Main


Description
Participants
Publications
Activities
Contact us
Links

Popular
science
material



NOClim I

MODULE B

Analyses of abrupt changes in the past

 

Principal investigator: Dr. Trond M. Dokken, Bjerknes Centre for Climate Research

Deputy : Svein Østerhus, Bjerknes Centre for Climate Research
Participants: From the Bjerknes Centre and the University of Bergen (UoB): Prof. Eystein Jansen, Dr. Carin Anderson Dahl, Dr. Mathias Moros, Dr. Helga Kleiven, Dr. Ulysses Ninnemann and one Ph.D student (from 2003) financed by the University of Bergen. National participants: Dr. Nalan Koc, Norwegian Polar Institute (NPI).
UK collaborators and link to the Rapid Programme: Profs. I. Nick McCave, and Harry Elderfield, University of Cambridge.

Main Objective
The main objective of module B in NOClim phase II is to unravel the chaining of events that occurred through known rapid changes in the MOC (Meriodional Ocean Circulation). The main method will be to investigate the relevant processes that led to large and abrupt amplitude climatic shifts, as well as those which could induce "climatic surprises". Within this module we wish to improve our understanding of thresholds and non-linearities in the complex cause-response properties of the changing MOC. The project focuses on the determination of changes in hydrographic structure and estimation of the transport of water masses in the past through the main gateways connecting the Nordic Seas and the North Atlantic.

Deliverables
·Calibrate proxy data collected from surface samples with present day hydrography
·Identify each event in all the cores that will be subject for detailed investigation and chemical/sedimentological/biological analysis
·Estimate the vertical temperature and salinity profiles through the main inflow and outflow areas between Iceland, the Faroes and Shetland during three major climate transitions
·Quantitative reconstructions of the transport of surface and deep water through the channels connecting the Nordic Seas and the North Atlantic during climate transitions.

1. BACKGROUND
Our ability to understand the potential for future abrupt changes in climate is limited by our lack of understanding of the processes that control them. The climate system seems to operate in quasi-stable modes, and may suddenly switch from one mode to another within a few decades. Series of operative modes that may be identified can be considered as stable on longer time scales, but recent evidence suggest that abrupt climate changes often occur when gradual causes push the earth system across a threshold. A few types of forcings external to the climate system are identified, and could operate as pacemakers of abrupt climate change. However, these forcings vary too slowly to be prime movers of abrupt change, but if the climate system itself exhibits a discontinuous response to continuous variations of some forcing parameters, then the possibility exists that external variations may determine the timing of events. It is therefore of major importance to identify candidates for feedback mechanisms in time and space that may respond to gradually changing external forcing mechanisms.
One unresolved question concerning rapid climate shifts in the past that may represent major feedback mechanisms, is in which way some minor forcing mechanisms result in significant changes in the Meridional Ocean Circulation (MOC). At present, a sudden change in the Thermohaline Circulation (THC) is the only well-developed theory for explaining abrupt climate changes. The global THC consists of: cooling-induced deep convection, brine rejection, and sinking at high latitudes, upwelling at lower latitudes, and horizontal currents feeding the vertical flows. In the North Atlantic/Nordic Seas, where much of the deep sinking occurs, the THC is responsible for the unusually strong northward heat transport, where part of this heat is imported from the Southern Hemisphere and the tropics. Much of this heat is given off to the atmosphere over the "Gulf Stream", from where it is transported northeastward by the atmosphere. This part of the heat loss is typical of all subtropical gyres and is not associated with the global overturn (Talley, 1999). The enhanced heat transport has been believed by many to contribute to the relative mildness of western European climate, particularly that of Scandinavia. However, the relative contributions to European climate of the THC, the wind driven circulation, atmospheric transport associated with land/ocean contrasts, atmospheric planetary waves, and so on, remain uncertain. In the experiments planned for this project, we expect to test: 1. the hypothesis that rapid climate transitions are always associated with changes in overturning rate (THC) in the Nordic Seas, and 2. whether changes in dense overflow water flux and properties are associated with changes in the inflow or heat flux of Atlantic Water to the Nordic Seas.

