Ocean Currents: Physical Drivers in a Changing World

By Robert Marsh and Erik van Sebille

Ocean Currents: Physical Drivers in a Changing World opens with a general introduction to the character, measurement, and simulation of ocean currents, leading to a physical and dynamical framework for understanding the wide variety of flows encountered in the oceans. The book comprises chapters covering distinct aspects of contrasting ocean currents: broad and slow, deep and shallow, narrow and swift, large scale and small scale, low latitudes and high latitudes, and moving in horizontal and vertical planes. Through this approach the authors cover a wide range of applications, from local to global, with considerable geographical context.

• Provides analyses of ocean observations and numerical model simulations, highlighting the pathways and drift associated with ocean currents, around the World Ocean, linked to online exercises for instructors and students that extend this perspective
• Presents applications to natural phenomena, showing how ocean currents shape marine ecosystems, helping researchers understand the distribution and adaptation of life in the oceans
• Addresses societal challenges, specifically how ocean currents disperse pollutants (e.g. plastic) from coastal sources and how the global ocean circulation is central to our changing climate, helping students and researchers develop an interdisciplinary approach to global environmental change

• Language: English
• Publisher: Elsevier Science
• Release date: Jun 30, 2021
• ISBN: 9780128160602

Robert Marsh

Professor Robert Marsh holds a Chair in Oceanography and Climate at the University of Southampton. With disciplinary expertise in Physical Oceanography, he has a wide range of experience across ocean and climate science, and specifically an in-depth knowledge of ocean currents. Examples of applied studies include the influences of ocean currents on sea turtle hatchlings, volcanic pumice and icebergs. He also co-pioneered the development and use of ocean, climate and Earth System models, and the water mass transformation framework that provides a novel perspective on physical and biogeochemical processes in the oceans. He has extensive experience of undergraduate and postgraduate oceanography teaching, both in the classroom and in the field. He is lead or co-author of around 100 peer-reviewed publications.

Book preview

Chapter 1: The restless ocean

Abstract

The ocean is a restless environment, where currents vary on a vast range of space and time scales, exerting profound influences on organisms, pollution, climate, and ultimately people. The ocean is also a hostile environment, a challenging place to make scientific observations. Over the last 70 years, however, a series of technological developments and international programmes have substantially improved both the quality and quantity of ocean observations. Observations of ocean currents in particular have been revolutionized by our ability to now measure flows from the surface to great depths, for long periods of time, in fine spatial detail, and at global scale. Alongside these measurements, computer simulations have kept pace, deepening understanding and facilitating prediction. Only with accurate measurement and sophisticated modelling of ocean currents can we explore intricate and complex drift around the global ocean, answering some of the most pressing environmental questions of our time.

Keywords

Ocean currents; Drift; Buoys; Dispersion; Current meters; Ocean models; Trajectories

Chapter Outline

1.1Ten big questions

1.2Organization vs. chaos

1.3Measuring the ocean—Challenges and international organization

1.4Measuring ocean currents

1.4.1Drift measurements

1.4.2Point measurements

1.4.3Profile measurements

1.4.4Radar measurements

1.5Estimating ocean currents

1.5.1Hydrography

1.5.2Bottom pressure recorders

1.5.3Satellite altimeters

1.5.4Electrical cables

1.6Computer simulation of ocean currents

1.7An ocean of scales

1.8Summary

References  

Unlabelled Image

Few people directly experience the powerful influence of our restless oceans in their daily lives, but public knowledge and perceptions are changing. While efforts to conserve marine environments and ecosystems have attracted broad support, the role of the oceans in climate change is now widely appreciated. Our seas and oceans increasingly demand the attention of policymakers, ranging from the United Nations, concerned with international treaties and regulations, to local authorities, concerned with management of ports and coastlines. For business and industry, our seas and oceans are integral to international trade, renewable energy, and fisheries.

The challenge to scientists, and to oceanographers specifically, is to measure, understand, and predict the character of our oceans. A fundamental aspect of this character is the constant motion of seawater—ocean currents. Ranging in strength from just a few centimetres per second in most places to 1–2 m per second (a brisk walking pace) in a few select locations, these currents connect widely separated locations—spanning regional seas, the basins that separate continents, and ultimately the entire global ocean. In recent years, this connectivity has been shown to be fundamentally important for many different marine ecosystems, and for the fate of our discarded material, such as our plastic waste. In addition, key environmental properties such as temperature and concentrations of dissolved substances are strongly controlled by ocean currents.

Glancing down from a window seat on a long-haul flight, the ocean appears static, and yet we know that it is in constant motion. So, why does the ocean move? In simplest terms, the ocean is pushed or pulled by forces associated with winds, Earth's gravity, and interactions between Earth, Moon, and Sun. Ocean currents are also related to spatial differences of seawater properties (water temperature and salt content), interactions of eddy fluctuations, and interior frictional stresses. Further complicating matters is the influence of Earth's rotation, leading to an evident turning of currents on time scales beyond a few hours. It is a major goal of physical oceanographers to quantify and understand how these forces and influences collectively drive the currents.

