Main


Description
Participants
Publications
Activities
Contact us
Links

Popular
science
material


NOClim I

MODULE A

Theory and modelling of the meridional oceanic heat transport

Principal Investigator: Helge Drange, Nansen Environmental and Remote Sensing Center (NERSC) and Dep. of Geophysics, Univ. of Bergen (GfI-UiB).

Project description by: Helge Drange (NERSC), Joe LaCasce (met.no), Arne Melsom (met.no), Ole Anders Nøst (Norwegian Polar Institute), Tore Furevik (GfI-UiB), Peter M. Haugan (GfI-UiB),

Major Objective
To examine the forcing, structure and sensitivity of the Atlantic Meridional Overturning Circulation in response to buoyancy forcing, internal mixing and wind driving, in idealized and realistic basin configurations

1. Background
The stability and variability properties of the Atlantic Meridional Overturning Circulation (AMOC) have been the focus of a series of observational and numerically based studies. There are several reasons for the general interest in the AMOC: Time series of temperature proxies from polar ice caps and marine sediments (Bond et al. 1993; Clark et al. 2002) indicate that the North Atlantic-European region has experienced changes in the surface temperature of 5-10 ?C on multi-annual to decadal time scales, the last warming taking place under the transition from the last glacial maximum to the Holocene. These transitions, known as Dansgaard-Oeschger (rapid warming) and Heinrich (rapid cooling) events (Alley 1998), have been linked to variations in the strength of the AMOC (Clark et al. 2002).
For the present day climate system, the Atlantic Ocean transports about 1.2 PW of heat poleward of 25 ºN, or 20-30% of the total heat flux carried by the atmosphere-ocean system at this latitude (Hall and Bryden 1982). The northward heat transport in the Atlantic Ocean is therefore one of the reasons for the anomalously mild climate over large parts of the North Atlantic and Europe (e.g. Rahmstorf and Ganopolski 1999). Consequently, changes in the AMOC have the potential to change the climate in this region, and this is considered as one of the major challenges in climate change (O'Neill and Oppenheimer 2002): How stable is the AMOC to human induced greenhouse warming, and is a weakening or even a (near) collapse of the AMOC likely to occur in the near future? Model results are by no means conclusive regarding this question (Cubasch et al. 2001).
Direct observations of the strength and variability of the AMOC have not been conducted, mainly due to the 3-D basin scale extent of the circulation. In the simplest description of the system, warm and saline surface waters of tropical-subtropical origin flow poleward. During the northward flow, heat is lost to the atmosphere, leading to a gradual increase of the density. At locations in the Greenland-Iceland-Norwegian (GIN) and Labrador Seas region (Dickson and Brown 1994), the surface water may become sufficiently dense during late winter, triggering convective mixing to intermediate or abyssal depths. These dense and cold water masses, together with subsurface water masses of polar origin flowing southward through the Fram Strait, constitute the major source waters of the southward deep-flowing branch of the AMOC.
There is observational-based evidence for a decreasing trend in the southward flow of dense waters from the Nordic Seas (Hansen et al. 2001). A freshening of the overflow waters is also observed (Dickson et al. 2002). Since these water masses are part of the AMOC, this may indicate that the strength or the structure of the circulation is changing. Furthermore, analyses of observed hydrography from the Atlantic sub-tropical and sub-polar gyres indicate that the AMOC-system exhibits strong variations on multi-annual to decadal time scales (e.g. Curry and McCartney 2001). It is therefore hard, if at all possible, to assess whether the observed changes of the properties of the overflow waters reflect natural variability modes, or are signals of human-induced changes in the climate system.
Using 3-dimensional atmosphere-ocean general circulation models (AOGCMs), simulations of future climate scenarios exhibit a 30-50% reduction, or a near stabilization of the AMOC at the end of the 21st century (Cubash et al., 200). Furthermore, climate simulations going beyond 2100, usually performed with models of intermediate complexity, show generally a reduced or even a complete shutdown of the AMOC (Schmittner and Weaver 2001).
The obtained differences in the model-predicted evolution of the AMOC may have a series of explanations. First, the fate of the AMOC may depend on the maximum atmospheric CO2 concentration and on the rate of CO2 increase (e.g. Stocker and Schmittner 1997; Clark et al. 2002) (the various emission scenarios become more uncertain, and consequently diverge, with time). Secondly, the simulated stability of the AMOC may depend on the actual model system (GCM versus model of simplified complexity), the horizontal and vertical resolution, the parameterisation of unresolved topographic features and horizontal/vertical mixing processes, the realism of the sea-ice component, the treatment of continental run-off, the ability for the model to describe the nature of the major natural climate variability modes, the use of heat and/or fresh water flux adjustments, the actual spatial temperature and salinity distributions in key regions (especially in the sinking regions of the North Atlantic Ocean), the basin scale vertical density distribution, and so on.
It is clear from the complexity and the still incomplete theoretical understanding of the system (e.g. Yang 1999; Jayne and Marotzke 2001; Johnson and Marshall 2002), that in-depth studies involving theoretical and observational analyses, together with (idealised and realistic) model studies, are required to significantly improve our understanding of the driving mechanisms and stability properties of the AMOC.
Improved knowledge of the present-day system has the potential to reduce the uncertainty in climate change predictions in general, and for the Atlantic-European climate system in particular. In addition, improved understanding of the major natural variability modes is a prerequisite for performing - if possible - multi-annual to decadal predictions of the Atlantic-European climate.
Work addressing the forcing, structure and stability of the modern AMOC have not received too much attention in Norway up to now, and only a few papers by Norwegian scientists exist on the topics (e.g. Mauritzen and Häkkinen 1999, Bentsen et al. 2002, Zhou et al. 2002). We propose to build such an expertise in NOClim Phase II by funding several national research institutes, and by a two-way visiting plan. As the available funding is limited, the work will be bridged with existing funding to ensure that the activities have a sufficient focus to meet the stated goals.

