Geen diatitel

1
The oceanic conveyor belt, cartoon or
reality? Hendrik M. van Aken Physical
Oceanography Department, Royal NIOZ
Things should be made as simple as possible, but
not any simpler. Albert Einstein
2
The THC as Oceanic Conveyor Belt was proposed by
Gordon as inter-ocean exchange, popularized by
Broecker as a continuous phenomenon, and
sanctified by the UNEP( IPCC), the OU(NL) and
the NIOZ feestcommissie.
3
Hydrographic evidence for the THC
4

• My thesis is that
• The Oceanic Conveyor Belt is a scheme of the deep
  overturning circulation or THC which is too
  simplified to allow physical understanding.
• We simply do not yet understand enough of the
  deep overturning circulation to allow sweeping
  statements about the stability or instability of
  this system.
• I will give some examples to illustrate my case.

5

• Many oceanographers state that the Oceanic
  Conveyor Belt is a major component of the climate
  system.
• Michael Vellinga (Hadley Centre) simulated the
  consequences of a break-down of the Conveyor Belt.

Climate effects?
1.25X1.25 DS-2

• Pier Vellinga (Institute for Environmental
  Studies, VUA) warned against the stagnation of
  this ocean current .

Flip-flop to catastrophe
6
10
Heat Flux Density a smoking gun?
The major heat loss by the ocean is found at
40N/S where the strong wind-driven western
boundary currents are found, not in the Arctic!
7
What drives the global THC? Different answers
have been formulated on this question. Here is a
pair.

• The traditional North Atlantic centred
  explanation Cooling at the sea surface in the
  North Atlantic Ocean increases the density of sea
  water. Because of its high density, the resulting
  cold water will sink and spread at large depths
  through the World Ocean.
• The non-US Euro centred alternative In the
  western tropical Pacific the sea surface height
  is 50 cm higher than in the North Atlantic. This
  drives a flow of warm water from the Pacific
  towards the North Atlantic.

By the way, what do we mean with drives?
8
The THC is ultimately driven (kinetic energy
transfer) by meridional pressure gradients due to
differences in density, generated by differential
heat (Q) and fresh water (P-E) fluxes to the
atmosphere.
BUT DONT FORGET the ARCTIC RIVERS!
Density differences are mainly caused by
meridional differences in surface fluxes of heat.
Q
P-E
9
Resulting surface density
Differences of the Atlantic surface density are
mainly related to temperature differences.
But pole to pole differences are caused by
salinity differences.
10
Convective mixing is not down-welling.
D?100 km ? 2-4x1
Due to the rotation of the earth direct
down-welling is prohibited, a rim current
develops, and rim eddies are required.
11
The THC often proceeds through narrow gaps in the
topography, O(25 km?0.25?), and as fast narrow
flows, O(100 km?1?) over the ocean bottom. Their
effects on the global THC are not known.
12
Cold deep water is also formed the
Antarctic(Weddell Sea, Ross Sea Adelie
Polynia). The spreading of this AABW is clearly
missing in the popular view of the Conveyor Belt.
13
THC as a heat engine
A stationary THC can only exist if the kinetic
energy generated by the density differences
balances the (turbulent) friction, that is, if
(just like a steam engine)
This can only be so if the descending water has
higher density than the up-welling
water. Differential surface heating cannot drive
a stationary THC!
Note the the Queens THC was not stationary!
14
Turbulent mixing brings low density warm water to
deeper levels with higher pressure and mimics a
deep heat source. Without oceanic turbulent
mixing no THC!
Probably turbulent mixing in the ocean mainly
derives its energy from internal tides. So the
moon may drive the climate too.
15
The meridional overturning flow on a rotating
earth is in near geostrophic balance. This
requires a zonal density gradient. At the equator
this gradient has to change sign. This
geostrophic adjustment is generally absent in the
Conveyor Belt scheme.
A highly simplified picture of the zonal density
structure in the Atlantic ocean. The symbols
indicate the mean flow directions.
16
With a meridional volume flux of NADW of 15 Sv (1
Sv 106 m3/s) over a layer of 2000 m thickness
in the subtropical North Atlantic (B ? 5000 km)
the mean velocity will be of the order of 1.5
mm/s. Because of the rotation of the spherical
earth this water can only well up (2-3 m/year)
if the water in the ocean interior flows towards
the poles ( 1 mm/s), fed from a fast Deep
Western Boundary Current (10 cm/s).
Stommel
17
The back-flow of the THC at shallower levels is
highly un-certain in pathway. Note that the near
surface flow is mainly wind-driven, not density
driven! The ITF) has a mean temperature of only
15ºC and is relatively fresh (S?34.5).
Pacific
Over 60 of the ITF water flows north-wards near
Africa, not south-wards. There it is strongly
heated and salinified.
Ind.Oc.
18

• The inflow into the Atlantic Ocean may go around
  Africa via the warm and saline Agulhas Rings.
  This process is mainly determined by the
  large-scale wind field, not by the density
  distribution.

19

• The inflow into the Atlantic also may come via
  the Drake Passage as cold and fresh Antarctic
  Intermediate Water.

20
Existing estimates with a simple box model
downplay the importance of the North Atlantic
source for the deep circulation in the World
Ocean.
A more detailed box model shows a combination of
many interacting loops. All with different
essential dynamics.
21
Note that a combination of different loops easily
can give an impression of a single large
circulation. That does not only apply to the
Mozambique Channel, but also to the Oceanic
Conveyor Belt.
Numerical simulations of the THC indeed show that
the loops and the modification of water are the
main features of the back-flow to the North
Atlantic, not the steady, broad conveyor belt
transporting warm water to the Nordic Seas.
22
How certain are we of the climate effects of the
warm back-transport?
The West-European climate is a marine climate
anyway, receiving seasonally stored heat from the
ocean due to the existence of the Rocky Mountains.
23
Some conclusions

• The emphasis on the North Atlantic as source and
  cause of the global THC, expressed by the
  conveyor belt scheme, shows a geographic bias
  which is probably more connected with the
  concentration of oceanographers around this ocean
  basin, than with the physical reality.
• The conveyor belt cartoon does not facilitate
  physical insight. Instead it makes it more
  difficult to identify and understand key physical
  processes.
• The strongly simplified conveyor belt cartoon
  ignores so many of these physical processes that
  it should not be used in discussions on possible
  climate interactions and tele-connections.