Ocean currents are many and varied, yet fundamental principles underpin a common dynamical framework that helps us to understand each in turn. Following this introductory chapter, we therefore outline this framework, along with a range of datasets and methods, necessary to reveal and explain how the ocean moves (Chapter 2). In the following 10 ‘topic chapters’ (Chapters 3–12), we introduce ocean currents across a wide range of geographical settings, in each case emphasizing an applied context and impact. In a closing Epilogue, we summarize and review the breadth of these impacts, and look forward to the exciting challenges that lie ahead for those who seek a deeper knowledge and understanding of ocean currents.

In this chapter, we start by posing 10 ‘big questions’, which provide a large part of the motivation for Chapters 3–12. We then highlight the tension between the organized and chaotic nature of ocean currents, emphasizing the intrinsic dispersion of material and properties. We next review the challenge of measuring ocean currents, with a focus on the challenge of organizing ocean observation at international scale. In a brief overview of observational approaches, we distinguish between direct measurement and estimation of ocean currents. Complementary to observations are increasingly realistic computer simulations, providing data that we use to trace flows throughout the ocean. We conclude with a reflection on the challenging range of time and space scales implicit in ocean currents, and on the co-development of technology and community activity that underpins our knowledge and understanding of the restless ocean.

1.1: Ten big questions

Associated with each applied context, and therefore featuring in each topic chapter, we pose a ‘big question’:

1.How does our plastic end up in some of the most remote places on Earth? (Chapter 3)

2.How do marine creatures experience and adapt to strong currents? (Chapter 4)

3.Why are the most productive fisheries found in the eastern subtropics? (Chapter 5)

4.How and where do destructive cyclones form over tropical oceans? (Chapter 6)

5.How do ocean currents connect the Arctic with the Atlantic and Pacific? (Chapter 7)

6.How does the ocean interact with the Antarctic ice sheet? (Chapter 8)

7.How does water and material move around shallow coastal and shelf seas? (Chapter 9)

8.How do the oceans connect to each other, and why does this matter? (Chapter 10)

9.How does water move at global scale? (Chapter 11)

10.How does the ocean circulation control climate? (Chapter 12)

These questions help to frame each topic chapter, in relation to particular types of ocean current and are illustrated in Fig. 1.1, introducing the visual imagery that we use throughout. The 10 panels combine a mixture of direct observations in (B, D), computer simulations (A, E–H), blends of observation and simulation (C, J), and a schematic (I). This mix of sources will be evident throughout subsequent chapters, and these are briefly discussed later in this chapter.

Fig. 1.1

Fig. 1.1 Visualizing the 10 big questions: (A) garbage patches ( van Sebille et al., 2015a); (B) passive drift of juvenile turtles ( Scott et al., 2014); (C) equatorward winds off California, driving upwelling and high productivity in June 1988; (D) the frequency of tropical cyclones; (E) climatological surface flows in and around the Arctic in September, from a high-resolution model hindcast; (F) simulated iceberg mass around Antarctica in an ocean model with interactive icebergs (Marsh et al., 2015); (G) tidal mixing fronts in the northwest European shelf in early July, at the boundary between stratified water offshore and mixed water inshore; (H) Agulhas leakage (in van Sebille et al., 2018); (I) schematic elements of global upper circulation, superimposed on high-resolution bathymetry; (J) January surface air temperature departure from zonal mean. Panel (D): Data from IBTrACS. Panel (J): Data are from the NCEP/NCAR Reanalysis 1, averaged over 1948–2018.

1.2: Organization vs. chaos

A traditional view is to identify, describe, and name individual currents around the World's oceans. In this approach, the ocean is an organized place, with predictable currents that may be drawn as smooth flows on nautical chart, or even on a globe. While we will respect this orthodox approach, we must stress the unpredictability of the flow. Averaging flows in time and space, currents do indeed emerge, but the instantaneous flow field is often much more complex. This inherent chaos is apparent in Fig. 1.1B andH, and more explicitly in Fig. 1.2 (van Sebille et al., 2015b), which illustrates how 10 pairs of drifting buoys move apart over a period of 7 months in the Southern Ocean, while all moving in the same eastward direction at roughly similar speeds. Each pair was deployed at the same location and time on a research expedition bound for the pack ice of Antarctica. Reporting positions hourly via satellite, drifter pairs stay close together at first, but their inevitable separation conforms to our understanding of chaotic ocean currents. This dispersion of everything carried by currents governs many important processes and can lead to some surprising phenomena, which we highlight in subsequent chapters.

Fig. 1.2

Fig. 1.2 Example of chaotic ocean drift in the Southern Ocean; pairs of buoys, deployed 15 m apart on the stern of a research ship, separated by up to 200 km in a month. From van Sebille, E., Waterman, S., Barthel, A., Lumpkin, R., Keating, S.R., Fogwill, C., Turney, C., 2015b. Pairwise surface drifter separation in the western Pacific sector of the Southern Ocean. J. Geophys. Res. Oceans 120, 6769–6781. https://doi.org/10.1002/2015JC010972.

