Ocean Sciences

OS Research

Eurec4a: Tales from the Tropics

Eurec4a: Tales from the Tropics

As many seagoing oceanographers find themselves on land for the foreseeable future, we’ve decided to share a tale of a research cruise to fill that ship-shaped void.

Back in January 2020, four research vessels ventured out into the Tropical North Atlantic as part of the Eurec4a and ATOMIC campaigns. Eurec4a’s aim: to investigate the couplings between clouds, circulation and convection and how these feed into climate change. Within this, vast webs of interlinking research studies were constructed. By the campaign’s end in February 2020, we had gathered data from kilometres up in the atmosphere right down to the ocean floor.

I found myself stepping aboard the German R/V Meteor in the middle of January, ready for my first experience of life at sea. Four of our six weeks aboard were spent studying an area east of Barbados between roughly 14°30’N and 12°N, dubbed ‘the tradewind alley’. Whilst much scientific focus was dedicated to the sky and cloud systems, a few groups of scientists onboard focused instead on the ocean. Collectively, we studied the biology and physics of the ocean, utilising research equipment ranging from the conventional CTD to the less conventional autonomous vehicle.

Elizabeth Siddle setting up a Seaglider on the R/V Meteor. (Credit: Callum Rollo)

The autonomous vehicles deployed from the R/V Meteor were two Seagliders and five Argo floats, complemented by an AutoNaut and Seaglider launched from Barbados by the University of East Anglia.  The AutoNaut was equipped to carry and release the Seaglider from beneath it, to preserve the Seaglider’s battery. The AutoNaut and three Seagliders teamed up over 11 days to patrol a 10×10 km box in line with the HALO aircraft’s flight circle and the R/V Meteor’s meridional transect, giving us a wealth of data in a small area. At the end of our time in tradewind alley, we recovered the 3 Seagliders to the R/V Meteor and the AutoNaut returned to Barbados under its own power.

Seaglider – A buoyancy-powered autonomous underwater vehicle which pumps oil in and out of an external bladder to vary its density, thus buoyancy. Can profile down to 1000 m depth.
Argo float – A profiling float that drifts at 1000 m depth. Every 10 days it completes a cycle of sinking to 2000 m then returning to the sea surface to relay data via satellite.
AutoNaut – A wave-propelled unmanned autonomous surface vessel designed to take meteorological and oceanographic measurements at the air-sea interface, piloted over the Iridium satellite network.

Whilst the Eurec4a data analysis is in its very early stages, we hope this will, in time, provide vital information to feed into oceanographic and climate system research. Personally, I hope to use this data to investigate the fluxes of heat and momentum at the air-sea interface. These fluxes will help us quantify the heat budget of the upper ocean mixed layer, which can feed into improving climate models.

Overall, the experiments conducted from the R/V Meteor and within the wider Eurec4a campaign were a great success. A huge thank you for everyone involved in making Eurec4a possible, with a special mention for the amazing crew of the R/V Meteor!

If you are interested in reading more about the Eurec4a project, see below:

  • Eurec4a project blog
  • Bony S, Stevens B, Ament F, Bigorre S, Chazette P, Crewell S, Delanoë J, Emanuel K, Farrell D, Flamant C, Gross S, Hirsch L, Karstensen J, Mayer B, Nuijens L, Ruppert JH, Sandu I, Siebesma P, Speich S, Szczap F, Totems J, Vogel R, Wendisch M, Wirth M (2017) EUREC4A: A Field Campaign to Elucidate the Couplings Between Clouds, Convection and Circulation. Surv Geophys. doi:10.1007/s10712-017-9428-0

How Climate Models helped uncover the mechanisms behind the North Atlantic Warming Hole

How Climate Models helped uncover the mechanisms behind the North Atlantic Warming Hole

One of the only regions that have been observed to cool over the past century is the North Atlantic cold blob just south of Greenland.

In our recent paper, we analyse the cold blob or “warming hole” and the processes that contribute to its creation and evolution. While sea surface temperature has been reliably observed, the underlying mechanisms of changing ocean circulation are only sparsely measured. Therefore, we used climate models in different configurations to analyse the role of the ocean circulation in driving the warming hole and identify additional contributing processes. Overall, climate models do a pretty good job reproducing the warming hole (Menary & Wood 2018), albeit with some variation among them.

Climate in general and North Atlantic climate in particular are affected by the phenomenon of natural variability. North Atlantic sea surface temperatures and ocean currents show considerable variations on a decadal timescale, which can mask or strengthen a (until now) fairly weak global warming signal
(Jackson et al. 2016). Therefore, an important question to consider is to what degree the warming hole and its drivers are just natural variability. It is hard to tackle this question using observations alone, since there is only one observed reality and disentangling natural trends and from those forced by greenhouse gas emissions is not always possible.

To overcome this, we used a so-called large ensemble of a climate model, in our case the MPI-ESM-1.1 Grand Ensemble (Maher et al. 2019), and investigated 100 simulations of the historical period. The simulations were run with the same boundary conditions but different initial conditions, resulting in 100 different realisations of the past climate. In these simulations we find that the warming hole varies in strength and position, but is consistently simulated in some form across all ensemble members (Fig. 2). Further, averaging the surface temperature trend over all ensemble members produces a clear cooling patch.

