Abstract preview

The abstracts below are arranged according to themes, and further subdivided into invited talks, oral presentations and posters.
Within each of these subsections the abstracts are arranged alphabetically by first author.

Theme 1: Characterising the AMOC - structure, variability, mechanisms and ocean response (66 abstracts)
Theme 2: Impacts of the AMOC on the atmosphere, cryosphere and land (15 abstracts)
Theme 3: AMOC state estimation, predictability and prediction (17 abstracts)
Theme 4: Novel approaches to pan-Atlantic observations, modelling, analysis and synthesis (16 abstracts)


Theme 1: Characterising the AMOC - structure, variability, mechanisms and ocean response

All sessions take place on Tuesday 21st and Wednesday 22nd July.   Invited talks (3)   Oral presentations (19)   Posters (44)  

Theme 1 invited talks:
Perspectives on ocean heat transport in the North Atlantic from the first decade of the RAPID-MOCHA array
William Johns1*, Jian Zhao1, Gerard McCarthy2, David Smeed2, Christopher Meinen3, Molly Baringer3, Eleanor Frajka-Williams2, Elaine McDonagh2, Brian King2, Darren Rayner2

* Presenting author

1) Rosenstiel School of Marine and Atmospheric Science, University of Miami, USA
2) National Oceanography Centre, UK
3) NOAA Atlantic Oceanographic and Meteorological Laboratory, USA

Since April 2004, continuous estimates of the oceanic meridional heat transport in the Atlantic have been derived from the RAPID-MOCHA-WBTS observing system along 26.5°N. Several improvements in the methodology for both the AMOC and heat transport calculation have been implemented recently, which have been applied retrospectively to the entire time series. These include improvements in the surface extrapolation of interior geostrophic velocities, adoption of the TEOS-10 equation of state, updated Gulf Stream temperature transport calibration, and weekly optimal interpolation of Argo and RAPID mooring data to estimate the interior temperature transport.

The mean values for the AMOC strength and northward heat transport from the 10-year time series (2004-2014) are 17.0 Sv and 1.24 PW, respectively. Both the AMOC strength and the heat transport have decreased in recent years compared to values observed prior to 2009; the 5-year means for the pentad 2009-2013 were 15.6 Sv (1.14 PW) compared to values of 18.7 Sv (1.34 PW) for the pentad 2004-2008. The decline in the heat transport of 0.2 PW between these periods is significant: it is equivalent to a net decrease in surface heat flux of ~7 W m-2 over the entire North Atlantic, and represents a net deficit in heat delivery to the North Atlantic of 1.0 PW during the last 5 years, nearly equivalent to one year's worth of the typical heat transport. Observations of ocean heat content (OHC) from Argo data show that the North Atlantic OHC reached a decadal peak in about 2007 and has since declined, consistent with the lower recent heat transport values recorded by the 26.5°N array.

More than 90% of the interannual variability that has occurred in the meridional heat transport is contained in the overturning component of the heat transport, while the gyre component has maintained a stable mean value. Both Ekman and Gulf Stream variability contribute to large short-term changes in the AMOC and heat transport, including occasional heat transport reversals, while the interannual variability of the heat transport is dominated by the geostrophic circulation and mostly by the mid-ocean heat transport. Analysis of GCMs and simpler forced dynamical models suggests that most of the interannual AMOC and heat transport variability can be explained by wind-forced changes in mid-ocean circulation associated with first baroclinic mode Rossby waves, that are excited by interannual wind stress curl anomalies in the central and western part of the basin.


Insights and impacts: the first 10 years of continuous observations of the Atlantic overturning circulation
Gerard D. McCarthy1*, William E. Johns2, Chris S. Meinen 3, Molly O. Baringer3, Darren Rayner1, Bengamin I. Moat1, Eleanor Frajka-Williams4, David A. Smeed1

* Presenting author

1) National Oceanography Centre, Southampton, UK
2) RSMAS, University of Miami, USA
4) University of Southampton, National Oceanography Centre, UK

The RAPID/MOCHA/WBTS is a joint UK-US project that has been measuring the Atlantic Overturning circulation (AMOC) at 26.5 N in the North Atlantic since 2004. Here we present some of the key results from the first 10 years of the program.

The first year’s measurements revealed a highly variable AMOC that encompassed all previous ship-based, hydrographic estimates of the AMOC, thus showing that a perceived decline could be encompassed in short-term variability. Seasonal variability in the AMOC was larger than expected with a 6 Sv range, with the largest single component derived from density fluctuations at the eastern boundary.

Interannual variability, far larger than that in the present state of the art climate models, was seen in 2009/10. A 30% reduction lasted 18 months, cooling the subtropical North Atlantic significantly and elevating sea levels in New York by 13 cm. The existence of continuous heat transport measurements enabled us to show that the main cause of the cooling was a reduction in ocean heat convergence rather than air-sea fluxes.

The winter of 2010/11 revealed a second consecutive winter of low AMOC: a double dip. Whether ocean re-emergence or the change in AMOC circulation was the cause of the SST tripole pattern pattern that emerged in the winter of 2010/11 is a topic of ongoing research. Nonetheless, this SST pattern was shown to be sufficient to push the atmosphere into a second consecutive negative wintertime North Atlantic Oscillation (NAO) and increased predictability of this negative NAO.

Most recently a multi-year decline in the AMOC has been observed. This 0.5 Sv/year decline is much larger than the long-term decline predicted due to anthropogenic climate change. The decline first reported on the 8.5-year timeseries has continued in the 10-year timeseries. The magnitude of the decline is so large as to suggest it may be decadal variability.

The Atlantic is a region with large multi-decadal climate signals. A decline in the AMOC is consistent with a declining phase of the Atlantic Multi-decadal oscillation of sea-surface temperatures that is predicted by a number of authors. Circulation proxies based on tide gauges have supported the hypothesis that ocean heat transport dominates the changes in ocean heat content on decadal timescales and eventually sea surface temperature variations. With continuous measurements of heat transport from the RAPID array we will be able to quantify this dynamic mechanism. On even longer timescales, continuous AMOC measurements will show whether the predicted decline in the AMOC due to anthropogenic climate change is occurring.


Reconstructing recent Atlantic overturning from surface wind and buoyancy forcing
Helen R. Pillar1,2, Patrick Heimbach3,4,5, Helen L. Johnson1*, David P. Marshall6

* Presenting author

1) University of Oxford, Earth Sciences, UK
2) University of Copenhagen, Niels Bohr Institute, Denmark
3) Massachusetts Institute of Technology, Earth, Atmospheric and Planetary Sciences
4) University of Texas at Austin, Institute for Computational Engineering and Sciences, USA
5) University of Texas at Austin, Jackson School of Geosciences, USA
6) University of Oxford, Atmospheric, Oceanic and Planetary Physics

The Atlantic Meridional Overturning Circulation (AMOC) carries a substantial amount of heat poleward in the North Atlantic and is projected to weaken over the next century in response to greenhouse gas emissions, with implications for the North Atlantic storm track, hurricane frequency, European climate, regional sea level, and global terrestrial and marine ecosystems. The AMOC is believed to be a key driver of multidecadal variations in North Atlantic sea surface temperatures and a potential source of regional climate predicability. The strength of the AMOC at 26°N has been continuously monitored since 2004 and exhibits large variability on all time scales. Here we investigate how much of the observed AMOC variability can be reconstructed by projecting observed atmospheric variability onto model-based estimates of AMOC sensitivity to surface wind, thermal and freshwater forcing over the preceding 15 years. We find that local, instantaneous wind forcing dominates the AMOC variability on short time scales, whereas subpolar heat fluxes dominate on interannual to decadal time scales. The reconstructed AMOC is able to reproduce most of the interannual variability observed by the RAPID-MOCHA array at 26°N, but not the apparent decadal trend, requiring the integrated response to subpolar heat fluxes over at least the past two decades.


Theme 1 oral presentations:
Temporal variability of North Atlantic anthropogenic carbon fluxes and their sensitivity to the strength of the meridional overturning circulation
Peter Brown1*, Elaine McDonagh1, Richard Sanders1, Brian King1, David Smeed1, Andrew Watson2, Ute Schuster2, Molly Baringer3, Chris Meinen3, Rik Wanninkhof4, Bill Johns5, Stuart Cunningham6

* Presenting author

1) National Oceanography Centre, Southampton, Marine Physics and Ocean Climate
2) University of Exeter, College of Life and Environmental Sciences, UK
3) NOAA-AOML Miami, Physical Oceanography Division, USA
4) NOAA-AOML Miami, Ocean Chemistry Division, USA
5) Rosenstiel School of Marine and Atmospheric Science, University of Miami, Division of Meteorology and Physical Oceanography
6) SAMS, Scottish Marine Institute,

The North Atlantic plays a critical role in the global carbon cycle both as a region of substantial air-sea carbon dioxide uptake and as a location for the transfer of CO2 away from atmospheric interaction on climatically-important timescales, as part of the global overturning circulation. While the spatial and temporal variation in surface fluxes is relatively well constrained, our understanding of the changing deep carbon distribution is restricted to sub-decadal repeat hydrographic sections. For anthropogenic carbon (Canth), the scale and variability of its meridional transport and interior redistribution is further limited to integrated multi-decadal basin-scale estimates. Our ability to predict the response of the ocean’s interior and surface carbon fluxes to a changing global climate (and by extension, the ocean’s continued ability to mitigate the rise in atmospheric carbon levels) is thus severely impacted.

Here, we present the first observation-derived high-resolution estimate of short-term meridional carbon transport variability and long-term trends across the subtropical North Atlantic. Historical hydrographic data-based estimates of Canth are used to generate predictive regressions that when applied to and combined with RAPID mooring and ARGO float-derived transport estimates, create a 10 day frequency interior ocean carbon flux time-series between 2004 and 2012.

The net anthropogenic carbon transport across this timeframe is found to be relatively independent of calculation method and robust at 0.22 PgC yr-1 northwards, with the poleward advection of high Canth loadings in shallow waters outweighing the predominantly southwards transports of low concentrations in the deep interior flow. Within these records, substantial seasonal, sub-annual and interannual variability in the transports is observed that is highly sensitive to the strength of the overturning circulation. While the multi-year decrease in meridional overturning circulation strength recently identified is expected to similarly impact Canth transports, this effect is currently seen to be masked by the northwards transport of increasing surface Canth levels. However, a comparison with historical estimates of the North Atlantic carbon sink reveals an intrinsic relationship between air-sea carbon uptake, ocean carbon transport and heat fluxes, which will become more important as the ocean responds to a changing global climate.


The importance of the Transition Zone to decadal AMOC variability
Martha Buckley1*, John Marshall2

* Presenting author

1) George Mason University, Department of Atmospheric, Oceanic
2) Massachusetts Institute of Technology, Department of Earth, Atmospheric

On short timescales variability of the AMOC primarily reflects that of the wind field, and meridional coherence is limited. However, on decadal timescales the AMOC exhibits meridionally coherent modes of variability, which are driven by a complex mixture of wind and thermohaline processes. Models differ substantially in the spatial patterns and dominant timescales of low-frequency AMOC variability, and isolating robust mechanisms of AMOC variability remains challenging. Despite this, we argue that the region near the Grand Banks where the Gulf Stream/North Atlantic Current (NAC) and the deep western boundary current cross over, henceforth called the Transition Zone, is a key region influencing large-scale decadal AMOC variability. It is here that we observe the Mann eddy, an intense anticyclone swirling to the southeast of the NAC. Variability in this key region is implicitly reflected in the AMOC indices commonly utilized by the modeling community. Processes that are important in creating buoyancy anomalies in the Transition Zone are expected to play an important role in AMOC variability. Such processes include local atmospheric forcing, advection of anomalies by mean currents, westward propagating (wind or buoyancy forced) baroclinic Rossby waves, anomalies resulting from large-scale ocean circulation changes (such as shifts of the Gulf Stream path), and anomalies advected/propagated from high latitudes. The complex ocean dynamics in transition zone likely explain why AMOC variability is so sensitive to model formulation, both between models and in the same model when changes are made to its resolution, overflow parameterizations, etc.


Dynamic Response of the North Atlantic Circulation to Rapid Ocean Heat Content Changes between 1990 and 2014
Stuart Andrew Cunningham1*, Clare Johnson1

* Presenting author

1) The Scottish Association for Marine Science, Physics and Technology, Scotland

Upper ocean heat content in the subpolar region of the North Atlantic varies on interannual to decadal timescales and with spatial variations between its sub-basins as large as the temporal variability. There is strong evidence for the role of these variations in forcing Atlantic hurricane frequency, location of the Inter-tropical convergence zone and other teleconnections. Ocean heat content can change through variations in the Atlantic overturning, and through local buoyancy and wind forcing. Here we show how the decadal and spatial variability in subpolar gyre heat content dynamically drives changes in the strength and structure (current speeds and transports) of the subpolar gyre. We quantify the changes to the horizontal and overturning circulations and the impact on subpolar gyre ocean heat flux, with a focus on the Labrador Sea and from Greenland to Scotland. Ocean heat content increases in the subpolar Atlantic from 1990 to 2014, decreasing the strength of the boundary currents and limiting the extent of the subpolar gyre in the Labrador and Irminger Basins. In the eastern subpolar gyre there is much less net change in heat content from 1990 to 2014, but the path of the North Atlantic Current through the Iceland Basin and Rockall Trough is altered and the European Slope Current strength is diminished. The zonal variation in heat content drives anomalies of overturning strength of about 1.5 Sv into the Labrador sea with a maximum at 600 m depth; between Greenland and Scotland the overturning anomaly is southward with a strength of 1.5 Sv at a depth of 1000 m. While there is little change to the net heat flux into the Labrador Sea the meridional heat flux between Scotland and Greenland is diminished by around 0.025 PW – approximately 10% reduction of the pre-1990’s canonical value.


North Atlantic Simulations in Coordinated Ocean-ice Reference Experiments phase II (CORE-II): Inter-Annual to Decadal Variability and Trends
Gokhan Danabasoglu1*, Stephen G. Yeager1, Who M. Kim2

* Presenting author

1) National Center for Atmospheric Research, USA
2) Texas A&M University, USA

Simulated inter-annual to decadal variability and trends in the North Atlantic for the 1958-2007 period from twenty global ocean – sea-ice coupled models are presented. These simulations are performed as contributions to the second phase of the Coordinated Ocean-ice Reference Experiments (CORE-II). The present study represents a continuation of our previous work which documented the mean states in the North Atlantic from the same models. A major focus here is the representation of Atlantic meridional overturning circulation (AMOC) variability and trends in the participating models. Relationships between AMOC variability and those of some other related variables, such as subpolar mixed layer depths, the North Atlantic Oscillation (NAO), and the Labrador Sea upper-ocean hydrographic properties, are also investigated. In general, AMOC variability shows three distinct stages. During the first stage that lasts until mid- to late-1970s, AMOC remains lower than its long-term (1958-2007) mean. Thereafter, AMOC intensifies with maximum transports achieved in mid- to late-1990s. This enhancement is then followed by a weakening trend until the end of our integration period. This sequence of low frequency AMOC variability is consistent with previous studies. Regarding strengthening of AMOC between about mid-1970s and mid-1990s, our results support a previously identified variability mechanism where AMOC intensification is connected to increased deep water formation in the subpolar North Atlantic, driven by NAO-related surface fluxes. The simulations tend to show general agreement in their representations of, for example, AMOC, sea surface temperature (SST), and subpolar mixed layer depth variabilities. In particular, the observed variability of the North Atlantic SSTs is captured well by all models. These findings indicate that simulated variability and trends are primarily dictated by the atmospheric data sets which include the effects of ocean dynamics. Despite these general agreements, there are many differences among the model solutions, particularly in the spatial structures of variability patterns. For example, the location of the maximum AMOC variability differs among the models between Northern and Southern Hemispheres.


From Days to Decades: Variability of the Subpolar DWBC Transports
Jürgen Fischer1*, Johannes Karstensen1, Martin Visbeck1, Rainer Zantopp1, Patricia Handmann1

* Presenting author

1) GEOMAR Helmholtz Centre for Ocean Research Kiel, Physical Oceanography, Germany

The Deep Western Boundary Current (DWBC) is a key element of the Meridional Overturning Circulation (MOC) in the subpolar North Atlantic (SPNA), and the Labrador Sea is the location where the North Atlantic Deep Water (NADW) constituents merge. Diverse pathways, underway modifications through thermohaline processes (e.g., entrainment and convection) and other forcing have modified the DWBC layers enroute. At the exit of the Labrador Sea a moored observatory (the “53°N-Array”) has been installed since 1997. With 17 years of data, this is one of the longest full ocean depth records of the boundary circulation worldwide. Transports derived for water mass layers show variability from days to decades with two frequency bands dominating the deep variability. The first variance maximum at 10-20d periods is due to topographic Rossby Waves (Fischer et al., 2015 – an international cooperation) tied to the steep Labrador Shelf break, but also found all along the western margin of the SPNA.
A second energy maximum is found at quasi-decadal time scales, especially in the deep overflow components. This is associated with a deep baroclinic current core hugging the continental slope at depth below 2000m. The overarching question is whether this Labrador Sea export can be interpreted as a fingerprint of the deep AMOC component on such timescales. This is investigated in conjunction with other long term subpolar observations and high resolution (VIKING20) model output. Presently the 53°N-Array is continued as part of the international OSNAP program.


Lagrangian pathways of temperature anomalies from the subtropical to the subpolar gyres in the North Atlantic
Nicholas Foukal1*, Susan Lozier1

* Presenting author

1) Duke University, Nicholas School of the Environment, USA

We explore both pathways and variability of the upper-layer throughput from the subtropical to the subpolar gyres of the North Atlantic in order to understand how, where, to what extent, and on what time scales thermal anomalies travel from one gyre to the other. Past studies with both modeled and observed Lagrangian floats have indicated that surface throughput between the two gyres is limited and that sub-surface pathways dominate. We translate these results to the oceanic heat transport by first using the modern satellite SST record (1981-2013) to reconcile the lack of inter-gyre connectivity at the surface with previous findings that showed an advection of SST anomalies from the subtropics to the subpolar gyre. We then launch synthetic trajectories in the FLAME model from the Florida Straits to trace the pathways of water particles along the Gulf Stream/North Atlantic Current. Along these time-varying pathways, we compare the Lagrangian integral time scales of the floats’ temperature anomalies to their travel time from the Florida Straits to the subpolar gyre in order to understand whether particles advect any memory of their anomalous temperature into the subpolar region. We then use backwards trajectories launched in the eastern subpolar gyre in FLAME to determine whether the strength of the inter-gyre throughput affects the temperature of the eastern subpolar gyre and its connection to the gyre index. Our results cast doubt on the idea that coherent water masses advect temperature anomalies from the subtropical to the subpolar gyres as well as demonstrate the importance of mixing to oceanic heat transport.


The North Atlantic contribution to global ocean warming
Sirpa Hakkinen1, Peter B Rhines 2*, Denise L Worthen 3

* Presenting author

1) NASA Goddard Space Flight Center, Code 615, USA
2) University of Washington, Dept of Oceanography, USA
3) NASA Goddard SFC/Wyle STE, Code 615, USA

Warming of the North Atlantic Ocean from the 1950s to 2012 is analyzed on neutral density surfaces and vertical levels in the upper 2000 meters. Three reanalyses and two observational datasets are compared. The net gain of 5 x 10^22 J in the upper 2000m is almost 30% of the global ocean warming over this period. Upper ocean heat content (OHC) is dominated in most regions by heat transport convergence without widespread changes in the potential temperature/salinity relation. The heat convergence is associated with sinking of mid-thermocline isopycnals, with maximum sinking occurring at potential densities σ0 = 26.4-27.3, which contain subtropical mode waters. Water masses lighter than σ0 =27.3 accumulate heat by increasing their volume, while heavier waters lose heat by decreasing their volume.
Spatially the heat content trend is non-uniform: there are large contributions in the western boundary current, northeastward to the North Atlantic Current, the Subtropical Mode Water/Gulf Stream recirculation gyre, and more broadly across tropical latitudes. Analysis of zonal behavior shows that the low latitudes, 0-30N are warming steadily while large multi-decadal variability occurs at latitudes 30-65N. Covariation of North Atlantic warming with the rest of the global ocean will be discussed. During the satellite altimetry era, sea level variability shows regions of high correlation with 0-700m heat content. The extraordinary, recent cold winters in eastern North America have been argued to arise from Pacific SST anomalies, and these involve significant heat storage.


Increasing transport of volume, heat and salt in two high-latitude branches of the AMOC during the last two decades
Bogi Hansen1*, Karin M. H. Larsen1, Hjálmar Hátún1, Svein Østerhus2

* Presenting author

1) Faroe Marine Research Institute, Faroe Islands
2) Uni Research Climate, Bergen,

The northernmost extension of the AMOC is formed by the exchanges across the Greenland-Scotland Ridge. Through the deep channels of the ridge, the overflow carries cold and dense water from the Nordic Seas into the Atlantic, which together with entrained water contributes the main part of the lower limb of the AMOC. In the top layers, the loop is closed by the inflow of Atlantic water to the Nordic Seas. Both the overflow and the Atlantic inflow are split into branches and two of these branches pass close to the Faroes. One is the Faroe Bank Channel (FBC) overflow, which transports about one third of the total overflow, and which before entrainment is the densest overflow branch. The other branch is the Faroe Current, which transports about half of the total Atlantic inflow. Since the mid-1990s, hydrographic properties and current velocity have been monitored regularly on sections crossing both branches. Combining the in situ measurements of the Faroe Current with altimetry, new time series for volume, heat, and salt transport in this Atlantic inflow branch have been generated for the period 1993-2013. In this period, the volume transport of the Faroe Current increased by 9±8% (95% confidence interval) while the heat transport (relative to 0°C) increased by 18±9%, partly due to the increased volume transport and partly due to a warming of about 1°C of the Atlantic water. At the same time, the Atlantic water in the Faroe Current became about 0.1 psu more saline. In the FBC-overflow, the bottom waters have also warmed, although only about one tenth of the Atlantic water warming, but the salinity has increased, as well, so that the density of the FBC-overflow increased rather than decreased. This is consistent with increased salt transport into the Nordic Seas by the Atlantic inflow. Relative to the average salinity of the FBC-overflow (34.93 psu), the salt transport of the Faroe Current more than doubled from 1993 to 2013 and the volume transport of the FBC-overflow seems to have increased by 9% since the start of monitoring in 1995, although the statistical significance is marginal. These results are consistent with measurements in other exchanges across the GSR, which show no indication of weakening, in contrast to reports from the Labrador Sea. The Faroese AMOC branches rather strengthened and have induced a considerable increase in the oceanic heat transport towards the Arctic.


