Simulation of Oxygen Isotopes in a Global Ocean Model
A. Paul*, S. Mulitza, J. Pätzold and T. Wolff
Universität Bremen, Fachbereich Geowissenschaften, Postfach 33 04 40,
D-28334 Bremen, Germany
*corresponding author (e-mail): [email protected]
Abstract: We incorporate the oxygen isotope composition of seawater δ18Ow into a global ocean
model that is based on the Modular Ocean Model (MOM, version 2) of the Geophysical Fluid
Dynamics Laboratory (GFDL). In a first experiment, this model is run to equilibrium to simulate the
present-day ocean; in a second experiment, the oxygen isotope composition of Antarctic Surface
Water (AASW) is set to a constant value to indirectly account for the effect of sea-ice. We check the
depth distribution of δ18Ow against observations. Furthermore, we computed the equilibrium
fractionation of the oxygen isotope composition of calcite δ18Oc from a paleotemperature equation
and compared it with benthic foraminiferal δ18O. The simulated δ18Ow distribution compares fairly
well with the GEOSECS data. We show that the δ18Ow values can be used to characterize different
water masses. However, a warm bias of the global ocean model yields δ18Oc values that are too light
by about 0.5 ‰ above 2 km depth and exhibit a false vertical gradient below 2 km depth. Our ultimate
goal is to interpret the wealth of foraminiferal δ18O data in terms of water mass changes in the
paleocean, e.g. at the Last Glacial Maximum (LGM). This requires the warm bias of the global ocean
model to be corrected. Furthermore the model must probably be coupled to simple atmosphere and
sea-ice models such that neither sea-surface salinity (SSS) nor surface δ18Ow need to be prescribed
and the use of present-day δ18Ow-salinity relationships can be avoided.
Introduction
Isotopes in the Hydrological Cycle
Measurements of the isotopic composition of water at the various stages of the hydrological cycle
has enabled the identification of different water
masses and the investigation of their interrelationships; measurements of the variations of the isotopic composition of proxy materials in climate archives such as ice and sediment cores has become
a most useful tool in paleoclimate reconstructions
(Gat 1996; Schotterer et al. 1996).
The isotopic species of the water molecule that
are of most interest to the geosciences are
1 2 16
H H O, 1H218O and 1H3H16O. A water molecule
containing the radioactive isotope 3H (tritium) or
either of the stable isotopes 2H (deuterium) and 18O
(oxygen-18) is heavier than the water molecule
1 2 16
H H O that is by far the most abundant in the
natural assembly of all isotopic species. Therefore,
water vapor evaporating from the sea surface is
depleted of heavy isotopes relative to ocean water,
while rain precipitating from a cloud is enriched
relative to the cloud moisture. Hence whenever a
water sample undergoes a phase transition (e.g.
evaporation or condensation), temperature-dependent isotope fractionation occurs (Fig. 1). In the case
of tritium, this effect is generally masked by mixing
water from different sources (Jouzel 1986). But in
the case of deuterium and oxygen-18, the isotopic
composition of precipitation shows a large variation
with latitude, height and continentality. The depletion of precipitation in deuterium and oxygen-18 is
most pronounced in the polar regions. The variation
in the isotopic composition of seawater is comparatively small and mainly determined by freshwater
input and mixing between water masses.
From FISCHER G, WEFER G (eds), 1999, Use of Proxies in Paleoceanography: Examples from the South Atlantic. Springer-Verlag
Berlin Heidelberg, pp 655-686
656
Paul et al.
The stable isotopes deuterium and oxygen-18
have been incorporated into a number of atmospheric models, e.g. the LMD model (Joussaume et
al. 1984; Joussaume and Jouzel 1993), the NASA/
GISS model (Jouzel et al. 1987), the ECHAM
model (Hoffmann and Heimann 1993; Hoffmann
1995) and the GENESIS model (Mathieu 1996). As
these four models account for the various
fractionation processes that occur at the sea surface and all succeeding stages of the hydrological
cycle, they provide a tool for calculating the isotopic
content of precipitation δP as well as of evaporation δE (Juillet-Leclerc et al. 1997). Prior to their
development there was no method to estimate δE
independently (Craig and Gordon 1965).
The δ value that is commonly used to express
the isotopic composition of a water sample is defined by
δ = ( Rsample / Rstandard − 1) ⋅1000‰ ,
(1)
where Rsample and Rstandard refer to the isotope
ratio 2H/1H or 18O/16O in the water sample and in
the standard. Thus positive δ values indicate an
enrichment of the heavy isotopic species relative
to the standard, and negative δ values indicate their
depletion. The generally used standard is the Standard Mean Ocean Water (SMOW, Craig 1961) or,
more recently, the Vienna Standard Mean Ocean
Water (V-SMOW, Gonfiantini 1978).
Fig. 1. Sketch of the global δ18O cycle. In our simulations, only the oceanic part has been considered so far.
Simulation of Oxygen Isotopes in a Global Ocean Model
The focus of this paper is on oxygen isotopes,
because oxygen-18 variations are directly related
to paleotemperature studies. It is in fact for this
reason that most of the oceanographic work to date
has concentrated on oxygen-18 and only few deuterium measurements have been performed (Craig
and Gordon 1965).
Applications in Present-Day Oceanography
The distributions of both salinity S and oxygen isotope composition of seawater δ18Ow or simply δw
are mainly controlled by the same two processes
precipitation P and evaporation E. Locally S and δw
of surface waters are linearly related. However, the
slope of the δw-S relationship varies between 0.1
for tropical surface waters and 1 for high-latitude
surface waters. This reflects the so-called temperature effect: in high latitudes, precipitation occurs at
lower temperatures and is more enriched in oxygen-18 than in low latitudes, which leads to a strong
depletion of the remaining cloud moisture. A slope
of 1 is observed for East Greenland surface waters, which are diluted by adding meltwater from
the permanent ice cap. Freshwater input from rivers can also modify the δw-S relationship. Hence S
and δw label the surface waters in different ways.
Mixing of water masses that originate from the sea
surface creates the characteristic vertical sections
known from GEOSECS (Birchfield 1987; Östlund
et al. 1987).
The isotopic composition of sea water is also
affected by the formation of sea ice, but the
fractionation effect is so small that the freezing
process leads to an increase in salinity, with essentially no observable influence on δw. This gives a
near-zero slope in a δw-S diagram (Craig and
Gordon 1965). As a result, the surface waters of
the Southern Ocean show a wide range of salinities,
although δw is nearly constant. For Antarctic Surface Water (AASW), Jacobs et al. (1985) report a
δw value of -0.31 ‰ with a standard deviation of
0.08 ‰.
In summary, in present-day oceanography δw
allows to distinguish between different freshwater
sources. Thus it is possible to investigate the influence of local P - E, river-runoff and glacial meltwater, sea ice formation and decay on a water mass
657
of interest (Weiss et al. 1979; Jacobs et al. 1985;
Bauch 1995; Toggweiler and Samuels 1995).
To give an explicit example, we reproduced the
detailed freshwater budget for the Ross Sea continental shelf constructed by Jacobs et al. (1985).