Simulated sea surface temperatures in the North Atlantic region for the first 24 ensemble members of the Grand Ensemble with historical forcing. Shown are trends in K per decade over the period 1870-2005. (Fig. 2; Credit: Paul Keil)

This means that the warming hole can indeed be attributed to global warming and is not just an expression of natural variability. Furthermore, examining a second set of 100 simulations with stronger CO2 forcing, we find that the warming hole is likely to strengthen and to emerge more distinctly from natural variability in the future.
We can use a similar approach to examine the driving processes of the warming hole. A change in the ocean circulation which is coupled to the heat transport into the region is likely responsible for a large part of the cooling (Rahmstorf et al. 2015, Menary and Wood 2018). It was thought that the incoming heat transport from the south associated with the AMOC (Atlantic meridional overturning circulation) is the crucial driver of the warming hole. However, in our large ensemble of the historical period, we find that the heat transport south of the warming hole, mainly associated with the AMOC, does show a decreasing trend in some simulations, there is no coherent trend over all 100 simulations (Fig. 3b).

While the AMOC is projected to slow down, and does so in our second set of simulations with stronger CO2 forcing (Fig. 3a), here it does not emerge from natural variability. Instead, the heat transport north of the warming hole shows a significant positive trend over all simulations, which means an increase of exported heat to the Arctic, rather than a decrease of heat import from the south. We are further able to demonstrate that this increase in heat transport into higher latitudes is mainly associated with a heat transport from the horizontal gyre circulation. Thus, for the formation of the warming hole, the role of the subpolar gyre heat transport is crucial (illustrated as blue line in Figure 1). For our observed reality, of which there only is one, this means that the high latitude heat transport changes are likely causing the observed warming hole, especially in its early stages. Nevertheless, due to its large natural variability, the AMOC might also be contributing to it. As global warming progresses, the impact of the AMOC on the warming hole will increase.

North Atlantic Ocean heat transport (OHT) changes in the Grand ensemble.
a, Evolution of the Atlantic heat transport anomalies relative to the preindustrial control simulation at 26.25° N (black solid line) and 63.75° N (black dashed line) in the 1pctCO2 experiment. Shading represents the 5th and 95th percentiles of the ensemble.
b, Ensemble mean linear trends of the Atlantic OHT in the historical ensemble from 1850 to 2005. Shading represents the 5th and 95th percentiles of the total heat transport trend (black line) of the ensemble. Linear trends of the AMOC and gyre components are shown in red and blue, respectively. (Fig. 3; Credit: Keil et al. 2020)

Climate models are far from perfect and continue to misrepresent some aspects of climate, and show considerable differences to the observations and amongst themselves. In our case for example, we have excluded a change in local vertical mixing as a driver for the warming hole, but this process plays an important role for the warming hole in some climate models (Menary and Wood 2018). Nevertheless, climate models can be a unique tool that helps us to understand the climate. We can even switch off the clouds and thereby erase any possible effect they might have on the warming hole (for more details see Keil et al. 2020).

The climate model helped us understand the mechanisms of changing ocean circulation, which are still measured sparsely over longer timescales. However, even if we had perfect measurements of the warming hole and its associated ocean circulation changes, we would still have trouble separating the climate change signal form the ever-present natural variability. With the 100 alternatives, but equally probable realities produced by the Grand Ensemble we were able to overcome this problem. This is just one aspect of how climate models let us construct artificial worlds, with which we explore the underlying mechanisms behind climate change.


References & Further reading:

Jackson LC, Peterson KA, Roberts CD, Wood RA (2016) Recent slowing of Atlantic overturning circulation as a recovery from earlier strengthening. Nat Geosci. doi:10.1038/ngeo2715

Keil P, Mauritsen T, Jungclaus J, Hedemann C, Olonscheck D, Ghosh R (2020) Multiple drivers of the North Atlantic warming hole. Nat Clim Chang. doi:10.1038/s41558-020-0819-8

Maher N, Milinski S, Suarez-Gutierrez L, Botzet M, Dobrynin M, Kornblueh L, Kröger J, Takano Y, Ghosh R, Hedemann C, Li C, Li H, Manzini E, Notz D, Putrasahan D, Boysen L, Claussen M, Ilyina T, Olonscheck D, Raddatz T, Stevens B, Marotzke J (2019) The Max Planck Institute Grand Ensemble: Enabling the Exploration of Climate System Variability. J Adv Model Earth Syst. doi:10.1029/2019MS001639

Menary MB, Wood RA (2018) An anatomy of the projected North Atlantic warming hole in CMIP5 models. Clim Dyn. doi:10.1007/s00382-017-3793-8

Rahmstorf S, Box JE, Feulner G, Mann ME, Robinson A, Rutherford S, Schaffernicht EJ (2015) Exceptional twentieth-century slowdown in Atlantic Ocean overturning circulation. Nat Clim Chang. doi:10.1038/nclimate2554

Carbon Brief article by Robert McSweeney (2020) Scientists shed light on human causes of North Atlantic’s ‘cold blob’.