Methods and tools
To develop Norwegian expertise in the field of idealized numerical model study of the AMOC, a suitable tool is required. The project also aims for more realistic simulations of the AMOC, which at the same time calls for a full primitive equation (PE) model. Given the demonstrated sensitivity to lateral and vertical diffusion in previous studies, it is believed that an isopycnic model (one with density as the vertical coordinate) is a natural choice. Specifying diapycnal fluxes is straightforward in an isopycnic model, thereby reducing spurious mixing which one usually finds in standard level (or z-coordinate) models. Norwegian scientists already have experience with the Miami Isopycnic Coordinate Ocean Model MICIOM (e.g., Furevik et al. 2002, Bentsen et al. 2002, Zhou et al. 2002). Therefore, MICOM - and the recent modification of MICOM named HYCOM - will be used as the major modelling tools in Module A. In addition, and for direct comparison with earlier studies, comparisons will be made with a dynamically simpler planetary geostrophic (PG) model. For the idealized model experiments, various features embedded into the HYCOM system will be switched off. For instance, the first task will be to adapt the HYCOM model to an idealised, hemispherical "wedge" geometry. In its current release the model is configured for the North Atlantic, but this version should be amenable to the idealized basin (E. Chassignet, pers. comm. 2002). This is the same configuration for the PG model, which will facilitate comparisons between the two.
The tasks outlined below represent a suite of experiments of increasing complexity. They are intimately related, and the results of one will benefit the others. As in many previous studies, the first tasks will focus on the AMOC in a flat-bottom ocean. Obviously the ocean is not flat, but in beginning this way, we will be able to confirm that our experiments are in line with previous studies. Moreover, we believe there are issues even in this simple environment which deserve more attention, specifically the question of model resolution.
Thereafter, we alter the model to make it more realistic. The first major step will be to add topography. There is substantial observational evidence for ocean currents being steered by topography, most frequently from subsurface float and surface drifter data because these indicate the actual motion of fluid parcels. By examination of the influence by topography and f/H (the ambient barotropic potential vorticity, or PV) on subsurface floats, LaCasce (2000) found a strong topographic influence throughout the water column in the North Atlantic. Similar indications of topographic control have been given by Fratantoni (2001) from surface drifters in the North Atlantic, and by Poulain et al. (1996) and Jakobsen et al. (2002) from surface drifters in the Nordic Seas.
How can topography affect the AMOC? First, topography can alter the vertical character of the circulation (e.g. Pedlosky, 1987). Second, topography can alter the stability of baroclinic mean flows, which can in turn affect variability (Winton, 1997). Third, topography can alter the path of currents (as suggested by the Lagrangian observations) and thereby affect atmospheric density fluxes. And fourth, sills and ridges could obviously block baroclinic flows, in turn altering the location and perhaps magnitude of convection.
After examining the topographic role, we turn to the winds. In the initial experiments, we will use either steady winds or no winds at all. This is because we wish to examine in isolation the affects of oceanic instability, without the added distraction of wind variations. In reality of course the atmosphere is fully variable, and its changes will be directly reflected in the ocean. So we next turn to variable forcing.
It is widely debated whether the ocean, atmosphere or both are responsible for climatic variations. Of the two, the ocean has the much larger heat capacity and longer adjustment times, so it was long thought that the ocean forced the long-term changes. In fact, idealized ocean models driven by stochastic (white noise) wind forcing exhibit decadal variability (e.g. Frankignoul et al., 1997). On the other hand, it has also been shown that a model atmosphere with fixed oceanic conditions can exhibit persistent climatic states (James and James, 1989) and even features resembling observed modes of variability, like the North Atlantic Oscillation (NAO) (Barnett, 1985). Nature is possibly somewhere in between. The atmosphere exhibits a red spectrum (Deser and Blackmon, 1993) which is quite different from that obtained with atmosphere-only models and also different from spectra found in ocean-only models with stochastic forcing. The question of the respective roles of atmosphere and ocean in climate change is still debated.
Atmospheric variations encompass both wind and buoyancy forcing. Changes in buoyancy forcing would directly impact the AMOC by altering the convected water and therefore the large-scale density gradients. But wind fluctuations should also affect the AMOC, by altering the upper ocean gyres and also possibly by changing the ocean vertical diffusivity (Munk and Wunsch, 1998). One could thus imagine idealized experiments in which one of these forcings was varied and the other held fixed.
But such experiments are obviously artificial. Consider just one aspect, the effect of mid-latitude sea surface temperature (SST) anomalies on the atmosphere. While this is still much debated (Frankignoul, 1985; Rahmstorf and Willebrand, 1995; Kushnir and Held, 1996; Elliott et al. 2001 and references therein), its impact on a coupled system would be quite different than that on an ocean-only system. Consider for instance: if a warm SST anomaly causes low level winds to intensify, those stronger winds could then alter the surface temperature. If the SST anomaly then intensified, the winds would grow, etc. Such feedback (seen in the idealized model of Goodman and Marshall, 1999) would produce unstable growth, perhaps causing this circulation mode to dominate all others.
Hereafter, we present more detail with regard to the tasks outlined above. We emphasize though that the goals are two-fold: to improve our own competence in the area of idealized models of the AMOC, and to enhance the understanding of fundamental dynamical elements related to it. Both should benefit the state of climate modelling in Norway, and - during the project period - put Norwegian scientists at the forefront of the topic.