1.3: Measuring the ocean—Challenges and international organization

The scientific method starts with measurement. There are many different ways to measure ocean currents, but oceanographers must contend with some major challenges in working at sea. The ocean is vast and diverse. Consequently, measurements in a just a few select locations may not be representative of the ocean as a whole. Ships are very expensive to operate, with costs for open-ocean vessels easily in the order of $40k per day. The ocean is hostile and unpredictable, with many working days—and sometimes scientific equipment—lost to bad weather. This equipment must be engineered very carefully to withstand the enormous pressures of the sub-surface ocean. The ocean is everywhere corrosive, with implications for all metal components in oceanographic equipment. The ocean is opaque to light and other types of electromagnetic radiation, so we have limited capability to measure beyond the immediate vicinity of most equipment. Most of the ocean belongs to nobody, with conflicting claims in the vicinity of some nations. Consequently, it is not always clear who can or should make which measurements, when and where.

In the face of so many measurement challenges, many smart methods have been developed and refined over the last century or so. Just as important, international cooperation has been developed and refined through a series of major programmes since the 1950s. The International Geophysical Year (IGY) spanning 1957–58 was an ambitious undertaking that included oceanography alongside 10 other earth sciences. In the realm of physical oceanography, IGY activities broadly spanned the Atlantic, Pacific, Indian, and Southern Oceans. It was already understood that the ocean moved in different ways at different depths. Oceanographers wondered whether and how the Atlantic might have changed since 1920’s and 1930’s surveys by the German Survey Ship Meteor, led by Georg Wüst, one of the great pioneers of oceanography. Mindful of their successors, IGY scientists paid particular attention to measurements that would be repeated in the following decades. Following soon after IGY during 1959–65, the International Indian Ocean Experiment (IIOE) addressed biological as well as physical oceanography. IIOE also provided an impetus to the development of marine science in the region, under the coordination of the new Intergovernmental Oceanographic Commission (IOC) of the United Nations Educational, Scientific and Cultural Organization (UNESCO).

Throughout the 1970s, the Geochemical Ocean Sections Study (GEOSECS) addressed the detection of dissolved chemicals throughout the ocean, from surface to abyss, in Atlantic, Indian, and Pacific oceans. Concentration patterns for different chemicals revealed how water moves across great distances on long (decadal to centennial) time scales, and the rate at which different bodies of seawater mix together. The 1990s saw a landmark achievement with the World Ocean Circulation Experiment (WOCE, 1990–97), which established a fully global survey of the oceans, revisiting many IGY locations to reveal a variety of changes since the late 1950s. By now, measurements from ships were complemented by satellite measurements of sea surface properties, and increasingly ambitious computer model simulations.

Overlapping with the latter years of WOCE, and integral to an ongoing synthesis of WOCE data, the Climate Variability and Predictability (CLIVAR) component of the World Climate Research Programme (WCRP) was launched in 1995, initially scheduled for 15 years, but currently ongoing. CLIVAR supports the international coordination of large-scale observing and modelling activities with a particular focus on the variability and predictability of climate, on seasonal to centennial time scales, which is determined to a large extent by changes in the ocean.

Despite the coordination of CLIVAR, it became apparent that a new framework specific to global ocean observation was needed in the years following WOCE, to optimize national commitments to oceanographic observation, standardize the suite of ocean measurements, and harmonize data sharing. The Global Ocean Ship-based Hydrographic Investigations Program (GO-SHIP) emerged in the 2000s. From 2007, a panel was formed to develop a new strategy for sustained global repeat hydrography, based around WOCE, integral to which are the standards and lines of communication necessary for effective data management and synthesis. With a ‘2012–23 Survey’ of GO-SHIP trans-ocean reference sections established (see www.go-ship.org), new measurements are available for comparison with those from WOCE, 20–30 years earlier.

Complementary to GO-SHIP, the decade-long GEOTRACES programme was launched in January 2010. GEOTRACES is providing the first coordinated global survey of dissolved chemicals. With more complete global sampling, we can better understand the sources, transports, and sinks of key chemical species that are vital for life. Also providing important tracers of ocean currents (and mixing) are the large-scale distributions of various trace elements and their isotopic variants that are detected at very low concentrations. The GEOTRACES sections (see geotraces.org) are rather more complex and convoluted than those of GO-SHIP, as the scientific objectives are specific to key locations, such as mid-ocean ridges and coastal zones.

1.4: Measuring ocean currents

We now turn specifically to the measurement of ocean currents. We distinguish the measurement of movement—of water and associated quantities/properties—through time and space, from the direct measurement of ocean currents in situ, instantaneously or through time. We start with the former case—measuring the drift due to ocean currents. We then review current measurements made at a fixed point, in profile, or with ground/ship-based radar.

