Elsevier

Journal of Marine Systems

Volume 195, July 2019, Pages 50-66
Journal of Marine Systems

Lagrangian pathways in the southern Benguela upwelling system

https://doi.org/10.1016/j.jmarsys.2019.03.008Get rights and content

Highlights

  • Fish larvae pathways in the Southern Benguela Upwelling System

  • A 300 miles along shore fast route for fish larvae in the Southern Benguela

  • Lagrangian pathways in the Southern Benguela Upwelling System

Abstract

The effect of ocean currents on fish eggs and larvae during their journey from spawning to nursery grounds in the Southern Benguela upwelling system is poorly understood. The survival and successful transport of fish eggs and larvae results from complex biological and physical processes. This study focuses on the advective processes, more specifically on how the dynamics and characteristics of the ocean currents shape the Lagrangian pathways in the Southern Benguela. A mesoscale eddy resolving interannual (1989–2010) simulation of the region, with a horizontal resolution of 7.5 km, is used to study processes impacting the connectivity between the western edge of the Agulhas Bank and the west coast upwelling region. A set of Lagrangian experiments are conducted with particles being released within the top 100 m of the water column along an across-shore transect off Cape Point (34 S). Transport success is given by the ratio of the number of particles that reach St Helena Bay (32 S) over the total number of particles released. The model results show a strong seasonal cycle in transport success which is governed by the complex three-dimensional structure of the along-shore jets, their variability, together with the wind-induced Ekman drift. Transport success is most efficient in spring when the Benguela Jet consists of one coherent intensified single-core branch that flows over the 300 m isobath, and when wind-induced Ekman transport favours the retention of particles within the jet. At this time of the year, the pathway leading to successful transport is located inshore, with 90% of particles released inshore the 300 m isobath being successfully transported to St Helena Bay in <15 days. This pathway is also characterized by low eddy kinetic energy values.

During the upwelling season (December–March), transport success becomes less efficient, and less sensitive to the initial across-shore position of the particles. The inshore route no longer dominates, as the majority of particles follow offshore pathways. The Benguela Jet shifts offshore and splits into several branches due to the shoaling of the poleward undercurrent. The entrainment of particles within the offshore branch of the jet is favored by the dominating offshore wind-induced Ekman transport. Particles trapped in the offshore branch get exposed to higher mesoscale variability. Their northward progression is slower, which leads to journeys generally exceeding 20 days.

This study shows that successful transport from the Agulhas Bank to the west coast upwelling region cannot be attributed to only a simple wind induced modulation of the jet. It explores how the seasonal modulation of the Benguela Jet, poleward undercurrent and offshore Ekman transport combine together with the turbulent off-shelf eddy field to set-up the characteristics of transport success.

Introduction

In the southern Benguela upwelling system (SBUS), anchovy spawn in austral spring over the Agulhas Bank, a 200 m deep large coastal plateau that extends south of the tip of the African continent (Hutchings, 1992). The nursery area1 is located off the west coast of southern Africa, 400 km north of the spawning area. This makes the SBUS distinctive from its northern Benguela counterpart and the other major eastern boundary upwelling systems (EBUS) for which spawning and nursery grounds overlap. In their early-stage, anchovy larvae have minimal swimming ability, which makes them behave as passive particles. Therefore, their dispersion and transport, from the Agulhas Bank to the nursery area, are directly dependant on the oceanic circulation (Pagès et al., 1991).

Hutchings (1992) presented a schematic picture of the major processes impacting the eggs and larvae journey from the western Agulhas Bank to the west coast upwelling region (Fig. 1). This scheme was based on knowledge of the coastal circulation and on numerous observations of eggs and larvae distribution (Armstrong et al., 1987; Bang and Andrews, 1974; Shannon, 1985; Shelton and Hutchings, 1982; Shelton and Hutchings, 1990). A coastal current flowing northward and following the shelf, namely the Benguela Jet, is considered as the major conveyor belt linking the western Agulhas Bank and the west coast upwelling area.