Observations of subpolar North Atlantic variability and overturning circulation from the Extended Ellett Line
N. Penny Holliday1*, Stuart A. Cunningham2, Stefan F. Gary2, Clare Johnson2, Matthew P. Humphreys3

* Presenting author

1) National Oceanography Centre, UK
2) Scottish Association for Marine Science, UK
3) University of Southampton, UK

Since 1996 the GO-SHIP section known as the "Extended Ellett Line" has been measuring the properties (temperature, salinity, carbonate chemistry) and overturning circulation between Iceland and Scotland. The section monitors the upper limb of the AMOC in the subpolar North Atlantic (SPNA): the northward flow of warm water into the Nordic Seas and Arctic, and eastern subpolar mode waters that travel cyclonically around the subpolar gyre to the Labrador Sea. Four decades of high quality ship-based measurements in the easternmost basin of the SPNA, the Rockall Trough, enhanced by data in the wider Iceland Basin, reveal long term salinity and carbonate chemistry variability that represents conditions across the SPNA and Nordic Seas. We will describe the observed variability in properties and discuss the uncertainties associated with the measurements. We compute the mean and variance of the overturning circulation at the section and consider the implications for present and future observing networks including OSNAP.


Interannual to decadal changes in the Western Boundary Circulation in the Atlantic at 11°S
Rebecca Hummels1*, Peter Brandt1, Marcus Dengler1, Jürgen Fischer1, Moacyr Araujo2, Doris Veleda2

* Presenting author

1) GEOMAR Helmholtz Centre for Ocean Research Kiel, Germany
2) DOCEAN Department of Oceanography UFPE Recife, Brazil

The western boundary current system off Brazil is a key region for variations of the Atlantic meridional overturning circulation (AMOC) and the southern subtropical cell. In July 2013 a mooring array was installed off the Brazilian coast at 11°S similar to an array installed between 2000 and 2004 at the same location. Here we present results from two research cruises and the first 10.5 months of moored observations in comparison to the observations a decade ago. The average transports of the North Brazil Undercurrent and the Deep Western Boundary Current (DWBC) have not changed between the two observational periods. DWBC eddies that are predicted to disappear with a weakening AMOC, are still present with similar characteristics. Upper layer changes in salinity and oxygen within the last decade are consistent with an increased Agulhas leakage, while at depths water mass changes are likely related to changes in the North Atlantic as well as tropical circulation changes.


Regional variability of freshwater in the North Atlantic in the RAPID/Argo era
Brian A. King1*, Elaine L. McDonagh1, Damien Desbruyeres1, N. Penny Holliday1

* Presenting author

1) National Oceanography Centre, Southampton, Marine Physics and Ocean Climate

The most striking interannual variability in the strength of the Atlantic Meridional Overturning Circulation since moored measurements began in 2004, was a reduction for a period of 18 to 24 months in 2009/10. This event, as measured by the 26.5N array, reduced the amount of heat and salt transported northwards by the AMOC, and was associated with a deficit of heat and salt in the region of the subtropical Atlantic north of the monitoring array. Our analysis of Argo data in the region north of 26.5N showed that the heat deficit recovered more quickly than the salt deficit. We have constructed time series of heat and freshwater flux across 26.5N, which are a synthesis of Argo measurements and the basin endpoint measurements made by the 26.5N moorings. By combining the spatial inventories determined from Argo data only, with the horizontal fluxes due to the AMOC, we identify the AMOC contribution to the freshwater budget north of 26.5N, and its persistence in time and spatial signature. We will show the vertical and horizontal distribution of changes in inventory, and the impact of air-sea exchange on the separate evolution of temperature and salinity. Finally, we will show the extent to which the subtropical and subpolar variability in freshwater are either correlated to each other or to variability in the AMOC measured at 26.5N.


The Meridional Overturning Variability Experiment at 16N (MOVE) in Relation to Other Latitudes
Matthias Lankhorst1*, Uwe Send1, Eleanor Frajka-Williams2, Jannes Koelling1

* Presenting author

1) Scripps Institution of Oceanography, USA
2) National Oceanography Centre, UK

MOVE has been measuring the NADW transport to represent the lower branch of the AMOC at 16N since January 2000. The most recent data retrieval from the mooring platforms took place in December 2014, extending the time series to almost 15 years duration. The data show significant variability on decadal time scales, and are highly correlated with AVISO patterns over the entire North Atlantic which establishes additional confidence in the variability observed at MOVE. Since we need to make assumptions for determining absolute transports, we compare the internal structure and its evolution between MOVE and RAPID, which is very similar in terms of dynamic height and shear profiles. A strong transition during 2009-2010 can serve as a test case during which we have bottom pressure observations, and also GRACE bottom pressure data, will be used to verify the absolute flow changes. A focus will be on how and where transport anomalies trace back to density anomalies. Another focus is to identify correlations of density and transport anomalies across latitudes.


A decade of Line W mooring observations of the Deep Western Boundary Current
Isabela A. Le Bras1,2*, Ruth Curry2, John M. Toole2

* Presenting author

1) MIT-WHOI Joint Program, Physical Oceanography, USA
2) Woods Hole Oceanographic Institution, Department of Physical Oceanography, USA

The Line W moored array, on the continental slope southeast of New England, measured the North Atlantic's Deep Western Boundary Current (DWBC) properties and velocity from 2004 to 2014. The DWBC is the primary branch of the Atlantic Meridional Overturning Circulation's (AMOC) cold limb, bringing North Atlantic Deep Water (NADW) equatorward along the continental slope. We separate NADW into neutral density classes based on their origin: Upper Labrador Sea Water (ULSW), Classical Labrador Sea Water (CLSW) from the Labrador Sea, and Overflow Waters (OW) from the Nordic Seas.

Building on the work of Pena-Molino et al. 2011, we analyze intermediate water properties at the central Line W mooring, for which there are observations starting in 2001. We find a continued trend of increasing planetary potential vorticity (PPV) in the CLSW range, reflecting a decrease in CLSW production in the Labrador Sea. The CLSW also warms (+0.01 degrees C per year) and becomes saltier (+0.001 per year) over the course of the record. These results are consistent with measurements in the Labrador Sea (Kieke et al. 2014) and indicate an approximate 10 year travel time from the Labrador Sea to Line W. Future work includes assessing the connectivity with measurements of the DWBC at 53N (Fischer et al. 2010).

We also present extended DWBC transport estimates calculated as in Toole et al. 2010, from the full moored array. The time mean transport of all NADW in the DWBC is 23.5 Sv with a standard deviation of 12.59 Sv. The CLSW portion of the DWBC has a mean transport of 7.73 Sv with a standard deviation of 4.07. The CLSW transport is decreasing at a rate of 4% a year, consistent with our water property findings from the single mooring analysis.

Transport estimates differ when data from all 6 moorings (available post 2008) are used instead of the original 5, raising questions about interpreting DWBC measurements from an array that ends in mid-ocean. Some of this difference is likely due to deep cyclones associated with Gulf Stream warm core rings. Andres et al. have recently shown that these cyclones alter the deep velocity structure along Line W in hydrographic sections, and may provide a mechanism for exchange between the DWBC and the interior. We plan to investigate this further using the mooring data.


Recent Variability in Water Mass Properties in the Labrador Sea and Scotian Rise Regions
John W. Loder1*, Igor Yashayaev1, Miguel A. Morales Maqueda0,2

* Presenting author

1) Fisheries and Oceans Canada, Bedford Institute of Oceanography, Canada
2) National Oceanography Centre, United Kingdom

Historical and recent moored, survey/vessel and Argo datasets are used to describe long-term variability in water mass properties of the lower limb of the Atlantic Meridional Overturning Circulation (AMOC) in the Labrador Sea (LS) and Scotian Rise (SR) regions. Focus is on the occurrence of moderate-to-deep convection and variability in intermediate, deep and abyssal waters in the LS, and on the properties of these components of the Deep Western Boundary Current (DWBC) on the SR. Variability in the upper and intermediate layers of the LS over the past century has been dominated by a multi-decadal variation with temperature and salinity peaks in the 1960s-1970s and early 2000s, apparently related to a combination of atmospheric forcing and larger-scale oceanographic variability in the subpolar North Atlantic. In recent years there has been weak cooling and freshening of the upper layer, possibly related to an increase in Greenland and Arctic meltwater, while during the past two decades, the intermediate layer has been intermittently (e.g., 2008, 2014) ventilated by deep convection, resulting in temporary cooling and freshening followed by slow recovery. In the LS’s abyssal layer, there has been an unprecedented increase in temperature and salinity between 2000 and 2010, related to changes in Denmark Strait Overflow Water (DSOW). On the SR, there has been overall warming at intermediate depths over the past half century, as well decadal-scale variability related in part to the changes in water mass properties (e.g. Labrador Sea Water) of the DWBC. There is also an indication of decadal-variability signals in the DSOW reaching the SR.


Variability of the North Atlantic ocean ventilation
Graeme A. MacGilchrist1*, David P. Marshall2, Helen L. Johnson1, Camille Lique1, Laura Jackson3, Richard A. Wood3, Matthew Thomas4

* Presenting author

1) University of Oxford, Earth Sciences, U.K.
2) University of Oxford, Atmospheric, Oceanic and Planetary Physics
3) Met Office, Hadley Centre, U.K.
4) Yale University, Geology and Geophysics, U.S.

In the subtropical gyres, only water subducted in late winter will remain in the ocean interior for longer than a seasonal cycle. The fast seasonal migration of outcropping density surfaces relative to the slow southward advection of water in the ocean interior means that water subducted before late winter has insufficient time to escape entrainment back into the mixed layer as the outcrop advances southward, a phenomenon denoted Stommel’s demon. We investigate whether a similar process operates on inter-annual timescales in the North Atlantic. Lagrangian analysis of a 1/4 degree ocean model reveals the ventilation age on interior density surfaces. We find that significant inter-annual variability exists and investigate its link to changes in the location of the late winter outcrop from year to year. Our results suggest that Stommel’s demon is a multi-year process that impacts ventilation on timescales longer than a seasonal cycle. This carries implications for the structure and properties of the main thermocline, nutrient cycling and the transport of atmospheric tracers into the ocean interior.


Possible consequences for the AMOC of anomalous Subpolar Mode Water formation in winter 2013/14
Robert Marsh1*, Jeremy Grist2, Erik van Sebille3

* Presenting author

1) University of Southampton, Ocean and Earth Science, UK
2) National Oceanography Centre, Southampton, Marine Systems Modelling
3) Imperial College London, Department of Physics, UK

Subpolar Mode Water (SPMW) is one of the precursors of North Atlantic Deep Water and thus an important component of the Atlantic Meridional Overturning Circulation (AMOC). Via a series of water mass transformations through mid-latitudes, progressively denser SPMW forms in a cyclonic sense around the subpolar gyre (SPG). Enhanced heat loss over the SPG in the winter of 2013/14 (W14) led to markedly enhanced formation of a particularly dense mode of SPMW, quantified in a water mass transformation framework. In W14, transformation rates in the eastern SPG peaked at around 20 Sv across a sigma-0 value of 27.3, compared to a long-term mean of ~12.5 Sv. Particle trajectory analysis is used to track the likely dispersal of newly-formed SPMW in two eddy-resolving ocean model hindcasts. We thus illustrate how the anomalous SPMW is most likely dispersing to the north and west. Timescales of dispersal and mixing are quantified, and the possible consequences for the AMOC are considered.


Circulation and Water Mass Variability in the South Atlantic
Renellys Perez1,2*, Rym Msadek3, Silvia Garzoli1,2, Ricardo Matano4, Christopher Meinen2

* Presenting author

1) University of Miami, CIMAS, USA
4) Oregon State University, CEOAS, USA

Most observational and modeling efforts on the Atlantic meridional overturning circulation (AMOC) have been focused on the North Atlantic and the Southern Oceans, which are the preferential sites for deep-water formation. There are, in contrast, fewer studies on the South Atlantic Ocean, which actively transforms AMOC-relevant water masses as they transit the basin. In this presentation, we characterize the natural modes of variability associated with the AMOC in the South Atlantic. We examine the variability of sea level anomalies, water mass properties, and meridional volume transport of the regional boundary currents (Brazil Current, North Brazil Current, Deep Western Boundary Current, and Benguela Current) and determine whether the primary mechanisms responsible for the variability of each of those fields are related to the mechanisms that govern the AMOC variability. Our analysis is based on models and observations. The model results include state-of-the-art eddy-permitting to eddy-resolving NOAA/GFDL climate simulations, ocean-only model simulations forced with CORE interannual forcing, and process-oriented numerical experiments using the Regional Ocean Modeling System. The observations include time series measurements of water mass properties and velocities inferred from moorings along 34.5°S, hydrographic transects, gridded temperature and salinity data sets, gridded sea level anomalies, sea surface temperature, surface currents, and winds obtained from satellite and satellite-in situ blended products. We will examine the extent to which a better representation of meso-scale features, which are key contributors to the variability of the South Atlantic circulation, can affect the representation of the AMOC in global climate models and hence yields a better comparison with observations.


Circulation of Dense Water Upstream and Downstream of Denmark Strait: A Review of Recent Observations and Modeling
R. Pickart1*, K. Vage2, M. Spall1, B. Harden1, W-J. von Appen3, I. Koszalka4, T. Haine4, H. Valdimarsson5, G.W.K. Moore6

* Presenting author

1) Woods Hole Oceanographic Institution, Physical Oceanography, USA
2) University of Bergen, Geophysical Institute, Norway
3) Alfred Wegener Institute, Helmholtz Centre for Polar and Marine Research, Germany
4) Johns Hopkins University , Earth and Planetary Sciences, USA
5) Marine Research Institute, Physical Oceanography, Iceland
6) University of Toronto, Physics, Canada

Most of the dense water that feeds the lower limb of the Atlantic Meridional Overturning Circulation (AMOC) flows through the Denmark Strait. Recently, a series of observational and modeling studies have led to a revised circulation scheme for the dense water approaching the strait, as well as new insights concerning the outflow of such water into the Irminger Sea. This talk assimilates some of these results into an updated view of the transformation, flow, and dynamics of dense water contributing to the AMOC via Denmark Strait. In contrast to earlier notions, a significant portion of the overflow water appears to originate from the Iceland Sea through a local overturning loop, then transported to the strait by the North Icelandic Jet. The remaining overflow water emanates from the East Greenland Current system, which is now known to include an interior branch. Not all of the dense water passing through Denmark Strait contributes to the deep overflow plume. Some of the water remains on the Greenland shelf and subsequently cascades onto the continental slope well downstream of the sill. This shelf-to-basin flux is driven by a combination of wind and offshore oceanic forcing, leading to the establishment of a permanent current on the upper slope known as the East Greenland Spill jet. It is argued that this spilling process, rather than convection in the Labrador Sea, constitutes the bulk of the intermediate AMOC.


Revisiting the assumptions behind the RAPID: how accurately does the MOCHA array measure the AMOC?
Bablu Sinha1*, Ben Moat1, Gerard McCarthy1, Adam Blaker1, Joel Hirschi1, Eleanor Frajka-Williams2, Simon Josey1

* Presenting author

1) National Oceanography Centre, Science and Technology, United Kingdom
2) University of Southampton, Ocean and Earth Sciences, United Kingdom

The RAPID array has given us an unprecedented insight into the variability of the AMOC on timescales ranging from days to years, and in combination with models and observational proxies is also shedding light on decadal variability and trends. The widespread use of the RAPID data in numerous scientific studies means it is essential to properly quantify the structural errors in the RAPID method, particularly from the point of view of vertical structure, which affects estimates of ocean heat transport. Previous work indicates that the assumption of a fixed level of no motion, and neglect of parts of the domain due to mooring coverage may introduce significant errors in calculation of the AMOC maximum volume transport and associated meridional heat transport (Duchez et al (2014), Roberts et al (2013), Elipot et al 2014) but there has been no systematic evaluation of the uncertainties due to these and other processes.

Absolute geostrophic currents are calculated using the NEMO-LIM2 ocean circulation/sea ice model. This allows us to identify and quantify errors which arise from the assumptions of geostrophy, a fixed reference level, spatially uniform mass compensation and sparse sampling. NEMO is run at 1/12 degree horizontal resolution, forced by the Drakkar Surface Forcing dataset which supplies surface air temperature, winds, humidity, surface radiative heat fluxes and precipitation. The model simulates the period 1978-2010. Whilst evaluated for the RAPID array at 26N, the method has wider geographical application, for example for the SAMOC and OSNAP arrays.

Preliminary results indicate that errors of order 1-2 Sv are expected at the depth of the maximum AMOC (~1000m), but that larger errors may arise at deeper depths. Errors are also likely to vary with time due to variations of the level of no motion on a variety of timescales.


Theme 1 poster presentations:
Denmark Strait Overflow Water observed at Line W: its evolution and mixing via deep cyclones
Magdalena Andres1*, John M. Toole1, Daniel J. Torres1, William M. Jr. Smethie2

* Presenting author

1) Woods Hole Oceanographic Institution, Physical Oceanography, USA
2) Columbia University, Lamont-Doherty Earth Observatory, USA

Shipboard velocity and property data from 18 transects across the North Atlantic Deep Western Boundary Current (DWBC) near 40˚N are analyzed to study the evolution of the Denmark Strait Overflow Water (DSOW) component of the DWBC and mixing between the DWBC and the interior. The transects were made between 1994 and 2014 and lie along Line W, which reaches from the continental shelf south of New England to Bermuda coincident with a Jason-2 satellite track. The shipboard data include measurements of velocity from lowered acoustic Doppler current profilers (LADCPs), CTDO profiles, and trace gas chlorofluorocarbon concentrations from bottle samples at discrete depths taken on the CTDO Rosette casts at 26 regular stations or a subset of these stations. In each transect, DSOW exhibits a distinct chlorofluorocarbon concentration maximum in the abyssal ocean (> 3000 m depth) along the sloped western boundary. Examination of the sea surface height (SSH) maps from satellite altimetry indicates that quasi-stationary meander troughs of the Gulf Stream path in the upper ocean were present at Line W during 5 of the 18 sections. For these 5 sections the LADCP velocity sections suggest the upper ocean trough is accompanied by a large cyclone in the deep ocean in the DSOW density layer. The occurrence of deep cyclones in conjunction with Gulf Stream troughs as inferred from the LADCP sections along Line W is consistent with previous observations (from 1988 to 1990) in the region from a moored array in the Synoptic Ocean Prediction (SYNOP) experiment. The SYNOP array suggested deep cyclones are present here about 35% of the time. The composite velocity section produced from the 5 Line W transects sampling through a Gulf Stream trough suggests that a typical cyclone reaches swirl speeds of greater than 30 cm/s at 3400 m depth and has a radius (distance between the center and the maximum velocity) of ~75 km. The tracer data suggest that these cyclones affect not only the deep velocity structure along Line W, but also provide a mechanism for water exchange between the DWBC and the interior.


Investigating the impact of CO2 on multi-decadal variability of the AMOC in the HadCM3 coupled climate model
Edward Armstrong1*, Paul Valdes1, Jo House1, Joy Singarayer2

* Presenting author

1) University of Bristol, School of Geographical Sciences, UK
2) University of Reading, Department of Meterology, UK

The strength and variability of the AMOC has important implications for Northern Hemisphere climate including precipitation, surface air temperatures and sea level. Understanding how the AMOC may respond to anthropogenic climate change is considered a crucial challenge facing the climate science community.

A number of modelling studies have shown that higher concentrations of CO2 will weaken the AMOC. However, understanding how the characteristics (i.e. frequency, amplitude) of decadal to centennial variability is still under debate and is highly dependent on the model used. It is crucial to understand the timescale and mechanisms of internal variability in order to differentiate impacts that are anthropogenically forced (i.e. a CO2 induced weakening) relative to those that are a natural response of the system. This improves the ability to predict the AMOC and climate on decadal timescales.

This study will investigate how increasing concentrations of atmospheric CO2 are expected to impact the long-term variability of the AMOC within the HadCM3 coupled climate model. We aim to answer two key questions; 1) how does CO2 impact the frequency and amplitude of variability, and 2) what mechanism(s) are responsible for this change. The model has been run on millennial time scales to provide a strong statistical signal at five different CO2 concentrations.


Meridional overturning circulations driven by surface wind and buoyancy forcing
Michael J. Bell1*

* Presenting author

1) Met Office, Climate Science, UK

A conceptual picture of the Meridional Overturning Circulation (MOC) has been developed using 2- and 3-layer models governed by the planetary geostrophic equations and simple global geometries. The picture has four main elements. First cold water driven to the surface in the South Atlantic north of Drake passage by Ekman upwelling is transformed into warmer water by surface heating. Second the model’s boundary conditions constrain the depths of the isopycnal layers to be almost flat along the eastern boundaries of the ocean. This results in warm water reaching high latitudes in the northern hemisphere where it is transformed into cold water by surface heat loss (the third element). The final element of the picture is the assumption that western boundary currents are able to close the circulations. The results from a set of numerical experiments for the upwelling limb in the Southern Hemisphere are summarised in a simple conceptual schematic. Analytical solutions have been found for the down-welling limb assuming the wind stress in the Northern Hemisphere is negligible. The expression for the depth of the isopycnal interface on the eastern boundary obtained by combining these solutions in a 2-layer model is similar to that obtained for the depth of the pycnocline by Gnandesikan (1999).

Bell, M. J. 2014 Water mass transformations driven by Ekman upwelling and surface warming in sub-polar gyres. Submitted to J. Phys. Oceanogr.

Bell, M. J. 2014 Meridional overturning circulations driven by surface wind and buoyancy forcing. Submitted to J. Phys. Oceanogr.


Interannual and decadal North Atlantic heat content variability in the CMIP5
Andrew Davis1*, Luanne Thompson1

* Presenting author

1) University of Washington, Oceanography, USA

Our work examines interannual and decadal North Atlantic heat content variability within the CMIP5 coupled climate model archive with focus on the NSF-NCAR CESM group. Within this set of models, Gulf Stream and North Atlantic Current variability dominates interannual temperature exchanges, necessitating approaches that can diagnose the horizontal and vertical propagation of anomalous temperature patterns.
Across the CESM there exist commonalities in spatial temperature variance. Historical runs also display a large degree of agreement in patterns of temperature trend. The dominant pattern of variance in historical runs is not only strongly interannual/decadal, but also contains much of the trend. This pattern is best described as a warming in the Gulf Stream extension, with a cooling in the subpolar gyre. This pattern is nearly barotropic, but intensifies at depth. This strongly suggests long-term trends and decadal variance in AMOC strength and thus necessitates an attribution to anthropogenic forcing.
The dominant modes of heat content variability are tested for robustness across model simulations, and are linked to Gulf Stream and North Atlantic Current regime shifts. Influence from the North Atlantic Oscillation (NAO) is also considered. Horizontal and vertical propagation of anomalies are diagnosed by employing complex EOF and Principal Oscillation Pattern (POP) analysis.
Results indicate that intrinsic interannual variance in Gulf Stream position and strength is primarily responsible for the observed large-scale pattern of heat content variability. This Gulf Stream variance is also tied to complementary alterations to the mean structure of the North Atlantic Current.