Edited by Meriel Bittner

Why do mesoscale eddies disappear at ocean western boundaries?

Why do mesoscale eddies disappear at ocean western boundaries?

In the ocean, mesoscale eddies are everywhere. In fact, about 90% of the ocean’s kinetic energy is contained within these ubiquitous, ~100 km swirling vortices. We know how these eddies form, through a process known as baroclinic instability, where density surfaces are tilted too steeply and become unstable. What remains unclear however, is what happens to this energy when eddies die. This knowledge gap is a key uncertainty in ocean/climate models that parameterise ocean eddies, like the state-of-the-art CMIP6 models used in the latest IPCC report.

What we do know however, is where these eddies typically go to die: ocean western boundaries. Using satellite-based altimetry we can observe westward propagating mesoscale eddies as they reach ocean western boundaries, like the east coast of the United States, and disappear. What happens to this eddy energy at western boundaries then becomes a question of the direction of the energy cascade. An eddy may become entrained within the larger scale flow features of the region in what is known as an inverse or reverse energy cascade (from mesoscale to large scale). Alternatively, an eddy may interact with the topography along the ocean boundary, forming smaller scale features that eventually dissipate as turbulence. This is what we call a direct or forward energy cascade.

The MeRMEED study region offshore of Great Abaco. The grey lines show location of our survey and the white markers show the location of existing moorings. (Fig. 2; credit: Gwyn Evans)

During my post-doc at the University of Southampton, I was part of the MeRMEED (MEchanisms Responsible for Mesoscale Eddy Energy Dissipation) project led by Eleanor Frajka-Williams. We set out to observe and quantify the strength of the direct energy cascade in mesoscale eddies at an ocean western boundary by measuring the turbulence generated as eddy flow interacts with topography. Our chosen study region was just offshore of Great Abaco, Bahamas, where we already have observations of possible eddy-topography interactions from mooring-based measurements and satellite altimetry (Fig. 2). Over the course of two years, the plan was to use a relatively small vessel to perform a short but intensive survey of the flow and turbulence along the steep and rough slope offshore of Great Abaco during several different eddies.

To understand how eddy flow-topography interactions can lead to an increase in turbulence and an eventual dissipation of eddy energy, we required detailed measurements of two key variables: the sub-surface velocity and the turbulent dissipation. For sub-surface velocity measurements, we used an Acoustic Doppler Current Profiler or an ADCP. Our chosen vessel, the R/V F.G. Walton Smith of the University of Miami (Fig. 1), came equipped with an ADCP mounted to the hull of the vessel, allowing us to perform a comprehensive survey of the eddy flow offshore of Great Abaco. To complement these measurements, we used a Vertical Microstructure Profiler or VMP to estimate turbulent dissipation (Fig. 3). The VMP is specialist piece of equipment that measures centimetre to millimetre scale changes in vertical gradients of velocity and temperature. By examining the spectra of these small-scale changes, we can then make estimates of the turbulent dissipation to tell us how much of the eddy flow’s kinetic energy is dissipated along the western boundary offshore of Great Abaco.

The VMP in action during a recover on the aft deck of the R/V F.G. Walton Smith handled expertly by colleagues from the National Marine Facilities group of the National Oceanography Centre. (Fig. 3; credit: Eleanor Frajka-Williams)

Our measurements across three different mesoscale eddies revealed a whole host of eddy flow-topography interactions that led to increased turbulence and dissipation of the eddy flow’s kinetic energy. For example, we observed several regions of strong vertical and horizontal gradients in velocity linked to interactions between the eddy flow and topography where we also observed increased turbulence. Turbulence was also stronger where the eddy flow is forced over bumps in the topography including the generation of internal gravity waves, which are linked to the presence of a hydraulic jump. Our work has therefore shown us that the interaction between eddies and the steep and rough slope at an ocean western boundary can lead to turbulent energy dissipation within the eddy flow. However, does the energy dissipation observed during our surveys explain the decay of eddy energy observed at western boundaries via satellite altimetry?

To answer this question, we again used satellite altimetry to track every eddy that entered our study region and “died”. We estimated the energy of these eddies and monitored the decay of their energy over time within our study region. Comparing this satellite-based estimate of eddy decay rate to the turbulent dissipation measured during our surveys, we find that these separate estimates agree very closely. This suggests that offshore of Great Abaco, eddy energy is lost through a forward cascade of energy, as the mesoscale eddy flow interacts with the steep and rough slope, forming smaller scale features that eventually lead to the generation of turbulence and an increase in the dissipation rate.

The MeRMEED project has shown us that the death of mesoscale eddies at ocean western boundaries can in some cases be explained through a forward cascade of energy, where eddy-topography interactions lead to smaller scale flow features that eventually dissipate as turbulence. The question remains, however: are our results relevant for other ocean western boundaries beyond this particular region offshore of Great Abaco? And further, how should these findings inform eddy parameterisations in the next generation of ocean/climate models? These remain open questions that will no doubt guide the future research within this field.