reached ~25 days post-hatch (nearing juvenile metamorphosis). The thresholds described here will prove useful when refining design and operational guidelines for hydropower and irrigation infrastructure to improve fish survival.

Survival and drifting patterns of grass carp eggs and larvae in response to interactions with flow and sediment in a laboratory flume.

A series of laboratory experiments was conducted to better understand the behavior of grass carp eggs and larvae in moving water in order to develop and implement new strategies for control and prediction of their dispersal and drift at early life stages. Settling velocity and density of a representative sample of eggs were estimated, and three trials of flume experiments with different flow conditions were conducted with live eggs in a temperature-controlled setting with a mobile sediment bed. In these trials, egg and larval stages were continuously analyzed over periods of 80 hours; and eggs and larvae interactions with the flow and sediment bed were monitored and characterized qualitatively and quantitatively. Survival rates were quantified after each trial, highlighting physical causes for increased mortality. Detailed flow analysis was correlated to the observed drifting and swimming behavior of eggs and larvae, to estimate distributions across the water depth, as well as traveling and swimming speeds. Evidence of the influence of mean and turbulent flow in the suspension and transport of eggs are reported, and swimming patterns of larvae at different developmental stages are described. These findings support the development of new strategies for monitoring the spread of grass carp eggs and larvae in rivers, and provide new inputs to predict conditions favorable for spawning and hatching, allowing for mitigation measures at early life stages, which are critical to control their dispersal.

Contemporary and historical river connectivity influence population structure in western brook lamprey in the Columbia River Basin

The western brook lamprey Lampetra richardsoni (WBL) is a small non-parasitic lamprey that inhabits rivers and streams from southern Alaska to northern California. WBL remain in fresh water throughout their entire life and show limited dispersal. Although adults may migrate short distances upstream to spawn, most movement likely occurs through passive drifting of larvae downstream. Genetic differentiation among populations is thus expected to be high, even within a single basin, but WBL population structure has received little attention. The present study examined population connectivity of WBL from 23 sites throughout the Columbia River Basin and coastal Washington, using eight microsatellite loci and cytochrome b sequence data. Although population structure generally corresponded to contemporary river connectivity, there were some cases where genetic patterns were better explained by historical connections. Microsatellite genetic differentiation among populations separated by < 570 km was moderate to high; FST values ranged from − 0.0026 to 0.7117 and averaged 0.2929. Tributary distance was the best predictor of FST, suggesting that most gene flow takes place in tributaries rather than through the mainstem of the Columbia River. As predicted, gene flow occurred primarily in a downstream direction, resulting in lower genetic diversity in upstream sites. WBL populations in these areas may be particularly vulnerable to local extinction. Therefore, whereas anadromous lamprey management efforts are focused on improving passage at mainstem dams, conservation of WBL will require protection of individual watersheds with particular emphasis on headwater areas.

Ontogenetic changes in swimming speed of silver carp, bighead carp, and grass carp larvae: implications for larval dispersal.

Bighead, silver, and grass carps are invasive in the waterways of central North America, and grass carp reproduction in tributaries of the Great Lakes has now been documented. Questions about recruitment potential motivate a need for accurate models of egg and larval dispersal. Quantitative data on swimming behaviors and capabilities during early ontogeny are needed to improve these dispersal models. We measured ontogenetic changes in routine and maximum swimming speeds of bighead, grass, and silver carp larvae. Daily measurements of routine swimming speed were taken for two weeks post-hatch using a still camera and the LARVEL program, a custom image-analysis software. Larval swimming speed was calculated using larval locations in subsequent image frames and time between images. Using an endurance chamber, we determined the maximum swimming speed of larvae (post-gas bladder inflation) for four to eight weeks post-hatch. For all species, larval swimming speeds showed similar trends with respect to ontogeny: increases in maximum speed, and decreases in routine speed. Maximum speeds of bighead and grass carp larvae were similar and generally faster than silver carp larvae. Routine swimming speeds of all larvae were highest before gas bladder inflation, most likely because gas bladder inflation allowed the fish to maintain position without swimming. Downward vertical velocities of pre-gas bladder inflation fish were faster than upward velocities. Among the three species, grass carp larvae had the highest swimming speeds in the pre-gas bladder inflation period, and the lowest speeds in the post-gas bladder inflation period. Knowledge of swimming capability of these species, along with hydraulic characteristics of a river, enables further refinement of models of embryonic and larval drift.

Evaluating flow management as a strategy to recover an endangered sturgeon species in the Upper Missouri River, USA

AbstractIn the Upper Missouri River, Fort Peck and Garrison Dams limit the length of free‐flowing river available to the endangered pallid sturgeon. These barriers restrict the upstream migration of adults and downstream larval dispersal. A one‐dimensional (1D) modelling framework is currently in use to evaluate reservoir operation alternatives and to simulate drift of dispersing free embryos for different flow regimes and reservoir stages. This paper presents the results of a large‐scale tracer experiment conducted in 2016 and associated modelling performed to evaluate flow management scenarios that might aid species recovery. Breakthrough curves from the tracer experiment were used to infer the 1D longitudinal dispersion coefficient from a parameter optimization procedure. Simulations generated using the calibrated 1D advection–dispersion model were compared with field observations of the passive tracer and with larval fish collected during a previous experiment in 2007. When used with the appropriate range of dispersion coefficients, the 1D modelling framework agrees well with the available direct measurements of larval drift distances. Although we cannot unequivocally state whether insufficient length of free‐flowing river alone is causing recruitment failure, given the current thermal regime and our understanding of pallid sturgeon development, the time required for pallid sturgeon to transition to the benthos and initiate feeding might exceed the duration of drift available given constraints of reservoir operations.