Meridional propagation of meridional volume transport anomalies in an idealised model
Edward W. Doddridge1*, David P. Marshall1

* Presenting author

1) University of Oxford, Physics, UK

An idealised reduced-gravity model is used to investigate the meridional coherence of meridional overturning circulation anomalies in the Atlantic ocean. The model is run at both eddy-resolving and non-eddying resolutions, the latter comparable with ocean circulation models currently used for climate prediction. Our idealized model produced meridionally coherent volume transport anomalies. A time-lagged autocorrelation of meridional volume transport anomalies shows southwards propagation at all latitudes.

Anomalies in the western boundary current of the subpolar gyre propagate advectively, but are compensated by changes in the gyre circulation. The zonally integrated transport propagates at the speed of a boundary wave, showing that previous idealised results are robust in the presence of a mean flow.


The Oleander Project: High-resolution observations of the dynamic ocean between New Jersey and Bermuda
Kathleen Donohue1*, Tom Rossby1, Charles Flagg2, Alejandra Sanchez-Franks 2

* Presenting author

1) University of Rhode Island, Graduate School of Oceanography, USA
2) Stony Brook University, School of Marine and Atmospheric Sciences, USA

Beginning in late 1992, high-horizontal resolution upper-ocean velocity has been sampled by an acoustic Doppler current profiler (ADCP) mounted in the hull of the container vessel CMV Oleander that operates on a weekly schedule between New Jersey and Bermuda. Observations made from the MV Oleander provide a unique perspective among the complement of North Atlantic in situ and remote sensing observations. The Oleander measurements yield a view of ocean-current variability over a very wide range of scales; measuring velocity at great detail down to 2 km scale, and also able to estimate upper layer fluxes out to O(1000) km scales. The route traverses the southwestern-most extension of the wind-driven subpolar gyre (Labrador Current) and part of the wind-driven subtropical gyre (the Gulf Stream and Sargasso Sea) in a region where the AMOC’s cold limb flow (Deep Western Boundary Current) and warm limb flow (Gulf Stream) are side-by-side rather than vertically stacked. Two decades of observations contribute to continued investigation of interannual and decadal variability. Contrary to recent claims of a Gulf Stream slow-down, results from the Oleander Project are unambiguous: direct observations show no long-term trend in Gulf Stream surface transport. Along the Oleander line, the Gulf Stream transport exhibits significant interannual variability. This variability does not co-vary with Florida Current transport, likely because of the presence of energetic wind-driven fluctuations north of Florida. It has been our working hypothesis that Slope Sea transport is related to both the position of the Gulf Stream and to the NAO. During low-NAO periods, when the Labrador Sea experiences little convection, it exports more water to the west through the Slope Sea and along the continental shelf, and the Gulf Stream shifts southward. Attempts to verify that Slope Sea transport is related to the position of the Gulf Stream and the NAO have proven difficult to verify with the direct measurements from the Oleander Project. This may be because despite low vertical shear in the Slope Sea, the existing measurements do not capture the upper limb of the Labrador Sea transport centered near 800 m depth. Looking to the future, we aim to enhance the observational system to include an acoustic Doppler current profiler (ADCP) pair. One ADCP will measure currents through the base of the thermocline to about 1200 m depth in the open ocean and the other ADCP will provide high vertical resolution profiling of the upper ocean and shelf waters. These measurements, as in the past, will be complimented by monthly XBT sections and continuous TSG coverage.


Recent changes in the Atlantic meridional overturning circulation: A thermohaline perspective
Dafydd G Evans1*, John Toole2, Gael Forget3, Jan D. Zika1, A. George Nurser4, Alberto C. Naveira Garabato1, Lisan Yu2

* Presenting author

1) University of Southampton, Ocean and Earth Sciences, UK
2) Woods Hole Oceanographic Institution, Physical Oceanography, USA
3) Massachusetts Institute of Technology, The Department of Earth, Atmospheric and Planetary Sciences
4) National Oceanography Centre Southampton, Physical Oceanography, UK

Interannual variability in the volumetric water-mass distribution within the North Atlantic subtropical gyre is described in relation to the recent reported variability in the Atlantic Meridional Overturning Circulation. We investigate the relative roles of buoyancy forcing at the sea-surface and local changes in the wind-driven circulation. We project the data into thermohaline coordinates as volumes of water defined by their temperature and salinity, using data from an Argo based climatology and a high-resolution ocean state estimate (ECCO). Regarding the reported Atlantic meridional overturning circulation changes, during the winters of 2009/10 the total subtropical gyre volume above the thermocline decreases while the volume below increases in compensation, a redistribution that is equivalent to a transport of 25 Sv (1 Sv==10^6 m^3s^-1) over 3 months. A comparison to two air-sea flux re-analyses products shows that this variability cannot be explained by anomalous diabatic cooling over the subtropical gyre, suggesting the volumetric redistribution is caused by changes in the transport divergence between 26 and 45N. In ECCO, we see a reduction in the zonal circulation of the subtropical gyre with a divergence of transport above the thermocline, compensated below the thermocline by an increase in the southward transport at 45N and a decrease at 26N. These changes are consistent with a short-term wind-driven change in the meridional position of the Subtropical Gyre. We relate this adiabatic variability to the relative roles of wind-driven meridional transport and vertical Ekman pumping.


How AMOC variations affect North Atlantic SSTs
Alexey Fedorov1*

* Presenting author

1) Yale University, Geology and Geophysics, USA

The AMOC role in climate critically depends on how AMOC variations affect the North Atlantic SST on decadal and longer timescales. Particular aspects of this relationship include the connection between AMOC variability and the Atlantic Multidecadal Oscillation (AMO), and the connection between AMOC and SST longer-term trends. Here, we use historical and control simulations of the CMIP5 dataset, together with numerical experiments with climate models, to investigate this relationship. We find that indeed the power spectra of variation in the AMOC volume transport and the basin-averaged North Atlantic SST share common spectral peaks in many models, and the two variables typically correlate with coefficients ranging from 0.2 to 0.8 (with the AMOC having a few-year lead). The estimated sensitivity of the North Atlantic SST given by the CMIP5 multi-model ensemble is about 0.3 degrees C per 1Sv of AMOC change; however, on average AMOC variations explain only one third of SST variance. The greatest impact of AMOC variations on SSTs is typically observed between 40 and 60N, but the spatial patterns vary greatly across the models as controlled by the models’ deep water formation regions and the Subpolar gyre circulation. Another issue is that, while the observed (relatively short) AMO record is dominated by a 50-70 year periodicity, climate models usually exhibit power spectra maxima at periods close to 20 years. In this study, we investigate both the mechanisms of the AMOC impacts and the leading causes of such large inter-model differences. In a related context, recent studies suggested that in the absence of AMOC measurements during the 20th century the North Atlantic SST index can provide a proxy for AMOC variability and trends. However, given these differences and the relatively small fraction of SST variance explained by the AMOC, our results caution against such an interpretation.


Estimates of the seasonal variability of volume, heat, and freshwater fluxes in the eastern subpolar North Atlantic
Stefan F. Gary1*, Stuart A. Cunningham1, Clare Johnson1, N. Penny Holliday2, Loic Houpert1

* Presenting author

1) Scottish Association for Marine Science, Department of Physics and Technology, UK
2) National Oceanography Centre, Marine Physics and Ocean Climate, UK

Recently, glider missions have supplemented the long-established ship-based observations along the Extended Ellett Line (EEL) hydrographic section between Scotland and Iceland. The EEL hydrographic section is situated to capture a large fraction of the volume, heat, and freshwater fluxes associated with the upper limb of the AMOC. Due to the weather, ship-based observations have primarily been in the summer and gliders present a renewed opportunity to investigate the seasonal variability of volume, heat, and freshwater fluxes in the upper layer of the AMOC in the eastern subpolar North Atlantic. First, we describe the quality control of glider data relative to ship-based observations and the merging of glider observations into a database of ship-based observations. Then, we present our estimates for the seasonal variability of volume, heat, and freshwater transports in the eastern subpolar North Atlantic. Finally, we interpret these fluxes in light of the larger scale circulation.


Monitoring the MOC in the South Atlantic: A 'SAMOC Initiative' update
Silvia L. Garzoli1,2*, Alberto Piola3, Sabrina Speich4, Edmo Campos5, Mike Roberts6, Renellys C. Perez1,2, Thierry Terre7, Christopher S. Meinen2

* Presenting author

1) University of Miami/Cooperative Institute for Marine and Atmospheric Studies, USA
2) NOAA/Atlantic Oceanographic and Meteorological Laboratory, USA
3) Departamento de Oceanografía, Servicio de Hidrografía Naval, and Departamento de Ciencias de la Atmósfera y los Océanos
4) Laboratoire de Météorologie Dynamique, École Normale Supérieure,
5) Oceanographic Institute, University of São Paulo,
6) Oceans and Coasts Research, Department of Environmental Affairs,
7) Laboratoire de Physique des Océans, Ifremer,

Variations in the Meridional Overturning Circulation (MOC) are known to have global implications to the climate system, however until recently most MOC observing programs have been focused in the North Atlantic. Recent model and data analyses have suggested that critical water mass changes to the upper and lower limbs of the MOC occur in the South Atlantic, and only limited latitudinal coherence has been found to date between the MOC observations made by the North Atlantic observing systems at different latitudes. As a result, a priority for the USAMOC Science Team has been the establishment of a MOC observing system in the South Atlantic, and recently the International CLIVAR panel endorsed a South Atlantic MOC (“SAMOC”) Initiative to both strengthen existing programs seeking to study the MOC in the South Atlantic and to encourage further expansion of the MOC observing system in the region. This presentation will summarize the present status of the international SAMOC observing system and present recent observation and modeling results developed through coordination by the international SAMOC Initiative.


Temporal variability of the South Atlantic Meridional Overturning Circulation between 20S and 35S
Gustavo Goni1*, Shenfu Dong2,1, Francis Bringas1

* Presenting author

1) National Oceanic and Atmospheric Administration, Physical Oceanography Division, USA
2) University of Miami, Cooperative Institute for Marine and Atmospheric Studies, USA

Satellite altimetry measurements are used to investigate the spatial and temporal variability of the Meridional Overturning Circulation (MOC) and Meridional Heat Transport (MHT) in the South Atlantic. Altimetry-derived synthetic temperature and salinity profiles between 20°S and 34.5°S are used to estimate the MOC/MHT, which compare well with estimates obtained from XBT measurements. Consistent with studies from XBTs and Argo data, both the geostrophic and Ekman contributions to the MOC exhibit strong annual cycles, and play an equal role in the MOC seasonal variations. The strongest variations on seasonal and interannual time scales in our study region are found at 34.5°S. The dominance of the geostrophic and Ekman components on the interannual variations in the MOC and MHT varies with time and latitudes, with the geostrophic component being dominant during 1993-2006 and the Ekman component dominant between 2006-2011 at 34.5°S.


Forced Multidecadal AMOC Variability Over the Past 1000 years
Paul R. Halloran1*, David J. Reynolds2, Ian R. Hall2, James D. Scourse3

* Presenting author

1) University of Exeter, Geography, UK
2) Cardiff University, School of Earth and Ocean Science, UK
3) Bangor University, School of Ocean Science, UK

The RAPID array has provided unprecedented information on AMOC variability on the timescales of days to years, but much of the societally relevant climate variability attributed to AMOC change is associated with its decadal to multidecadal behaviour (e.g., Knight et al., 2006, Delworth and Mann, 2000). Attempts to get a handle on multidecadal AMOC variability to-date have been based on the assumptions that most of the North Atlantic SST variability over the observed period results from AMOC variability (e.g. Mann et al., 2009), but recently this assumption has been challenged (Booth et al., 2012). If multidecadal variability in North Atlantic SSTs is not reflecting AMOC variability, we must question whether that AMOC variability exists, and if it does, what it looks like and how it is generated.

Here we present results from the first annually-resolved reconstruction of seawater density from North Iceland (Reynolds et al., submitted), and demonstrate that multidecadal variability in this record closely matches the PMIP3 Greenland-Iceland-Norway Seas multimodel mean seawater density variability over the past 1000 years. We demonstrate that this density timeseries is closely related to the models’ AMOC variability, and can be attributed almost exclusively to volcanic and solar forcing.

From these results we can conclude: (1) Multidecadal AMOC variability over the past 1000 years was externally forced; (2) Since this variability is not seen in any individual models’ AMOC time-series, models either generate spurious natural AMOC variability, or do not respond strongly enough to the natural forcings; (3) The Atlantic Multidecadal Oscillation does not exist as a mode of natural AMOC variability in the real world. Each of these points has important implications for decadal and centennial climate prediction, as well as our basic understanding of the climate system.

- J. Knight, C. Folland and A. Scaife., Climate impacts of the Atlantic Multidecadal Oscillation, GRL, 2006
- T. L. Delworth and M. E. Mann 
Observed and simulated multidecadal variability in the Northern Hemisphere, Climate Dynamics, 2000
- M.E. Mann, Z. Zhang, S. Rutherford et al., Global Signatures and Dynamical Origins of the Little Ice Age and Medieval Climate Anomaly, Science, 2009
- B.B.B Booth, N. Dunstone, P.R. Halloran et al., Aerosols implicated as a prime driver of twentieth-century North Atlantic climate variability, Nature, 2012
- (Reynolds et al., submitted)


Decadal Variations of the Atlantic Meridional Overturning Circulation as simulated by the VIKING20 Model
Patricia Handmann1*, Jürgen Fischer1, Martin Visbeck1, Erik Behrens2, Lavinia Patara1

* Presenting author

1) GEOMAR Helmoltz Centre for Ocean Research Kiel, Physical Oceanography, Germany
2) NIWA National Institute of Water and Atmospheric Research, Wellington,

Time series of observed deep circulation transports and water mass properties in the subpolar North Atlantic are beginning to be long enough to investigate multiannual to decadal variability of the deep water. Simultaneously high resolution ocean circulation models (1/20° resolution VIKING20 model) can be used to compare observations with model simulation. The models also allow to diagnose the deep water circulation processes more completely and to relate local to basin scale signals.
At the exit of the Labrador Sea the pathways of different originating water masses meet. These built the complex combination of the North Atlantic Deep Water (NADW). The lower part of NADW is formed by water masses entering the subpolar basin over the Greenland-Scotland ridge. Iceland-Scotland Overflow Water (ISOW) from the eastern sills has the longest pathway and joins the densest deep water component from Denmark Strait (DSOW) after crossing the Mid-Atlantic-Ridge through Charlie-Gibbs Fracture Zone (CGFZ); together, they form the Lower NADW. The upper component of the NADW is composed of Labrador Sea Water (LSW), which is formed and modified through deep convection in the Labrador Sea.
Using 60 year long time series of North Atlantic water masses and currents produced by the Viking20 model driven by the reanalysis CORE2 forcing, a comparison of transport variability of observed and modeled data will be presented at three locations: Deep flow at the exit of the Labrador Sea at 53°N; upper layer transports between New Jersey and Bermuda (OLEANDER section) and between the southern tip of Greenland and Portugal (OVIDE section). Another point of interest is the propagation of the transport signal through Labrador Sea. Is the model reproducing the observed long-term behavior of the different components in phase and amplitude? Do the results permit identification of the processes leading to these variations in transport variability? Finally, is it possible to extend the observed variability pattern over the observed time span (15 years) to the total time range of the model simulations (60 years)?


Upstream Sources of the Denmark Strait Overflow Water: Results from a Year-Long Moored Array
Benjamin Harden1*, Robert Pickart1, Héðinn Valdimarsson2, Kjetil Våge3, Laura de Steur4, Steingrímur Jónsson5,2, Andreas Macrander2, Eli Børve3, Lisbeth Håvik3

* Presenting author

1) Woods Hole Oceanographic Institution, Physical Oceanography Department, USA
2) Marine Research Institute, Iceland
3) University of Bergen, Geophysical Institute and Bjerknes Centre for Climate Research, Norway
4) Norwegian Polar Institute, Norway
5) University of Akureyri, Iceland

We present the first results from a densely instrumented mooring array upstream of the Denmark Strait sill, extending from the Iceland shelfbreak to the Greenland shelf. The array was deployed from September 2011 to August 2012, and captured the vast majority of overflow water denser than 27.8 kg/m3 approaching the sill. The mean transport of overflow water over the length of the deployment was 3.55 Sv. Of this, approximately 0.6 Sv originated from below sill depth, revealing that aspiration takes place in Denmark Strait. We confirm the presence of two main sources of overflow water: one involving the East Greenland Current and the other via the North Icelandic Jet. Using an objective technique based on the hydrographic properties of the water, the transports of these two sources are found to be 2.55 Sv and 1.00 Sv, respectively. We further partition the East Greenland Current source into that carried by the shelfbreak jet (1.50 Sv) versus that transported by a separated branch of the current (1.05 Sv). Over the course of the year the total overflow transport through the array varies considerably less than the flux in each of the branches, demonstrating that compensation takes place among the pathways. This is especially true for the two East Greenland Current branches whose transports vary out of phase with each other on weekly time scales. We argue that wind forcing plays a role in this partitioning.


Mechanisms of MOC hysteresis in a GCM and the importance of hosing scenarios
Laura C Jackson1*, Robin S Smith2, Richard Wood1

* Presenting author

1) Met Office, Hadley Centre, UK
2) Reading University, UK

A previous paper (Hawkins et al, 2011) described the first time in which a hysteresis of the Meridional Overturning Circulation has been found in a general circulation model (FAMOUS). They showed that, for a given input of fresh water into the north Atlantic, there existed two possible states: one with a strong overturning in the north Atlantic and the other with a reverse Atlantic cell.

In this study we investigate the mechanisms behind the hysteresis. We assess the changes in surface fluxes and advection that lead to nonlinear MOC changes and the connection between the Atlantic and Pacific overturning circulations. The formulation of the hosing scenario, and in particular how the input of fresh water into the Atlantic is compensated, is found to be very important.


Size matters: another reason why the Atlantic is saltier than the Pacific
Catherine S. Jones1, Paola Cessi1*

* Presenting author

1) University of California, San Diego, Scripps Institution of Oceanography

The meridional overturning circulation (MOC) is studied in a model forced by longitude-independent surface wind-stress, freshwater flux and temperature relaxation, in a domain of simplified geometry comprising two basins connected by a circumpolar channel occupying the southernmost region. The only asymmetry between the two basins is that one is twice as wide as the other. The model exhibits one stable state with deep sinking in the northern hemisphere of the narrow basin and upwelling in the wide basin and the circumpolar channel. A state with sinking in the wide basin and upwelling in the narrow basin (and channel) can be forced to occur but is unstable. This preference for sinking in the narrow basin is illustrated using the extension of Gnanadesikan (1999) model for the upper branch of the MOC to two basins connected by a re-entrant channel, coupled to one-dimensional advection-diffusion equations for temperature and salinity. The two basins are coupled at the northern edge of the channel by a geostrophic exchange flow caused by the difference in depth of the thermocline at the eastern boundaries of each basin. Two states, one with sinking in the narrower basin, and one with sinking in the wider basin, are compared. The meridional transport per unit width and the associated salt advection are larger when sinking occurs in the narrow basin and upwelling in the wider basin, explaining why the narrow basin is saltier and deep water formation is favored there, even in the absence of longitudinal differences in the atmospheric forcing.


Heat and Freshwater Convergence Anomalies in the Atlantic Ocean Inferred from Observations
Kathryn Kelly1, LuAnne Thompson2*, Kyla Drushka1

* Presenting author

1) University of Washington, Applied Physics Lab, USA
2) University of Washington, School of Oceanography, USA

Observations of thermosteric (TSL) and halosteric sea level (HSL) from hydrographic data, liquid water equivalent (LWE) from GRACE and altimetric sea surface height (SSH) are used to infer meridional heat transport (MHT) and freshwater convergence (FWC) anomalies for the Atlantic Ocean. An "unknown control" version of a Kalman filter in each of eight regions extracts smooth estimates of heat transport convergence (HTC) and FWC from discrepancies between the response to monthly surface heat and freshwater fluxes and observed heat and freshwater content. The model is run for 1993-2012. Estimates of MHT anomalies are derived by summing the HTC from north to south and adding a spatially uniform, time-varying MHT derived from observed values near 40N. Estimated anomalies in MHT are comparable to those recently observed at the RAPID/MOCHA line at 26.5N. Coherent anomalies of Atlantic MHT are shown to correlate with the Antarctic Oscillation, as well as with estimates of the Agulhas Leakage. In combination with changes in the tracks of Agulhas eddies between high and low MHT periods these results depict large-scale wind-forced changes in Atlantic Ocean circulation that include the Agulhas region. FWC estimates in the Atlantic Ocean (67N to 35S) have a minimum in 2003-2004 (anomaly of -0.1 Sv) with subsequently increasing values (to 0.1 Sv) with error estimates of about 0.1 Sv. The FWC anomaly increases over the latter part of the record at a time when MHT decreases (freshwater convergence as the AMOC decreases), indicative of a positive feedback between the MOC and FWC.


Introduction to OSNAP: Overturning in the Subpolar North Atlantic
Susan Lozier1*, Feili Li1

* Presenting author

1) Duke University, Earth and Ocean Sciences, United States

OSNAP (Overturning in the Subpolar North Atlantic) is an international program designed to provide a continuous record of the full-water column, trans-basin fluxes of heat, mass and freshwater in the subpolar North Atlantic. The OSNAP observing system, deployed in the summer of 2014, consists of moored instruments and gliders along a line extending from southern Labrador to the southwestern tip of Greenland across the mouth of the Labrador Sea (OSNAP West), and from the southeastern tip of Greenland to Scotland (OSNAP East). The observing system also includes subsurface floats in order to trace the pathways of overflow waters in the basin and to assess the connectivity of currents crossing the line. In this poster we will describe the OSNAP objectives, and the observation and modelling components of the program.


A relation between the volume transports of the Atlantic Meridional Overturning Circulation and Antarctic Circumpolar Current
David Marshall1*, Helen Johnson2

* Presenting author

1) University of Oxford, Department of Physics, United Kingdom
2) University of Oxford, Department of Earth Sciences, United Kingdom

A simple heuristic model is developed to explain the relative volume transports of the Atlantic Meridional Overturning Circulation (17.5 Sv) and Antarctic Circumpolar Current (137 Sv) in terms of three depth scales: the e-folding depth of the global stratification, the depth of maximum overturning streamfunction and the depth of Drake Passage. For realistic values, the model is able to explain the factor 8 difference in the magnitudes of these currents.