From the difference between the δ18O content
of shelf water and its source they estimated that
36 cm a-1 of glacial meltwater is introduced to the
shelf. They found that surface and deep waters
move onto the shelf with an average δ18O content
of -0.22 ‰. On the shelf, its δ18O content will be
lowered by marine precipitation, glacial meltwater
and the net freezing of sea ice, such that the average δ18O content of shelf water becomes -0.42 ‰:
(1) Marine precipitation at a rate of 15 cm a-1 with
an δ18O content of -16.5 ‰ will lower the δ18O
content of shelf water by 0.005 ‰ a-1. (2) Further
depletion of 0.025 ‰ a-1 will result from the addition of glacial meltwater at a rate of 36 cm a-1 with
an δ18O content of -36 ‰. (3) The decrease in
shelf water δw from sea ice freezing will be only
0.002 ‰ a-1, due to the near-zero slope of the freezing line. The net change in δ18O by -0.20 ‰ a-1 and
the glacial addition of meltwater at a rate of 36 cm
a-1 require a shelf water residence time of 6.2 a.
An important application of a detailed freshwater budget is the study of the role of sea ice in the
formation of Antarctic Bottom Water (AABW)
(Toggweiler and Samuels 1995; Stössel et al. 1996).
In the example of Jacobs et al. (1985), the onshelf
flow has an average salinity of 34.455 and the
offshelf flow has an average salinity of 34.589,
which leaves a salinity enrichment of 0.134 units.
This salinity enrichment corresponds to the brine
rejected during the formation of sea ice at a rate of
37 cm a-1, with a residual salinity of 4.1, from Antarctic Surface Water (AASW) with a salinity
of 34.1. Furthermore salt must be drained from
58 cm a-1 of sea ice to balance 15 cm a-1 of marine
precipitation and 36 cm a-1 of glacial meltwater.
Toggweiler and Samuels (1995) pointed out that on
the one hand brine rejection causes a net salinity
enrichment of only 0.134 units, which is much
smaller than the value proposed by Broecker (1986)
on the basis of open ocean GEOSECS δw measurements, but on the other hand brine rejection is
essential in compensating the effects of marine precipitation and glacial meltwater on the shelf.
658
Paul et al.
Applications in Paleoceanography
In contrast to present-day oceanography, applications in paleoceanography do not rely on the isotope content of seawater, but on that preserved in
benthic and planktic foraminifera: the equilibrium
fractionation of the oxygen isotope composition of
calcite δ18Oc or simply δc. For example, Broecker
(1986, 1989) used benthic and planktic foraminifera
to constrain tropical SSTs and to determine the
salinity contrast between the tropical Atlantic and
Pacific Oceans. Duplessy et al. (1991) present a
reconstruction of Last Glacial Maximum (LGM)
sea surface salinity (SSS) derived from planktic
foraminifera. A number of studies address the vertical structure of the ice-age ocean (e.g., Birchfield
1987; Zahn and Mix 1991; Labeyrie et al. 1992).
Lehman et al. (1993) (in a two-dimensional ocean
model) and Mikolajewicz (1998) (in a three-dimensional ocean model) employ an idealized δw tracer
in order to compare meltwater-induced changes in
the thermohaline circulation with the geologic
record.
The calculation of paleosalinity from δw requires
linear δw-salinity relationships that are temporally
invariant on short (monthly to seasonal) as well as
on long (geological) timescales. Rohling and Bigg
(1998) critically review this basic assumption and
find that it is unwarranted for many regions of the
global ocean because of local changes in the freshwater budget and sea-ice coverage that are
advected into the ocean interior. This conclusion
is supported by the estimation of surface δw from
an atmospheric model (Juillet-Leclerc et al. 1997)
and the simulation of surface δw in a global ocean
model using full flux boundary conditions at the sea
surface (Schmidt 1998).
We feel attracted to the idea of Zahn and Mix
(1991) to interpret the wealth of oxygen isotope data
in terms of water mass changes. To this end, we
also incorporated δw into an ocean model, but we
prescribe δw at the surface and employ restoring
boundary conditions, with the advantage that the
ocean model can be run to equilibrium. The depth
distribution of δw can then be compaired with the
observations. Since δw and T are simulated simultaneously, δc can be determined from a paleo-
temperature equation and compared with benthic
foraminiferal δ18O.
Here we show the results of a simulation of the
present-day ocean and discuss to what degree δw
and δc can be used to characterize different water
masses. We find that δw can indeed serve to trace
the formation and spreading of deep and bottom
waters.
With respect to δc we face a fundamental problem: The decrease of temperature with depth acts
against the decrease of δw with depth. Hence below 2 km depth in the present-day Atlantic Ocean,
the δc values attain a nearly constant level and do
not allow to separate deep and bottom waters (cf.
Zahn and Mix 1991, Figs. 4 and 5; however, in the
glacial Atlantic Ocean there might be a small offset of 0.2 ‰ between deep and bottom waters). A
technical problem that can probably be solved by
more realistic heat flux and mixing
parameterizations is the warm bias of our ocean
model. This warm bias yields δc values that are too
light by about 0.5 ‰ above 2 km depth and exhibit
a false vertical gradient below 2 km depth.
Our ultimate goal is a simulation of the
paleocean, e.g., at the Last Glacial Maximum
(LGM). This probably requires the ocean model to
be coupled to simple atmosphere and sea-ice models such that neither sea surface salinity (SSS) nor
surface δw need to be prescribed and the use of
present-day δw-salinity relationships can be
avoided.
Ocean Model
Governing Equations
The ocean model we employ is based on the Modular Ocean Model (MOM, version 2) of the Geophysical Fluid Dynamics Laboratory (GFDL)
(Pacanowski 1996). It is governed by the primitive
equations. These equations are derived from the
basic laws of physics, particularly the conservation
equations for momentum, mass and energy. They
are formally written in spherical polar coordinates:
longitude λ increasing to the east, latitude φ increasing to the north and height z increasing upward.
Simulation of Oxygen Isotopes in a Global Ocean Model
The coordinate z is measured from the sea-level
geo-potential surface a = 6367.456 km, parallel to
the local direction of gravity (Gill 1982). Seven variables specify the physical state of the ocean: the
longitudinal, latitudinal and vertical velocity components u, v and w, pressure p, density ρ, potential
temperature θ and salinity S.
Discussions of the governing equations may be
found in the books by Washington and Parkinson
(1986) and McGuffie and Henderson-Sellers (1996)
at an introductory level, and in the book by Landau
and Lifschitz (1987) and the articles by Veronis
(1973), Cane (1986) and Semtner (1986) at an
advanced level. Several approximations are used
to simplify the conservation equations for momentum (Navier-Stokes equation) and mass (continuity equation):
•the thin shell approximation (on account of the
shallowness of the ocean z << a)
•the hydrostatic approximation (because ocean
motions are slow, the gravitational force and the
vertical gradient of pressure are balanced very
well)
659
The first term on the left hand side of Eq. (2) is
the time rate of change of the longitudinal velocity
component, the second term is the advection of the
longitudinal velocity component with the flow field,
the third term is a metric term that is due to the
curvature of the Earth and the fourth term is the
Coriolis acceleration resulting from the Earth’s rotation; the first term on the right hand side is the
longitudinal component of the pressure gradient
force and the second and third term derive from
the vertical and horizontal frictional forces. The
terms in Eq. (3) have an analogous meaning.