Biophysical modeling of survival and dispersal of Central and Eastern Baltic Sea flounder (Platichthys flesus) larvae

The period of larval drift into a suitable nursery area is considered to be of great significance for recruitment variability in flatfish. Here, a hydrodynamic model coupled with a Lagrangian particle tracking technique was utilized to study the drift from the first feeding larval stage until time of settlement of Central and Eastern Baltic flounder (Platichthys flesus), originating from spawning in the Baltic Sea deep basins, the Arkona- and Bornholm basin (central Baltic Sea), and the Gdansk deep and Gotland basin (eastern Baltic Sea). We examined the spatio-temporal dynamics of the probability to settle in preferred nursery habitat by detailed drift model simulations. The study suggests that the majority of larvae (89% and 74% for Central- and Eastern Baltic flounder, respectively) drift towards coastal areas and settle at metamorphosis ≤20 km from a sandy habitat enabling further migration to a preferred nursery area, i.e. larval drift seems not to be a major bottleneck in recruitment of flounder spawning in the Baltic Sea deep basins. The drift model results suggest that Central Baltic flounder utilize nursery areas mainly in the central and western Baltic, and in the Kattegat, whereas Eastern Baltic flounder mainly utilize the coast in the central and eastern Baltic. Thus, the two stock components seem to use different nursery areas following settlement. Further, in accordance with the “nursery size hypothesis”, the model demonstrates that larvae from the Bornholm basin, utilizing areas with extensive distribution of preferred nursery habitat, display the highest relative successful transport to nursery grounds until settling (72% of successfully settled larvae), suggesting that spawning in the Bornholm Basin is of great importance for stock recruitment of deep basin spawning Baltic flounder.

Species‐specific fish larvae drift in anthropogenically constructed riparian zones on the Vienna impoundment of the River Danube, Austria: Species occurrence, frequencies, and seasonal patterns based on DNA barcoding

AbstractAs a result of river regulations over several centuries, followed by restoration measures in recent decades, most of the River Danube shoreline is man‐made, primarily riprap, but some reconstructed gravel banks and riparian side arms. We investigated the effects of these different structures on fish larval dispersal over a 20‐km stretch in Vienna via the use of drift nets. The habitats examined were created 18 years ago when the impoundment of the Danube hydropower station Vienna/Freudenau was constructed. About 15,000 fish larvae were trapped, and a subsample was determined to species level by DNA barcoding. In total, 26 different species were detected, including 10 species that are endangered or in danger of extinction. When species composition was considered, cyprinids become dominant at sites downstream of gravel bars, whereas in riprap sections, the majority of the larvae consist of invasive Gobiidae. Side arm habitats provide spawning and nursery grounds for additional species. Furthermore, clear species‐related seasonal patterns were observed with peak densities and multiple spawning periods of some species being recorded. The largest peak of Percidae occurred in the first half of May, followed by Cyprinidae at the end of May and Gobiidae in mid‐June.

Juvenile lake sturgeon monitoring and determinants of year‐class strength in the Rupert River, mid‐northern Québec, Canada

In 2007, Hydro-Québec began the construction of the Rupert Diversion in conjunction with the Eastmain-1A and Sarcelle powerhouses. The partial diversion of the Rupert River became operational in 2009. Mitigation measures to preserve lake sturgeon (Acipenser fulvescens) habitat downstream of the diversion include an instream flow, weir and spurs to maintain water levels, and fish passage channels and spawning grounds. An environmental follow-up was done in the reduced-flow section of the Rupert. The baseline status was established from 2007 to 2009 and follow-up studies were conducted from 2010 through 2012, and in 2014 and 2016. Besides presenting results from Hydro-Québec's environmental monitoring, analyses were performed to search for determinants of year-class strength. The results of the lake sturgeon monitoring activities indicate that the abundance of juveniles ≤8-year-old in the reduced flow section of the river remained similar or increased. Although larval production increased in post diversion conditions, cohort strength tended to decrease as did juvenile growth. Year-class strength was positively correlated with spring and summer flow. Also, a significant, strong negative correlation was found between estimated larval abundance and water temperature during larval drift.