The North Atlantic subpolar circulation in an eddy-resolving global ocean model
Alice Marzocchi1,2*, Joel J.-M. Hirschi1, N. Penny Holliday1, Stuart A. Cunningham1,3, Adam T. Blaker1, Andrew C. Coward1

* Presenting author

1) National Oceanography Centre Southampton , UK
2) University of Bristol, School of Geographical Sciences , UK
3) Scottish Association for Marine Science, UK

The subpolar North Atlantic represents a key region for global climate, but most numerical models still have well described limitations in correctly simulating the local circulation patterns. Here, we present the analysis of a 30-year run with a global eddy-resolving (1/12°) version of the NEMO ocean model. Compared to the 1° and 1/4° equivalent versions, this simulation more realistically represents the shape of the Subpolar Gyre, the position of the North Atlantic Current, and the Gulf Stream separation. Other key improvements are found in the representation of boundary currents, multi-year variability of temperature and depth of winter mixing in the Labrador Sea, and the transport of overflows at the Greenland–Scotland Ridge. However, the salinity, stratification and mean depth of winter mixing in the Labrador Sea, and the density and depth of overflow water south of the sill, still present challenges to the model. This simulation also provides further insight into the spatio-temporal development of the warming event observed in the Subpolar Gyre in the mid 1990s, which appears to coincide with a phase of increased eddy activity in the southernmost part of the gyre. This may have provided a gateway through which heat would have propagated into the gyre's interior.


Hydrographic Structure of Overflow Water Passing Through the Denmark Strait
Dana Mastropole1*, Robert Pickart1, Héđinn Valdimarsson2, Kjeti Våge3, Carolina Nobre1, Kerstin Jochumsen4, Detlef Quadfasel4, Gerd Krahmann5, Bert Rudels6, James Girton7

* Presenting author

1) Woods Hole Oceanographic Institution, United States
2) Marine Research Institute, Iceland
3) University of Bergen, Norway
4) Universität Hamburg, Germany
5) Helmholtz Center for Ocean Research Kiel, Germany
6) Finnish Meteorological Institute, Finland
7) University of Washington, United States

Denmark Strait Overflow Water (DSOW) constitutes the densest portion of North Atlantic Deep Water, which feeds the lower limb of the Atlantic Meridional Overturning Circulation (AMOC). As such, it is critical to understand the manner in which DSOW is transferred across the Greenland-Scotland Ridge from the Iceland Sea to the North Atlantic Ocean. The goal of this study is to characterize the hydrographic structure of the different DSOW constituents at the sill, before the water descends into the Irminger Basin. We use temperature and salinity (T/S) data from 117 shipboard crossings in the vicinity of the sill, collected between 1990 and 2012. The mean section reveals circulation components consistent with what is known about the upstream flow paths. The individual realizations indicate that weakly stratified "boluses" of DSOW frequent the sill and contribute the densest water to the overflow. Using an objective definition to identify boluses, we characterize their structure, size, and location and relate them to the T/S modes found at the sill. Lastly, using an historical hydrographic data set from the Nordic Seas, we make inferences regarding the origin of the boluses.


Atlantic Multidecadal Variability in a multi-model ensemble of CMIP5 simulations: an assessment of its spectral characteristics and its non-stationary behaviour
Irene Mavilia1,2*, Alessio Bellucci1, Panos Athanasiadis1, Silvio Gualdi1,3, Rym Msadek4, Yohan Ruprich-Robert4

* Presenting author

1) Euro-Mediterranean Center on Climate Change (CMCC), Italy
2) Ca’ Foscari University, Italy
3) Istituto Nazionale di Geofisica e Vulcanologia (INGV), Italy
4) NOAA/Geophysical Fluid Dynamics Laboratory, USA

The Atlantic Multidecadal Variability (AMV) is a coherent pattern of variability of the North Atlantic Sea Surface Temperature field affecting several components of the climate system in the Atlantic region and the surrounding areas. AMV is thought to be the surface signature of the Atlantic meridional overturning circulation (AMOC) variability. Our current knowledge of the AMV is based on a relatively short observational record, which severely limits our understanding of the mechanisms involved, as well as the characterisation of the low-frequency tail of the variability spectrum. In order to quantify accurately the contribution of anthropogenic forcings to the observed climatic changes, it is essential to understand better the natural climate variability occurring at long time scales. Here, the behaviour of the AMV is examined in a set of multi-century CMIP5 pre-industrial climate simulations performed with different Coupled General Circulation Models (CGCMs). Only the longest simulations (minimum 500-year long) of the CMIP5 archive have been used. In these simulations the AMV exhibits a non-stationary behaviour, which is objectively assessed. In some of the models a shorter time scale mode (∼20 years) seems to alternate with a longer time scale mode (∼60 years) with a gradual shift from one to another across different epochs, involving also their co-existence. A multi-model analysis of the AMV allows us to investigate similarities and differences across an ensemble of state-of-the-art climate models and to identify the prominent simulated mechanisms of air-sea interaction at mid-latitude and over decadal and longer time scales. The relationship between the detected AMV behaviour and the variability of the AMOC is also examined. The non-stationary behaviour identified in most models suggests that the character of the observed AMV may undergo significant changes in the future. This ongoing analysis will provide further insight into the dynamics of the AMV variability.


Property variability in Florida Straits: Contributions from the North and South Atlantic
Elaine McDonagh1*, Molly Baringer2, Chris Meinen2

* Presenting author

1) National Oceanography Centre, UK

Florida Straits contains most of the northward flowing water associated with the Atlantic Meridional Overturning Circulation at 26.5°N. Determining the temperature and salinity of the water that flows through the Strait has a large impact on the calculation of heat and freshwater fluxes across the whole subtropical North Atlantic. Previous studies have shown an apparent systematic increase in the transport-weighted salinity in Florida Straits from the approximately pentadal repeat hydrographic sections at this latitude. Here we examine the higher frequency variability from observations to determine if this apparent trend is actually aliasing of a signal. We also examine the variability in the water properties in Florida Straits, particularly the partition between waters of South Atlantic Origin, found on the western side of the Straits, and waters recirculated from the North Atlantic Subtropics found on the eastern side of the Straits.

Analysis of Pegasus data collected in 1982-1984 shows that the variability of the properties in the Straits on the western side (water of South Atlantic origin) is different to that on the eastern side (water of North Atlantic origin). On the western side of the Straits the transport-weighted salinity is strongly correlated with the transport, on the eastern side of the Straits variability in the transport-weighted salinity is dominated by a seasonal signal. Although the transport of both the western and the eastern side of the Straits increases with an increasing total transport, the transport on the western side of the Straits is anti-correlated with that on the eastern side of the Straits.

We extend this analysis to more recent data collected in the Florida Straits. In addition we put the Florida Straits variability into the context of larger scale changes in the North and South Atlantic as well as determining the impact of the variability on the heat and freshwater fluxes across 26.5°N.


Monitoring Florida Current transport at 27°N using pressure gauges
Christopher S. Meinen1*

* Presenting author

1) NOAA/Atlantic Oceanographic and Meteorological Laboratory, Physical Oceanography Division, United States of America

The continuous daily estimates of the Florida Current/Gulf Stream volume transport made via submarine cable between Florida and Grand Bahama Island are a critical component of the Meridional Overturning Circulation (MOC) observing array at 26.5°N. Furthermore, with more than 30 years of daily estimates, the Florida Current transport record is one of the longest ocean transport records in existence and it represents an important time series for validating and testing ocean and coupled ocean-atmosphere numerical models. While the NOAA program that funds the cable observations, the Western Boundary Time Series project, is fairly securely funded, the project is dependent upon an out-of-service telephone cable that will, in all likelihood, break at some unknown time in the future. Results will be presented on tests of a possible backup system for the cable – pressure/tide gauges maintained on either side of the Straits of Florida near the cable termini. Six years of data from 2008-2014 suggest that a pair of pressure gauges captures some of the key variations in the Florida Current transport, although the overall correlation is not as good as one might hope (r = 0.76). Correlations do not improve significantly after the removal of high frequency (periods less than 90 days) variations, which is consistent with the barotropic and baroclinic components of the flow in the Straits of Florida varying independently of one another at seasonal and shorter time scales. Some implications for future observing of the Florida Current will be discussed.


North Atlantic heat transport at 26.5°N: New insights from synthesis of RAPID array observations with output from a high resolution global ocean circulation model
B. I. Moat1*, S. A. Josey1, B. Sinha1, D. A. Smeed1, A. T. Blaker1, W. E. Johns2, G. McCarthy1, J. J-M. Hirschi1, D. Rayner1, E. Frajka-Williams3, A. Duchez1

* Presenting author

1) National Oceanography Centre, UK
2) Rosenstiel School of Marine and Atmospheric Science, USA
3) University of Southampton, UK

The RAPID/MOCHA/WBTS project has been measuring the Atlantic Overturning circulation (AMOC) at 26.5 N in the North Atlantic since 2004. We compare ocean heat transport (OHT) at 26.5 N derived from the NEMO-LIM2 ocean circulation/sea ice model with the RAPID measurements. NEMO is run at 1/12 degree horizontal resolution, forced by the Drakkar Surface Forcing dataset which supplies surface air temperature, winds, humidity, surface radiative heat fluxes and precipitation. The model simulates the period 1978-2010.

The basin wide model mean OHT underestimates the RAPID observations (0.76±0.35PW vs 1.26±0.37 PW), but shows very similar seasonal to interannual variability, including the major reduction of the AMOC observed in 2009-10. The lower value from the model is mainly due to underestimates of the northwards Florida Current (0.18 PW) and overestimates of the southwards mid ocean current between the Bahamas and Africa (0.33 PW). The model accurately reproduces the observed correlation between the OHT and the AMOC.

On seasonal to interannual timescales the model simulates large extremes in OHT, varying from a peak of 2.6PW in 1985 to near zero or even negative values (e.g. 1998). We perform composite analysis to determine the oceanic and atmospheric conditions necessary to achieve these extreme values of OHT and consider the potential consequences for regional weather and climate on these timescales.


Reconciling two alternative mechanisms behind bi-decadal AMOC variability
Pablo Ortega1,2*, Juliette Mignot2,3, Didier Swingedouw4, Florian Sévellec1,2, Eric Guilyardi1,2

* Presenting author

1) LOCEAN Laboratory-IPSL, Université Pierre et Marie Curie, France
2) NCAS Climate/Department of Meteorology, University of Reading, United Kingdom
3) Climate and Environmental Physics and Oeschger Centre for Climate Change Research, University of Bern, Switzerland
4) EPOC, Université de Bordeaux, France

Understanding the preferential timescales of variability in the North Atlantic, usually associated with the Atlantic meridional overturning circulation (AMOC), is essential for the prospects of decadal prediction. However, the wide variety of mechanisms proposed from the analysis of climate simulations, potentially dependent on the models themselves, has stimulated the debate of which processes take place in reality. One mechanism receiving increasing attention, identified both in idealised models and observations, is a westward propagation of subsurface buoyancy anomalies, that impact the AMOC through a basin-scale intensification of the zonal density gradient, enhancing the northward transport via thermal wind balance. In this study, we revisit a control simulation from IPSL-CM5A-LR, characterised by a strong AMOC periodicity at 20 years, previously explained by an upper ocean-atmosphere-sea ice coupled mode with a direct effect on convection activity South of Iceland. Our study shows that this mechanism interacts constructively with the basin-scale mode in the subsurface. This constructive feedback may explain why bi-decadal variability is so intense in this coupled model as compared to others.


Understanding boundary density variability at 26N and its relation to the geostrophic transports in the North Atlantic
Irene Polo1*, Keith Haines1, Jon Robson1, Rowan Sutton1, Magdalen Balmaseda2, Chris Roberts3

* Presenting author

1) University of Reading, Department of Meteorology, UK
2) European Centre for Medium-range Weather Forecasts, UK
3) Met Office Hadley Centre, UK

We have investigated the density variability at 26N because the AMOC has been monitored since 2004 in the framework of RAPID. Ocean-only NEMO at both, 1 degree and 0.25 degree horizontal resolutions has been used for the period 1958-2010. 1degree resolution uses the model version and forcing fluxes are those used for the ECMWF ORAS4 while 0.25 degree resolution uses the model version from Met office and CORE2 forcing. For understanding important drivers of the zonal density gradient, the NEMO experiments are defined in such a way that only one of either the wind stress or the buoyancy forcing has inter-annual variability, while the other has a seasonally varying climatology. We then seek to distinguish the main forcing-related signals which are imprinted in the density vertical profiles at the eastern and western boundaries at 26N. We present the results of the variability modes from this set of experiments and the relationship between boundaries. Buoyancy-forced density variability has vertical structure and time-scales substantially different from wind-forced variability. On sub-annual to inter-annual (<3 years) time- scales, variations in the density gradient are driven largely by local wind-forced coastal upwelling at both the western and eastern boundaries. These density variations are felt in the upper 1500m at the western and eastern boundaries simultaneously. Inter-annual variations (3-7 years) related to winds occur due to excitation of Rossby waves in the central Atlantic, which propagate westward to interact with the western boundary. On decadal time scales (7-13 yrs), buoyancy forcing related to the North Atlantic Oscillation dominates variability in the AMOC. This decadal density variability is related to freshwater changes over the Labrador Sea. Anomalies propagate along the coast and are felt at both boundaries up to 3500m, with the eastern boundary lagging from 7 months to 3 years. Two speeds of propagation are found at different levels: one fast response at ~1300m and a slow response at ~3000m. Similar propagating features are found in the control experiment when truncation at 800m is applied to the calculation of density variability modes at western boundary. The fact that the boundaries are linked when the buoyancy is the main forcing could be a useful tool to assimilate low frequency AMOC signals. We plan to use these modes of variability to condition a data assimilation experiment. Additional results using observations from the RAPID-array at 26N for the period 2004-2014 are also presented. Density variability at the western boundary show two main modes at inter-annual time-scales associated with different wind forcing patterns: The leading mode has anomalous density around 1000m and is related to the geostrophic transport, while the second mode shows deeper density anomalies and is related to Eastern boundary Ekman pumping. Results from observations are discussed with caution considering the short-period of study.


Time series of AMOC components in the subpolar North Atlantic
Monika Rhein1*, Christian Mertens1, Tilia Breckenfelder1, Achim Roessler1, Dagmar Kieke1, Reiner Steinfeldt1, Birgit Klein2, Claus Böning3, Arne Biastoch3

* Presenting author

1) MARUM, Bremen University, Germany
2) BSH, Hamburg, Germany
3) GEOMAR, KIel, Germany

The subpolar North Atlantic is characterized by an inflow of the warm and saline North Atlantic Current (NAC), its transformation in colder and denser water, and a southward export of cold deep water. Thus it is a key area for the Atlantic Meridional Overturning Circulation. By combining results from moored instruments at 47°N and along the Midatlantic Ridge with Argo data, satellite altimetry and the output of a high resolution ocean model, long-term transport time series are constructed to quantify the circulation and to study changing transports and pathways for main AMOC components under different NAO states.


Predictability of regional sea level and ocean circulation in a decadal forecasting system
Chris D. Roberts1*, Nick Dunstone1, Leon Hermanson1, Matt Palmer1, Doug Smith1

* Presenting author

1) Met Office, Hadley Centre, UK

A number of studies have emphasized the strong links between the variability of regional sea level and large scale modes of ocean circulation, particularly in the North Atlantic. Recent work has demonstrated that some aspects of regional sea level are predictable on seasonal time scales using a global climate model initialized with an appropriate ocean state (Miles et al. 2014). However, the predictability of regional sea level on multi-annual time-scales has yet to be evaluated. Here, we present an initial assessment of the predictability in time-mean dynamic sea level using five-year hindcasts from the latest configuration of the UK Met Office Decadal Prediction System (DePreSys3), a coupled ocean-atmosphere-sea ice model with an eddy-permitting ocean resolution. Hindcasts are initialized from a model assimilation of a full-depth ocean analysis and skill is evaluated against satellite altimetry and tide-gauge reconstructions of sea surface height.

Miles et al. "Seasonal prediction of global sea level anomalies using an ocean–atmosphere dynamical model." Climate Dynamics 43.7-8 (2014): 2131-2145.


Changes in poleward transport between Greenland and Scotland from 1999-2002 to the present time.
Thomas Rossby1*, Corinna Schrum2, Henrik Søiland3

* Presenting author

1) University of Rhode Island, Graduate School of Oceanography, USA
2) University of Bergen, Geophysics Institute, Norway
3) Marine Research Institute, Bergen, Norway

In the fall of 2012 we resumed a program to measure currents across the northeast Atlantic from the Royal Arctic Line container vessel Nuka Arctica; now equipped with a 75 kHz hull-mounted acoustic Doppler current profiler (ADCP) that can profile to about 800 m, about twice as deep as the earlier 150 kHz ADCP. The Nuka Arctica operates between Nuuk, Greenland and Aalborg, Denmark. From her repeat transits across the Irminger Sea and Iceland Basin we can determine quite accurately currents and transports in selected areas. One of the striking features to emerge from 1999-2002 program was the fundamental role of the Reykjanes Ridge in organizing the poleward flow of warm, salty North Atlantic water through these seas with near-equipartion of transport west and east of the ridge. Preliminary results from the restarted operations through last fall indicate a decrease in poleward flow on the western side of the Reykjanes Ridge from about 8-9 Sv to 6-7 Sv in the top 400 m. Most of the decrease is concentrated to the topographically constrained flow along the western ridge flank whereas the circulation in the interior of the Irminger Sea is relatively unchanged. This decrease in poleward transport in the Irminger Sea may be consistent the weakening subpolar gyre that has been noted by many in recent years.

The poleward flow through the Iceland Basin appears to have increased a comparable amount suggesting a shift in transport away from the Irminger/Labrador Sea branch towards the Iceland Basin branch. The estimation uncertainties here are larger due to the longer integration distances, but we expect these can be reduced as the database increases. This will be important because we also monitor the inflow into the Nordic Seas with a similar vessel-mounted ADCP on the Norröna, a ferry that operates between Iceland, the Faroes and Denmark. Any change over time in flux divergence between these two lines must balanced by a change either in densification rates in the Iceland Basin (sinking) or in cross-ridge flow into the Irminger Sea (see talk by Childers et al). Both the Nuka Arctica and Norröna ADCP programs are ongoing with a view towards developing a more quantitative handle on interannual variability and longer-term trends in transport through the region.


The role of the MOC in forming gyre-scale heat content anomalies
Vassil M. Roussenov1*, Richard G. Williams1, M. Susan Lozier2, Doug Smith3

* Presenting author

1) University of Liverpool, School of Environmental Sciences, UK
2) Duke University, Nicholas School of the Environment, USA
3) Met Office, Hadley Centre, UK

North Atlantic climate variability on decadal time scales is often characterised by basin-scale changes in sea surface temperature, which are generally attributed to coherent changes in the meridional overturning circulation (MOC). However, this view is inconsistent with striking gyre-scale contrasts in ocean heat content over the basin: in the periods of warmer subtropics the subpolar gyre is cool, and vice versa. We explore how the gyre-scale changes in heat content are mechanistically controlled using dynamical assimilations of historical temperature and salinity data over the last 60 years. The heat content anomalies are usually associated with thermocline and overturning anomalies, a warmer heat content associated with a deeper thermocline and often with a weaker MOC anomaly. The tendency of the subtropical heat content is primarily controlled by the heat transport convergence, dominated by the Ekman component of the MOC. Stronger Trade winds enhance the influx of heat from the tropics, augmented by the air-sea heat fluxes, leading to a deeper thermocline. The tendency of the subpolar heat content does not though directly link to the air-sea fluxes, but instead is controlled by the convergence in heat transport, dominated by the geostrophic component. Density increases in the Labrador Sea induced by atmospheric forcing are followed by a density increase along the western boundary leading to enhanced overturning at the subtropical/subpolar interface, which then drive a warming of the subpolar gyre. Hence, the heat content anomalies for each gyre are formed via different mechanisms involving the MOC, either directly via the winds and air-sea heat fluxes over the subtropical gyre or indirectly via atmospheric-induced changes in boundary density over the subpolar gyre.


Subpolar Atlantic convection and heat budget in observations and models
Oleg Saenko1*, Igor Yashayaev2, Paul Myers3, Gregory Smith4

* Presenting author

1) Environment Canada, Canadian Centre for Climate Modelling and Analysis, Canada
2) Fisheries and Oceans Canada, Bedford Institute of Oceanography, Canada
3) University of Alberta, Department of Earth and Atmospheric Sciences, Canada
4) Environment Canada, Meteorological Research Division, Canada

Deep convection in the subpolar Atlantic Ocean, such as in the Labrador Sea and Greenland Sea, is an important component of the Atlantic meridional overturning circulation. It is preconditioned by weak stratification, cyclonic circulation and surface buoyancy loss. The loss of heat to the atmosphere is thought to be resupplied by narrow boundary jets, although the details of this process are not well understood. Limited observations and high-resolution model simulations indicate that eddies play a key role in the subpolar Atlantic heat budget, by removing heat from the boundary regions and supplying it to the areas of deep convective mixing – a process not represented in low resolution ocean-climate models. Furthermore, in some places, such as off the west coast of Greenland, the boundary jets themselves appear to be partly maintained by the eddies, through inverse barotropic instability. Sensitivity experiments, based on a high-resolution global model, show that under warmer surface climate conditions the cyclonic circulation, heat transport convergence and convection all weaken in the Labrador Sea, whereas they intensify in the Greenland Sea.


An observations and model based analysis of meridional transports in the South Atlantic
Claudia Schmid1*, Sudip Majumder2,1, George R. Halliwell1

* Presenting author

1) NOAA/AOML, Physical Oceanography Division, USA
2) CIMAS/University Miami, USA

An analysis of transports of the Atlantic Meridional Overturning Circulation (AMOC) in the South Atlantic Ocean based on three-dimensional velocity fields derived from Argo data and AVISO sea surface heights collected in the years 2000-2014 reveals its spatial and temporal variability. Because these velocity fields end at 2000m the deeper layers need to be padded with climatology. The uncertainty of the results is quantified by sub-sampling the output fields from several models with data assimilation at the resolution of the observation-based fields. Models used for this purpose include the high-resolution HYCOM (HYbrid Coordinate Ocean Mode) and NCODA (Navy Coupled Ocean Data Assimilation) global reanalysis model, the Simple Ocean Data Assimilation (SODA) model and the NCEP (National Center for Environmental Prediction) GODAS (Global Ocean Data Assimilation System) model. Sensitivity to the wind field used in the analysis will be assessed using different wind fields. The relative importance of the eastern boundary, western boundary and interior transports to total variability will be studied.


Move transport correlations with Nordic Seas overflows in models and observations in response to stochastic subpolar wind stress curl
Uwe Send1, Matthias Langhorst1, Nuno Serra2, Rolf Käse2*

* Presenting author

2) University of Hamburg, CEN, Germany

According to Köller et al. (2010) we perform spectral analysis in a multi-decadal UVicESCM imulation. while in the original simulation the focus has been laid on the response of seaice melt to the breakdown of the subpolar gyre, we here look at the processes that link variations of the MOVE transport to those in the Nordic Seas overflow.

For pink noise wind stress forcing similar to that of NAO-variations, there exist three prefered frequency bands in the MOVE transport, with periods near 10, 20 and 100 years. Since observations do not allow resolution of these periods except for the first, we take data from a high-resolution North Atlantic model driven by the NCEP atmospheric reanalysis and compare the spectra and correlations in the 20y-band. Then the actual observational data are compared to the realistic model output over the time of observations.