The hydrostatic balance and continuity equations
become:
∂p
= − ρg
∂z
1
∂w
=−
∂z
a cos φ
(4)
 ∂u ∂

 ∂λ + ∂φ (cos φv )


(5)
•the neglect of terms involving w in the horizontal momentum equations (on the basis of scale
analysis)
Eq. (4) simply relates the vertical pressure gradient (on the left hand side) to the gravitational force
(on the right hand side). Eq. (5) expresses the vertical gradient of the vertical velocity (on the left
hand side) in terms of the longitudinal gradient of
the longitudinal velocity and the latitudinal gradient
of the latitudinal velocity (on the right hand side).
In Eqs. (2)-(5) t is time, r0 is the reference density, Av is the vertical eddy viscosity coeffcient and
g is the gravitational acceleration. Furthermore, L
denotes the advection operator,
With respect to these approximations, the horizontal momentum equations are:
L(α ) =
•the Boussinesq approximation (since seawater
is nearly incompressible, variations in density are
neglected, except where buoyancy effects are
concerned)
u v tan φ
∂u
∂ p ∂  ∂u
1
A
+ Fu ,
+ L(u ) −
− fv = −
+
a
∂t
ρ 0 a cos φ ∂λ ∂ z  v ∂ z 
(2)
∂v
u 2 tan φ
1 ∂ p ∂  ∂ v
 Av
+Fν .
+ L(v ) −
+ fu = −
+
∂t
ρ 0 a ∂ φ ∂ z  ∂ z 
a
(3)
1
∂
(uα ) + 1 ∂ (cos φvα ) + ∂ (wα ),
a cos φ ∂λ
a cos φ ∂φ
∂z
(6)
where α is a generic scalar variable, and f is the
Coriolis parameter,
f = 2Ω sin φ ,
(7)
with Ω being the angular velocity of the Earth.
Finally, Fu and Fv represent the horizontal eddy
stress divergence,
660
Paul et al.
,
(8)
(
)
 1 − tan φ v 2 sin φ∂u / ∂λ 
F v = ∇ ⋅ (νh∇v ) + νh 
+
,
a2
a 2 cos 2 φ 

(9)
2
Here Ah is the horizontal eddy viscosity coefficient, and
∇⋅s =

1  ∂sλ ∂
+ ( sφ cosφ ) 

α cosφ  ∂λ ∂φ

 1 ∂α 1 ∂α 

,
∇α = 
 a cos φ ∂λ a ∂φ 
(10)
(12)
where the first term on the left hand side is the
time rate of potential temperature change and the
second term represents the advection of potential
temperature with the flow field; the first term and
second term on the right hand side denote the vertical and horizontal diffusion of potential temperature. A similar conservation equation holds for salinity S:
∂S
∂  ∂S 
+ L( S ) =  Dv  + ∇ ⋅ ( Dh∇S ) .
∂t
∂z  ∂z 
ρ = ρ ( T , S , p) .
(14)
In our ocean model we use a polynomial approximation to the equation of state developed by
the Joint Panel on Oceanographic Tables and Standards (UNESCO 1981).
In order to satisfy the viscous stability criterion,
the horizontal viscosity and diffusivity coefficients
are tapered towards the poles (Weaver and Hughes
1996; NCAR Oceanography Section 1996). Consequently, extra metric terms are included in the
momentum equations, such that pure rotation does
not generate a viscous stress (Wajsowicz 1993).
These extra metric terms are
(11)
are the horizontal divergence and gradient operators, where s is a generic vector variable. The conservation of heat is expressed in the conservation
equation for potential temperature θ:
∂θ
∂  ∂θ 
+ L(θ ) =  Dv  + ∇ ⋅ ( Dh∇θ ) ,
∂t
∂z  ∂z 
The equation of state for seawater relates
seawater density ρ to in situ temperature T , salinity S and pressure p:
(13)
Here Dv and Dh are the vertical and horizontal
eddy diffusivity coefficients.
 u tan φ
∂ v  ∂ Ah  1
∂v
v sin φ  ∂ Ah
1



−
+
+ 2
2
a cos φ ∂ λ  ∂ φ  a 2 cos φ ∂ φ a 2 cos 2 φ  ∂ λ
 a
(15)
in the longitudinal momentum equation (2) and
 v tan φ
1
∂ u  ∂ Ah  u sin φ
1
∂ u  ∂ Ah



− 2
+
+
2
a cos φ ∂ λ  ∂ φ  a 2 cos 2 φ a 2 cos φ ∂ φ  ∂ λ
 a
(16)
in the latitudinal momentum equation (3).
Model Resolution, Land-Sea Mask and
Bathymetry
The model resolution, land-sea mask and
bathymetry are the same as in the coarse-resolution ocean model of Large et al. (1997). The only
exception is that two deep vertical levels are added,
such that the maximum model depth is now 5900
m, and that there are 27 vertical levels in total,
monotonically increasing in thickness from 12 m
near the surface to 450 m near the bottom. The
longitudinal resolution is 3.6°. The latitudinal resolution is 1.8° near the equator (to better resolve the
equatorial currents), increases away from the equator to a maximum of 3.4°, then decreases in the midlatitudes as the cosine of latitude (to maintain
Fu = ∇
Simulation of Oxygen Isotopes in a Global Ocean Model
horizontally isotropic grid boxes) and is finally kept
constant at 1.8° poleward of 60° (to prevent any
further restrictions on the model time step).
The continental outlines of the model are shown
in Fig. 2. While the 9.4°- wide Drake Passage
closely matches its actual width, the Indonesian
Channel, the Florida Straits and the connections to
the Sea of Japan are much wider than in reality.
Furthermore the Bering and Gibraltar Straits and
the Red Sea outflow are closed.
The bathymetry used in the present study is
taken from the ocean model of Large et al. (1997)
and is fairly realistic, e.g. the maximum depths
of the Denmark Strait and Faroe-Iceland ridge are
734 m and 903 m. Regions deeper than 5000 m are
filled in from the ETOPO5 topography data
(NCAR Data Support Section 1986).
661
Model Viscosity and Diffusivity, Filtering
and Additional Tracer
Since the horizontal resolution is coarser than the
Rossby radius of deformation at all latitudes, the
effects of mesoscale eddies are only parameterized
in terms of enhanced viscosity and diffusion. The
horizontal viscosity is Αh = 2.5 x 105 m2s-1. The
vertical viscosity is Av = 16.7 x 10-4 m2 s-1. Following Bryan and Lewis (1979), depth-dependent horizontal and vertical diffusivities are employed.
Hence
Dh( z ) = Dbh + ( Dsh − Dbh) exp( − z / 500m) , (17)
such that the horizontal eddy diffusion Dh decreases
from Dsh= 0.8 x 106 m2s-1 in the top layer to Dbh =
0.4 x 106 m2s-1 in the bottom layer. Furthermore
Fig. 2. Continental outlines and bathymetry of the model. Contour interval is 0.5 km.
662
Dv( z ) = D* +
Paul et al.