According to Nelson (1989), the Benguela Jet is a narrow current (∼20 km) that flows northward along the South African west coast between Cape Point (34.15° S, 18.4° E) and Cape Columbine (33° S, 17.5° E). The jet can sustain high velocities, ranging from 25 cm·s−1 to 75 cm·s−1 (Bang and Andrews, 1974; Gordon, 1985). It lies over a poleward undercurrent (Nelson and Hutchings, 1983), that flows over the shelf edge and occasionally reaches the surface. The jet varies in position and strength, and responds rapidly to changes in wind stress magnitude and direction. It follows the seasonal cycle of the upwelling (Armstrong et al., 1987) and intensifies in summer when coastal upwelling of cold waters contributes to reinforce the across-shore density gradient (Veitch et al., 2006; Veitch et al., 2017). In winter, when the wind direction reverses, this across-shore density gradient weakens, but persists due to the intrusions of warm Agulhas waters from the South (Twatwa Mhlongo et al., 2005).

Hutchings' (1992) scheme for Lagrangian transports of fish eggs and larvae has been widely used to set-up numerous Lagrangian modelling studies in the Southern Benguela (Mullon et al., 2003). Huggett et al. (2003), followed by Parada et al., 2003, Parada et al., 2008, carried out pioneering work demonstrating that the dominant anchovy spawning patterns in the Southern Benguela could be reproduced when combining a 3D hydrodynamic model with a Lagrangian particle-tracking tool. They also quantified the impact, on eggs and larvae transport success, of parameters such as the initial location, density and patchiness of eggs, as well as behavioral processes such as diurnal vertical migration, mortality due to lethal temperatures or predation. However, neither the local pathways nor the physical processes responsible for the transport of particles were studied in detail, despite the fact that the success of their experiments was obviously linked to the capacity of the hydrodynamical model to reproduce the main characteristics of the currents in the region.

The modelling works of Veitch et al. (2006) and Blanke et al. (2009) provided insights on the importance of the Benguela Jet for the connectivity between the western Agulhas Bank and the west coast upwelling area. More recently, in a detailed modelling study of the shelf edge currents, Veitch et al. (2018) described the bifurcation of this Jet between Cape Point and Cape Columbine into a weak narrow branch flowing over the shelf edge and a strong offshore branch flowing in the northwestward direction. A bifurcation of the jet, as well as other physical processes such as induced offshore Ekman transport and occasional entrainment by Agulhas rings, have been suggested as important physical processes impacting the alongshore transit of fish eggs and larvae (Hutchings et al., 1998; Skogen et al., 2003; Veitch et al., 2006; Garavelli et al., 2012).

This work builds upon these recent physical modelling studies to investigate in details how the horizontal and vertical structure of the shelf edge jets and the presence of mesoscale turbulence impact the along-shore trajectory of passive particles in the SBUS. The approach consists in a set of Lagrangian particles tracking experiments based on a robust hydrodynamical model capable of simulating the major physical processes influencing the along-shore transport of particles, i.e. a realistic seasonal coastal circulation, a realistic mesoscale eddy field, Agulhas leakage bringing warm Agulhas waters offshore of the Benguela shelf.

While our study remains motivated by questions related to the dynamics of anchovy reproductive patterns and to recruitment variability, it distinguishes from the others as it focuses on the physical transport processes that impact the survival of individuals at early stages, whereas previous studies mainly focused on the sensitivity to biological processes (Lett et al., 2015). The Lagrangian experiments are built upon the pioneering work of Huggett et al. (2003) to focus exclusively on the northward alongshore portion of the hypothesized trajectory of anchovy larvae from the Agulhas Bank to St Helena Bay, i.e. the portion between Cape Point and St Helena Bay. If the Benguela Jet is assumed to play a dominant role in the transport of fish larvae, the success of this route should be in theory more predictable than the cross-shelf transport in the vicinity of the turbulent Agulhas retroflection.