Insights into the Mesoscale Circulation of the Rockall Trough
Toby Sherwin1*, Dima Aleynik1, Mark Inall1

* Presenting author

1) Scottish Association for Marine Science, Physics and Technology, UK

The Rockall Trough is the nearest region of truly deep oceanic water west of the British Isles, and it is one of the main conduits for warm salty water that is transported to high latitudes as well as the source of oceanic water feeding onto the shallow European Shelf. A full picture of decadal variations of the circulation in the AMOC, and of the Sub-Polar Gyre in particular, requires a proper understanding of the mesoscale variability of this eastern boundary region.
Annual scientific cruises have monitored the temperature and salinity of the Trough since 1975. Since 1995 satellite altimeters have measured small fluctuations in level of the sea surface to determine variations in the European slope current that is an important northward transport path on the eastern side of the Trough, and to quantify the intensity of large horizontal eddies that mix its ocean waters. In recent years remotely operated underwater gliders have begun to supplement these observations with regular crossings of the Trough to report detailed in situ measurements of currents, temperature and salinity from the surface to 1000 m.
This talk describes the results of the first such glider mission which involved eight crossings over the winter of 2009/2010. Its principal findings are that:
i. Much of the surface and deeper meandering current field in the central Rockall Trough is driven by deep eddies that have migrated into the Trough from both its northern and southern entrances.
ii. Surface currents appear to be much stronger during the autumnal period of seasonal surface stratification than in late winter when the upper Trough is mixed to a depth of 600 m.
iii. In late 2009, during a period of unusually large eddy activity, a chance arrangement of some deep circulations caused a westward deflection of the slope current that resulted in a large quantity of slope water being carried to, and thereby warming, the upper 500 m of the western side of the Trough.
iv. Limitations to the altimeter observations the sea surface are identified. By combining them with glider measurements it is shown that they do not pick out the mean flow in narrow slope currents either side of the Rockall Trough.
Novel measurement techniques invariably result in new scientific understanding. The insights derived from this first underwater glider mission in the Rockall Trough confirm that it is a dynamically active region but that much of this activity is driven by deep hidden processes and results in significant fluctuations in the observed mean conditions. There are important implications here for OSNAP in the interpretation of satellite altimeter observations, the understanding of the causes of variability in the results of regular ocean monitoring programmes, and in the establishment and interpretation of the results of ocean modelling exercises.


The fate of baroclinic energy reaching the Atlantic western boundary at 26N
Zoltan Szuts1*, Kim Martini2

* Presenting author

1) University of Washington, Applied Physics Lab, USA
2) University of Washington, Joint Institute for the Study of Atmosphere and Ocean, USA

Baroclinic motion in the form of Rossby waves and eddies propagates westward, but what happens when this energy reaches a western boundary? Theory and idealized numerical models have proposed hypotheses - reflection to short Rossby waves or transformation to boundary waves - but they have not been tested with observations for lack of a sufficiently large dataset. Here, we use the decade-long mooring observations made by the Atlantic Meridional Overturning Array at 26N to infer energy pathways at the western boundary from subsurface moorings. Four moorings measure velocity and density within 2 Rossby radii (100 km) of the boundary, with other moorings further offshore measuring just density. We consider the distribution of kinetic energy, potential energy, and energy flux in terms of frequency, vertical dimension (depth or vertical mode), and distance from the boundary. If reflection of Rossby waves is a linear process in time, then energy at low frequencies should transfer from potential to kinetic upon reflection and the first vertical mode should remain dominant approaching the boundary. This is not seen: instead, low frequency and first mode motion is less energetic close to the boundary. This suggests the dominance of scattering from the boundary in frequency and in the vertical. A different spatial pattern is seen for higher frequencies, however, implying different boundary processes than at low frequencies. Periods shorter than 10 days show less energy loss approaching the boundary in potential and zonal kinetic energy, while meridional kinetic energy is roughly uniform. Southward energy flux at all moorings implies equator-propagating boundary waves, with most variance at periods of 10-100 days. Even a small amount of energy transfer from the large amount of westward propagating energy offshore to boundary motion can be significant for energy budgets. In addition, understanding the vertical distribution of energy loss at all frequencies is important for determining the relative impact on circulation in the upper ocean (subtropical gyre) or the lower ocean (deep overturning return flow). This vertical distribution also depends on tidal and near-inertial internal waves, which have unresolved patterns of seasonality, cross-basin propagation, and boundary interactions that the moored array is unique in addressing.


AMOC variability and “ringing” related to oceanic damped oscillatory modes
Matthew D. Thomas1*, Alexey V. Fedorov1, Florian Sevellec2, Les Muir1

* Presenting author

1) Yale University, Geology and Geophysics, USA
2) University of Southampton, National Oceanography Centre, UK

Climate models suggest that the AMOC can undergo pronounced decadal to multi-decadal oscillations, though the dynamics responsible for this variability are not yet well understood. Recent linear sensitivity analyses of an ocean general circulation model (Sevellec and Fedorov 2013, 2015) have highlighted the potential importance for AMOC variability of an oceanic damped mode of interdecadal period that is related to westward propagating large-scale temperature anomalies. We here investigate this mode by analyzing the output of the CMIP5 coupled climate model archive and by conducting numerical experiments with two climate models. We find that about half of the CMIP5 models exhibit this oscillatory mode (Muir and Fedorov 2015), leading to a pronounced AMOC “ringing” that occurs during specific intervals lasting a few decades to centuries. These results raise the question of what causes AMOC ringing in the models, whether differences in ocean physics or atmosphere-ocean coupling are important, and whether the ocean requires a specific or optimal type of forcing in order to induce this mode. We are now investigating these questions with experiments performed on two of the CMIP5 coupled models, one that exhibited a pronounced AMOC ringing, GFDL CM2.1, and one that did not, CESM 1.2. By initializing the two models with identical temperature and/or salinity perturbations applied to the initial upper ocean density field, we explore whether a ringing AMOC mode can be induced in both models and assess the conditions and dynamics necessary for ringing.


The fate of Mediterranean Outflow Water in the North Atlantic and its mixing with Labrador Sea Water
Erik Van Sebille1*, Robert Marsh2, Gerard McCarthy3, Jesus Peña-Izquierdo4, Josep Pelegrí4

* Presenting author

1) Imperial College London, Grantham Institute & Department of Physics, United Kingdom
2) University of Southampton, United Kingdom
3) National Oceanographic Centre, United Kingdom
4) Institut de Ciències del Mar, Spain

Mediterranean Outflow Water spreads into the North Atlantic Ocean below the thermocline, forming a tongue that slowly moves westward until it reaches the North American coast. Meanwhile, Labrador Sea Water formed at high latitudes in the North Atlantic flows southward along the American coast in the Deep Western Boundary Current, and flows on similar density levels as the Mediterranean Outflow Water.

The two water masses mix at some point, to jointly cross the Equator in the Deep Western Boundary Current. The question is how and where this mixing happens. In particularly, it is unclear how variability in the strength of formation the two water mass relates and impacts on the southward export of mid-depth water. This is important because, while the two water masses have the same density, they have vastly different temperature and salinity properties. Better knowledge of the mixing processes will allow for an improved understanding of AMOC dynamics.

Here, we show results from eddy-resolving models where we track the water flowing from both the southern tip of the Labrador Sea and the Strait of Gibraltar, using virtual Lagrangian particles. We analyse the trajectories of these particles to see where and how the two water masses mix, and reconcile this with information on the modeled and observed salinity field in the North Atlantic. We show what the role of meso-scale eddies is, and how the interaction of the two water masses ultimately sets the thermohaline structure of the mid-depth North Atlantic Ocean.


Abyssal volume and heat transport through the Samoan Passage: A 16-month timeseries of the deep Pacific Overturning Circulation based on recent observations
Gunnar Voet1*, Matthew H. Alford1, James B. Girton2, Glenn S. Carter3, John B. Mickett2, Jody M. Klymak4

* Presenting author

1) UC San Diego, Scripps Institution of Oceanography, United States
2) University of Washington, Applied Physics Laboratory, United States
3) University of Hawaii, United States
4) University of Victoria, Canada

The flow of dense water through the Samoan Passage accounts for the majority of the bottom water renewal in the North Pacific, thereby making it an important element of the Meridional Overturning Circulation. Here we report recent measurements of the flow of dense waters of Antarctic and North Atlantic origin through the Passage. A sixteen month long moored time series of velocity, temperature and salinity of the abyssal flow was recorded between 2012 and 2013. This allows for an update of the only prior volume transport time series from the Samoan Passage from 1992/1993. While highly variable on various time scales, the overall pattern of the abyssal flow through Samoan Passage was remarkably steady. The time-mean northward volume transport of about 5.5 Sv in 2012/2013 stayed within the uncertainty of +/- 1 Sv of the earlier measurements. However, the abyssal current significantly warmed on average by 1 mdeg/year over the past two decades. This implies increased heat transports into the deep North Pacific.


Perspectives on the future of AMOC
I.Neil Wells1*

* Presenting author

1) University of Southampton, Department SOES, U.K.

The MOC has been measured for over 10 years at 26°N at 10 day intervals, and during that time we have learnt that the MOC can change by 30% of its mean value on inter-annual time scales. Understanding of this record is essential in the coming years, using the observations and modelling. Examples of how the observations of heat content, MOC and surface heat heat flux have been used during this 10 year period will be discussed. Furthermore ideas on how coupled atmosphere-ocean models and ocean-only models may be used to suggest and possibly clarify the mechanisms for the observed changes in the MOC.


Interior Wave Pathways for the AMOC variability propagation
Jiayan Yang1*

* Presenting author

1) Woods Hole Oceanographic Institution, Department of Physical Oceanography, USA

Two mechanisms connect AMOC variations over the whole Atlantic basin: slow advection and fast waves. The western boundary has been considered as the predominant pathway for both mechanisms. The concept that the deep western boundary current (DWBC) is the predominant advection pathway, however, has been challenged by evidence that interior advection pathways exist. This study seeks to examine whether interior pathways also exist for topographic Rossby waves and whether they play any significant role in communicating forcing between the subpolar and subtropic basins. Our results using a two-layer model and ECCO4 simulation strongly support the existence of such pathways and show that the interior wave pathways play an important role in basin-wide AMOC adjustments to external forcing.


Changes in AMOC variability under increased CO2
Laure Zanna1*, Douglas MacMartin2, Eli Tziperman3

* Presenting author

1) University of Oxford, Atmospheric Physics, UK
2) California Institute of Technology, Computing and Mathematical Sciences, US
3) Harvard University, Earth and Planetary Sciences, US

It is shown that under a 4xCO2 scenario in the GFDL ESM2M general circulation model (GCM), the mean Atlantic Meridional Overturning Circulation (AMOC) weakens, and the multi-decadal variability is significantly altered compared to the preindustrial control simulation. In the preindustrial simulation, the model exhibits a peak in the power spectrum of both AMOC and northwards heat transport between 26 and 45°N at around 20 years, resulting from thermohaline variability. In the 4xCO2 simulation, the variability is significantly reduced at 26 and 45°N and the only significant spectral peak is found at high-latitudes near 60°N.

Understanding the differences between the control and greenhouse scenario in AMOC variability at different latitudes is crucial for our ability to predict future climate change in the North Atlantic, and in order to further elucidate the physics underlying AMOC variability. In particular, it has been hypothesized based on idealized models that increased CO2 would shift the AMOC closer to a bifurcation point, which would increase the variability, rather than decrease it as found in the present GCM.

The variability is characterized in each simulation by creating composite maps constructed from years of AMOC maxima and minima. Transfer function analysis demonstrates that the shift in variability characteristics is predominantly due to a shift in the internal ocean dynamics rather than a change in the atmospheric (stochastic) forcing. In the preindustrial simulation, the changes in vertical velocity associated with the AMOC oscillation are spatially localized east of the Grand Banks, and are shown to be cross-isopycnal. However, in the 4xCO2 simulation, the same region is characterized by a stronger stratification, no convective activity and does not exhibit variability in the vertical velocity. We thus suggest that an important factor in the reduction in mid-latitude AMOC variability in the model is the increased stratification east of the Grand Banks, associated with the shallower and weaker mean overturning in the high-CO2 simulation. The high-latitude variability emerging in the 4xCO2 simulation is related to the advection of anomalies by the subpolar gyre, and has a small impact on mid-latitude heat transport.


A simple model of the response of the Atlantic to the North Atlantic Oscillation
Xiaoming Zhai1*, Helen L. Johnson2, David P. Marshall3

* Presenting author

1) University of East Anglia, School of Environmental Sciences, UK
2) University of Oxford, Department of Earth Sciences, UK
3) University of Oxford, Department of Physics, UK

The response of an idealised Atlantic ocean to wind and thermohaline forcing associated with the North Atlantic Oscillation (NAO) is investigated both analytically and numerically in the framework of a reduced-gravity model. The NAO-related wind forcing is found to drive a time-dependent "leaky" gyre circulation that integrates basin-wide stochastic wind Ekman pumping and initiates low-frequency variability along the western boundary. This is subsequently communicated, together with the stochastic variability induced by thermohaline forcing at high latitudes, to the remainder of the Atlantic via boundary and Rossby waves. At low frequencies, the basin-wide ocean heat content changes owing to NAO wind forcing and thermohaline forcing are found to oppose each other. The model further suggests that the recently reported opposing changes of the meridional overturning circulation in the Atlantic subtropical and subpolar gyres between 1950-1970 and 1980-2000 may be a generic feature caused by an interplay between the NAO wind and thermohaline forcing.


Breaking the linkage between the Labrador Sea Water production and its export to the subtropical gyre
Sijia Zou1*, Susan Lozier1

* Presenting author

1) Duke Unviersity, Earth and Ocean Sciences, United States

The linkage between the Labrador Sea Water (LSW) production and its export into the subtropical gyre primarily derives from Eulerian-based studies that show LSW property anomaly in the Labrador Basin leading the property anomaly at 26°N by about 10 years. However, few studies have shown a direct relationship between the LSW productivity and its export variability to the subtropical basin. In the present study, we launch Lagrangian floats in the LSW with an eddy-permitting ocean circulation model (ORCA025) per year (1961-2004) across AR7W and track them for 40 years. No significant correlation is found between the number of floats launched in the Labrador Basin (LSW production) and the number of floats exported to the subtropical gyre (LSW export). Moreover, the simulated floats take an unexpectedly long time to be exported: 22 ± 10 years to the subtropical gyre and 30 ±8 years to 30°N. This is in contrast to the 10 years’ transit time of LSW properties in the Eulerian frame, indicating different mechanisms that control the LSW volume export and property propagation.



Theme 2: Impacts of the AMOC on the atmosphere, cryosphere and land

All sessions take place Thursday 23rd July.   Invited talks (2)   Oral presentations (6)   Posters (7)  

Theme 2 invited talks:
Atmospheric influence of the Atlantic Meridional Overturning Circulation
Guillaume Gastineau1*

* Presenting author

1) Sorbonne Universités, UPMC/CNRS/IRD/MNHN, LOCEAN/IPSL

The North Atlantic Oscillation (NAO) dominates the European and North American climate variability during the cold season, for time scales ranging from 10 days to several decades. At decadal time scale, the SST also has an influence, as a warming in the subpolar Atlantic Ocean leads to a negative NAO in winter. Such SST influence might involve the ocean dynamics, as climate models show that the Altantic meridional overturning circulation (AMOC) is followed by a supolar Atlantic basin warming accompanied by a similar atmospheric response. The atmospheric changes seem to be driven by the diabatic heat flux in the main eddy development region, over the Gulf Stream/North Atlantic current region. But the stratosphere or the tropical Atlantic SST forcing may also play an important role.
To further establish the causality links between the ocean and the atmosphere, ensembles of atmosphere-only simulations are designed. The atmospheric model simulations use prescribed SST and sea-ice anomalies that follow an intensification of the AMOC in the coupled model IPSL-CM5A-LR. We confirm that the main influence is due to warm subpolar Atlantic SST anomalies north of 30°N and the associated upward heat flux which are responsible for a decrease of the lower-tropospheric baroclinicity in the region of maximum eddy growth. But we also found that the positive sea-ice anomalies over the Arctic associated with a larger AMOC further amplify the atmospheric circulation anomalies, as they act to warm the stratosphere one month before the tropospheric circulation anomalies.


Impact of AMOC on the Low Frequency Variability of Summer Arctic Sea Ice Extent
Rong Zhang1*

* Presenting author


Satellite observations reveal a substantial decline in September Arctic sea ice extent (SIE) since 1979, which has played a leading role in the observed recent Arctic surface warming and has often been attributed, in large part, to the increase in greenhouse gases. The observed decline trend and future projections of ice-free summer by some climate models forced with increasing anthropogenic greenhouse gases bring up the potential for trans-Arctic shipping in the near future. However, the detail mechanisms causing the low frequency variability of summer Arctic SIE are still unclear. The most rapid decline in summer Arctic sea ice actually occurred during the recent global warming hiatus period. The CMIP5 multi-model mean response to changes in anthropogenic radiative forcings exhibits much less decline in September Arctic SIE but stronger warming in global mean surface temperature than that observed over the recent hiatus period, implying that natural variability might have played an important role in the observed recent decline in September Arctic SIE.

In this study, it is shown that AMOC and the associated Atlantic heat transport into the Arctic have played a significant role in the low frequency variability of summer Arctic SIE using the GFDL coupled climate model. At low frequency the March Barents Sea SIE anomaly is dominated by the anti-correlated anomaly in the Atlantic heat transport into the Arctic, thus is also significantly correlated with September Arctic SIE anomaly. The observed March Barents Sea SIE has a very similar normalized decline trends as the observed September Arctic SIE since 1979, consistent with an increasing trend in the Atlantic heat transport into the Arctic. The estimated increase in the Atlantic heat transport into the Arctic since 1979 is consistent with the strengthening of AMOC over this period as implied by AMOC fingerprints, and could have contributed substantially to the observed summer Arctic SIE decline. If the AMOC and the associated Atlantic heat transport into the Arctic were to weaken in the near future due to internal variability, there might be a hiatus in the decline of September Arctic SIE, and a delay in attaining a summer ice-free Arctic. This plausible scenario with enormous social and economical impacts cannot be ignored.

This study also shows that at multidecadal/centennial time scales, changes in poleward atmosphere heat transport across the entire Arctic Circle are compensating to and dominated by AMOC induced Atlantic heat transport anomalies into the Arctic, i.e. a stronger AMOC and associated enhanced Atlantic heat transport into the Arctic ocean leads to both reduced summer Arctic SIE and reduced poleward atmosphere heat transport into the Arctic. Hence variations in the atmosphere heat transport across the Arctic Circle provide a negative feedback to September Arctic SIE variations at multidecadal/centennial time scales.


Theme 2 oral presentations:
Historical analogues to the recently observed minima in the Atlantic meridional overturning circulation
Adam Blaker1*, Joel Hirschi1, Gerard McCarthy1, Bablu Sinha1, Sarah Taws2, Robert Marsh2, Andrew Coward1, Beverly de Cuevas1

* Presenting author

1) NOC, Marine Systems Modelling, UK
2) University of Southampton, Ocean Earth Sciences, UK

Observations of the Atlantic meridional overturning circulation (AMOC) by the RAPID 26°N array show a pronounced minimum in the northward transport over the winter of 2009/10, substantially lower than any observed since the initial deployment in April 2004. It was followed by a second minimum in the winter of 2010/2011. We demonstrate that ocean models forced with observed surface fluxes reproduce the observed minima. Examining output from five ocean model simulations we identify several historical events which exhibit similar characteristics to those observed in the winter of 2009/10, including instances of individual events, and two clear examples of pairs of events which happened in consecutive years, one in 1969/70 and another in 1978/79. In all cases the absolute minimum, associated with a short, sharp reduction in the Ekman component, occurs in winter. AMOC anomalies are coherent between the Equator and 50°N and in some cases propagation attributable to the poleward movement of the anomaly in the wind field is observed. We also observe a low frequency (decadal) mode of variability in the anomalies, associated with the North Atlantic Oscillation (NAO). Where pairs of events have occurred in consecutive years we find that atmospheric conditions during the first winter correspond to a strongly negative Arctic Oscillation (AO) index. Atmospheric conditions during the second winter are indicative of a more regional negative NAO phase, and we suggest that this persistence is linked to re-emergence of sea surface temperature anomalies in the North Atlantic for the events of 1969/70 and 2009/10. The events of 1978/79 do not exhibit re- emergence, indicating that the atmospheric memory for this pair of events originates elsewhere. Observation of AO patterns associated with cold winters over northwest Europe may be indicative for the occurrence of a second extreme winter over northwest Europe.


The impact of the North Atlantic Oscillation on 20th century climate through its influence on the Atlantic Meridional Overturning Circulation
Thomas L. Delworth1*, Fanrong Zeng1, Liping Zhang1,2

* Presenting author

1) Geophysical Fluid Dynamics Laboratory/NOAA, USA
2) Princeton University, USA

The impact of the North Atlantic Oscillation (NAO) on the Atlantic Meridional Overturning Circulation (AMOC) and large-scale climate is assessed using simulations with three different climate models. Perturbation experiments are conducted in which patterns of anomalous fluxes corresponding to the NAO are added to the model ocean; in companion experiments no such fluxes are added. Differences between the experiments illustrate how the model ocean and climate system respond to the NAO. A positive phase of the NAO tends to strengthen the AMOC by extracting heat from the subpolar gyre, thereby increasing deepwater formation, horizontal density gradients, and the AMOC.

The flux forcings have the spatial structure of the observed NAO, but the amplitude of the forcing varies in time. The temporal variation of the imposed fluxes is one of the following types: (a) sudden switch on of the flux forcing, (b) vary the amplitude of the flux forcing sinusoidally in time with distinct periods varying from 2 to 200 years, (c) vary the flux forcing to match the observed time sequence of the NAO over the 20th and early 21st centuries. The first two types of experiments are idealized, while the third attempts an assessment of the impact of NAO-induced AMOC anomalies in the observed record. In the idealized experiments we show that the response of the AMOC to NAO variations is small at short time scales, but increases up to the dominant time scale of internal AMOC variability (20-30 years for the models used). The amplitude of the response of the AMOC, and associated oceanic heat transport, is approximately constant as the timescale of the forcing is increased further. In contrast, the response of other properties, such as hemispheric surface air temperature or Arctic sea ice, continues to increase as the time scale of the forcing becomes progressively longer. The larger response of temperature and sea ice is associated with an increased impact of radiative feedback processes at progressively longer time scales. The impact of the NAO on the AMOC and climate is a function of the dominant timescale of internal AMOC variability, as well as the background mean state. In the experiments using the observed sequence of the NAO we estimate the contribution of NAO-induced AMOC anomalies to hemispheric temperature variations in the 20th and early 21st centuries. We show that NAO-induced AMOC variations contributed substantially to multidecadal warming and cooling of the Northern Hemisphere, including cooling from the 1960s through the 1980s, and warming from the 1980s through the 2000s.