Cr
π
[
]
arctan λ ( z − z* ) ,
(18)
where D* = 0.61 x 10-4 m2 s-1, Cr = 1.0 x 10-4 m2
s-1, λ = 1.5 x 10-3 m-1 and z* = 1000 m (Weaver
and Hughes 1996), such that the vertical viscosity
Dv is about 0.3 x 10-4 m2 s-1 at the surface and 1.1
x 10-4 m2 s-1 in the deep ocean. To increase the permissible time step, the flow variables are Fourierfiltered north of 75° N, and the horizontal viscosity
Dh is tapered north of 81.9° N. The oxygen isotope composition of seawater δw is incorporated into
the ocean model as a passive tracer that does not
influence the ocean circulation.
Experimental set-up
In the control experiment (Experiment A), the acceleration technique of Bryan (1984) is used to
accelerate the approach to equilibrium in the deep
ocean. Thus, in the first stage, the tracer time step
is ten times larger in the deep ocean than at the
surface. Furthermore the surface tracer time step
is ten times larger than the time step of 3504 s for
the momentum and barotropic streamfunction
equations. In the second stage, the integration is
continued with equal time steps of 3504 s for all
equations at all depths. The length of the accelerated phase is 96 momentum years, 960 surface
tracer years and 9600 deep tracer years, as in the
experiments of Large et al. (1997). The length of
the synchronous phase is 15 years, which is the
minimum length recommended by Danabasoglu et
al. (1996). In the sensitivity experiment (Experiment
B), the length of the accelerated phase is 20 momentum years, 200 surface tracer years and 2000
deep tracer years, and the length of the synchronous phase is 15 years.
Experiment A is started from a set of initial
conditions that corresponds to a state of rest with
January-mean distributions of potential temperature
and salinity from Levitus (1982). The converged
state at the end of the accelerated phase of Experiment A is then used as the set of initial conditions for Experiment B.
The set of boundary conditions at the ocean
surface consists of mid-month fields of the zonal
and meridional wind stress components, the sea
surface temperature (SST), the sea surface salinity (SSS) and the sea surface oxygen isotope composition. The surface wind stress fields are obtained
from the NCEP reanalysis data covering the four
years 1985 through 1988 (Kalnay et al. 1996), and
the surface temperature and salinity fields are derived from the climatologies of Shea et al. (1990)
and Levitus (1982), as described by Large et al.
(1997).
The surface δw forcing field is computed from
the mid-month fields of the sea surface salinity using
a set of seven δw -salinity relationships, in a manner similar to Fairbanks et al. (1992). The δw and
salinity data are taken from GEOSECS (Östlund
et al. 1987), Mackensen et al. (1996), Bauch (1995)
and Veshteyn et al. (1974). The set of δw -salinity
relationships is depicted in Fig. 3 and summarized
in Table 1.
The surface temperature, salt and δw fluxes that
enter the tracer equations are calculated from the
usual restoring boundary conditions given by
FT = −
Fδw = −
H
τ
H
τ
(T − T ) ,
*
(δ
w
− δ w* ) .
Fs = −
H
τ
(S − S ) ,
*
(19)
Here τ = 50 d is a relaxation time scale relative
to the upper ocean depth H = 50 m, and T*, S* and
δ*w denote the prescribed SST, SSS and surface
δw values. The fraction of a grid cell covered by
sea ice is diagnosed from the climatologic SST and
subject to a strong temperature restoring, with a
time constant τice= 6 d (cf. Large et al. 1997). Two
experiments are performed: In Experiment A, a
slope of a = 1.030 and an abscissa of b = -35.80 ‰
is used to compute the isotopic composition of
AASW. In Experiment B, a constant value of 0.31‰ is used. This corresponds to a zero slope
of the δw -S relationship and indirectly accounts for
the sea-ice effect.
Simulation of Oxygen Isotopes in a Global Ocean Model
663
Fig. 3. δw -salinity relationships for various regions of
the global ocean, based on the GEOSECS data. The Tropics extend from 25°S to 25°N. The Arctic Ocean includes
the North Atlantic Ocean north of 60°N. The term
Antarctic Zone is used to describe the Southern Ocean
surface waters south of the 2°C-isotherme. The δw -salinity relationship for the mid-latitudes is applied to the
remaining regions of the global ocean.
664
Paul et al.
Region
Arctic Ocean
a
1.010
b
-34.75
Tropical Atlantic
Tropical Pacific
Tropical Indian
Midlatitudes
Antarctic Zone
0.180
0.285
0.180
0.497
1.030
0.000
-5.63
-9.50
-5.74
-17.05
-35.80
-0.31
Note
Experiment A
Experiment B
Source
Veshteyn et al. (1974), Bauch
(1995)
Östlund et al. (1987)
Östlund et al. (1987)
Östlund et al. (1987)
Östlund et al. (1987)
Mackensen et al. (1996)
Jacobs et al. (1985)
Table 1. Slopes a and abscissas b of the δw -S relationships. In the last column, the sources of the data used for
calibration are given.
Results
Barotropic Streamfunction
The vertically integrated volume transport is presented in Fig. 4. Here we use a streamfunction map
rather than a vector map because a streamfunction
map is easier to visualize on a global scale. The flow
is directed along the streamlines, with the sense of
rotation clockwise for positive values, and anticlockwise for negative values of the streamfunction. The flow rate through any line intersecting two adjacent streamlines is given by the
contour interval.
The spatial features of the vertically integrated
mass transport are determined by the wind forcing and the bathymetry. Because of the identical
winds and the almost identical bathymetry, all gyre
patterns and transport magnitudes are very similar
to the solutions of Large et al. (1997). The major
currents and passage throughflows are reproduced
reasonably well.
In the Northern Hemisphere, the maximum
transport in the Gulf Stream is 21 Sv. This is lower
than that observed, which is about 40 Sv at 30°N,
upstream of the recirculation regime (Knauss
1969), or 32.2±3.2 Sv for the Florida Current alone
(Larsen 1992). The maximum transport in the
Kuroshio is low, too. The Atlantic inflow into the
Nordic Seas is 4 Sv, about half the observational
estimate by Gould et al. (1985).
In the Southern Hemisphere, we obtain for the
Agulhas Current a maximum transport of 62 Sv.
This is in good agreement with the estimates of
Gordon et al. (1987) and Stramma and Lutjeharms
(1996). The 25 Sv-flow through the Mozambique
Channel is significantly stronger than the 5 Sv observed by Stramma and Lutjeharms (1996). The
flow in the Benguela Current is 12 Sv. According
to Garzoli and Gordon (1996), the observed transport at 30°S in the upper kilometer of the ocean is
13 Sv, with considerable seasonal variations. The
maximum Brazil Current transport amounts to 30
Sv, on the high side of the estimate of Peterson and
Stramma (1991).
A distinct feature of the vertically integrated
mass transport is the Antarctic Circumpolar
Current (ACC). The flow through Drake Passage
is 225 Sv. An observational estimate for this flow
is 130±20 Sv (Whitworth and Peterson 1985). Thus
it is clearly overestimated in our model. As demonstrated by Danabasoglu and McWilliams (1995),
the ACC flow through the Drake Passage depends
on the horizontal or isopycnal diffusion coefficients,
showing an enhancement in transport with decreased diffusivity. Furthermore, the ACC transport is larger in the case of horizontal diffusion than
in the case of isopycnal diffusion. In our model
the vertically averaged horizontal diffusivity is
0.43x103 m2s-1. Extrapolating from the two solutions of Danabasoglu and McWilliams (1995) with
horizontal diffusion we find a flow of about 250 Sv.