1.4.1: Drift measurements

Perhaps the simplest observation of an ocean current is the resulting drift. Consider a corked bottle deployed from a boat. The bottle may contain a message that records the time and location of deployment. The bottle is buoyant and floats at the sea surface. In time, it drifts elsewhere. If the bottle is retrieved and we open it to discover where the boat deployed it, then we might simply draw an arrow from the point of deployment to the point of retrieval, we have a direction. If we also read the time of ‘departure’ and combine that with our current location and time, we can work out a straight-line distance travelled and simply divide this by time elapsed to get a speed. With speed and direction, we have a time-averaged current vector. Of course, objects in the ocean do not drift in arrow-straight lines, but inevitably meander along unique trajectories. An accurate record of the bottle trajectory would thus reveal a more detailed picture of ocean currents varying in both time and space.

Since the late 1970s, oceanographers have organized the measurement of near-surface flow with drifting buoys. Apart from technical challenges in batteries and satellite communication, these oceanographers faced challenges about how deep reaching to make these buoys. Nowadays, there are essentially two types of buoys: those that aim to measure the currents near the absolute surface of the ocean where ocean waves and winds can exert a substantial influence, and buoys that measure the currents averaged over the upper few tens of metres. Basic design and deployment of the latter type is illustrated in Fig. 1.3. These surface buoys are drogued to drift with the currents at a target depth around 15 m (Fig. 1.3A), to avoid drifting with currents very close to the surface. The drogue then acts as a sea anchor, and the buoy trajectory is representative of the currents that transport most heat and nutrients in the upper ocean. However, for buoyant material (seaweed, debris, plastics, etc.), it is precisely the surface drift that we need to measure and understand. In that case, drifters need to be as flat as possible, without sticking too high above the water because then they would sail, rather than drift. Technological development of very thin buoys is moving at a rapid pace, with the latest iteration of drifters (the Stokes drifters from Florida State University) being only a few centimetres thick.

Fig. 1.3

Fig. 1.3 (A) Drogued drifting buoy in vertical profile; (B) drifter deployment off the Bahamas. Photos: Chris Meinen.

Throughout this book, we will mostly be referring to the trajectories of drogued buoys, and then specifically the SVP (Surface Velocity Program) type. With the advent of accurate satellite tracking in the 1970s, drifter technology developed rapidly within the SVP (Niiler et al., 1995). Drift measurements since 1979 have been coordinated via the Global Drifter Program (GDP). However, while these buoys all start out with a drogue, they often lose their drogue during storms and heavy swell, when the connection between drogue and buoy snaps. So even the dataset of the GDP is a combination of drogued and un-drogued drifters.

A complex pattern of ocean currents emerges from analysis of the GDP dataset. As a regional example, Fig. 1.4 shows all GDP drifter tracks for the southwest Pacific. Some order is apparent in the chaos, such as off the coasts of eastern Australia and New Zealand, where currents are constrained by coastlines and steep ocean bottom slopes, but chaos is otherwise dominant.

Fig. 1.4

Fig. 1.4 All southwest Pacific trajectories of GDP buoys. From van Sebille, E., 2014. Adrift.org.au—a free, quick and easy tool to quantitatively study planktonic surface drift in the global ocean. J. Exp. Mar. Biol. Ecol. 461, 317–322.

Deeper ocean drift is a more challenging measurement, but of great importance for understanding ocean currents at global scale. The primary challenge is to target flows at great depths, for which a drogue is quite impractical. The key to ‘parking’ an ocean drifter at such depth is to neutralize its buoyancy, or density (for more details on density, see Chapter 2). If a drifter is more buoyant (or less dense) than surrounding water, it will rise; if it is less buoyant (or denser), it will sink; if it is exactly as buoyant as the surrounding water—i.e. the same density, or with neutral buoyancy—it will remain in situ (at constant depth). Seawater gets gradually denser with depth, for more than one reason. One reason is the sheer weight of water above any given depth, applying a pressure to the slightly compressible seawater; but water also gets denser as temperature falls or as salt concentration (salinity) increases. The combination of pressure, temperature, and salinity thus determines seawater density, and precise hydrographic measurements indicate how density consequently varies, up and down the water column, and from place to place.

Density variations are much greater in the vertical direction than horizontally, so if a ‘float’ can be made as buoyant as the water at a given depth, it will subsequently drift away at approximately that depth. The open ocean is typically 4000–6000 m deep, and target depths for following deep currents are typically in the range 1000–3000 m. Having achieved neutral buoyancy, the second challenge is to obtain information on position, ideally at regular intervals. This is a difficult problem, as communication from great depth demands innovative technology. One method relies on ocean acoustics, as sound waves can potentially travel great distances in the ocean. The travel time of sound signals may thus be used to fix the location of a sub-surface float.