Climate impacts of an AMOC shutdown in a high resolution GCM
Laura C Jackson1*, Ron Kahana1, Tim Graham1, Mark Ringer1, Tim Woollings2, Jennifer Mecking3, Richard Wood1

* Presenting author

1) Met Office, Hadley Centre, UK
2) University of Oxford, Atmospheric Physics, UK
3) University of Southampton, Ocean and Earth Science, UK

We present the impacts from a hypothetical slowdown in the Atlantic Meridional Overturning Circulation (AMOC) in a state-of-the-art global climate model (HadGEM3), with particular emphasis on Europe. This is the highest resolution coupled global climate model to be used to study the impacts of an AMOC slowdown so far. Many results found are consistent with previous studies and can be considered robust impacts from a large reduction or collapse of the AMOC. These include: widespread cooling throughout the North Atlantic and northern hemisphere in general; less precipitation in the northern hemisphere midlatitudes; large changes in precipitation in the tropics and a strengthening of the North Atlantic storm tracks.

The focus on Europe, aided by the increase in resolution, has revealed previously undiscussed impacts, particularly those associated with changing atmospheric circulation patterns. Summer precipitation decreases (increases) in northern (southern) Europe and is associated with a negative summer North Atlantic Oscillation (NAO) signal. Winter precipitation is also affected by the changing atmospheric circulation, with localised increases in precipitation associated with more winter storms and a strengthened winter storm track. Stronger westerly winds in winter increase the warming maritime effect while weaker westerlies in summer decrease the cooling maritime effect. In the absence of these circulation changes the cooling over Europe's landmass would be even larger in both seasons.


Interdecadal modulation of ENSO by the Atlantic Multidecadal Variability
Yohan Ruprich-Robert1*

* Presenting author

1) Princeton University, AOS, USA

In this study, the climate impacts of the AMOC decadal variability have been investigated through its hypothesized Sea Surface Temperature (SST) fingerprint: the so-called Atlantic Multidecadal Variability (or AMV). We performed experiments based on the GFDL CM2.1 model in which SSTs in the North Atlantic sector are restored to the observed AMV pattern, while other basins are left fully coupled. In order to explore and isolate the potential AMV impacts and maximize the signal-to-noise ratio, we use an ensemble simulation of 100 members that are run for 11 years. We find that a positive AMV leads to the set-up of a negative Pacific Decadal Variability pattern that is characterized by cold anomalies in the tropical Pacific and warm anomalies in the western Pacific mid-latitudes. The response of the tropical Pacific ocean to a positive phase of the AMV is similar to that observed during a development of La Niña event. However, the AMV-induced anomalies are not stationary and our results suggest that the AMV can modulate ENSO on interdecadal timescale. We investigate the frequency of occurrence of El Niño and La Niña events and find that the probability of La Niña events increase by about 15% during the first 3 years of the simulation and is followed by an increase of El Niño events for the following 2 years. We propose a physical mechanism to explain this modulation of ENSO by the North Atlantic variability.


Connections between AMOC and Air-Sea Interaction in the Gulf Stream and the North Atlantic Current
LuAnne Thompson1*, Kathryn Kelly2, JulieAnn Koehlinger1

* Presenting author

1) University of Washington, School of Oceanography, United States
2) University of Washington, Applied Physics Laboratory, United States

Using observationally derived quantities, we show that changes heat content in Gulf Stream and North Atlantic Current are controlled by changes in AMOC, and that in turn, the heat content changes controls surface fluxes. In addition, those fluxes directly impact the atmosphere through changes in mid-level cloud cover. These results are consistent with previous analyses that have shown that on interannual times scales and longer in the Gulf Stream region, ocean heat transport convergence is balanced locally by surface heat flux from the ocean to the atmosphere; and that mechanistically, mid-level cloud cover over the Gulf Stream in winter are forced by surface fluxes there. Here we expand the above analysis to the entire western basin using satellite sea surface height as a proxy for upper ocean heat content, turbulent air sea fluxes, and along with estimates of AMOC and meridional heat transport at 26.5N and 41N. We show that sea level in the Gulf Stream precedes surface heat flux with a warmer Gulf Stream leading to heat flux out of the ocean; this relationship is also found in the North Atlantic Current. In addition, changes in AMOC at precede changes in heat content at those locations. We build on previous studies showing the linkages between seasonal changes in cloud cover over the Gulf Steram to investigate the year-to-year changes in cloud cover. Using three different cloud cover products, we find that turbulent heat fluxes are positively correlated with mid-level cloud fraction and increased sea level in the Gulf Stream in summer predict a larger amount of mid-level clouds. In general, on interannual time scales, sea surface temperature shows much weaker relationships with cloud cover, AMOC, and surface fluxes than does sea level.

The atmosphere controls air-sea interaction in the Florida Current as it travels along the Eastern seaboard as well as the transition region between the Labrador Current and the North Atlantic Current. In addition, the North Atlantic Oscillation is shown to be strongly correlated with turbulent fluxes in the Florida Current and precedes changes sea level there. As found in previous studies, changes in the North Atlantic Oscillation also precede changes in the path of the Gulf Stream.

Taken together, these observationally derived relationships suggest increases in the AMOC increase the heat content near the Gulf Stream and the North Atlantic current. Much of this increased heat content is fluxed to the atmosphere, increasing cloud cover. The Gulf Stream is the locus of these interactions on interannual times scales, with a second center of action in the North Atlantic Current.


A Holocene and last 1000 yr perspective on variability in the deep currents of the AMOC: An exceptional twentieth century slowdown of the AMOC?
David J.R. Thornalley1,2*, Delia Oppo2, Paola Moffa-Sanchez3,4, Ian R. Hall4, Lloyd Keigwin2

* Presenting author

1) University College London, Department of Geography, UK
2) Woods Hole Oceanographic Institution, Department of Geology and Geophysics, USA
3) Rutgers, Department of Marine and Coastal Sciences, USA
4) Cardiff University, School of Earth and Ocean Sciences, UK

Several proxy and modelling studies suggest that there may have been considerable change in the operation the AMOC during the present interglacial. Yet despite its importance for regional and global climate, the history of the AMOC is poorly constrained. Improving our knowledge of past longer term AMOC variability will contribute to our general understanding of the dynamics of ocean circulation and the role it may play in causing or amplifying climate variability on multi-decadal to millennial timescales. It also enables a longer term perspective on the current state of, and recent trends in the AMOC.

We present Holocene grain-size records in depth transects from Blake Outer Ridge and Cape Hatteras, sampling the full-depth range of the Deep Western Boundary Current (DWBC), part of the lower limb of the AMOC. These records complement a depth-transect of grain-size records sampling the Iceland-Scotland (I-S) overflow, showing Holocene variations that reflect deglacial meltwater forcing in the early Holocene and insolation-forced trends from the middle-to-late Holocene (Thornalley et al., 2013, Climate of the Past). We will also present detailed grain-size records for the last 1,000 years, both in a depth transect of cores off Cape Hatteras, and from cores in the Iceland Basin, sampling the I-S overflow. Our extensive datasets enable us to provide a coherent synthesis of changes in the flow strength of key components of the AMOC on multi-decadal to millennial timescales.

Specific questions to be addressed include: How well coupled are Holocene trends in Iceland-Scotland overflow and the DWBC? How did I-S overflow and the AMOC vary during the last millennium, including the last ~150 years since the end of the Little Ice Age? Initial results suggest a long-term anti-phasing of the Nordic overflows, wherein mid-late Holocene weakening of the I-S overflow has been compensated for by a strengthening of Denmark Strait overflow. We will also report on pronounced centennial-millennial scale reductions in the inferred flow strength at sites bathed by Labrador Sea Water (LSW). Emerging results for the last millennium also indicate an exceptional slowdown in the flow of ISOW and LSW during the twentieth century, providing paleoceanographic support for the findings of Rahmstorf et al (2015, Nature Climate Change)


Theme 2 poster presentations:
North Atlantic Cooling: Anatomy, Causes and Impacts
Dan Hodson1*, Jon Robson1, Rowan Sutton1

* Presenting author

1) University of Reading, NCAS-Climate, Department of Meteorology

North Atlantic Sea Surface Temperatures underwent a rapid cooling during the 1960s, a period when the SSTs outside the North Atlantic showed a long-term warming trend. Several hypotheses for the origin of this observed cooling exist - changes in anthropogenic aerosol forcing or changes in heat convergence due to changes in the Atlantic Meridional Overturning Circulation, being leading contenders. It is clear that the Atlantic did not cool uniformly over this period. We therefore ask - how did the cooling evolve, and does this reveal anything about the likely causes?

Through analysis of multiple observational datasets, it is demonstrated that the cooling proceeded in several distinct stages. Cool anomalies initially appeared in the mid-1960s in the Nordic Seas and Gulf Stream extension, before spreading to cover most of the subpolar gyre. Subsequently, cool anomalies spread into the tropical North Atlantic before retreating, in the late 1970s, back to the subpolar gyre. There is strong evidence that changes in atmospheric circulation, linked to a southward shift of the Atlantic ITCZ, played an important role in the event, particularly in the period 1972–76. Theories for the cooling event must account for its distinctive space–time evolution. The authors’ analysis suggests that the most likely drivers were 1) the ‘‘Great Salinity Anomaly’’ of the late 1960s; 2) an earlier warming of the subpolar North Atlantic, which may have led to a slowdown in the Atlantic meridional overturning circulation; and 3) an increase in anthropogenic sulphur dioxide emissions.

The climate impacts of such a cooling event cannot be determined from observations alone. Atmosphere models driven by fixed SSTs provide one route to examining the impacts, however such experiments cannot capture any potential atmosphere-ocean coupled feedbacks that could amplify and persist the atmospheric response. Such a coupled response can be examined using an atmosphere model coupled to a mixed layer ocean model. Here we present results from experiments using such a model to explore the regional and global climate impacts of North Atlantic cooling. These experiments have relevance to the current climate system, in view of the evidence of a possible slowdown in the AMOC happening now.


Attribution of multidecadal climate trends in observations and models
Sergey Kravtsov1*, Marcia Wyatt2, Judith Curry3, Anastasios Tsonis1

* Presenting author

1) University of Wisconsin-Milwaukee, Mathematical Sciences, USA
2) University of Colorado - Boulder, Geosciences, USA
3) Georgia Tech., Earth and Atmospheric Sciences, USA

We analyzed long-term trends and multidecadal variability in a network of well-known climate indices based on the sea-surface temperature and sea-level pressure fields, in observations and CMIP5 model simulations. Various ensemble-mean estimates of the forced variability were derived from 18 independent ensembles of these simulations, each using a single model with fixed physics package and an identical forcing history. It was shown that the residual intrinsic variability time series over the total of 116 simulations considered are statistically independent. The 18 estimates of the forced signal were further used for semi-empirical attribution of the observed climate variability. The model uncertainty results in a fairly broad range of the “intrinsic” multidecadal variability so estimated. Furthermore, the observed surface temperature “intrinsic” residuals exhibit the levels of multidecadal variance overwhelmingly exceeding those of the simulated intrinsic multidecadal variability. This may reflect the climate models’ consistently underestimating the true amplitude of the forced multidecadal variability, and/or be indicative of the lack of multidecadal intrinsic dynamics in these models.


Sea-level fluctuations show Ocean Circulation impact on Atlantic Multidecadal Variability
Gerard D. McCarthy1*, Ivan D. Haigh2, Joel J.-M. Hirschi1, Jeremy P Grist1, David A. Smeed1

* Presenting author

1) National Oceanography Centre, Southampton, UK
2) University of Southampton, National Oceanography Centre, UK

We present observational evidence that ocean circulation controls the decadal evolution of heat content and consequently sea-surface temperatures (SST) in the North Atlantic. Positive (negative) phases of the Atlantic multidecadal oscillation (AMO) are associated with warmer (cooler) SSTs. Positive phases of the AMO have been linked with decadal climate fluctuations including increased summer precipitation in Europe; increased northern hemisphere land temperatures, fewer droughts in the Sahel region of Africa and increased Atlantic hurricane activity. It is widely believed that the Atlantic circulation controls the phases of the AMO by controlling the decadal changes in heat content in the North Atlantic. However, due to the lack of ocean circulation observations, this link has not been previously proven. 
We present a new interpretation of the sea-level gradient along to the east coast of the United States to derive a measure of ocean circulation spanning decadal timescales. We use this to estimate heat content changes that we validate against direct estimates of heat content. We use the longevity of the tide gauge record to show that circulation, as interpreted in sea-level gradient changes, drives the major transitions in the AMO since the 1920’s. 
We show that the North Atlantic Oscillation is highly correlated with this sea-level gradient, indicating that the atmosphere drives the circulation changes. The circulation changes are essentially integrated by the ocean in the form of ocean heat content and returned to the atmosphere as the AMO. 
An additional consequence of our interpretation is that recently reported accelerations in sea-level rise along the US east coast are consistent with a declining AMO that has been predicted by a number of authors.


Statement of the problem on the mutual adjustment of the atmosphere and ocean in 'fast' and 'slow' time
Dmitry M. Sonechkin1*

* Presenting author

1) P.P. Shirshov Oceanology Institute, Russian Academy of Sciences, Russia

It is generally accepted to believe that atmosphere drives the ocean dynamics in the fast timescales of days and decades, and vice versa, the ocean significantly affects long-term (seasonal) and super long-term (interannual and longer) variations of the atmosphere. The evident reason of this believing consists of an enormous difference in the heat capacity of these two subsystems of the global climate system. However, according to my knowledge, nobody gave a rigorous mathematical justification of this believing. The aim of my report is to pose the problem of the atmosphere and ocean mutual adjustment with using the ratio of the atmosphere and ocean heat capacities as “the small parameter”. It turns out that the adjustment of the ocean to the atmospheric influences in the “fast” timescales is a certain “regular” problem. It means that the respective oceanic responses to the atmospheric influences are stable, and so predictable in principle. Instead, the adjustment of the atmosphere to the oceanic influences in the “slow” timescales is a kind of the so-called “singular” problem. The zero-order guess of this problem solution (when the value of the above “small parameter” is assumed to equal to zero) consists of a steady state of the atmospheric subsystem (pseudo) randomly chosen from a number of such steady states existing. According to the well-known results for the energy-balance climate models, it is almost for certain that the steady state being chosen turns out to be unstable to small disturbances of the initial atmospheric state. Therefore, this zero-order guess can not be observed in reality, and instead, significant deviations of the atmosphere from this steady state should take place. The signs of these deviations can be arbitrary. In practice, it means that the first-order guess of the problem (when the value of the “small parameter” is taken into consideration in a linear sense) has to be used. The essence of this guess is determined by free atmospheric variations unadjusted to the oceanic state. The moral of this result is that the use of long-lived oceanic anomalies for very long-term predictions of the atmospheric behavior is an ill-posed problem.


Climate sensitivity to ocean sequestration of heat and carbon
Ric Williams1*, Philip Goodwin2, Andy Ridgwell3

* Presenting author

1) University of Liverpool, School of Environmental Sciences, UK
2) University of Southampton, Department of Ocean and Earth Sciences, UK
3) University of Bristol, School of Geographical Science, UK

Ocean ventilation is a crucial process leading to heat and anthropogenic carbon being sequestered from the atmosphere. The rate by which the global ocean sequesters heat and carbon has a profound effect on the transient global warming. Based on theory for an idealised atmosphere and ocean (Goodwin et al., 2015), the relative rates by which heat and carbon are sequestered affects whether transient global warming depends nearly linearly on the cumulative carbon emissions; this dependence is referred to as the Transient Climate Response to Emissions, the TCRE. As the ocean drawdown of heat declines in time, the sensitivity of surface warming to radiative forcing increases in time. On the other hand, as the ocean drawdown of carbon increases in time, the sensitivity of radiative forcing to carbon emissions decreases in time. The sensitivity of surface warming to cumulative carbon emissions (the TCRE) depends on the product of the sensitivity of surface warming to radiative forcing and the sensitivity of radiative forcing to cumulative carbon emissions. Thus, the sensitivity of surface warming to cumulative carbon emissions (the TCRE) only weakly varies with time. This response is illustrated with model integrations with an Earth System Model (GENIE), configured as a coupled atmosphere and ocean with an active meridional overturning: there are partly opposing changes in time of the sensitivity of surface warming to radiative forcing and sensitivity of radiative forcing to carbon emissions. The response is likely to be modified in climate models with additional climate forcing from non-CO2 greenhouse gases and aerosols, and possibly by a more realistic representation of ocean circulation.

Goodwin, P., R.G. Williams and A. Ridgwell, 2015. Sensitivity of climate to cumulative carbon emissions due to compensation of ocean heat and carbon uptake. Nature Geoscience, 8, 29-34, doi:10.1038/ngeo2304.


AMOC and ocean heat content changes in the past two decades-wind and buoyancy forcing
Lin Xiaopei1*, Liu Hao1, Yang Jiayan2, Zhao Jian3, Xie Shang-Ping4

* Presenting author

1) Ocean University of China, Physical Oceanography Lab, China
2) Woods Hole Oceanographic Institution, Department of Physical Oceanography, United
3) Columbia University, Lamont-Doherty Earth Observatory, United
4) Scripps Institution of Oceanography, United

The water mass that fills the global ocean is so enormous for its capacity to store heat that it can effectively buffer the pace of global warming. In fact, an increased storage of heat and thus a warming in the subsurface ocean has been attributed to a slow down in the warming trend on earth’s surface in the last 15 years, a phenomenon also known as the global warming hiatus. The subsurface warming in the North Atlantic Ocean is particularly pronounced and strong related with AMOC variability. No mechanism has been decisively pined down for these changes in the past two decades even though the deep convection in the subpolar basin has been suspected to play a role. In this study we demonstrate that it is the surface wind stress that is primarily responsible for the AMOC and subsurface ocean heat content changes in the North Atlantic Ocean. The wind-stress curl has been anomalously negative in the Subpolar North Atlantic Ocean in the last 2 decades, which has resulted in a deepening in the thermalcline depth and an increased ocean heat content in upper 1500m. This added heat storage in subsurface offset the slowdown in surface warming and contributed to the hiatus in surface warming.


Influence of AMOC on Sea Levels in the North Atlantic
Jianjun Yin1*, Paul Goddard1, Stephen Griffies2, Shaoqing Zhang2

* Presenting author

1) University of Arizona, Department of Geosciences, USA
2) Geophysical Fluid Dynamics Labratory, NOAA, USA

The AMOC influences the sea levels in the North Atlantic through different mechanisms and on various time scales. We will discuss and review the recent progress in this research field. In particular, we will case study the sea level signals associated with the 2009-2010 AMOC downturn. In addition to the interannual time scale, we will also present long-term sea level change patterns in the North Atlantic and their linkage to the AMOC variability and change. Outstanding questions about AMOC/sea level research will be discussed.



Theme 3: AMOC state estimation, predictability and prediction

All sessions take place Thursday 23rd July.   Invited talks (3)   Oral presentations (4)   Posters (10)  

Invited talks:
Comparison of the Atlantic Meridional Overturning Circulation between 1960 and 2007 in six ocean reanalysis products
Alicia Karspeck1*

* Presenting author

1) National Center for Atmospheric Research, USA

The mean and variability of the Atlantic Meridional Overturning Circulation (AMOC), as represented in six ocean reanalysis products, are analyzed over the period 1960-2007. Particular focus is on multi-decadal trends and interannual variability at 26.5degreeN~and 45degreeN. For four of the six reanalysis products, corresponding reference simulations obtained from the same models and forcing datasets but without the imposition of subsurface data constraints are also included for comparison. An emphasis is placed on identifying general characteristics of the reanalysis representation of AMOC relative to their reference simulations without subsurface data constraints. The AMOC as simulated in these two sets are presented in the context of results from the Coordinated Ocean-ice Reference Experiments, phase II (CORE-II), wherein a common interannually varying atmospheric forcing data set was used to force a large and diverse set of global ocean-ice models.

Relative to the reference simulations and CORE-II forced model simulations it is shown that i) the reanalysis products tend to have greater AMOC mean strength and enhanced variability [in better agreement with observational estimates at 26.5degreeN]; ii) the reanalysis products have enhanced multi-decadal trends [from 1975-1995 and 1995-2008] in the midlatitudes; iii) the reanalysis products have less agreement in year-to-year AMOC changes.


Predicting near-term AMOC variations and their associated impact
Daniela Matei1*

* Presenting author

1) Max Planck Institute for Meteorology, Ocean in the Earth System, Germany

Over the past decade a rich body of literature attributed a major role to AMOC variations in driving significant decadal scale or longer climate variations or setting the preconditions to abrupt strong climatic shifts over the North Atlantic region. Moreover, AMOC variations itself were found to be potential predictable from a couple of years to a decade in advance. However, most of these inferences were based on coupled model simulations or model-based proxy reconstructions and therefore, the results were model dependent. Benefiting from the recently emerged field of initialized decadal climate predictions and the already a decade-long direct AMOC observational estimates from the RAPID-MOCHA program, we are now in the position to make the first evaluations of the real AMOC predictability and its potential associated climate impact.

By using various suites of initialized decadal hindcast/prediction experiments performed with the MPI-M coupled models, I will investigate whether state-of-art couple models are able to skillfully predict subseasonal-to-interannual AMOC fluctuations and how important is an accurate initialization of AMOC state for extending the climate predictive potential over the North Atlantic/European sector up to a decade or more ahead. A special focus will also be on the interplay between heat transport changes associated with AMOC and gyre variability in setting the evolution and predictability of strong climatic changes over the North Atlantic Subpolar Gyre region, such as the mid1990s warming and the currently forecasted cooling tendency.


Predicting the subpolar North Atlantic up to a decade ahead
Jon Robson1*

* Presenting author

1) University of Reading, NCAS-Climate, Department of Meteorology

Research in Decadal Climate Prediction has suggested a key role for the AMOC in recent changes in North Atlantic, particularly in the 1990s transition and associated shifts in surface climate. However, there are reasons to be somewhat sceptical about the previous predictions; in particular, they tend to have low resolution atmosphere and ocean components, and thus may not represent all the processes that are important to simulate Atlantic Multi-decadal Variability. As evidence now suggests that the AMOC has weakened over the past decade, and could weaken still further, the need to understand the AMOC’s climatic impact, whether ongoing AMOC trends will continue, and whether these events are predictable, is more pressing than ever.

In this talk I will introduce new results which further quantify the state-of-the-art of climate prediction for the North Atlantic region, with a focus on the role of the ocean. I will introduce results from new high-resolution climate predictions made with the UK HiGEM model, which exhibit substantial skill at predicting the subpolar North Atlantic, and some initial results from an international multi-model comparison of predictions of the North Atlantic subpolar gyre. Finally, I’ll present observational evidence that a cooling of the North Atlantic Ocean is already underway and appears, to a large extent, consistent with a reduction in ocean poleward heat transports. These ongoing changes in ocean heat content, coupled with state-of-the-art observations and climate prediction systems, offer a fantastic opportunity to improve our understanding of the role of the AMOC in climate.