2. SCIENCE PLAN
We plan to study the behaviour of the MOC during specific rapid climate changes that took place with different initial conditions, where forcing mechanisms were different in each case studied, and where the climate transition is much more rapid than the apparently gradually-changing forcing. Based on this, we will identify changes in the main oceanographic variables: temperature, salinity, density and current speed that occur through the abrupt climate shifts. This will include identification of the degree to which the climate shift is linked to variations of the MOC, and if so identify possible feedbacks, in accordance with the overall science plan in NOClim.

Identification of climate transitions
The plan is to reconstruct oceanographic variables across three different climate transitions that are all easily detectable in the paleo-record, and where each climate transition represents a transition from different initial conditions. As an example, the Younger Dryas - Holocene transition represents an extreme case of climate change, where the climate was dramatically changed from almost full glacial conditions to a warm climate within a few decades. The two other climate transitions that are chosen are from the Holocene, where abrupt events with substantial climate changes took place when physical conditions on the earth were more similar to today. These represent climate transitions which are much less dramatic than the Younger Dryas transition. Understanding the causes of both types of abrupt climate change is essential for assessing the mechanisms leading to rapid climate changes.
·The transition from the Younger Dryas cold period to the present-day warm period (Holocene). In a direct continuation of the NOClim phase I project we can identify the response of the MOC to strong freshwater forcing.
·The Neoglacial transition, c. 5000-3000 BP, represents a climate transition from a relatively warm initial condition. The most significant change is seen in the hydrological cycle where tropical and subtropical areas changed from relatively humid climate to very dry conditions. This transition is identified as a cooling in the North Atlantic and in the Nordic Seas. The climate changes were probably forced by slow changes in summer insolation due to the orbital parameters, yet the response as recorded in many sedimentary archives was rapid. A change in the MOC is proposed as one possible mechanism, but this is as yet unidentified.
·The transition from the `Medieval Warm Period' into the `Little Ice Age', which may have been caused by solar variability. A feedback including the MOC has been implicated in this climate shift, but no hard evidence exists to support or falsify this contention.
Areas of investigation
We will focus on the variability in the inflow and outflow of water masses across channels that connect the Atlantic Ocean and the Nordic Seas, in particular the Faroe-Shetland Channel (section S), the Faroe Bank channel (section W) and along one transect north of Iceland-Faroe Ridge (section N) (Fig. 1). One transect North of the Faroe-Shetland channel along the inflow axis of Atlantic Water will also be included in this study (Fig. 1). These transects will all cover thewater massses entering the Nordic Seas that has its origin in the North Atlantic Current (NAC), and will also cover the main outflow route of deep and intermediate water generated in the Nordic Seas (Figs. 2 and 3).

Figure 1: The main transects to be studied in module B. The black lines represent sections also covered with CTD profiles since 1994, and orange circles and rectangles represent ADCP mooring stations. The red line indicates the transect along the inflow axis of Atlantic Water.


Figure 2: The main surface currents in the North Atlantic and the Nordic Seas

 

Figure 3: The water balance for the Arctic Mediterranean.

At present, about 3.5 ±0.5 Sv of Atlantic Water enters the Nordic Seas between Iceland and the Faroes, and the same amount inflows between the Faroes and Shetland. This is about 80% of the total volume flux entering the Nordic Seas and the Polar Ocean through all the gateways connecting these basins, including the Bering Strait (Fig. 3).
All these transects have been subject to detailed CTD measurements during the last decades, and since 1994 permanent ADCP mooring stations have been covering the section lines. The modern hydrographic data set will serve as a unique calibration data set for paleo-proxy data that will be applied for reconstructing the temperature, salinity and density. Also through the Nordic World Ocean Circulation Experiment (WOCE) project, there exist reasonably well-constrained flux estimates for the water masses flowing through these gateways.