The first neutrally buoyant float for tracking water movements at great depth was invented in the mid-1950s by British oceanographer John Swallow, and famously used to make the first direct observations of strong equatorward flows deep beneath the Gulf Stream (Swallow and Worthington, 1961). The body of the Swallow float comprised a pair of thinned aluminium tubes, each of length 3 m. Before deployment, Swallow floats were prepared on the deck, in a tank of water with known density, to ensure neutral buoyancy for that density, i.e. that the float neither rises nor sinks. In the preparations, further adjustments must be made for the compression (under target depth pressure) of both seawater and the float itself. After being lowered on a wire to the target depth, to ensure no leaks of the saline fluid required for neutral buoyancy, the float could be fully deployed. Hydrophones deployed over the side of the ship would then listen for signals from an acoustic transmitter housed on the float. By obtaining a series of fixes for position and depth, deep currents could be inferred from drift of the float. It was with measurements thus obtained over a period of around 2 weeks, March–April 1957, that the deep flows were first directly observed.

Subsequent to the pioneering work of Swallow and Worthington, the Sound Ranging And Fixing (SOFAR) channel, located at depths of around 1000 m throughout much of the ocean, was exploited using SOFAR floats that emitted sound pulses for remote reception. By the 1990s, SOFAR floats were superseded by Range and Fixing of Sound (RAFOS) floats (Rossby et al., 1986). RAFOS floats, moving at depth, listen for sound signals from multiple sub-surface beacons, much like modern-day GPS. They are programmed to return to the surface at mission end, whereupon they transmit these positional fixes to shore via satellite. It was in requiring less battery power on the floats themselves that RAFOS technology superseded that of SOFAR. In a pioneering mid-1990’s experiment, Bower and Hunt (2000) deployed 26 RAFOS in deep flows at two levels beneath the Gulf Stream, off the east seaboard of the United States. Drifting for up to 2 years, the floats revealed complex pathways through the region as dense water moves either along the continental slope or mixes into the basin interior.

1.4.2: Point measurements

Consider now the time-varying ocean current at a fixed location, such as a mooring, or a stable platform such as an Ocean Weather Ship. Such a scenario is typical of the fixed point, or Eulerian, measurements obtained with modern current meters. Fixed-point current meters work on various different operating principles. The most commonplace Rotor Current Meters (RCMs) operate on mechanical principles. In coastal or estuarine environments, we use RCMs that are constructed from titanium and polymers, to resist corrosion, limit dimensions, and minimize weight, such as illustrated in Fig. 1.5. Flow past the instrument drives the impeller (yellow component in Fig. 1.5) to rotate, generating a voltage in proportion to the rate of rotation. The current meter bodily aligns with the water flow, and an on-board compass records this orientation. Current speed and direction are thus measured at regular intervals and this data is logged during a period of deployment. Tethered to wire, these lightweight instruments may be lowered through the water column to obtain a current profile or maintained at a fixed depth to obtain a time series.

Fig. 1.5

Fig. 1.5 Lightweight Current Meter (Valeport Model 105), owned and used by the University of Southampton. Photos courtesy of author (Marsh).

In the deep ocean, current meters need to meet two criteria: resilience to extreme environmental conditions (corrosion and high pressure) and the capability to internally record a large amount of data with minimal battery power (for month- to year-long deployments). The development of such RCMs began in Norway in the late 1950s. By the early 1970s, RCM pioneer Ivar Aanderaa had achieved 5-year deployments (1968–72) in the Weddell Sea, off Antarctica. With remarkably long current records recovered, this showcased the longevity of his design, establishing the Aanderaa brand of current meters still used worldwide.

Deployed for months to years, as pioneered by Aanderaa, current meters have provided important information on the strength and often unidirectional character of key flows throughout the World Ocean. At locations known to play an important role in the large-scale ocean circulation, these flows can be monitored over the long term. One such location is in the Denmark Strait, between Iceland and Greenland, continuing along the continental rise to the east of Greenland. Flows here are strongest near the seabed and best monitored with moored current meters. Over a 1-year mooring deployment, Dickson and Brown (1994) thus recorded steady flows towards the southwest, with speeds in the range 20–40 cm s− 1, a vital branch of the Atlantic circulation that we return to in Chapters 7 and 11.

1.4.3: Profile measurements

Currents may also be measured in vertical profile using acoustic methods. These measurements are based on small variations in the reflection of sound waves from particulate matter drifting horizontally at different depths in the water column—a Doppler shift in the reflected sound wave, similar to the siren of a speeding emergency vehicle that is experienced by a stationary roadside observer. The Acoustic Doppler Current Profiler (ADCP), developed in the 1980s, both transmits and receives sound signals. Doppler shifts in the reflected signals are combined and converted into current speeds and directions, relative to the ADCP. The 1200 Hz ADCP shown in Fig. 1.6 is suitable for use in shallow water environments.

Fig. 1.6

Fig. 1.6 The Teledyne RDI Workhorse Sentinel ADCP, operating at 1200 Hz—another University of Southampton instrument that sees plenty of active service. Photos courtesy of author (Marsh).