A first step towards elucidating the role of the jet consists of identifying the different physical processes that control the seasonal cycle of this northward alongshore connectivity between Cape Point and St Helena Bay, just north of Cape Columbine. The Lagrangian experiments aim at addressing the following: (1) the identification of the dominant pathways that ensure this connectivity; (2) their characteristics and robustness; (3) the ocean dynamics that link to the alongshore transport success seasonal variability. Unlike the climatologically forced model of (Veitch et al. (2018), our ocean model is forced by 6-hourly surface forcing (wind, heat and freshwater fluxes) over the 1989–2010 period. The use of a 20 year simulation forced with a realistic non filtered wind field that conserves its full interannual, seasonal and intraseasonal variability gives robustness to our conclusions regarding the role of each individual process on the transport success of particles.

The outline of the paper is as follows. The numerical model and the Lagrangian experiments designed for the purpose of this study are presented in Section 2. This section also includes an evaluation of the model's realism by comparing simulated and observed sea surface temperature (SST) and eddy kinetic energy (EKE). Section 3 characterizes the seasonal cycle of the alongshore connectivity in terms of duration of travel and initial positions. The Lagrangian pathways taken by particles are identified in Section 4 while Section 5 investigates the link between the observed Lagrangian dispersal patterns and the across-shore structure of the ocean circulation, mainly the alongshore currents and the mesoscale turbulent eddy field. Finally, a summary and discussion of our results highlight the complexity of the across-shore structure of alongshore jet currents in the region.

Section snippets

Numerics

This study uses the Regional Ocean Modelling System (ROMS; Shchepetkin and McWilliams, 2005) in its Coastal and Regional Ocean Community (CROCO) version (Debreu et al., 2012; Penven et al., 2006). ROMS is a free-surface, terrain-following σ-coordinate model with split-explicit time stepping and with Boussinesq and hydrostatic approximations. The advection scheme is third-order upstream biased, which reduces dispersion errors, essentially enhancing precision for a given grid resolution (

Characteristics of the alongshore connectivity

The following section intends to characterize the seasonal cycle of the alongshore connectivity between the SARP line and St Helena Bay in terms of: i) transport success, ii) time taken by successful particles to travel northward, iii) initial vertical and across-shore position of successful particles.

Fig. 7 plots the monthly climatology of the alongshore wind stress at the SARP line together with transport success TSm and its standard deviation.

Both monthly alongshore transport success and its

Alongshore Lagrangian pathways

In order to get a general view of the routes taken by Lagrangian particles, maps of particles' spatial distribution (plumes) are plotted according to their age τ (in days) on the horizontal grid of the “BENGR15” simulation. The age of particles is zero when they are released. For a series of release events, we count all particles aged τ found within the water column of the grid cell located at position (x, y) in Cartesian coordinates. After dividing this by the total number of particles released

Characteristics of the ocean surface circulation

The Lagrangian experiments show that a “fast” inshore route and a “slow” offshore route co-exist and bring particles from Cape Point to St Helena Bay. The following section investigates how the Eulerian oceanic circulation of the model relates to these identified pathways.

Maps of monthly climatological surface currents show a nearshore equatorward jet, namely the Benguela Jet (Fig. 13). This jet can be observed all-year round, with a marked seasonal variability. In October, it is intensified

Discussion and conclusions

Lagrangian experiments, based on a 22 years (1989–2010) interannual simulation of the oceanic circulation in the SBUS, are used to describe the alongshore connectivity between Cape Point and St Helena Bay. Hutchings (1992) had previously pointed out the existence of an alongshore corridor followed by fish eggs and larvae that were spawn on the Agulhas Bank. A series of Lagrangian studies had then linked this corridor with the existence of the Benguela Jet, a shelf edge equatorward current (

Acknowledgements

The authors acknowledge the funding of N. Ragoasha's PhD by the South-African National Research Foundation (NRF, South Africa) and the French Institute for Research and Sustainable Development (IRD, France). This work was also supported by the French national program LEFE/INSU under the project's name Benguela Upwelling Innershelf Circulation (BUIC). This work was granted access to the HPC resources of [TGCC/CINES/IDRIS] under the allocation 2017- [DARI n° A0020107443] attributed by GENCI

References (54)

  • J. Veitch et al.