Oral presentations:
The role of the AMOC in multi-annual predictions of the Atlantic subpolar gyre in Met Office models
Leon Hermanson1*, Nick Dunstone1, Rosie Eade1, Niall Robinson1, Adam Scaife1, Doug Smith1

* Presenting author

1) Met Office, Hadley Centre, UK

Decadal variability in the North Atlantic and its subpolar gyre (SPG) has been shown to be predictable in climate models initialized with the concurrent ocean state. Numerous impacts over ocean and land have also been identified. We use three versions of the Met Office Decadal Prediction System to provide a multimodel ensemble forecast of the Atlantic Meridional Overturning Circulation (AMOC) and related impacts. We present evidence that the ensemble forecast is able to skilfully predict these quantities over recent decades. The recent cooling trend in the SPG is predicted to continue until at least the end of 2017 due to a decrease in the SPG heat convergence related to a slowdown of the AMOC. We also consider the role of the AMOC in predictability of the SPG in a new relatively high resolution decadal prediction system (approximately 0.25 degree in the ocean, approximately 60km in the atmosphere). The overflows over the Greenland-Scotland ridge are too weak in this model leading to deep Labrador Sea waters that are too light and a high-latitude AMOC that is shallower than previous models. Consequently, predictions initialized with observed deep water densities do not predict the evolution of the AMOC as well as in previous versions of the model with smaller biases. This highlights the importance of a realistic model background state and variability for making decadal predictions.


Exploring the influence of increased atmospheric resolution on AMOC-related predictability
Rym Msadek1*, Yohan Ruprich-Robert2, Tom Delworth3

* Presenting author

2) Princeton University, USA

Several models have shown that climate anomalies like those observed following the mid-1990s North Atlantic subpolar gyre warming could be predicted one to few years in advance when the ocean and more specifically the AMOC is initialized from observational estimates (Yeager et al. 2013, Robson et al. 2013, Msadek et al. 2014). These initialized decadal prediction experiments were all based on rather coarse-resolution coupled models, both for their ocean and atmospheric components. The atmosphere plays an important role in the decadal variability of the AMOC through changes in wind and surface fluxes. The quality of the simulation of the near-surface oceanic and the atmospheric climate can be considerably improved as the atmospheric resolution is increased from ˜2º like in the GFDL CM2.1 model to 0.5º like in the GFDL-FLOR model (Jia et al. 2014, Wittenberg et al. 2014). Here we explore the impact of increased atmospheric resolution on the decadal predictability associated with the AMOC by comparing two suites of initialized experiments based on the GFDL CM2.1 and FLOR coupled models, which share the same ocean and sea ice models but differ in their atmospheric component. The AMOC in the CM2.1 and FLOR models exhibits the same mechanism of decadal variability but has a slightly longer time scale in FLOR. We explore the extent to which increased atmospheric resolution impacts the predictive skill associated with the AMOC in the GFDL models when the models are initialized from the same ocean-sea ice initial conditions. We focus on specific periods during which the AMOC was shown to play a role in driving North Atlantic changes and remote climate anomalies like during the mid-1990s or during the recent 2004-2012 AMOC decline (Smeed et al. 2013).


Reconstructing extreme AMOC events through nudging of the ocean surface
Pablo Ortega1,2*, Eric Guilyardi1,2, Didier Swingedouw3, Juliette Mignot1,4, Sebastien Nguyen1

* Presenting author

1) LOCEAN Laboratory-IPSL, Université Pierre et Marie Curie, France
2) NCAS Climate/Department of Meteorology, University of Reading, United Kingdom
3) EPOC, Université de Bordeaux, France
4) Climate and Environmental Physics and Oeschger Centre for Climate Change Research, University of Bern, Switzerland

Rapid and large changes in the Atlantic meridional overturning circulation (AMOC) can strongly impact climate at the global scale. These rapid fluctuations emerge frequently in models as a result of internal climate variability, but in the real world, the lack of long-enough continuous observations has prevented their identification. Indirect estimates of past AMOC variability can be obtained through the use of climate models, following different assimilation techniques. However, the fidelity of these products can only be partially validated with the limited in situ measurements available. To overcome some of these limitations, we here follow a perfect model approach with the IPSL-CM5A-LR model to assess the performance of several nudging techniques towards different sets of surface variables (i.e. sea surface temperature and salinity relaxation, wind stress restoring) in reconstructing the simulated AMOC variability. The motivation to use only surface nudging comes from the longer well-observed surface ocean variables when compared to the sub-surface. We first use the standard 2-months relaxation time scale for surface restoring, classically used for ocean-only simulations. A specific focus is made on the representation of an extreme positive peak in the target control simulation used as “surrogate reality”. Our analysis highlights the sensitivity to the initial conditions, and recommends the use of an ensemble of nudged simulations to guarantee a correct estimate of uncertainty. All the standard nudging approaches used here succeed in reproducing the timing of the extreme AMOC peak, but underestimate its amplitude. A careful analysis of the AMOC precursors reveals that this underestimation comes from a deficit in the formation of the dense water masses in the main regions of convection. This issue is largely corrected in an improved nudged simulation that uses a varying relaxation term, proportional to the mixed layer depth. This development improves the restoring of surface temperature and salinity in the regions of convection, and eventually the representation of AMOC variability, preventing unphysical restoring fluxes elsewhere. This is therefore a promising nudging strategy that applied to the real world can help to better constrain the recent AMOC variability over the last few decades.


Predicted growth of Atlantic sea-ice in the coming decade
Stephen G. Yeager1*, Haiyan Teng1, Gokhan Danabasoglu1

* Presenting author

1) NCAR, Climate and Global Dynamics, USA

There is little doubt that we will see a decline in Arctic sea-ice cover in this century in response to anthropogenic warming, and yet internal climate variations are expected to generate considerable spread in the multi-year trends in Arctic sea-ice extent for many more years into the future. Variations in the strength of the Atlantic Meridional Overturning Circulation (AMOC), in particular, would appear to play an important role in modulating rates of sea-ice loss because of the associated variations in heat transport into the high latitude North Atlantic. We present evidence that the extreme negative trends in Arctic winter sea-ice extent in the late 1990s were a predictable consequence of the preceding decade of persistent positive winter North Atlantic Oscillation (NAO) conditions and associated spin-up of the thermohaline circulation (THC). Initialized forecasts made with the Community Earth System Model decadal prediction system indicate that relatively low rates of North Atlantic Deep Water formation in recent years will result in a continuation of a THC spin-down that began more than a decade ago. Consequently, projected 10-year trends in winter Arctic winter sea-ice extent seem likely to be much more positive than has recently been observed, with the possibility of actual decadal growth in Atlantic sea-ice in the near future.


Poster presentations:
Predictability of Atlantic Ocean Heat Content
Martha Buckley1*, Tim Delsole1

* Presenting author

1) George Mason University, George Mason University, Department of Atmospheric

Due to the inherently chaotic nature of the atmosphere, prospects for decadal climate prediction lie in the interaction of the atmosphere with slower parts of the climate system, including the ocean. We propose a new diagnostic to estimate the interannual predictability of the ocean—the ocean heat content integrated over the maximum climatological mixed layer depth. This quantity, which we henceforth denote as H, is a measure of the heat contained in the layers that interact with the atmosphere seasonally and is highly coherent with SST on interannual timescales. We estimate the predictability of H in the Atlantic using a multivariate regression model derived from control runs of climate models. We relate the geographic variations and inter-model differences in predictability of H to differences in mixed layer depths, damping timescales, and the relative importance of atmospheric forcing and ocean dynamics in setting H.


High resolution ocean model simulations with interactive icebergs to estimate AMOC sensitivity to increased Greenland ice sheet melt
Alan Condron1*

* Presenting author

1) University of Massachusetts Amherst, Geoscience, USA

Results from several high-resolution model simulations assessing the impact of increased Greenland ice sheet (GrIS) melt on AMOC strength will be presented. A global ocean model (MITgcm) is integrated at an eddy permitting (1/6 deg.; 18-km) resolution and coupled to a new dynamic-thermodynamic iceberg model (MITberg) to more accurately simulate ocean-cryosphere interactions. MITberg uses a multilevel keel model to realistically simulate ocean drag below the waterline, solves all relevant melt terms, and includes a parameterization to simulate the fracturing of overhanging ice caused by wave erosion. The model was spun-up for 25 years with 405 Gt/yr (~0.013 Sv) of ice calved from 33 major ice streams along the edge of the GrIS. At any one time ~6700±430 icebergs are present in the North Atlantic and are mainly confined to narrow coastal boundary currents (Labrador, EGC, WGC). In the Control simulation, an average of 458 icebergs drift south of 48°N each year in the Labrador Current with peak fluxes occurring in late-spring/early summer. These results are in remarkable agreement (r=0.5) with historical observations (annual mean 474) collected by the International Ice Patrol at this latitude since 1900. Ocean circulation changes resulting from increased GrIS freshwater forcing are assessed by instantaneously increasing GrIS discharge to 3160 Gt/yr (~0.1 sv) (Exp.1) and linearly increasing ice discharge by 27.5 Gt/yr^2 to reproduce several recently observed increases in GrIS melt (Exp.2). In Exp.1 there is a 10-fold increase in the total number of icebergs in the North Atlantic (~50,000-60,000 per yr) and a ~8-fold increase (up to ~3700) in the number of icebergs drifting south of 48N each year. An increase in iceberg activity of this magnitude would present a considerable hazard to commercial shipping and oil platforms operating near the Grand Banks of Newfoundland and in the northwest Atlantic. After 10 years the Labrador Sea is ~1.5 psu fresher and the eastern subpolar North Atlantic 0.3 psu fresher, while AMOC shows a 3.5Sv (18.6%) reduction. Results from a longer (30-year) integration of Exp.1, as well as Exp.2, will be presented at the meeting.


Remedying excessive numerical diapycnal mixing in the GO5.0 NEMO configuration
Alex Megann1*, Dave Storkey2

* Presenting author

1) Marine Systems Modelling, National Oceanography Centre, Southampton
2) Met Office, Exeter, UK

If numerical ocean models are to simulate faithfully the upwelling branches of the global overturning circulation, they need to have a good representation of the diapycnal mixing processes which contribute to conversion of the bottom and deep waters produced in high latitudes into less dense watermasses. It is known that the default class of depth-coordinate ocean models such as NEMO and MOM5, as used in many state-of-the art coupled climate models and Earth System Models, have excessive numerical diapycnal mixing, resulting from advection across coordinate surfaces.

The GO5.0 configuration of the NEMO ocean model, on an “eddy-permitting” 0.25° global grid, is used in the current UK GC1 and GC2 coupled models. Megann and Nurser (2015) have shown, using the isopycnal watermass analysis of Lee et al (2002), that spurious numerical mixing is substantially larger than the explicit mixing prescribed by the mixing scheme used by the model. It will be shown that increasing the biharmonic viscosity by a factor of three tends to suppress small-scale noise in the vertical velocity in the model. This leads to changes in the overturning circulation, especially when expressed as a function of density: these are associated with a significant reduction in the numerical mixing in GO5.0, and we shall show that this leads to large-scale improvements in model biases.


Testing the impact of CMIP5 model biases on the simulation of North Atlantic decadal variability
Matthew B. Menary1,2*, Daniel L. R. Hodson2, Jon I. Robson2, Rowan T. Sutton2, Richard A. Wood1, Jonathan Hunt3

* Presenting author

1) Met Office Hadley Centre, UK
2) NCAS-Climate, University of Reading, UK
3) University of Oxford, UK

Instrumental observations, palaeo-proxies, and climate models suggest significant decadal variability within the North Atlantic subpolar gyre (NA SPG). However, a combination of a poorly sampled observational record and a diverse range of model behaviours have left the precise nature of this variability unclear. Here, we analyse an exceptionally large ensemble of 42 present-generation climate models to test whether NA SPG mean state temperature and salinity biases systematically affect the representation of decadal variability. We find that biases in the Labrador Sea co-vary and influence whether density variability is controlled by temperature or salinity changes. Ocean horizontal resolution is a good predictor of the biases and affects whether models choose northern or southern feedbacks within the NA SPG. Despite these relationships, we find no link to the spectral characteristics of the variability. Our results suggest that the mean state and evolution of anomalies within the NA SPG are not independent.


Decadal prediction skill with surface nudging
Juliette Mignot1,2,3*, Javier Garcia-Serrano1, Didier Swingedouw4, Agathe Germe1, Sébastien Nguyen1, Pablo Ortega1,5, Eric Guilyardi1,5, Sulagna Ray1

* Presenting author

2) Climate and Environmental Physics, Physics Institute, University of Bern
3) Oeschger Center for Climate Change Research, University of Bern, Switzerland
4) EPOC-CNRS, Université de Bordeaux, France
5) NCAS-Climate, University of Reading, UK

Predictability of the Atlantic meridional circulation and related temperature variations is investigated in the IPSL system, initialized through surface nudging. Two decadal prediction ensembles, based on the same climate model (IPSL-CM5A- LR) and the same surface nudging initialization strategy are analyzed and compared with a focus on upper-ocean variables in different regions of the globe. One ensemble consists of 3-member hindcasts launched every year since 1961 while the other ensemble benefits from 9 members but with start dates only every 5 years. Analysis includes anomaly correlation coefficients and root mean square errors computed against several reanalysis and gridded observational fields, as well as against the nudged simulation used to produce the hindcasts initial conditions. The last skill measure gives an upper limit of the predictability horizon one can expect in the forecast system, while the comparison with different datasets highlights uncertainty when assessing the actual skill. Results provide a potential predictability skill (verification against the nudged simulation) beyond the linear trend of the order of 10 years ahead at the global scale, but essentially due to the non-linear response to external forcing (i.e. volcanoes). At regional scale, we obtain 1 year in the tropical band, 10 years at extratropical latitudes in the North Atlantic and 5 years at tropical to subtropical latitudes in the North Atlantic, for both sea surface temperature (SST) and upper-ocean heat content. Actual prediction skill (verification against observational or reanalyzed data) is overall more limited and less robust. Even so, large actual skill is found in the extratropical North Atlantic for SST. The interplay between initialization and internal modes of variability limits the actual predictability skill of the AMOC.


Atlantic Meridional Overturning Circulation (AMOC) geostrophic transport: Comparison of the RAPID time series with hydrography from 2004 to 2011
Neela Morarji1,2*

* Presenting author

1) National Oceanography Centre, Southampton, Physical Oceanography
2) University of Southampton, Ocean and Earth Science, UK

The AMOC has been measured at 26°N by infrequent hydrographic sections and, since 2004, continuous measurements by the RAPID-MOCHA monitoring system. During the first 4 years of RAPID-MOCHA observations the overturning circulation was, on average, 2.7 Sv stronger than in the subsequent 4 years (Smeed et al., 2014). However, transport inferred from the 2004 and 2010 hydrographic sections yield contrasting results with stronger overturning in 2010 (Atkinson et al 2012). In this presentation we examine the possible causes of this apparent discrepancy between the RAPID-MOCHA monitoring array and the hydrographic sections. Not only do these two estimates of the overturning circulation use different data but also different methods. We examine the assumptions and approximations used in each method and show that when applied to the same data the two methods provide consistent results. The primary reason for the trend calculated from hydrographic sections differing from that calculated with the RAPID-MOCHA time series is the large temporal variability at short timescales that masks the long-term trend when comparing the two hydrographic sections.


Freshwater advection into the Labrador Sea
Lena Schulze1*, Eleanor Frajka-Williams1, Sheldon Bacon2

* Presenting author

1) University of Southampton, OES, UK
2) National Ocanography Center, UK

Deep convection in the Labrador Sea forms the sinking limb of the Atlantic MOC. While the amount of heat lost to the atmosphere is an important factor to the strength of convection, changes in surface buoyancy, such as additional freshwater, are also critical and can lead to a suppression of deep mixing. With increasing melt rates from Arctic and Greenland ice, freshwater could therefore impact the formation of deep water. Whether the freshwater reaches regions of deep convection is still debated. Using a Lagrangian approach with particles in a NEMO 1/12 degree ocean model, we investigate how and where freshwater from the Greenland boundary currents reaches the central Labrador Sea or if it is simply advected south via the boundary currents. At the surface, particles are advected into the central basin along the entire boundary of the Labrador Sea, while only certain key regions are important in deeper layers. This suggests that Ekman transport as well as eddies are an important factor in transporting freshwater to the basin. Furthermore the key regions in the exchange between the boundary and basin and therefore the freshwater flux, have changed in the last decade.


Bidecadal North Atlantic ocean circulation variability controlled by timing of volcanic eruptions
Didier Swingedouw1, Pablo Ortega2,3*, Juliette Mignot3,4,5, Eric Guilyardi2,3, Valérie Masson-Delmotte6, Paul Butler7, Myriam Khodri3, Roland Séférian8

* Presenting author

1) EPOC, CNRS - Université de Bordeaux, France
2) NCAS-Climate, University of Reading, UK
4) Climate and Environmental Physics, University of Bern, Switzerland
5) Oeschger Center of climate change research, University of Bern, Switzerland
7) School of Ocean Sciences, Bangor University, UK
8) CNRM-GAME, Météo-France - CNRS, France

While bidecadal climate variability has been evidenced in several North Atlantic paleoclimate proxy records, the drivers of this variability remain poorly understood. Here, we show that the subset of CMIP5 historical climate simulations that produce such bidecadal variability exhibit a robust synchronisation with a maximum in Atlantic meridional overturning circulation (AMOC) strength, 15 years after the 1963 Agung eruption. The mechanisms at play involve salinity advection from the Arctic and explain the timing of Great Salinity Anomalies observed in the North Atlantic in the 1970s and 1990s. Simulations as well as Greenland and Iceland paleoclimate records indicate that coherent bidecadal cycles were excited following five Agung-like volcanic eruptions of the last millennium. Over the last decades, climate simulations and a conceptual model reveal that destructive interference caused by the Pinatubo 1991 eruption may have led to lower AMOC variability in the 2000s. Our results imply a long-lasting climatic impact of volcanic eruptions around the North Atlantic, and multi-decadal climate predictability following the next Agung-like eruption.


On the predictability of North Atlantic ocean state
Florian Sévellec1*, Alexey V. Fedorov2

* Presenting author

1) University of Southampton, Department Ocean and Earth Science - National Oceanography Centre Southampton, UK
2) Yale University, Department of Geology and Geophysics, USA

This study investigates the decadal predictability of the ocean climatic state in the North Atlantic in an ocean-forced context. To assess this oceanic predictability, we compute Linear Optimal Perturbations (LOPs) in a realistic Ocean General Circulation Model in a 2° global configuration (NEMO-ORCA2) and estimate the maximum impact of small disturbances on ocean dynamics. Our calculations of LOPs involve a maximization procedure based on Lagrangian multipliers in a non-autonomous context. As the metrics of the ocean state we use four different measures: the Meridional Volume Transport (MVT), the Meridional Heat Transport (MHT), the mean Sea Surface Temperature (SST), and the Oceanic Heat Content (OHC), all in the North Atlantic. We show that the four metrics are dramatically different in regard to predictability. Whereas the SST and OHC can be modified only by relatively large-scale anomalies, the MVT and MHT are strongly affected by small-scale anomalies as well (acting along the basin eastern and western basin boundaries and changing the East-West density difference across the Atlantic). This suggests that MVT and MHT are much less predictable than SST and OHC. It is only when MVT and MHT are averaged over climatically relevant timescales (e.g. 30 years) that the four metrics have comparable predictability. This result stresses the need for long-term measurements of the Atlantic Meridional Overturning Circulation intensity and its associated heat transport in order to have climatically relevant data. Our study also suggests that initial errors of a few milli-Kelvins can lead on a decadal timescale to an error of 0.1 K in North Atlantic mean sea surface temperature estimates. This transient error growth is maximal after about 17 years and can be interpreted as a decadal predictability barrier.


The Day Before Tomorrow: How might the Atlantic Meridional Overturning Circulation have changed?
Simon Tett1*, Toby Sherwin2, Gabi Hegerl1

* Presenting author

1) University of Edinburgh, School of Geosciences, UK
2) Scottish Marine Institute, Scottish Association for Marine Science, UK

Mass transports from seven different ocean reanalysis (DePreSys, MOVE-CORE, SODA-2.2.4, GFDL-CM2.1-ECDA, K7ODA, GECCO2 and ORA-S4) were computed across three lines across the North Atlantic (Greenland-Scotland Ridge (GSR), 41N and 26N) and these compared with observational estimates. GSR transport was computed by computing total transport with respect to temperature and finding the maximum transport across the ridge. Using uncertainty from observational error and random variability in the reanalysis Bayes weights were used to compute multi-reanalysis mean transports. Uncertainties in the multi-reanalysis mean were computed through bootstrapping. As the Bayes weights are dominated by only two ocean reanlayis (ORA-S4 and SODA-2.2.4) these uncertainties need to be interpreted cautiously. A similar analysis was done for many CMIP5 simulations. Neither reanalyses nor simulations show significant trends relative to the uncertainty in the multi-reanalysis or multi-model mean respectively. However, uncertainty in the reanalysis is small enough on decadal scales to examine decadal variability in the AMOC. The ocean reanalysis have decadal variability with a magnitude of about 1 Sv that is coherent across much of the North Atlantic with maximum northward transports circa 1970, 1985 and 1995.



Theme 4: Novel approaches to pan-Atlantic observations, modelling, analysis and synthesis

Poster session Thursday 23rd p.m.; oral session Friday 24th a.m.   Invited talks (3)   Oral presentations (4)   Posters (9)  

Invited talks:
Progress toward AMOC Estimation using Deepgliders
Charles C. Eriksen1*

* Presenting author

1) University of Washington, School of Oceanography, USA

In preparation for use in estimating the Atlantic Meridional Overturning Circulation (AMOC), long-range autonomous underwater Deepgliders are being used to continuously survey the full open deep ocean water column near the Bermuda Atlantic Time Series (BATS) site 90 km southeast of Bermuda through winter and spring 2014 & 2015.

Deepgliders have carried out 3-month missions in the vicinity of the BATS site from March to June 2014 and again from January through present, 2015. These missions surveyed an 80 km corner-to-corner bow-tie pattern centered on the BATS site, repeating the pattern fortnightly. So far, more than 100 dive cycles to depths between 4000 and 4850 m in the repeat survey have been carried out. These have produced more than 200 profiles of temperature, salinity, and dissolved oxygen at 0.5-3 m resolution from the top meter of the water column to within 30 m of the ocean bottom, all transmitted in near-real time via satellite telemetry. Collocated monthly shipboard CTD casts have been carried out simultaneously with Deepglider dives to evaluate Deepglider data quality. The current mission is using battery energy at a rate that implies endurance of more than one year.

In addition to gathering hydrographic sections, Deepgliders estimate barotropic (full ocean depth averaged) current. Such currents have been persistently to the southwest at speeds in the range 0.05-0.10 m/s in the BATS region, except during the passage of submesoscale anticyclones, when they are stronger. The eddies, distinguished by cool, fresh water nearly saturated in dissolved oxygen between ~1000 m and ~2000 m depth, are of presumed subpolar origin. Two such eddies have been observed passing through the BATS region in 6 months of observation so far.