2. Tasks
Each task constitutes a self-contained entity aimed at assessing the importance for the forcing, variability and stability of the AMOC. The results of each task represent a significant contribution to the overall objective of Module A, and are relevant for the overall aim of NOClim Phase II. Each task will have a small set of milestones for targeting the progress of work in time as detailed in Table 6 in the formal application form.

2.1 Task descriptions
Task A.1 Establish an idealized model to study the AMOC (months 1-12) (met.no)
The initial work will be oriented toward reproducing prior results, albeit with a model (HYCOM) not used before in this context. Comparisons will be made with a dynamically simpler planetary geostrophic (PG) model (Samelson and Vallis, 1997).
The first task will therefore be to adapt the HYCOM model to an idealised, hemispherical "wedge" geometry. The same configuration will be used for the PG model, facilitating comparisons. Companion experiments will be performed with the HYCOM and PG models, with boundary conditions consistent with previous studies. This will give us confidence that we can reproduce the same circulations. In addition, no one has to our knowledge performed a side-by-side comparison of these two models. Such a comparison will be useful to quantify the shortcomings of the PG model, and also to identify the parameter ranges where the two models agree. Knowing how well the PG system mimics the full system will be useful for knowing when to use the PG model in simplified climate experiments, as well as for developing analytical models.
Milestone A.1.1: Idealized set-up of HYCOM (months 1-3); A.1.2: Perform companion experiments (months 3-8); A.1.3: Analyse results (months 8-12); A.1.4: Write technical report/article (months 8-12).