    Shelf-edge jet currents in the southern Benguela: a modelling approach

    J. Mar. Sys.

    (2018)
  • D.A. Armstrong et al.

    Physical and biological features across an upwelling front in the southern Benguela

    S. Afr. J. Mar. Sci.

    (1987)
  • N. Bang et al.

    Direct-current measurements of a shelf-edge frontal jet in southern Benguela system

    J. Mar. Res.

    (1974)
  • A. Beckmann et al.

    Numerical simulation of flow around a tall isolated seamount. Part i: problem formulation and model accuracy

    J. Phys. Oceanogr.

    (1993)
  • B. Blanke et al.

    Ocean variability over the Agulhas Bank and its dynamical connection with the southern Benguela upwelling system

    Journal of Geophysical Research: Oceans

    (2009)
  • X. Capet et al.

    Eddies in eastern boundary subtropical upwelling systems

  • X. Capet et al.

    Mesoscale to submesoscale transition in the California current system. Part I: flow structure, Eddy flux, and observational tests

    J. Phys. Oceanogr.

    (2008)
  • J.a. Carton et al.

    A reanalysis of ocean climate using simple ocean data assimilation (SODA)

    Mon. Weather Rev.

    (2008)
  • R.M. Castelao et al.

    The role of wind stress curl in jet separation at a cape

    J. Phys. Oceanogr.

    (2007)
  • C.W. Fairall et al.

    Cool-skin and warm-layer effects on sea surface temperature

    J. Geophys. Res.

    (1996)
  • J.L. Fowler et al.

    Transport of anchovy and sardine eggs and larvae from the western Agulhas Bank to the west coast during the 1993/94 and 1994/95 spawning seasons

    S. Afr. J. Mar. Sci.

    (1998)
  • J. Gan et al.

    A modeling study of shelf circulation off northern California in the region of the Coastal Ocean Dynamics Experiment 2. Simulations and comparisons with observations

    J. Geophys. Res.

    (2002)
  • J. Gan et al.

    A modeling study of shelf circulation off northern California in the region of the Coastal Ocean Dynamics Experiment: response to relaxation of upwelling winds

    J. Geophys. Res.

    (2002)
  • L. Garavelli et al.

    Modeling the dispersal of Cape hake ichthyoplankton

    J. Plankton Res.

    (2012)
  • A. Gordon

    Indian-Atlantic transfer of thermohaline water at the Agulhas retroflection

    Science

    (1985)
  • J. Gula et al.

    Submesoscale cold filaments in the gulf stream

    J. Phys. Oceanogr.

    (2014)
  • J. Huggett et al.

    Weekly variability of clupeoid eggs and larvae in the Benguela jet current: implications for recruitment

    S. Afr. J. Mar. Sci.

    (1998)
  • Cited by (8)

    • Particle trajectories in an eastern boundary current using a regional ocean model at two horizontal resolutions

      2022, Journal of Marine Systems
      Citation Excerpt :

      They found that the vertical distribution of eggs did influence retention, and at intermediate depths of 20–50 m, eggs avoided offshore Ekman drift and deep cold water. Ragoasha et al. (2019) investigated the physical mechanisms influencing egg and larval dispersal in the Benguela upwelling system. The researchers found that the combination of the Benguela jet, poleward undercurrent, and offshore Ekman transport combined with the off-shelf eddy field set up characteristics for particles transported and retained to nursery areas (Ragoasha et al., 2019).

    View all citing articles on Scopus
    View full text