From the studies of Danabasoglu and McWilliams
(1995) and Large et al. (1997) we conclude that
we can obtain a lower and more realistic ACC
transport by turning to an isopycnal diffusion
parameterization.
Simulation of Oxygen Isotopes in a Global Ocean Model
Meridional Overturning Streamfunction
The zonally integrated meridional overturning
streamfunction is presented in Figs. 5a-c. Figure
5a shows the total transport of all the ocean basins
in one section, whereas Figs. 5b and c show the
total transport broken down among the Atlantic and
Indo-Pacific basins, restricted to those latitudes
where the ocean basins are limited zonally by continental boundaries. The sections that represent the
meridional overturning streamfunction are read in
the same way as the barotropic streamfunction map
(Fig. 4).
The mode of ocean circulation most important
for ventilating the deep sea is the vertical overturning (Toggweiler et al. 1989). It is usually thought
665
of as a manifestation of the thermohaline circulation, although in the upper ocean the upwelling and
downwelling induced by the surface wind stress are
the dominant processes. In Fig. 5a two shallow cells
are depicted that straddle the equator. These cells
are driven by the divergence of the Ekman transport under the easterly trade winds. Figures 5a and
b also indicate downwelling in mid-latitudes that
produce another set of overturning cells, which is
linked to wind-induced divergences in the subpolar
regions. The overturning cell in the Northern Hemisphere is much weaker and shallower than the socalled Deacon cell in the Southern Hemisphere.
This is due to a partial cancellation between the
northern cell and the thermohaline circulation
(Toggweiler et al. 1989).
Fig. 4. Annual-mean barotropic streamfunction. Contour interval is 10 Sv below a vertically integrated volume
transport of 60 Sv and 20 Sv above. The dashed lines represent negative contour levels and indicate anti-clockwise circulation.
666
Paul et al.
Between 1500 m and 3500 m the meridional
overturning is dominated by the southward flow of
North Atlantic Deep Water (NADW), which is
produced at a rate of 21 Sv. As much as 7 Sv of
this transport upwells north of the Equator, only 14
Sv of it is exported into the Southern Ocean. An
estimate for the total southward transport of
NADW at 24°N is 17±4 Sv (Roemmich and
Wunsch 1985), which is somewhat larger than the
15-Sv flow in the model.
There is a northward flow of Antarctic Bottom
Water (AABW) in all three ocean basins, compensated by a return flow above 4000 m depth. In the
Atlantic, the deep inflow at the bottom amounts to
3 Sv.
Potential Temperature and Salinity Distributions
The annual and zonal mean distributions of potential temperature and salinity for the Atlantic Ocean
are presented in Fig. 6. Figure 7a depicts the corresponding deviation of potential temperature from
the climatology of Levitus et al. (1994). All water
masses are up to 4°C warmer than the observations, except for the AABW that is colder than the
observations by about 0.5°C. The warm bias is
largest in the thermocline, which is also more diffuse than observed. A warm bias and a diffuse
thermocline are two widespread problems in ocean
modelling (Danabasoglu and McWilliams 1995).
Fig. 5. Annual-mean meridional overturning streamfunction. Contour interval is ~2Sv except for the Atlantic where
it is ~1Sv for negative levels. The negative contour levels are represented by dashed lines that indicate anti-clockwise circulation. (a) For the present global ocean. (b, right) For the present Atlantic Ocean. (c, right) For the present
Indo-Pacific Ocean.
Simulation of Oxygen Isotopes in a Global Ocean Model
667
668
Paul et al.
Fig. 6. Annual and zonal mean tracer distributions for the present Atlantic Ocean. (a) Potential temperature.
Contour interval is 2°C. (b) Salinity. Contour interval is 0.25.
Simulation of Oxygen Isotopes in a Global Ocean Model
669
Fig. 7. Annual and zonal mean tracer distributions for the present Atlantic Ocean. Difference between simulation
and climatology. (a) Potential temperature. Contour interval is 0.5°C. (b) Salinity. Contour interval is 0.05.
670
Paul et al.
Figure 7b illustrates the deviation of salinity from
the climatology of Levitus and Boyer (1994) for the
Atlantic Ocean. It reveals that the tongue of Antarctic Intermediate Water (AAIW) is rather broad
and much too salty in the model ocean. The AABW
is too fresh by up to 0.30 units.
Seawater Oxygen Isotope Composition and
Benthic Foraminiferal δ 18O
In Fig. 8 the GEOSECS δw along a cross section
in the Atlantic Ocean is compared with the annual
and zonal mean δw from Experiments A and B. Experiment A yields a heavy δw value of AABW,
whereas in Experiment B the δw of AABW is close
to the GEOSECS data. Both experiments lead to
a broad tongue of AAIW that is also seen in the
salinity distribution. The δw of AAIW is heavier than
observed by 0.1-0.2‰. Figure 9 extends the comparison of the GEOSECS and the simulated δw to
the Pacific Ocean. In the upper 3000 m the agreement is reasonably good. The δw of AABW is too
light by about 0.1‰. Unfortunately, there are no
GEOSECS δw data between 30°S and 15°N.
Anticipating an application of the global ocean
model in a paleoceanographic setting, Fig. 10
presents core top benthic foraminiferal δ18O compiled from 169 Atlantic core sites. To facilitate a
comparison with the model ocean, the equilibrium
fractionation of δc is computed for Experiment B.
Two cases are studied: In the first case (Fig. 11a)
the model temperature is inserted into the
paleotemperature equation of Erez and Luz (1983).
In the second case (Fig. 11b) the climatologic temperature of Levitus et al. (1994) is used (cf. Fig. 4
of Zahn and Mix (1991) who employed the
GEOSECS temperature and δw data to estimate
δc). In the real ocean the decrease of δw with depth
is offset by a corresponding decrease of temperature. Below about 2000 m water depth, a roughly
constant level of about 3‰ is reached, such that
NADW and AABW can be seen in δw , but not in
δc (Fig. 11b). This is also suggested by the Atlantic core top benthic foraminiferal δ18O shown in Fig.
10. The warm bias in the model temperature introduces a false vertical gradient into the simulated
δc distribution (Fig. 11a).
Combined T-S-δ 18O Diagrams
Figure 12 shows combined T-S-δ18O diagrams for
the GEOSECS data and Experiments A and B. The
GEOSECS data were taken from three stations in
the North Atlantic and five stations in the South Atlantic. The ocean model output was sampled at the
same geographic positions and depths. Fig. 12 also
depicts the temperature-salinity ranges of Mediterranean Water (MW), AAIW, NADW and AABW,
as defined by Emery and Meincke (1986) (see
Table 2).
The GEOSECS serial points all plot on curves
that hook into the cold AABW (Fig. 12a). The two
other end members are the relatively fresh AAIW
(stations 56 and 67) and the relatively salty NADW
(stations 29, 37 and 115). Finally the influence of
MW is reflected in the T-S curve of station 115.