In a variety of configurations, ADCPs can be mounted on a ship's hull, on an autonomous glider (a small submersible vehicle that can survey the ocean for days–months), moored (tethered to a cable or anchored to the seabed), or lowered from a ship with the CTD package (see Section 1.5.1). Movement of hull-mounted (shipborne) or glider-mounted ADCPs, relative to the measured current, must be taken into account, to determine an absolute current. ADCPs have the advantage over a single current meter of quickly measuring time series of current profiles, in the upper ocean (hull-mounted or glider-mounted), in the mid or deep ocean (mounted on the seabed or on a mooring), or throughout the water column (lowered from a ship).

An example of shipboard ADCP current measurements is shown in Fig. 1.7, illustrating the upper ocean response to a major storm in the South Atlantic which took place during 9–10 January 1993 (to which we return in Chapter 8). The individual ‘sticks’ represent the strength (length) and direction (angle) of ocean currents averaged hourly and in 8-m depth intervals. Over 4 days (10–13 January), the ADCP revealed rotating currents in the upper 50–100 m that gradually increased in strength and depth extent. The gaps in data on 14 January are a consequence of extreme waves, which degraded the quality of the reflected signals received by the ADCP.

Fig. 1.7

Fig. 1.7 ‘Stick plot’ time series of upper ocean currents from in situ (hull-mounted) ADCP, during RRS Discovery research cruise in the South Atlantic, January 1993 ( Marsh, 1995). The longest sticks correspond to current speeds of 50 cm s− 1.

Alongside RCMs, moored ADCPs are integral to modern programmes designed to monitor the large-scale ocean circulation. Perhaps the best example is an Atlantic-wide array of instruments spanning the Atlantic at 26°N—the RAPID array—stretching from the United States to Africa, which has been in place since spring of 2004. The RAPID moorings, including ADCPs, must be deployed and recovered every 1–2 years. Fig. 1.8 shows one such deployment.

Fig. 1.8

Fig. 1.8 A modern mooring operation: the ADCP in (A) is deployed along with buoyancy floats (B), as part of the RAPID array of moorings that collectively provide measurements necessary to monitor currents across the Atlantic at 26°N—from RRS Discovery cruise in November 2012; in this example, the ADCP is part of a ‘Western Boundary’ mooring located on the continental rise near the Bahamas. Photographs courtesy of Ben Moat (National Oceanography Centre, UK).

1.4.4: Radar measurements

In more recent decades, it has been possible to observe surface currents using active radar. Transmitted high-frequency (HF) radio waves (from a ground- or ship-based source) are reflected from the sea surface and received back at the transmitter, as sketched in Fig. 1.9. Based on the return pulse or waveform, horizontal patterns of ocean currents can be inferred. In some ways, these HF radar systems are similar to the Acoustic Doppler profilers, except that they use electromagnetic radiation (in the radar part of the spectrum) rather than sound. By working out the Doppler shift of the radar signal reflected on the sea surface, the speed at which the surface moves away or towards the antenna can be computed. And if more than one antenna is used, the direction of the movement can also be computed. However, a complicating factor in HF radar is that it detects movement associated with both the current and waves. As the transport by waves is very different from the transport by currents (see Section 3.6), these effects have to be separated.

Fig. 1.9

Fig. 1.9 The principles of HF radar, involving a transmitted pulse that is scattered from the ocean surface waves; Doppler shifts in the scattered pulse, received at the radar station, can be interpreted as a consequence of coastal currents moving water (and the waves) towards or away from the radar antenna.

In coastal configurations, HF radar can be used, for example, to observe background currents such as the Rhine outflow, characterized as a relatively fresh plume to the northeast, off the coast of Holland in the southern North Sea (Souza et al., 1997).

1.5: Estimating ocean currents

While direct measurements are ideal, oceanographers have long made use of proven theories from fundamental physics. We thus obtain proxy ocean currents using hydrographic data (temperature and salinity), sea surface height variations, or even induced voltages across a major current. In the following sections, we briefly explain the measurements and instruments used in the various approaches.

1.5.1: Hydrography

In simple terms, if cold water is adjacent to warm water, the density difference (from cold dense water to warm light water) sets up a force which drives a flow in proportion to the difference. Somewhat counter-intuitively, this flow is typically perpendicular (at right angles) to the orientation of density difference, for reasons that will be fully explained in Chapter 4. To obtain the measurements needed to determine these density differences, we lower a group of instruments—a conductivity cell, thermistor, and pressure sensor—through the water column, to measure conductivity (C), temperature (T) and pressure, or depth (D)—from which we derive the shorthand ‘CTD’. These co-located CTD measurements are recorded at regular high-frequency intervals (several times per second). From conductivity, we obtain the salinity values that we need to calculate density, accounting for pressure.