 


Experiment: Reconstructing velocity and transport using geostrophic methods
Traditionally the discipline of paleo-oceanography has been somewhat disconnected from the field of physical oceanography. One obvious reason is that within paleo-studies it has not been possible to produce data that could be placed in the context of modern observations of the physical properties of water masses in the ocean for quantitative reconstructions of water mass flow. During the recent years, good proxies for determining certain properties of seawater in the past have been developed, and the resolution of paleo data sets has been significantly improved. Chemical, physical and biological proxies in combination can be used for reconstructions of temperature, salinity, density and relative current-strength of certain water massses, and also to distinguish between different water massses. This knowledge, however, has not been put into the context of physical oceanography, in terms of quantifying the transport of water masses. To secure a strong connection between the different disciplines, the module leader (T. Dokken) and deputy leader (S. Østerhus) are chosen to represent the paleo-oceanography and physical oceanography community, respectively.
Here we propose to use a newly-developed method for estimating transport of water masses in the past (Lynch-Stieglitz et al. 1999 a and b) via the reconstruction of the density field of the water column through major gateways connecting the Nordic Seas and the Atlantic Ocean. We can construct vertical density profiles based on temperature and salinity reconstructions from cores collected from a range of depths from either side of the channels. The combination of well calibrated Mg/Ca measurements and oxygen isotope measurements will give parallel temperature and salinity estimates (e.g. Elderfield and Ganssen, 2000, Martin et al. 2002). The vertical shear of large-scale ocean currents can be reconstructed from density gradients within the ocean using a geostrophic approach. The vertical distribution of transport, and the net transport relative to an assumed reference level (level of no motion), can be calculated through a section defined by only two vertical density profiles. This approach can be used for both the Faroe-Shetland channel and the Faroe Bank channel by mapping the temperature and salinity from each side of channel to produce two profiles estimating the density structure in the water column at specific times. The method we are proposing to use here is very similar to the method Lynch-Stieglitz and co-workers (1999) used in the Florida Strait for estimating the transport of surface water during the last glacial maximum. The main procedure for creating these vertical profiles is illustrated in Figure 4.
This method will not determine an absolute vertical profile, but will give an integrated signal of density over the entire slope side. Where the surfaces of constant density are nearly horizontal, the constructed vertical density profile will be similar to a true vertical profile, but where density isolines are very tilted this may bias the data. However, these results depend to a large extent on the depth interval of the pycnocline (or thermocline). In this area there is a very sharp pycnocline (and thermocline) and the distance between the core locations defining the main density (temperature) transition is rather short, and the water masses below and above the main pycnocline are rather homogenous, and very different from each other. Therefore the constructed vertical profile will closely image a true vertical profile.
Once we have a pair of density profiles we can make the following observations about the density field itself including T/S properties and slopes of the density surfaces. We then proceed to compute the Fofonoff potential energy anomaly (PEA) for each profile (Fofonoff, 1962). The PEA method, as do all geostrophic methods, requires an assumption of a level of no motion in order to give absolute flow. However, under all circumstances the method will give us vertical velocity shear. To the extent that it is the exchange we want, this shortcoming in the dynamic method may not be so serious. As experience is accrued, one can expect to improve the estimates. For example, in the Faroe-Shetland channel, we have warm water flowing in, and cold water flowing out. In the Faroe Bank Channel, the proportions are quite different, with principally an outflow. The requirement of continuity of the deep-water component through both sections will help to constrain the system.