Additional Deepglider deployments at Bermuda are planned to include transects to and from the continental slope offshore the U.S., crossing the western boundary current system (Gulf Stream and Deep Western Boundary Current). A pair of Deepgliders is planned to be deployed offshore Great Abaco Island, Bahamas, as part of the RAPID-MOCHA array. Preliminary results from these missions will be presented at the meeting, should they have become available.


Using carbon isotopes to reconstruct AMOC changes during the last deglaciation
Andreas Schmittner1*

* Presenting author

1) Oregon State University, College of Earth, Ocean

A synthesis of carbon isotope data (d13C) from deep sea sediments is used together with model simulations to estimate AMOC changes during the initial phase of the last deglaciation. This phase was characterized by ice rafting and cold conditions in the North Atlantic (Heinrich Stadial Event 1, HS1). The reconstructions show decreases in d13C in the North Atlantic with largest values at intermediate depths at high latitudes. The amplitude decreases further south and deeper in the water column. This pattern agrees well with model simulations in which the AMOC was collapsed, but the simulations starting from pre-industrial conditions overestimate the amplitude of the d13C changes. In the model atmospheric CO2 increases and d13C of atmospheric CO2 declines as a response to the AMOC shutdown, consistent in amplitude and rate with ice core measurements. Sensitivity of these results to different initial conditions will be explored.


How overturning circulation sets ocean heat uptake and why this matters for transient climate change
Jan D. Zika1*

* Presenting author

1) University of Southampton, National Oceanography Centre

Heat transport between the surface and deep-ocean strongly influences transient climate change. Mechanisms setting this transport are investigated using coupled climate models and by projecting ocean circulation into the temperature-depth diagram. In this diagram, a `cold cell' linked to Antarctic Bottom Water cools the deep ocean and is balanced by a `warm cell' linked to the Atlantic Meridional Overturning Circulation which warms the deep ocean. With anthropogenic warming, the cold cell collapses while the warm cell continues to warm the deep ocean. Changes in the advective vertical heat flux dominate over changes in the diffusive vertical heat flux. Simulations with increasingly strong warm cells, set by their mean Southern Hemisphere winds, exhibit increasing deep-ocean warming and hence weaker transient surface warming in response to the same anthropogenic forcing. It is argued that the accurate description of the mean overturning circulation – including its strength and the temperature of its upwelling and downwelling branches - is key to simulating transient climate change.


Oral presentations:
Reconstructing Atlantic and global overturning from meridional density gradients
Edward D. Butler1, Kevin I. C. Oliver1*, Joel J.-M. Hirschi2, Jennifer V. Mecking1

* Presenting author

1) University of Southampton, National Oceanography Centre Southampton, UK
2) National Oceanography Centre, UK

Numerous attempts have been made to scale the strength of the meridional overturning circulation (MOC), principally in the North Atlantic, with large-scale, basin-wide hydrographic properties. In particular, various approaches to scaling the MOC with meridional density gradients have been proposed, but the success of these has only been demonstrated under limited conditions. Here we present a scaling relationship linking overturning to twice vertically-integrated meridional density gradients via the hydrostatic equation and a “rotated” form of the geostrophic equation. This provides a meridional overturning streamfunction as a function of depth for each basin. Using a series of periodically forced experiments in a global, coarse resolution configuration of the general circulation model NEMO, we explore the timescales over which this scaling is temporally valid. We find that the scaling holds well in the upper Atlantic cell for multi-decadal and longer timescales, accurately reconstructing the relative magnitude of the response for different frequencies and explaining over 85% of overturning variance on timescales of 64 to 2048 years. Despite the highly nonlinear response of the Antarctic cell in the abyssal Atlantic, between 76% and 94% of the observed variability at 4000m is reconstructed on timescales of 32 years (and longer). The scaling law is also successfully applied in the Indo-Pacific. These results indicate that meridional density gradients and overturning are linked via meridional pressure gradients, and that both the strength and structure of the MOC can be reconstructed from hydrography on multi-decadal and longer timescales provided that the link is made in this way. This has gives us the potential to reconstruct the Atlantic MOC over much of the 20th Century, and can contribute to long-term monitoring efforts.


Using satellite observations to broaden our spatial view of AMOC variability
Eleanor Frajka-Williams1*

* Presenting author

1) University of Southampton, Ocean and Earth Sciences, UK

Climate simulations predict a slowing of the AMOC in the coming years, while present day observations from boundary arrays demonstrate substantial variability on weekly- to interannual timescales. These arrays are necessarily limited to individual latitudes. How well does satellite altimetry replicate transbasin, full-depth ocean transports? Can we use satellite altimetry to extend our estimate of AMOC variability back in time (at 26N)? Do the spatial patterns of SSH variability help to broaden our view of AMOC strength beyond individual latitudes? This analysis complements in situ observational efforts to measure the MOC at multiple latitudes.


Atlantic BiogeoChemical (ABC) Fluxes: adding carbon and nutrients to the RAPID array, a contribution to NERC’s RAPID-AMOC progamme
Elaine McDonagh1*, Richard Sanders1, Peter Brown1, Brian King1, Darren Rayner1, Andrew Yool1, Sue Hartman1, Sinhue Torres-Valdes1, Val Byfield1, David Smeed1, Mark Moore2, Nick Bates2,8, Andrew Watson3, Paul Halloran3, Ute Schuster3, Marie-Jose Messias3, Tim Smyth4, Stefano Ciavatta4, Giorgio Dall'Olmo4, Rosa Barciela5, Molly Baringer6, Shenfu Dong6, Chris Meinen6, Dennis Hansell7, Becky Garley8, Steven Emerson9

* Presenting author

1) National Oceanography Centre, UK
2) University of Southampton, UK
3) University of Exeter, UK
4) Plymouth Marine Laboratory, UK
5) UK Met Office, UK
6) NOAA Atlantic Oceanographic and Meteorological Laboratory, US
7) University of Miami, US
8) Bermuda Institute of Ocean Sciences, Bermuda
9) University of Washington, US

ABC fluxes is a six-year project running from October 2014 to September 2020 funded by NERC’s RAPID-AMOC programme. Here we present an overview of the project.

The North Atlantic Ocean plays a pivotal role in the global carbon cycle, by storing anthropogenically mobilised carbon and by supporting the downward flux of organic matter. Our understanding of how lateral oceanic fluxes in the subtropics contribute to these processes is largely based on hydrographic sections occupied every 5 years at 24.5°N, a sampling programme that is inadequate to resolve and understand the role these transfers play in regulating these processes. Detailed time series of physical fluxes at 26.5°N from the RAPID array suggest that variability in these transfers will be occurring on a range of timescales, which, once measured, will likely modify our understanding of the role the North Atlantic subtropical gyre plays in the global carbon cycle.

ABC fluxes is addressing these issues by deploying new instruments on the RAPID array at 26.5°N, making biogeochemical measurements on the bi-monthly NOAA cruises in Florida Straits and deploying oxygen-enabled Argo floats. ABC fluxes has purchased pCO2 sensors and Remote Access (RAS) samplers that will be deployed on the tall mooring in the array in October 2015. ABC fluxes will sample its first of 22 Florida Straits cruises in May 2015. The first 12 moored oxygen sensors (sampling the western boundary) will be recovered in October 2015 and 24 new sensors will be deployed at that time to make observations in the western boundary, Mid-Atlantic ridge and eastern boundary. In December 2015 we will deploy 16 oxygen-enabled Argo floats from the GO-SHIP repeat hydrographic section at 24.5°N

ABC fluxes will calculate time series of nutrient and carbon, including anthropogenic carbon, fluxes across 26.5°N. We adopt a hierarchical approach, successively using existing observations, then new oxygen observations and ultimately direct observations of the carbon and nutrient fields from sensors and samplers in order to identify the added value each successive stage of ABC observations provides. We interpret our direct flux calculations as contributions to the North Atlantic budget in conjunction with other observations and models to assess how oceanic fluxes control the strength and variability of the role the North Atlantic plays in the global carbon cycle. In other abstracts at this meeting Pete Brown presents ABC fluxes calculations of Anthropogenic carbon fluxes using existing observations, Tim Smyth presents ABC fluxes in a pan-Atlantic biogeochemical context, and Paul Halloran details the modeling aspects of ABC fluxes.


Can we monitor how far the AMOC is from a threshold?
Richard A. Wood1*, José M. Rodriguez1, Robin S. Smith2, Ed Hawkins 2

* Presenting author

1) Met Office, Hadley Centre, UK
2) University of Reading, NCAS-Climate, UK

Palaeoclimatic and modelling evidence suggests that the AMOC has the potential to exhibit threshold and hysteresis behaviour in response to changes in the fresh water budget of the Atlantic basin. The sign of the fresh water transport by the AMOC into the Atlantic basin (‘Fov’) is an indicator of whether the current AMOC is in a monostable or bistable regime, and this is sometimes interpreted as a measure of the stability of the AMOC. However we show here, using a hierarchy of models of varying complexity, that Fov does not have a simple relationship to the distance of the AMOC from a threshold.

Our model hierarchy suggests that the distance of the AMOC from a stability threshold is controlled by low order dynamics that can be estimated using large scale emergent properties of the ocean (such as large regional mean temperatures and salinities). These properties can in principle be estimated from observations, opening up the possibility of observational monitoring of the AMOC threshold.

Under increasing greenhouse gases, we show that the AMOC threshold is sensitive to a number of changes in the ocean climate state. Some of these changes (e.g. an intensification of the atmospheric hydrological cycle) are thought to be robustly modelled by the current generation of global climate models; however others (e.g. changes in gyre fresh water transport) are more model-dependent. While this modelling uncertainty remains, observationally-based monitoring of the AMOC threshold may be the best way to manage the risks arising from possible AMOC ‘surprises’.


Poster presentations:
AtlantOS: Optimizing and Enhancing the Integrated Atlantic Ocean Observing System
The Atlantos Consortium1, Johannes Karstensen1*

* Presenting author

1) represented by: GEOMAR Helmholtz Centre for Ocean Research Kiel, Germany

We will introduce AtlantOS (www.atlantos-h2020.eu) - an EU/Horizon2020 project started on 1. April 2015. The AtlantOS vision is to improve and innovate Atlantic observing towards a more systematic, cost effective and user-driven and international accepted Integrated Atlantic Ocean Observing System (IAOOS). AtlantOS will work follow the “Framework of Ocean Observing” to achieve a transition from a loosely coordinated set of existing ocean observing activities, producing fragmented, often mono-disciplinary data, to a sustainable, efficient, and fit-for-purpose AOOS.
The research and innovation activities focus on: (1) Defining requirements and systems design, (2) Improving the innovation and readiness of observing networks and data systems, (3) Engaging stakeholders around the Atlantic.

AtlantOS will strengthen Europe's contribution to the Global Ocean Observing System (GOOS), a major component of the Group on Earth Observations' (GEO), its Global Earth Observation System of Systems (GEOSS), and specifically on the emerging "Oceans and Society: Blue Planet" initiative. AtlantOS contributes to “Blue Growth” by merging new information needs relevant to key sectors such as transport, tourism, fisheries, marine biotech, resource extraction and energy with existing requirements. AtlantOS significantly contributes to trans-Atlantic cooperation (in particular in the framework of the “The Galway Statement on Atlantic Ocean Cooperation”) by integrating existing observing activities established by European, North and South American, and African countries and by filling existing gaps to reach an agile, flexible IAOOS and associated ocean information systems around the Atlantic.
In particular, work package (WP5) will investigate the Regional Atlantic observing with a focus on climate (incl. AMOC) and ecosystem research – in the North Atlantic subpolar gyre and in the South Atlantic subtropical gyre.


Multiple equilibria as a possible mechanism for decadal variability in the North Atlantic Ocean
Andreas Born1,2*, Juliette Mignot1,2,3, Thomas Stocker1,2

* Presenting author

1) Climate and Environmental Physics, Physics Institute, University of Bern
2) Oeschger Center for Climate Change Research, University of Bern, Switzerland

Decadal climate variability in the North Atlantic has received increased attention in recent years, because modeling results suggest predictability of several years. However, the applicability of these results in the real world is challenged by an incomplete understanding of the underlying mechanisms. Here, we show that recent attempts to reconstruct the decadal variations in one of the dominant circulation systems of the region, the subpolar gyre (SPG) are not always consistent. A coherent picture is partly recovered by a simple conceptual model solely forced by reanalyzed surface air temperatures. This confirms that surface heat flux indeed plays a leading role for this type of variability as has been suggested in previous studies. Performance of this conceptual model is tested against a statistical stochastic model. Results suggests that large variations in the SPG correspond to the crossing of a bifurcation point that is predicted from idealized experiments and an analytical solution of our model. Hysteresis and the existence of two stable modes of the SPG circulation shape its response to forcing by atmospheric temperatures. The identification of the essential dynamics and the reduction to a minimal model of SPG variability provides a quantifiable basis and a framework for future studies on decadal climate variability and predictability.


Directly Measured Currents and Estimated Transport Pathways of Atlantic Water between 59.5N and the Iceland-Faroes-Scotland Ridge
Katelin Childers1*, Charles Flagg1, Thomas Rossby2, Corinna Schrum3,4

* Presenting author

1) Stony Brook University, School of Marine and Atmospheric Science, USA
2) University of Rhode Island, Graduate School of Oceanography, USA
3) University of Bergen, Geophysical Institute, Norway
4) Bjerknes Center for Climate Research, Norway

Using vessel-mounted acoustic Doppler current profiler (ADCP) data from four different routes between Scotland, Iceland and Greenland, we map out the mean flow of water in the top 400 m of the northeastern North Atlantic. The poleward transport east of the Reykjanes Ridge decreases from ~8.5 to 10 Sv (1 Sverdrup = 10^6 m3/s) at 59.5 to 61°N to 6 Sv crossing the Iceland-Faroe-Scotland Ridge (IFSR). The two longest ~1200 km transport integrals have ~1.4 to 0.94 Sv uncertainty, respectively. The overall decrease in transport can in large measure be accounted for by a ~1.5 Sv flow across the Reykjanes Ridge into the Irminger Sea north of 59.5°N and by a ~0.5 Sv overflow of dense water along the Iceland Faroe Ridge. A remaining 0.5 Sv flux divergence is at the edge of detectability, but if real could be accounted for through wintertime convection to greater than 400 m and densification of upper ocean water.


Modelling in ABC fluxes: Using carbon cycle and ecosystem models to interpret and add value to RAPID observations
Paul R. Halloran1*, Andrew Yool2, Stefano Ciavatta3, Rosa Barciela4, Elaine McDonagh2

* Presenting author

1) University of Exeter, Geography, UK
2) National Oceanography Centre, UK
3) Plymouth Marine Laboratories, UK
4) Hadley Centre, Met Office, UK

The Atlantic BiogeoChemical (ABC) Fluxes project aims to quantify carbon and nutrient fluxes across 26.5°N, and use this information to help constrain North Atlantic biogeochemistry and carbon cycling. ABC Fluxes is instrumenting the RAPID array with biogeochemical sensors, conducting field campaigns to place these observations in context, and interpreting and adding value to these observations through the use of biogeochemically capable 3D ocean models.

The modelling component of ABC Fluxes will:

1) Evaluate the degree to which adding biogeochemical sensors to the RAPID array will allow the constraint of 26.5°N biogeochemical fluxes. We are determining the section’s biogeochemical representativeness by attributing components of modeled carbon, nitrate and dissolved oxygen, and Total Alkalinity fluxes to each physical transport process and assessing the variability in components, and assessing the sampling strategies using the CMIP5 models, and NEMO 0.25° simulations.

2) Evaluate the added value provided by RAPID array dissolved inorganic carbon and Total Alkalinity observations in monitoring the North Atlantic carbon sink. Pilot CMIP5 analysis indicates that AMOC strength and 26.5°N section Total Alkalinity flux data allow constraint of North Atlantic carbon sink change. We are testing the hypothesis that assimilation of 26.5°N Total Alkalinity and dissolved inorganic carbon into a NEMO-FOAM physical ocean reanalysis facilitates accurate reanalysis/monitoring of the North Atlantic CO2 sink,

3) Evaluate the value added to North Atlantic biological flux estimates by RAPID array observations. We are estimating North Atlantic biological fluxes and their uncertainty in the ERSEM biogeochemical model run within a physical reanalysis produced using RAPID physics observations, and through comparison with 26.5°N biogeochemical observations.


State and variability of the North Atlantic water mass formation and its relation to the AMOC in the NorESM-1 climate model
Detelina Ivanova1,3*, Mats Bentsen2,3, Mehmet Ilicak2,3, Yanchun He1,3, Yongqi Gao1,3

* Presenting author

1) Nansen Environmental and Remote Sensing Center , GC Rieber Climate Institute, Norway
2) Uni Research Ltd, Uni Climate, Norway
3) Bjerknes Centre for Climate Research, Norway

The Atlantic meridional overturning circulation (AMOC) is closely related to the the water mass formation in the North Atlantic. The latter strongly depends on the air-sea buoyancy flux and heat and freshwater transports. Accurate representation of these important water mass formation processes is still a challenge for the present climate models. Here we use the water mass transformation rate (WMTR) function as an integrated measure of the air-sea interaction and deep water formation thus serving as both validation metric of the air-sea fluxes and the water mass formation rates in the fully coupled models. Following Speer and Tziperman (1992), we examine the mean state and variability of WMTR in the Norwegian Earth System Model (NorESM) featuring an isopycnal coordinate ocean component (MICOM). We compare fully-coupled NorESM simulations results to observational estimates based on NOC1.1a product and CORE2 forced NorESM simulations. We investigate further their relationship to the AMOC intensity and variability. The WMTR analysis showed deficiencies in the model mode water formation such as lack of or formed in rather lighter density classes and with smaller rates Subtropical Mode Water, and excessive production of Subpolar Mode Water. The application of salinity restoring in the forced simulations seems to reduce these errors. These water mass biases can be partly related to the vigorous AMOC in the NorESM which is found to be too intense (~30Sv) in the fully coupled model solution compared to the RAPID observations (~18Sv), and which we found to be significantly correlated at 45N to the oscillations in the Labrador Sea Water and North Atlantic Deep Water volumes. We explored the sensitivity of the model representation of these processes to the isopycnal mixing and mixed layer parameterizations.

"This work has been supported by the Research Council of Norway through project ORGANIC Overturning circulation and its implications for global carbon cycle in coupled model (no. 239965)"


Spurious initialization in near term AMOC predictions
Jürgen Kröger1*, Jin-Song von Storch1

* Presenting author

1) Max-Planck-Institut für Meteorologie, Ozean im Erdsystem, Germany

A new assimilation procedure is proposed aiming at suppressing spurious initialization in the coupled forecast system for decadal predictions developed in the framework of the German BMBF (Bundesministerium für Bildung und Forschung) project MiKlip (Mittelfristige Klimaprognosen).

Ocean reanalysis products are commonly used for initialization in climate predictions. Inherent to the state estimates of these products is an imprint of their underlying dynamical models which may have a negative impact on the skill of predictions. Applying the state estimates only in regions where the estimates are better constrained by real observations may lead to better skill.

To test this hypothesis, two sets of ensemble hindcasts are performed where the common full area initialization is compared to a reduced area initialization with the latter set applying a masking in the assimilation procedure that mimics the recent world ocean's data coverage (including the network of ARGO observations). Except for the tropical South Atlantic, multi-year AMOC hindcasts with reduced area initialization outperform those with full area initialization almost everywhere.


Nonlinear optimal perturbation of the Atlantic meridional overturning circulation: A new ocean modelling tool for predictability studies
Simon A. Müller1*, Florian Sévellec1

* Presenting author

1) University of Southampton, Ocean and Earth Science / National Oceanography Centre Southampton, United Kingdom

We present initial-condition perturbations of the background state of a global ocean general circulation model that are optimal with respect to an optimisation metric that can be directly related to the Atlantic meridional overturning circulation (AMOC). Previous studies utilised the computation of linear optimal perturbations (LOP) of initial conditions in order to analyse aspects of variability and predictability of the AMOC. This method involves Lagrangian multipliers and the use of the adjoint version of the ocean model. We have generalised this approach to nonlinear optimal perturbations (NOP). For this purpose we have developed an iterative method that takes advantage of the computation of LOP to provide close approximations of optimal perturbations for the fully nonlinear system. As a next step, we demonstrate its applicability in the ORCA2 configuration of NEMO version 3.4. Further, more specific applications of the NOP computation are explored in the context of AMOC variability and predictability.


Developing a telemetry system for the RAPID 26N moorings.
Darren Rayner1*, Miguel Charos-Llorens2, Stephen Mack2, Jonathan Campbell3

* Presenting author

1) National Oceanography Centre, Marine Physics and Ocean Climate, UK
2) National Oceanography Centre, Ocean Technology and Engineering, UK
3) Campbell Ocean Data (formerly of NOC), UK

In 2013 the RAPID-WATCH programme commissioned development of a new system for returning data from instruments on a mooring. Previous attempts of developing a system at the start of the RAPID programme proved unviable for the RAPID 26M array due to difficulties experienced with maintaining a surface expression on the tall (up to 5000m deep) moorings.
The new system does not have a permanent surface expression and instead uses data pods that are released at pre-defined times to relay the data stored on them to shore. A data pod system has previously been developed at NOC in the form of the MYRTLE-X lander. For this development we have built on this existing capability and our experience gained through inductive mooring telemetry by linking a mooring to the lander via an acoustic modem.
Data are collected from the instruments by a central hub (referred to as the buoy controller) on the mooring via inductive telemetry. The buoy controller then compresses the data and transmits it through the acoustic modem to the lander. The lander writes these data to the data pods which are intended to surface and transmit the data to shore using the Iridium satellite network.
Here we present details of the development along with the results of field trials in both the sheltered waters of a Scottish loch and in deeper water offshore of Gran Canaria.


Requirements for a real-time array to monitor the AMOC at 26°N
David Smeed1*, Darren Rayner1, Gerard McCarthy1, Ben Moat1, Eleanor Frajka-Williams2

* Presenting author

1) National Oceanography Centre, UK
2) University of Southampton, National Oceanography Centre, UK

One of the most striking features so far observed in the RAPID-MOCHA-WBTS time series was the downturn in 2009-2010 when the AMOC at 26°N was approximately 30% less than the long-term average. Subsequent studies linked this downturn to a reduction in heat content of the subtropical north Atlantic and other climate impacts. This suggests that it would be useful to obtain more frequent updates to the RAPID time series than the current system allows (since 2012 data are retrieved once every 18 months). Here we consider the value of using telemetry technology on a subset of moorings in the 26°N array to obtain seasonal updates to the AMOC time series. An analysis is presented that quantifies the impact on the estimated AMOC of each of the different measurements within the array. From this the accuracy with which the AMOC could be estimated using only data from telemetered moorings is determined.


spacer spacer spacer spacer spacer spacer spacer spacer spacer spacer spacer spacer spacer
nerc logo logo logo logo logo