Task A.2 Quantify role of applied model resolution (months 12-18) (met.no)
Here the question of model resolution will be addressed. Recently, Marotzke (1997; with a PE model) and Samelson (1998; with a PG model) examined the effect of localized vertical density mixing. Vertical diffusivity plays an important (and possibly controlling) role in determining the strength of the overturning. The observations however show unequivocally that vertical mixing is highly inhomogeneous in the ocean (Munk and Wunsch, 1998 and refs. therein), and this spatial variation should impact overturning; the results of Martozke and Samelson confirm this. Confining diffusion to the lateral boundaries, for example, causes all the upwelling and downwelling to occur there, yielding a boundary-intensified circulation quite different from classical perceptions (Stommel and Arons, 1960).
The studies of Marotzke and Samelson were made with coarse resolution models (grid spacing of several degrees of latitude and longitude). It is of interest to assess the effect of higher resolution, with correspondingly narrower boundary layers at the walls. In previous works on resolution dependency (Bryan, 1991; Fanning and Weaver, 1998), the thermohaline circulation becomes unstable at higher resolution (if the horizontal diffusivity is simultaneously decreased). Instability means time variability, and so is of interest for climatic variations.
Pertinent questions include: does the model become unstable? If it does, how does this instability manifest itself (for instance through the instability of a boundary current)? What is the role of eddies, if any, in heat transport? And is the spectrum of the response smooth, or are there dominant frequencies? Besides being of numerical interest, these higher resolution simulations are relevant to the ocean, with its localized mixing and very large Reynolds number.
Relation to Overall aim: Tasks A.1 and A.2 will address the Overall aim in Module A by establishing basic experience that is needed for understanding the AMOC, and examining the role of phenomena on various spatial scales. Milestone A.2.1: Perform model integrations with increased resolution (months 12-14); A.2.2: Analysis of results (months 14-18); A.2.3: Write report/article (months 14-18).