A somewhat different approach has to be applied for the section north of the Faroe Islands, tracking the inflow over Iceland-Faroe Ridge. Here we do not have two slope sides bracketing the inflow current, which can be used to define two vertical density profiles in the edges of the stream. The iso-lines for temperature and salinity, typically defining the lower limit of Atlantic Water in a cross-stream profile through the main axis of inflowing Atlantic Water, are presently rather deep (500m) on the continental margin side toward the Faroe Islands, and outcrop to the surface northward, defining a frontal zone (the Arctic Front). By studying cores along a transect downslope perpendicular to the main stream, we can reconstruct the depth of Atlantic Water influence on the slope, back in time. At the same time, we can estimate the position of the Arctic Front at the surface, by studying planktonic foraminifera further northward along the same transect perpendicular to the main stream. Consequently, for every climate transition defined in this study, we can define the slope and changes in the slope of any given isotherm, and calculate the transport of inflowing Atlantic Water.

Finally, we will map the density gradient in the water column from the Faroe Bank Channel through the Faroe-Shetland Channel, and upstream along the axis of inflowing Atlantic Water, for the same time slices as defined above. This is a similar approach, in a paleo-perspective, to that presented from observational data by Hansen et al. (2001), which showed that the density structure of the water column is related to the volume flux of cold, dense overflow water from the Nordic Seas into the Atlantic Ocean.

Relevant additional questions to be addressed are:
·What are the patterns of environmental variability associated with abrupt climate change in the tropics and high latitudes?
·What are the teleconnections between the tropics and the Northern Hemisphere high latitudes?
·What is the role of freshwater cycling in abrupt climate change?
·What feedback processes are dominant, and what are their roles in causing the persistence of abrupt climate changes?

3. DATA MATERIAL AND CRUISE PLAN
Some data already exist from all these transects described above. Existing data confirms that we will be able to detect all the climate transitions planned for our experiment. Some cores are available at the University of Bergen and some cores will be shared with our collaborating partners at the University of Cambridge. However, this study requires a large data set of good quality cores, and during a four years programme we will need to arrange three cruises. This study also requires good surface samples for calibration of proxy data with observational data.
One cruise is planned in 2004 within the UK programme `RAPID', and two cruises will be arranged by the University of Bergen/Bjerknes Centre. Research cruises are planned to be arranged in June 2003 with the research vessel Håkon Mosby (20 days), and in June 2004 with the new research vessel in Bergen, G.O. Sars (20 days).


4. INTERNATIONAL COLLABORATION
Downstream variability, along the route of deep water flow from the Nordic Seas into the Atlantic Ocean, will be tracked by a study of current-sensitive sedimentological proxies which will be an indirect measure of the current strength of the outflowing deep water. The latter study will be conducted in close collaboration with Professor Nick McCave´s group at Cambridge University, planned within the RAPID Programme. Professor McCave will be the principal investigator for all proxy development in the Faroe Bank channel.

At the Bjerknes Centre an ICP-AES lab will be established in early 2003 for measurements of Mg/Ca ratios in foraminifera. Development and refinement of the Mg/Ca method is currently being performed in an ongoing EU-project (CESOP). This will establish common procedures and protocols for preparation and calibration, supervised by Prof. Harry Elderfield at the University of Cambridge, which has one of the foremost laboratories in the world for this type of geochemistry.

Another EU-project (PACLIVA) will start in the autumn of 2002, focussing on climate variability of the past 2 millennia in the North Atlantic. The same UK partners are part of PACLIVA and data for the "Medieval Warm Period" into the "Little Ice Age" (MWP-LIA transition) will be forthcoming from PACLIVA.

5. CONNECTION TO OTHER CO-ORDINATED PROJECTS FROM THE RESEARCH COUNCIL
The project design of module B is very different from any projects described in the proposal for the next phase of NORPAST. However, this project may benefit from some of the results from NORPAST. There will therefore be contact between different components of marine research in NORPAST and module B in NOClim phase II.