Alongside the CTD are several standardized Niskin bottles that sample water at selected depths, for further measurement of water properties. These bottles are arranged in a cylindrical rosette. The rosette of sampling bottles and CTD instruments is securely housed in a frame, an assemblage collectively known as a CTD package, such as the examples in Fig. 1.10. In deep ocean deployments, the rosette typically comprises 24 bottles (Fig. 1.10A), with a reduced number (typically 6 bottles) used in shallower environments such as estuaries (Fig. 1.10B). Adjacent CTD profiles provide the necessary density differences, in vertical profile, to estimate the variation of ocean currents with depth.

Fig. 1.10

Fig. 1.10 CTD packages at sea: (A) full suite of 24 bottles, here deployed from RRS Discovery during April 2004 in the subtropical North Atlantic; (B) small suite of 6 bottles, used for sampling water properties in Tamar estuary, Plymouth, UK during July 2018. Panel (A): Photo courtesy Prof Harry Bryden. Panel (B): Photo permission of author, Marsh.

CTD measurements of temperature and salinity are relatively sparse, obtained through research cruises to some remote ocean basins on only rare occasions. For comprehensive time and space sampling of temperature and salinity around the global ocean, we now rely on robotic floats. These are Argo floats, developed in the 1990s and commonplace since around 2000, as illustrated in Fig. 1.11. Within the Argo programme, the international community currently maintains a global array of currently almost 4000 floats (see www.argo.ucsd.edu). Each float is typically ‘parked’ at a target depth of 1000 m, descending to 2000 m and ascending to the surface every ~ 10 days to relay a vertical profile of temperature and salinity. Each float has an operational lifetime of almost 4 years. A vital component of the modern climate observing system, Argo data enables indirect estimates of ocean currents throughout the upper 2000 m of the World Ocean.

Fig. 1.11

Fig. 1.11 Deployment of an Argo float off the Bahamas in 2011. Photo courtesy of the author, van Sebille.

1.5.2: Bottom pressure recorders

Bottom pressure recorders (BPRs) record variations of the downward pressure, at the seabed, proportional to variations in sea level far above, accurate to around 1 mm. With two or more BPRs, or one BPR and tide gauge at a nearby coast or island, we can obtain a horizontal gradient, or slope, of sea level, and from this we can calculate the currents perpendicular to that slope (and in proportion to it). We return to the relation between flows and surface slopes in Chapters 4 and 8.

Initially deployed for around 1 month to measure sea level fluctuations associated with the tides, BPRs were developed for deep ocean use through the 1970s. BPR durability improved with the availability of low power electronic devices and better-performing pressure sensors. Deep ocean deployments in excess of 1 year became possible in the early 1980s, and 5-year deployments were achieved by the 1990s (Spencer and Vassie, 1997). These long-term deployments have enabled oceanographers to infer the variability of full-depth ocean currents on a wide range of time scales, from hours to years. A good example of BPR deployment for this purpose is in the Southern Ocean, where variations of the Antarctic Circumpolar Current (ACC) are clearly related to variations of sea level. For ACC monitoring, BPRs are most effectively deployed in Drake Passage, where the ACC is constricted to flow between South America and Antarctica.

1.5.3: Satellite altimeters

Another way to estimate ocean currents associated with sea level fluctuations exploits the efficient reflection of microwave radiation from the sea surface. Radar pulses continuously emitted from satellites orbiting the Earth, reflected back to space, can be detected by a receiver (or altimeter) aboard the same satellite. The time taken for emitted pulses to be received is a consequence of the distance travelled. Time differences correspond to longer travel paths, attributed in turn to small changes of sea level. The inferred sea surface height anomaly is currently accurate to within 2 cm. Fig. 1.12 illustrates this principle schematically. By resolving ‘bumps’ (and ‘hollows’) in the ocean surface, we can use established theory to estimate the currents associated with sea surface slopes.

Fig. 1.12

Fig. 1.12 The principles of satellite altimetry: microwave pulses emitted from an orbiting satellite reflect from a swathe of undulating ocean surface—here, the radar pulses are reflected from a ‘hump’ in the ocean surface, associated with a rotating eddy feature (in section); ocean currents, into and out of the page, are associated with sea surface slopes identified by successive altimetric measurements of sea surface height.

The development of satellite altimetry began in the 1980s. Following launch of the pioneering TOPEX/POSEIDEN satellite in 1992, a near-global field of sea surface height was achieved for the first time. Subsequent satellite missions (ERS-1, ERS-2, Jason-1, Jason-2, Envisat) have ensured continuous measurement since 1992. Extensive processing of the altimetry data is necessary, a service provided to the international community through the Aviso + system (www.aviso.altimetry.fr).

1.5.4: Electrical cables

The salinity of seawater is primarily attributed to sodium and chloride ions, which possess positive and negative charge, respectively. The Earth meanwhile generates a magnetic field, which can be regarded as constant on time scales of interest to oceanographers measuring ocean currents. As moving seawater carries charged particles (sodium and chloride ions) through this steady magnetic field, an electric field is generated. Placing a conductor in this environment, a voltage may be induced by the electric field, proportional to the rate of movement of seawater. This principle underpins a method for measurement of the collective transport by ocean currents through a gap traversed by an electrical cable. Fluctuations in the voltage can be calibrated to provide a continuous estimate of the combined transport associated with currents in cross-section. One location where this method has been put to impressive use is in the Florida Straits, where a disused submarine telephone cable runs for around 150 km, from just north of Miami to the island of Grand Bahama. Since 2004, the Florida Straits cable measurements have been an important component of RAPID observing system (see Section 1.4.3).