These four water masses are also seen in the δ18O
values, which proves that they can indeed be used
to characterize different water masses. However,
there can be no one-to-one correspondence, e.g.
there can be water types that are similar in temperature T and isotopic composition δ18O, but differ in salinity S, conceivably because of a sea ice
effect (stations 85 and 91 in the Southern Ocean).
Table 2 also gives the δw statistics for the serial
points from all GEOSECS stations in the Atlantic
that fall into the temperature-salinity ranges of the
four water masses MW, AAIW, NADW and
AABW: MW has the highest δw mean value
(0.36‰), AABW the lowest (-0.13‰). The δw
mean value of NADW (0.20‰) is between that
of AAIW (0.00‰) and AABW. The 1σ intervals
do not overlap except for AAIW and AABW. This
reflects that AAIW and AABW both have sources
in the Southern Ocean, the lower δw mean value
of AABW being due to the addition of glacial meltwater.
The influence of MW is missing from the ocean
model output simply because the Mediterranean is
not included in the ocean model domain (Figs. 12b
and c). The other water masses are found in the
ocean model, but the deficiencies in their representation are manifest in the temperature-salinity
ranges that differ significantly from those given by
Emery and Meincke (1986): AAIW and NADW
are too warm, and AABW is too cold and fresh
Simulation of Oxygen Isotopes in a Global Ocean Model
671
Fig. 8 (a, b). Seawater oxygen isotope composition for the present Atlantic Ocean in units of parts per mil versus
SMOW. Contour interval is 0.1 ‰. (a) GEOSECS δw along a cross-section in the Atlantic Ocean. Crosses indicate
data points. (b) Annual and zonal mean δw from Experiment A.
672
Paul et al.
Fig. 8 (c, d). Seawater oxygen isotope composition for the present Atlantic Ocean in units of parts per mil versus
SMOW. Contour interval is 0.1 ‰. (c) Annual and zonal mean δw from Experiment B. (d) Difference between
Experiment B and Experiment A.
Simulation of Oxygen Isotopes in a Global Ocean Model
673
Fig. 9. (a, b) Seawater oxygen isotope composition for the present Pacific Ocean in units of parts per mil versus
SMOW. Contour interval is 0.1 ‰. (a) GEOSECS δw along a cross section in the Pacific Ocean. Crosses indicate
data points. (b) Annual and zonal mean δw from Experiment A.
674
Paul et al.
Fig. 9. (c, d) Seawater oxygen isotope composition for the present Pacific Ocean in units of parts per mil versus
SMOW. Contour interval is 0.1 ‰. (c) Annual and zonal mean δw from Experiment B. (d) Difference between
Experiment B and Experiment A.
Simulation of Oxygen Isotopes in a Global Ocean Model
Temperature range (°C)
Salinity range (‰)
Oxygen-18 range (‰ SMOW)
Oxygen-18 mean value (‰ SMOW)
Oxygen-18 standard deviation (‰ SMOW)
Oxygen-18 number of samples
MW
2.6-11.0
35.0-36.2
0.26-0.54
0.36
0.08
19
AAIW
2.0-6.0
33.8-34.8
-0.29-0.18
0.00
0.09
34
NADW
1.5-4.0
34.8-35.0
0.05-0.31
0.20
0.06
64
675
AABW
-0.9-1.7
34.64-34.72
-0.27-0.04
-0.12
0.07
77
Table 2. Temperature, salinity and oxygen-18 characteristics of Mediterranean Water (MW), Antarctic Intermediate Water (AAIW), North Atlantic Deep Water (NADW) and Antarctic Bottom Water (AABW). The temperature
and salinity ranges are as defined by Emery and Meincke (1986). The oxygen-18 statistics is computed for the serial
points from all GEOSECS stations in the Atlantic that fall into these temperature-salinity ranges.
Fig. 10. Benthic foraminiferal oxygen isotope composition for the present Atlantic Ocean in units of parts per mil
versus PDB (normalized to Uvigerina). Contour interval is 0.25 ‰. The δc data are compiled from 169 core sites.
Some data are unpublished, but most of them are taken from the literature (Broecker 1986; Birchfield 1987; Curry et
al. 1988; Zahn and Mix 1991; Labeyrie et al. 1992; Bickert 1992; McCorkle and Keigwin 1994).
676
Paul et al.
Fig. 11. Equilibrium fractionation δc computed from the paleotemperature equation of Erez and Luz (1983). (a) Temperature taken from Experiment B. (b) Temperature taken from climatology. Contour interval is 0.25 ‰.
Simulation of Oxygen Isotopes in a Global Ocean Model
(cf. Figs. 6-8). Experiments A and B differ mainly
in the δw values of AAIW and AABW. In comparison to the GEOSECS data (Fig. 12a) the δw of
AABW is too heavy in Experiment A, but turns out
to be too light in Experiment B, at least at station
85 in the Southern Ocean.
Isotope Content of the Net Surface
Freshwater Flux
The surface salt and δw fluxes that are calculated
from Eq. (19) can be related to the net surface fresh
water flux P - E through
Fs = −
S0
ρ0
(P − E ) ,
(20)
Fδw = −
1
ρ0
δ P − E (P − E ) ,
677
(21)
where S0 = 34.8 is the reference salinity and δP-E
is the isotope content of the net surface freshwater flux. From Eqs. (20) and (21) we can derive
the following expression for δP-E:
δ P−E = −S 0
Fδw
FS
(22)
Thus a slope of 0.5 (i.e. Fδw:FS = 1) would correspond to a δw value of the net surface freshwater flux of -17.4 ‰, which is the value assumed by
Mikolajewicz (1998). The δP-E distributions shown
in Fig. 13 reveal a rich meridional structure with
Fig. 12. (a) Combined T-S-δ18O diagrams. (a) GEOSECS data.
678
Paul et al.
Fig. 12. (b, c) Combined T-S-δ18O diagrams. (b) Experiment A. (c) Experiment B.
679
Simulation of Oxygen Isotopes in a Global Ocean Model
high positive values in the subtropics, where evaporation dominates over precipitation, small negative
values in the equatorial trough region and high negative values in the mid- and high latitudes, where
precipitation dominates over evaporation. Excess
evaporation concentrates the less volatile isotopic
molecule 1H218O along with salinity, in agreement
with the positive δP-E values (Craig and Gordon
1965).
The prescribed surface δ18O field shows meridional gradients across the boundaries of the different regions given in Table 1, in particular in the
North Atlantic at 25°N and in the Southern Ocean.
These meridonal gradients are reflected in the δw
surface fluxes. But since the restoring is not very
stiff (t = 50 d), they are smeared out in the uppermost model layer by advection and diffusion processes, such that the simulated surface δw fields look
rather smooth (Fig. 14). In comparison to Experiments A, Experiment B shows a less depleted net
surface fresh water flux in the Southern Ocean,
along with a much smaller meridional gradient.
Discussion
The advantage of restoring boundary conditions
over full flux boundary conditions is that the ocean
model can be run to equilibrium without flux adjustments. In a sense our approach is complementary
to the one of Schmidt (1998). Table 3 summarizes
the resulting hydrography of the present-day water masses and compares it to the observations. The
regional means display the same problem as the
zonal mean distributions shown in Figs. 6a, Fig. 7a:
The entire ocean is too warm, except for the bottom North Atlantic Ocean and Pacific Oceans. In
the North Atlantic Ocean in particular the AABW
influence is too strong. The ocean as a whole is also
too fresh.