Cast away: drift as the main determinant for larval survival in western fire salamanders (Salamandra salamandra) in headwater streams

AbstractUnidirectional drift is amongst the most relevant population regulators in riverine animal populations. Drift occurs randomly, but it can also be a behavioural response to abiotic and biotic stressors, or the result of catastrophic events such as heavy rain. In this study, we investigate the spatiotemporal distribution of larvae of the fire salamander (Salamandra salamandra) in three headwater streams in the Eifel mountains, Western Germany. We marked 568 salamander larvae to conduct capture–recapture estimates of abundance, and followed their individual fate throughout the season. Population density of larval salamanders along the stream showed both strong temporal and spatial variation. Random drift of salamander larvae was particularly rare, and drift did not result in a size structuring of larval populations in the field. Yet, catastrophic events of high summer precipitation were the best predictor for pulse‐wise restructuring of the salamander larvae's density distribution. Strong discharge coincided with a severe decrease in population density, and thus appears to be the most relevant cause of mortality for salamander larvae. Furthermore, we find strong indications that drift has the potential to influence this species’ evolutionary processes, as well as its ecological prevalence.

Regionalization in the distribution of larval fish assemblages during winter and autumn in the Gulf of California

Larval fish composition and abundance were analyzed during February-March (winter) and November-December (autumn) of 2005 to investigate which environmental conditions and mesoscale oceanographic processes affect the distribution of larval fish assemblages in the Gulf of California, and if these variables and processes are useful to delimit specific regions of ichthyofaunal distribution. The overall pattern was mostly a latitudinal gradient of the abundance of two faunistic complexes of larval fish assemblages (temperate northern and tropical southern complexes) strongly associated with the spatial distribution of sea surface temperature. The integration of these results, with previous investigations in the region, lets us confirm that the northern and southern larval fish assemblages extend and contract their distribution ranges according to the latitudinal displacement of the 21°C sea surface isotherm. We identified this to be a valuable criterion to establish the southern distribution limit of the temperate larval fish assemblage, and the 18°C as the northern limit of the tropical larval fish assemblage. However, this general pattern is maintained only in absence of mesoscale oceanographic features (upwelling and eddies) that increase larval drift from the coast to the central region of the Gulf, particularly during the November-December period when longitudinal gradients were stronger than in February-March.

Modelling benthic invasion by the colonial gastropod Crepidula fornicata and its competition with the bivalve Pecten maximus. 2. Coupling the 0D model of colony-forming species to a connectivity matrix for a realistic distributed simulation of benthic invasion

The anthropogenic introduction in U.K. waters of the north-American marine gastropod Crepidula fornicata (Linné, 1758), commonly called slipper limpet, and its consecutive spreading has led in less than a century to the invasion of a part of benthic grounds along the North-European coasts. Competition for space has hampered the maintenance of the native scallop Pecten maximus, whereas dredge clogging has drastically limited scallop fishing, especially in the Western English Channel. In order to assess the possible future distribution and abundance of both species (Crepidula and Pecten), an original model of slipper limpet chains joined to a simple year-class model of the scallop (Ménesguen and Grégoris, 2017) has been coupled to a connectivity matrix summarizing the annual dispersion of emitted larvae in a realistic marine domain. This distributed model can explore the dynamics of both populations at the century scale and has been applied at two different geographical scales: a local one (the bays of Brest and Douarnenez, Western Brittany, France) and a regional one (Bay of Biscay and English Channel). First eigenvectors of the connectivity matrix are used to delineate the main retention areas. For each species taken alone, simulations starting with a few animals in different spots do converge in less than a century towards the same geographic distribution, compatible with the actual field distribution. This suggests that hydrodynamic patterns of larval drift create a strong attractor for these populations; it is reached by different routes of colonization, depending on the initial inoculation. The non-linear interaction created by the lack of space when the populations have filled the benthic area seems to be stronger for scallop populations than for slipper limpets, because of the permanent capacity of Crepidula beds to fix some larvae on the top of existing colonies. Whereas the scallop abundance experiences a rather large limit-cycle, with a dominant 11 years period and a lot of harmonics, the slipper limpet has commonly a stable steady state abundance. Competition between both populations tends to lower the mean scallop abundance and to lengthen and damp its fundamental period of oscillations. Simulated distributions can be locally improved by linking the larval survival to a distributed environmental stressor, as the salinity for scallop or a metallic contamination for the slipper limpet in the case of the bay of Brest.

Modeling the drift of European (Anguilla anguilla) and American (Anguilla rostrata) eel larvae during the year of spawning

The distribution of the leptocephalus larvae of European (Anguilla anguilla) and American (Anguilla rostrata) eels collected during recent Sargasso Sea surveys was used to model larval drift. The drift trajectories of individual larva were back-calculated to the estimated time of spawning, using current data from two global oceanographic assimilation models. The results of both models give the same overall result; widespread spawning extended in time from December to March. The drift was also calculated forwards for approximately 1 year. The forward drift modelling showed that most leptocephali remained in the area south of the Subtropical Frontal Zone. One conclusion is that the majority of leptocephali remain trapped and possibly die in the retention area. A small proportion of leptocephali are entrained into the Gulf Stream system. An implication is that the spawning success may be highly sensitive to oceanographic and climatic factors that alter the dispersion of leptocephali out from the retention area. An alternative interpretation is that the surveys were made too late after the peak spawning period and that the core spawning area was missed.