1.6: Computer simulation of ocean currents

Given the vastness of the ocean and the associated expense involved in ocean observations, computer simulations have proved to be an important asset in advancing our knowledge and understanding of ocean currents. Ocean sampling has grown through time, as illustrated in Figs 1.13 and 1.14 for salinity, but coverage is patchy, both horizontally and vertically. In particular, high latitudes are poorly sampled, across the vast Southern Ocean in particular (Fig. 1.13), and observations are biased towards the surface (Fig. 1.14). Despite arrival of the Argo programme around the year 2000, observations below 2000 m continue to be very limited in number and coverage. In addition to these sampling biases, many observations are also biased towards summertime and periods of fine weather. Seagoing oceanographers are simply unable to make observations in the most severe or difficult of conditions. Even in good weather, observations are expensive to acquire. Even if observations are plentiful, we need a way to predict future changes, and even to undertake counterfactual experiments.

Fig. 1.13

Fig. 1.13 Decadal sampling of ocean surface salinity: (A) 1950s, (B) 1970s, (C) 1990s, (D) 2000s. From Skliris, N., et al., 2014. Salinity changes in the world ocean since 1950 in relation to changing surface freshwater fluxes. Clim. Dyn. 43, 709–736.

Fig. 1.14

Fig. 1.14 Decadal sampling of ocean salinity in the upper 3000 m, 1950–2010: (A) in the southern hemisphere; (B) in the northern hemispheres. From Skliris, N., et al., 2014. Salinity changes in the world ocean since 1950 in relation to changing surface freshwater fluxes. Clim. Dyn. 43, 709–736.

For these combined reasons, we need computational models of the ocean, which have become increasingly sophisticated over the last 60 years, with the growth in computing power alongside the development of innovative computing technologies. Such models provide simulated flow fields, with information available at all locations within the model domain. Physical space is resolved horizontally at a scale that is typically constant in increments of latitude and longitude. These increments define the model resolution. Fig. 1.15 shows the surface ocean current speed in the North Atlantic, averaged over 5 days of an ocean model simulation, with horizontal resolution of 1/12°. This map reveals an overall pattern, with the strongest currents (highest speeds) emerging from the Gulf of Mexico to follow the US eastern seaboard before extending eastward across the Atlantic at around 40°N. This broad pattern can be traced basin-wide, a spatial scale of several thousand kilometres. Superimposed on this flow are more complex meandering and eddying flows associated with smaller spatial scales of around 100 km. The great advantage of simulated currents over measured currents is that we can obtain information everywhere in three dimensions (not just at the surface) and at regular time intervals.

Fig. 1.15

Fig. 1.15 Current speed (cm s − 1 ) in the North Atlantic, from recent global simulation at a horizontal resolution of 1/12° (the ORCA12 configuration of the NEMO ocean model—see Section 2.5.5 ).

With the growing power of computers, we are thus able to explore the evolution of ever more realistic virtual ocean currents in three dimensions, in relation to a wide range of phenomena and applications. One branch of analysis has involved the development of particle trajectories that record the model ocean currents in a Lagrangian and, ultimately statistical, sense. The example in Fig. 1.16 illustrates 1-year particle trajectories that were computed for a study of sea turtle hatchling drift. Basin-scale flow patterns are thus revealed, upon which is superimposed variability associated with the meanders and eddies. In Chapter 2, we explain how ocean models work, in relation to fundamental principles, and how we further use particle trajectories to analyse the drift associated with simulated ocean currents.

Fig. 1.16

Fig. 1.16 Example trajectories computed with model ocean currents; trajectories start from 42 coastal locations; 1000 trajectories are obtained per start location, starting at random times; each trajectory is 1 year in duration and is identified with a random colour (in Scott et al., 2014).

1.7: An ocean of scales

So, to recap, what do we see when we measure ocean currents? The ocean is forced to move at large scale, up to the width of an ocean basin, or thousands of kilometres. All the kinetic energy of motion is eventually lost, or dissipated, on very small scales, from centimetres to molecular dimensions. Time scales of variability correspondingly range from years to seconds. Different measurements more or less capture the range of space and time scales experienced by ocean currents. Drift measurements implicitly include all scales visited in space and experienced through time by a water parcel that has been accurately tracked through the ocean. Point measurements are inevitably restricted in space (to a chosen location or a limited set of locations), although accurate time averages may be obtained. In contrast to measurements, models can never capture all relevant scales in space and time. Each model has a lower limit to the space and time scale of simulated currents, and only global ocean models will capture the largest spatial