Task A.3 Examine role of topography (months 18-48)
To address this issue, we propose to extend the study in Tasks A.1 and A.2 by adding topographic perturbations to the idealized geometry. A flat bottom control run will be made, forced by a (steady) density flux and an idealized wind stress; then we will methodically alter the topography. Possible choices include a continental slope (e.g. Winton, 1997), to assess model stability, and a subcontinent (e.g. Greenland), to see whether the model develops multiple convection regions (like the Labrador and Nordic Seas). We would also examine an idealized ridge (like the mid-Atlantic ridge) and an idealized sill (like the Iceland-Scotland sill). And eventually we would consider a topographic configuration like that in the North Atlantic and Nordic seas.
In all cases, we would gauge alterations in the AMOC, in the current paths, in convection and in stability. In addition, we will examine closely variation in the bottom velocities (without which there is no topographic influence); as noted, there remain questions about the abyssal flow even in flat bottom simulations. For topographic steering to take place there must be geostrophic velocities near the bottom. Over steep topography these velocities will follow contours of f/H. Turning of the velocity with depth will only occur if vertical velocities is induced in the water column, or if the density is not materially conserved in the flow (see Stommel and Schott, 1977; Schott and Stommel, 1978). The flow may turn unhindered across a level of no motion (the turning is then 180 degress, see Hughes and Killworth, 1995), but this means that the flow is still following bathymetric contours. Because the possibility of turning with depth is limited also the surface flow is observed to follow topographic contours (Poulain et al. 1996; LaCasce 2000; Fratantoni 2001; Jakobsen et al. 2002).
One question to be investigated is why the Gulf Stream system in the Atlantic is topographically steered (That it is topographically steered is quite obvious from the drifter analysis of Fratantoni, 2001). Among the processes that may drive the near bottom flow in these regions, there are two possibilities to be studied:
a) In a classical circulation picture the dense water that sinks in the polar regions and spreads southwards along the bottom. Are the bottom geostrophic velocities forced this way able to explaing the topographically steered Gulf stream system? This will be investigated with the PG model including a simplified topography. The aim will be to investigate if the southward spreading dense water may cause the northward flowing warm water to be topographically steered.
b) The results of Nøst and Isachsen (submitted to JMR 2002) suggests that the potential vorticity input by winds in the Nordic Seas and Arctic Ocean (on long time scales) is transmitted to the bottom forcing the near bottom currents and causing strong topographic steering. Could this happen in the Gulf stream system in the Atlantic Ocean? One possibility is that mixing caused by baroclinic instabilities in the Gulf Stream system may lead to fluxes of potential vorticity that connect the surface wind forcing with the near bottom layers. The topography itself may also influence the baroclinic instabilities (see Pedlosky 1962, "Baroclinic instability in two layer systems") and the effect of this will also be studied. This may cause bottom geostrophic velocities and topographic steering of the whole water column. This problem will be addressed by studying the effect of baroclinic instabilities over a sloping bottom using simplified process models. This may be models we develop ourselves or a simplified setup of an existing model. The results of these studies will be attempted to include in the PG model used in 1a.
If there is topographic steering, a ridge at the sea floor may effectively block transport across it. However, diapycnal mixing or possibly the beta effect may lead to a turning of the direction of the current with depth (see Stommel and Schott, 1977; Schott and Stommel, 1978; Hughes and Killworth, 1995). The processes leading to turning of the velocity vector with depth should be studied in order to understand where the northward flowing warm water near the surface is topographically steered, and where it may flow across f/H contours. This will be done studying an idealized underwater ridge, with either self developed process models or simplified setup of an existing model. Answers to the following questions will be sought: How will the beta effect and the mixing of water masses across the ridge affect the turning angle of the current with depth? Can different mixing parameterizations affect this turning angle? On what conditions will the blocking of the currents be most/least efficient?
The AMOC is probably to a high degree controlled by topographic steering. By understanding the processes forcing the near bottom currents and the processes controlling the blocking of baroclinic currents we may be able to understand and predict possible changes to the system caused by a changing forcing fields. In the end of the project we should study possible changes in the forcing fields that could increase or decrease the effect of topography. The change in the forcing fields should be selected from the results of the studies described above. These study will be done using a PG model.
The work described above is highly relevant for the work in A4 and A2, as topography may be very important in localising mixing and diffusion processes. Mixing and instabilities may be important for bottom velocities and topographic steering as well as baroclinic currents above topography. The work described may then increase our understanding on how the AMOC depends on mixing processes.
Relation to Overall aim: Task A.3 will address the Overall aim in Module A by shifting our focus from understanding the general dynamics of the MOC to a setting that resembles the Atlantic Ocean. Milestone A.3.1 (Melsom and LaCasce): Perform model integrations with topographic perturbations (months 18-24); A.3.2 (Melsom and LaCasce): Analyse results (months 20-30); A.3.3 (Melsom and LaCasce): Write report/article (months 24-30). A.3.4 (Nøst): Forcing of near bottom flow (months 12-18). A.3.5 (Nøst): Blocking of baroclinic currents (months 12-30). A.3.6 (Nøst): Long term changes in the AMOC (months 14-36).

Task A.4 Assess role of horizontal and vertical mixing parameterisations (months 1-48, NERSC)
Several GCM simulations show that the natural variability modes of the AMOC, and also the threshold between on and off modes of the AMOC, are critically dependent on the actual strength and spatial distribution of the vertical and horizontal mixing (Bryan 1987; Schmittner and Weaver 2000; Scott and Marotzke 2001). Consequently, Schmittner and Weaver (2000) argue that it is extremely difficult to assign a probability of future abrupt change in the AMOC, a question of potential key importance for the North-Western Europe.
In this task, a theoretical approach will be taken to assess the role of vertical and horizontal mixing as controlling agents for the AMOC. Thermodynamic and advection-diffusion considerations of the ocean circulation will be applied following the numerical experiments presented by e.g. Scott and Marotzke (2001). Special emphasis will be paid to mixing at thermocline and abyssal depths, and to boundary mixing versus planetary geostrophic vorticity balance in the ocean interior. The theoretical work will be accomplished by analyses of available hydrographic observations, microstructure and tracer release measurements of mixing in the ocean (e.g., Ledwell et al. 1993; Ledwell et al. 2000), and results from numerical ocean models of various complexities and resolutions.
The work will be carried out in close collaboration with Jochem Marotzke at the Southampton Oceanography Centre. Marotzke will visit Bergen for 2-3 weeks each year in the period 2003-2006. Likewise, Helge Drange (NERSC) will visit Marotzke and possibly the University of Reading (Rowan Sutton and co-workers) for 2-4 weeks each year.
Relation to Overall aim: Task A.4 addresses the Overall aim in Module A by assessing the AMOC dependence and stability to vertical and horizontal mixing. The obtained results will be of direct use to Task A.2 and A.3. Milestone A.4.1: Review existing literature (months 1-6); A.4.2: Develop theoretical framework for mixing at thermocline and abyssal depths (months 1-30); A.4.3: Develop theoretical framework for boundary mixing versus boundary mixing versus planetary geostrophic vorticity balance (months24-42); A.4.4: Writing of three papers (months 18, 30, 48).