The present ocean model is best compared with
the case H1 ocean model of Danabasoglu and
McWilliams (1995), which exhibits a very similar
depth distribution of the deviation from climatology
(note that in Fig. 7 we show the errors for the Atlantic Ocean that are somewhat larger than those
for the global ocean). The only exception is the
AABW that is slightly too cold in the present ocean
model and slightly too warm in the case H1 ocean
model. The reason for this is probably that
Danabasoglu and McWilliams (1995) did not employ a strong temperature restoring under sea ice.
We find that a surface δw forcing field can be
generated such that a global ocean model reproduces the GEOSECS δw data fairly well, both along
cross sections in the Atlantic and Pacific oceans
and in terms of regional means. The too light
bottom water δw values in Experiment A partly
reflect the too strong AABW influence and partly
result from the too light surface δw values in the
Southern Ocean. Experiment B in comparison to
Temperature (°C)
Model
Data
Salinity (‰)
Model
Data
δw (‰ SMOW)
Model
Data
>1
2-4
>4
North Atlantic
4.95
34.87
4.66
2.9
1.99
2.4
34.93
34.86
34.60
0.28
34.92
34.89
0.24
0.29
0.07
0.26
0.21
2-4
>4
Pacific
2.75
1.08
34.44
34.41
34.67
34.70
-0.04
-0.07
0.00
-0.01
Whole-basin mean
Globe
5.34
34.59
34.74
0.08
0.08
Water depth (km)
1.7
1.1
Table 3. Hydrography of present-day water masses. The observed temperature, salinity and δw are taken from
GEOSECS as given in Zahn and Mix (1991) and Labeyrie et al. (1992).
680
Paul et al.
Fig. 13. Isotope content δP-E of the annual mean net surface fresh water flux. (a) As diagnosed from Experiment A.
(b) As diagnosed from Experiment B.
Simulation of Oxygen Isotopes in a Global Ocean Model
Fig. 14. Annual mean surface δw. (a) As simulated in Experiment A. (b) As simulated in Experiment B.
681
682
Paul et al.
Experiment A shows the effect of a homogenous
surface δw distribution in the Southern Ocean on
the surface δw fluxes and the zonal mean δw distribution. The zero slope of the δw -S relationship
indirectly accounts for the sea-ice effect. As a
result, the bottom water δw values are relatively
more enriched, which illustrates that δw can be
used to constrain the role of sea-ice in AABW
formation (cf. Section 1). The apparantly better fit
of Experiment B to the GEOSECS δw data results
from a partial cancellation of errors in the representation of AABW.
The effect of the warm bias on δ18Oc is depicted
in Fig. 11. In the upper 2 km where this bias is largest the δc values are too light by about 0.5 ‰.
Below 2 km depth the simulated δc values exhibit
a vertical gradient that is seen neither in the
GEOSECS nor in the benthic foraminiferal δ18O
data. In order to simulate this data successfully, the
bias towards warm temperatures must be reduced
significantly.
A warm bias and a diffuse thermocline are two
widespread problems in ocean modelling, which can
probably be overcome by using diffusion of tracers along isopycnals rather than the physically unjustifiable horizontal diffusion of tracers
(Danabasoglu and McWilliams 1995). The formulation by Gent and McWilliams (1990) includes an
additional tracer transport by mesoscale eddies,
which also eliminates the effect of the Deacon cell
on the zonal mean tracer distributions. Finally, the
ACC transport through Drake Passage can probably take a more realistic value due to the isopycnal
form stress that supports more drag with weaker
flow.
The isopycnal transport parameterization taken
by itself still leads to the formation of too cold and
fresh AABW which can possibly be corrected for
by a more realistic heat flux parameterization
(Large et al. 1997) or coupling to an atmospheric
energy balance model and a sea ice ocean model
(Fanning and Weaver 1996; Mikolajewicz 1998).
The equilibrium fractionation of the oxygen isotope composition of calcite δc is a good water mass
tracer above 2 km depth. Below 2 km depth the δc
values attain a nearly constant level in the presentday Atlantic Ocean and do not allow the separation of NADW and AABW. The near constancy
of δc below 2 km depth in the present-day Atlantic
Ocean is an important feature in itself that must be
simulated correctly before the wealth of isotope
data can be interpreted in terms of water mass
changes.
In the present-day North Atlantic, water masses
deeper than 1 km have considerably higher salinity and δw than in the global ocean. This salinity and
δc excess persisted during glacial times, although
NADW is probaly produced at a lower rate (Boyle
and Keigwin 1987, Duplessy et al. 1988). Zahn and
Mix (1991) find that the LGM δc values from deep
water Atlantic sites (between 2 and 4 km depth)
are on average 0.2 ‰higher than those from
Atlantic Bottom Water sites (below 4 km depth) see the depth distributions of benthic foraminiferal
δ18O of Zahn and Mix (1991) (their Fig. 6) and of
(Labeyrie et al. 1992) (their Fig. 3-b). A meridional cross section shows that the deep water δc
maximum is most pronounced north of 10°N (Zahn
and Mix 1991, Fig. 7). It can be traced through the
north-east Atlantic up to Rockall Plateau where it
is reflected in high δc values between 1.4 and 2.3
km depth (Jung 1996).
The LGM δc values from Pacific Deep Water
sites are similar to those from Atlantic Bottom
Water sites. Zahn and Mix (1991) conclude that if
the measured gradients could be taken at face
value, and if the slope of the δw -S relationship were
assumed to be larger at the LGM than at present,
then Atlantic bottom waters and Pacific deep waters would be allowed to have a common source,
again in the Southern Ocean.
A clear δc signal can only be seen in the coldest
water at the very bottom of the glacial Atlantic
Ocean. If the zone of mixing between NADW and
AABW rises by a few hundred meters, this can
probaly not be detected in the depth distribution of
δc. Therefore it would be desirable to incorporate
a tracer such as δ13C into the ocean model that
could reveal additional information on the vertical
structure of the deep ocean.
In computing the equilibrium fractionation
of δ18Oc we used the paleotemperature equation
of Erez and Luz (1983), which is calibrated
between 14°C and 30°C with cultured planktic
foraminifera. Zahn and Mix (1991) found that this
paleotemperature equation gives the best fit to the
Simulation of Oxygen Isotopes in a Global Ocean Model
benthic foraminferal δ18O values from northeast
Atlantic core tops at water depths >2 km. Many
empirically derived paleotemperature equations
have been published during the past several decades (Bemis et al. 1998). In particular, Shackleton
(1974) used core top data to calibrate a δ18O-temperature relationship for the benthic foraminifer
Uvigerina. At low temperatures it yields larger isotopic differences between calcite and water δ18Oc
- δ18Ow than the one by Erez and Luz (1983). In
contrast, Bemis et al. (1998) derived a new
paleotemperature equation that is also based on
cultured planktic foraminifera and which at low
temperatures yields smaller isotopic differences
(cf. their Fig. 5). A smaller isotope fractionation
effect corresponds to a saturation of δc at a larger
depth and vice versa. However, our conclusions
would not change if we had chosen any of the other
paleotemperature equations. In particular, the striking difference between Fig. 11a and b would not
be eliminated. With our choice of the
paleotemperature equation, these figures are directly comparable to Figs. 4 and 7 of Zahn and Mix
(1991).