6. MILESTONES
As mention above, this study relies very much on obtaining a large data set of good quality. It will require manpower and laboratory facilities, and accurate measurements of Mg/Ca and oxygen isotopes. The study will go forward in the following steps:
·Core collection, sampling and initial measurements of surface samples (year 1)
·Calibration of the proxies required for precise estimation of key hydrographic variables by comparing analyses from surface sediments with present day hydrography in the region (year 2)
·Identification of time slices (year 2)
·Measurement of proxy data for every time slice (year 2 and 3)
·Calculation of the geostrophy for all sections and every time slice (year 2, 3 and 4)
·Main conclusions (year 4)

A more detailed plan is illustrated in the table below:

Activities

2003

2004

2005

2006

Cruise preparation

xxx---------

xxx---------

----xxx-----

 

Core selection

xxx---------

xxx---------

 

 

Core collection

----xxx-----

----xxx-----

 

 

Isotope measurements

--xxxxxxxx-

xxxxxxxxxxx-

xxxxxxxxxxx-

 

Setup of Mg/Ca facilities in Bergen

xxxxxx------

 

 

 

Calibration of Mg/Ca measurements with Cambridge

----xxxx----

x

x

 

Mg/Ca measurements

------xxxxxx

xxxxxxxxxxx-

xxxxxxxxxx--

 

Foraminifer based temperature reconstr.

-xxxxxxxxxxxx

xxxxxxxxxxx-

xxxxxxxxxxx-

 

Diatom based temperature reconstr.

----xxxxxxx-

xxxxxxxxxxx-

xxxxxxxxxxx-

 

Transformation of proxy data into SST, SSS and density data

-------xxxx-

xxxxxxxxxxx-

xxxxxxxxxxx-

xxxx

Calibrate and test proxy data with modern observations

 

xxxxxxxxxxx-

 

 

Calculate geostrophy

 

------xxxx--

xxxxxxxxxxx-

xxxxxx

PhD activity

----xxxxxxxx

xxxxxxxxxx-

xxxxxxxxxxx-

xxxxxxxxxx-

Reports/evaluation

-----------x

-----------x

-----------x

-----------x

Publication

 

---------xxx

xxxxxxxxxxx-

xxxxxxxxxxx-

7. BUDGET

Not available

8. REFERENCES

·Elderfield, H.M. and Ganssen, G. (2000): Past temperatures and d18O of surface ocean waters inferred from foraminiferal Mg/Ca ratios, Nature, 405, 442-445.
·Fofonoff, N.P. (1962): Dynamics of ocean currents. In The Sea: Ideas and Observations on Progress in the Study of the Sea, vol. 1, Physical Oceangraphy (ed. M.N. Hill), p. 323-395, John Wiley, New York.
·Hansen, B., Turrell, W.R. & Østerhus, S. (2001): Decreasing overflow from the Nordic Seas into the Atlantic Ocean through the Faroe Bank channel since 1950. Nature, 411, 927-930.
·Lynch-Stieglitz, J., Curry, W.B. & Slowey, N. (1999): A geostrophic transport estimate for the Florida Current from the oxygen isotope composition of benthic foraminifera. Paleoceanography, 14, 360-373.
·Lynch-Stieglitz, J., Curry, W.B. & Slowey, N. (1999): Weaker Gulf Stream in the Florida Strait during the Last Glacial Maximum. Nature, 402, 644-648.
·Martin, P.A., Lea, D. W., Rosenthal, Y., Shackleton, N., Sarnthein, M. and Papenfuss, T. (2002) Quaternary deep sea temperature histories derived from benthic foraminiferal Mg/Ca. Earth and Planetary Science Letters, 198, 193-209.
·Talley, L.D. (1999): Some aspects of ocean heat transport by the shallow, intermediate and deep overturning circulations. In Mechanisms of Global Climate Change at Millennial Time Scales. Geophysical Monograph Series. (Eds. Clark, Webb and Keigwin). American Geophysical Union, 112, 1-22.
·Østerhus, S., Turrell, W.R., Hansen, B., Lundberg, P. & Buch, E. (2001): Observed transport estimates between the North Atlantic and the Arctic Mediterranean in the Iceland-Scotland region. Polar Research, 20, 169-175.

Top of Page