3. Links to other modules of NOClim II
Module A will develop Norwegian expertise in the theoretical and numerical understanding of the variability and stability of the AMOC. It is linked to Modules B and C since both paleo and instrumental observations are needed for the theoretical and modelling work, and since the expertise will provide a dynamic understanding to the observations.

4. Links to other co-ordinated projects, free projects under KlimaProg, and to the Polar Climate research initiative
The activity proposed in this description will be an integrated part in a broader range of modelling work at the newly awarded Bjerknes Centre of Excellence in Climate Research. Also, the activity proposed under RegClim will be useful as the simulations there will be actively used and disseminated in Task A.4. Likewise, the findings from Module A will be used to improve the model and analyses tools in RegClim. NOClim Phase II and RegClim Phase II are therefore complementary activities.

5. Other national and international collaboration
The participating groups in Bergen, Oslo and Tromsø have continuous contact and collaboration with a series of international groups and projects. Projects of particular relevance are the EU-funded PRISM, PREDICATE, NOCES and TRACTOR projects. Particular collaboration exists with the Southampton Oceanography Centre, the University of Reading, Meteo France, and the University of Miami.

6. Scientific publications
We plan to prepare and submit at least 8 manuscripts to peer-reviewed journals over the four-year period, distributed over Tasks A.2-A.4.

7. Expected results of interest for other research
To build top international expertise in the study of the stability and variability of the AMOC. This expertise is a prerequisite for assessing the climate development in our region, both on land and in the ocean. Such an expertise is also needed to ensure balanced information to the public, and to act as a scientifically based and credible information source for politicians and the government administration.

8. Need for computer resources
The proposed activity is moderately demanding in terms of computer resources, and it is assumed that the required resources will be made available through the Norwegian High Performance Computing Consortium.

9. Budget
Not available

10. References
Alley, R. B. 1998. Paleoclimatology: Icing the North Atlantic. Nature, 392, 335-337
Barnett, T.P. (1985): Variations in near-global sea level pressure. J. Atmos. Sci., 42, 478-501.
Bentsen, M., H. Drange, T. Furevik and T. Zhou (2002), Variability of the Atlantic thermohaline circulation in an isopycnic coordinate OGCM, Clim. Dynam., submitted
Bond, G., W. Broecker, S. Johnsen, J. McManus, L. Labeyrie, J. Jouzel, and M. G. Bonani, 1993: Correlations between climate records from North Atlantic sediments and Greenland ice. Nature, 365, 143-147
Bryan, K. (1991): Poleward heat transport in the ocean: a review of a hierarchy of models of increasing resolution. Tellus, 43, 104-115.
Cessi, P. (2000): Thermal feedback on wind-stress as a contributing cause of climate variability. J. Climate, 13, 232-244.
Clark, P. U., N. G. Pisias, T. F. Stocker, and A. J. Weaver, 2002: The role of the thermohaline circulation in abrupt climate change. Nature, 415, 863-869
Cubasch, U. et al. (2001): Projections of future climate change. In J. T. Houghton, Y. Ding, M. Nogua, D. Griggs, P. V. Linden, and K. Maskell, Eds., Climate Change 2001: The Scientific Basis, pages 525-582. Cambridge University Press.
Curry, R. and M. S. McCartney (2001): Ocean gyre circulation changes associated with the North Atlantic Oscillation. J. Phys. Oceanogr., 31, 3374-3400
Deser, C. and M.L. Blackmon (1993): Surface climate variations over the North Atlantic Ocean during winter: 1900-1989. J. Climate, 6, 1743-1753.
Dickson, B., I. Yashayaev, J. Meincke, B. Turrell, S. Dye, and J. Holfort, M. (2002): Rapid freshening of the deep North Atlantic Ocean over the past four decades. Nature, 416, 832-837
Dickson, R. R. and J. Brown, 1994: The production of North Atlantic Deep Water: Sources, rates, and pathways. J. Geophys. Res., 99, 12,319-12,341
Fanning, A.F. And A.J. Weaver (1998): Thermohaline variability: the effects of horizontal resolution and diffusion. J. Climate, 11, 709-715.
Frankignoul, C. (1985): Sea surface temperature anomalies, planetary waves and air-sea feedbacks in the middle latitude. Rev. of Geophys., 23, 357-390.
Frankignoul, C., P. Muller and E. Zorita (1997): A simple model of the decadal response of the ocean to stochastic wind forcing. J. Phys. Oceanogr., 27, 1533-1546.
Fratantoni, D. M. (2001): North Atlantic surface circulation during the 1990's observed with satellite-tracked drifters. J. Geophys. Res., 106 , 22,067-22,094.
Furevik, T., M. Bentsen, H. Drange, I. K. T. Kindem, N. G. Kvamstø and A. Sorteberg (2002): Description and validation of the Bergen Climate Model: ARPEGE coupled with MICOM, Clim. Dyn., submitted
Goodman, J. And J. Marshall (1999): A model of decadal middle-latitude atmosphere-ocean coupled modes. J. Climate, 12, 621-641.
Hall, M. M. and H. L. Bryden, 1982. Direct estimates and mechanisms of ocean heat transport. Deep-Sea Res., 29, 339-359
Hansen, B., W. R. Turrell, and S. Østerhus (2001): Decreasing overflow from the Nordic seas into the Atlantic Ocean through the Faroe Bank channel since 1950. Nature, 411, 927-930
Jakobsen, P.K., M.H. Rikergard, D. Quadfasel and T. Schmidt (2002): The near surface circulation in the northern North Atlantic as inferred from Lagrangian drifters: variability from the mesoscale to interannual. J. Geophys. Res. (submitted).
James, I.N. And P.M. James (1989): Ultra-low frequency variability in a simple atmospheric model. Nature, 342, 53-55.
Jayne, S. and J. Marotzke (2001): The dynamics of ocean heat transport variability. Rev. Geophys., 39, 385-411
Johnson, H. L. and D. P. Marshall (2002): A theory for the surface Atlantic response to thermohaline veriability. J. Phys. Oceanogr., 32, 1121-1132
Kushnir, Y. And I.M. Held (1996): Equilibrium atmospheric response to North Atlantic SST anomalies. J. Climate, 9, 1208-1220.
LaCasce, J.H. (2000): Floats and f/H. J. Mar. Res., 58, 61-95.
Ledwell, J. R., A. J. Watson and C. S. Law (1993): Evidence for slow mixing across pycnocline from a open-ocean tracer release experiment. Nature. 364, 701-703.
Ledwell, J. R., E. T. Montgomery, K. L. Polzin, L. C. St. Laurent, R. W. Schmitt and J. M. Toole (2000). Evidence for enhanced mixing over rough topography in the abyssal ocean. Nature. 403. 179-182.
Marotzke, J. (1997): Boundary Mixing and the Dynamics of Three-Dimensional Thermohaline Circulations. J. Phys. Oceanogr., 27, 1713-1728.
Mauritzen, C. and S. Häkkinen (1999): On the relationship between dense water formation and the ``Meridional Overturning Cell'' in the North Atlantic Ocean. Deep-Sea Res., 46, 877-894
Munk, W. and C. Wunsch (1998): Abyssal recipes II: Energetics of tidal and wind mixing. Deep-Sea Res., 45, 1977-2010.
Munnich, M., M. Latif, S. Venze and E. Maier-Reimer (1998): Decadal oscillations in a simple coupled model. J. Climate, 11, 3309-3319.
O'Neill, B. C. and M. Oppenheimer (2002): Dangerous climate impacts and the Kyoto Protocol. Science, 296, 1971-1972
Pedlosky, J. (1987): Geophysical Fluid Dynamics, Springer-Verlag, 710 pp.
Poulain, P-M., A. Warn-Varnas and P.P. Niiler (1996): Near-surface circulation of the Nordic seas as measured by Lagrangian drifters. J. Geophys. Res., 101, 18237-18258.
Rahmstorf, S. and A. Ganopolski (1999): Long-term global warming scenarios computed with an efficient coupled climate model. Climate Change, 43, 35-367
Rahmstorf, S. and J. Willebrand (1995): The Role of Temperature Feedback in Stabilizing the Thermohaline Circulation. J. Phys. Oceanogr., 25, 787-805.
Samelson, R. M. (1998): Large-scale circulation with locally enhanced vertical mixing. J. Phys. Oceanogr., 28, 712-726.
Samelson, R. M., and G. K. Vallis (1997): Large-scale circulation with small diapycnal diffusion: the two-thermocline limit. J. Mar. Res., 55, 223-275.
Schmittner, A. and A. J. Weaver (2001): Dependence of multiple climate states on ocean mixong parameters. Geophys. Res. Lett., 28, 1027-1030
Scott, J. R. and J. Marotzke (2001): The location of diapycnal mixing and the meridional overturning circulation. J. Phys. Oceanogr. submitted
Stommel, H. and A.B. Arons (1960): On the abyssal circulation of the world ocean---I. Stationary planetary flow patterns on a sphere. Deep-sea Res., 6, 140-154.
Sura, P., F. Lunkeit and K. Fraedrich (2000): Decadal variability in a simplified wind-driven ocean model. J. Phys. Oceanogr., 30, 1917-1930.
Yang, J. (1999): A Linkage between Decadal Climate Variations in the Labrador Sea and the Tropical Atlantic Ocean. Geophys. Res. Lett., 26, 1023-1026
Zhou, T., T. Furevik, H. Drange and M. Bentsen (2002): Interannual Scale Adjustment of the North Atlantic Thermohaline Circulation to the Atmospheric Forcing in a Global Air-Sea Coupled Model, J. Climate, submitted
Winton, M. (1997): The damping effect of bottom topography on internal decadal-scale oscillations of the thermohaline circulation. J. Phys. Oceanogr., 27, 203-208.

Appendix - Description of PhD project module A

The PhD project will be devoted to study the role of horizontal and vertical mixing on the AMOC. The work will start by performing a literature review on the problem, including the papers by Bryan (1987), Winton (1997), Fanning and Weaver (1998), Schmittner and Weaver (2000), Jayne and Marotzke (2001), Scott and Marotzke (2001), Johnson and Marshall (2002), Nilson and Walin (2001), and Bentsen et al. (2003). This activity will be performed in close collaboration with Jochem Marotzke at SOC, Southampton.

The second step will be to apply theoretical considerations to assess the role of vertical and horizontal mixing as controlling agents for the AMOC. Thermodynamic and advection-diffusion considerations of the ocean circulation will be applied following the numerical experiments presented by e.g. Scott and Marotzke (2001). Special emphasis will be paid to mixing at the thermocline and at abyssal depths, and to boundary mixing versus planetary geostrophic vorticity balance in the ocean interior. The theoretical work will be accomplished by analyses of available hydrographic observations, microstructure and tracer release measurements of mixing in the ocean (e.g., Ledwell et al. 1993; Ledwell et al. 2000), and results from numerical ocean models of various complexities, resolutions and forcings. The latter analyses will be carried out in close collaboration with the BCM modelling group at the Bjerknes Centre of Climate Research in Bergen (Furevik et al. 2003).

Tentative work plan:
Month 1-6: Review of existing literature

Month 7-24: Develop theories of the role horizontal and vertical mixing have on the AMOC; writing of a peer-refereed paper.

Month 18-33: Analyse results from OGCMs of various complexities, resolutions and forcings, and compare with the theoretical findings; writing of two peer-refereed papers.

Month 34-36: Finalize PhD-work.