Figure 13 shows the δP-E distribution, which
reveals a rich meridional structure, quite different
from the constant value assumed by Lehman et al.
(1993) and Mikolajewicz (1998).
Our ultimate goal remains a simulation of the
paleocean and an interpretation of the wealth of
isotope data in terms of water mass changes. In
the case of the LGM the proxy data coverage may
be dense enough to generate the surface forcing
fields from a combination of SST reconstructions,
planktic foraminiferal δ18O data and atmospheric
model output (e.g. Herterich et al. this volume). In
any case, the simulated equilibrium fractionation of
the oxygen isotope composition of calcite δc can
be directly compared with the benthic foraminiferal
δ18O data. Thus the simulation of oxygen isotopes
in an ocean model has the potential to investigate
and validate circulation regimes very different from
the present one.
There is of course a problem in the calculation
of paleosalinity from δc. It requires linear δw -salinity relationships that are temporally invariant over
long (geological) timescales (see Section 1). Hence
upon going to the past it is desirable to avoid the
683
restoring boundary condition on salinity. For this
reason, and for paleostudies of more different periods than the LGM, the ocean model must probably be coupled to a prognostic sea-ice model, or
to both a simple atmosphere and a prognostic seaice model.
A promising application beyond the glacial equilibrium state is the deglaciation phase with rapidly
changing δc values. Lehman et al. (1993) (in a twodimensional ocean model) and Mikolajewicz (1998)
(in a three-dimensional ocean model) indeed employed an idealized δw tracer in order to compare
meltwater-induced changes in the thermohaline
circulation with the geologic record.
Conclusions
In conclusion,
· a surface δ18Ow forcing field can be generated such that a global ocean model reproduces the
GEOSECS δ18Ow data fairly well, both along cross
sections in the Atlantic and Pacific oceans and in
terms of regional means,
· a successful simulation of benthic foraminiferal δ18O, however, requires a correction of
the warm bias of the present global ocean model,
possibly by using an isopycnal transport parameterization,
· δ18Ow can be used to characterize different
water masses,
· but since δ18Oc does not resolve the zone of
mixing between NADW and AABW, it would be
desirable to employ another tracer such as δ13C
that could reveal additional information on the vertical structure of the deep paleocean,
· for paleostudies in general, the global ocean
model must probably be coupled to simple atmosphere and sea-ice models, such that neither SSS
nor surface δ18Ow need to be prescribed and the
use of present-day δ18Ow-salinity relationships can
be avoided.
Acknowledgments
We thank Dr. G. Edward Birchfield and Dr.
Christopher D. Charles for their help during the
early stages of this work. We are grateful to Dr.
684
Paul et al.
Harmon Craig for providing us with the GEOSECS
δ18Ow data. We also thank Dr. Martina Pätzold for
supplying their unpublished δ18Ow data and Sven
Stregel for preparing Fig.1. Figs. 2, 4-11 and 13-14
were produced with the help of FERRET program,
which is a product of NOAA/PMEL. Fig. 12 was
prepared using GMT, i.e. the Generic Mapping Tool
developed by Dr. Paul Wessel and Dr. Walter
Smith. A. P. is very grateful to Ron Pacanowski
who shared with him the latest version of the MOM
code and helped him to compile and run it at the
University of Bremen. Thanks also to the NCAR
oceanography section and to Nancy Norton in particular who provided the NCAR ocean model
(NCOM) bathymetry and surface forcing fields.
Finally he thanks Dr. Andreas Manschke for his
continued technical support. We acknowledge the
thorough reviews by Dr. Uwe Mikolajewicz and
Dr. Thomas Stocker that improved the original
manuscript very much. This research was
funded by the Deutsche Forschungsgemeinschaft
(Sonderforschungsbereich 261 at Bremen University, Contribution No. 198). Data are available under www.pangaea.de/Projects/SFB261.
References
Bauch D (1995) The Distribution of δ18O in the Arctic
Ocean: Implications for the Freshwater Balance of
the Halocline and the Sources of Deep and Bottom
Waters. Ber Polarforsch Bremerhaven 159, pp 1-144
Bemis BE, Spero HJ, Bijma J, Lea DW (1998)
Reevaluation of the oxygen isotopic composition of
planktonic foraminifera: Experimental results and
revised paleotemperature equations. Paleoceanography 13: 150- 160
Bickert T (1992) Rekonstruktion der spätquartären
Bodenwasserzirkulation im Östlichen Südatlantik
über stabile Isotope benthischer Foraminiferen. Ber
Fachber Geowiss Univ Bremen 27, pp 1-205
Birchfield GE (1987) Changes in deep-ocean water δ18O
and temperature from the last glacial maximum to the
present. Paleoceanography 2: 431-442
Boyle EA, Keigwin L (1987) North Atlantic thermohaline
circulation during the last 20,000 years: Link to highlatitude surface temperature. Nature 330: 35-40
Broecker WS (1986) Oxygen isotope constraints on
surface ocean temperatures. Quat Res 26: 121-134
Broecker WS (1989) The salinity contrast between the
Atlantic and Pacfic oceans during glacial time.
Paleoceanography 4: 207-212
Bryan K (1984) Accelerating the convergence to equilibrium of ocean climate models. J Phys Oceanogr
14: 666-673
Bryan K, Lewis LJ (1979) A water mass model of the world
ocean circulation. J Geophys Res 84: 2503-2517
Cane MA (1986) Introduction to ocean modeling. In:
O’Brien JJ (ed) Advanced Physical Oceanographic
Modelling. NATO ASI Series C 186, Reidel,
Dordrecht, pp 5-21
Craig H (1961) Standard for reporting concentrations of
deuterium and oxygen-18 in natural waters. Science
133: 1833-1834
Craig H, Gordon LI (1965) Isotopic oceanography: Deuterium and oxygen-18 variations in the ocean and the
marine atmosphere. In: Tongiorgi E (ed) Proceedings
of the Third Spoleto Conference. Consiglio
Nazionale Delle Ricerche, pp 9-130
Curry WB, Duplessy JC, Labeyrie LD, Shackleton NJ
(1988) Changes in the distribution of δ13C of deep
water CO2 between the last glaciation and the
holocene. Paleoceanography 3: 317-341
Danabasoglu G, McWilliams JC (1995) Sensitivity of the
global ocean circulation to parameterizations of
mesoscale tracer transports. J Climate 8: 2967-2987
Danabasoglu G, McWilliams JC, Large WG (1996).
Approach to equilibrium in accelerated global oceanic models. J Climate 9: 1092-1110
Duplessy JC, Shackleton NJ, Fairbanks RG, Labeyrie LD,
Oppo D, Kallel N (1988) Deepwater source variations
during the last climatic cycle and their impact on the
global deepwater circulation. Paleoceanography 3:
343-360
Duplessy JC, Labeyrie LD, Juillet-Leclerc A, Maitre F,
Duprat J, Sarnthein M (1991) Surface salinity reconstruction of the North Atlantic Ocean during the last
glacial maximum. Oceanol Acta 14: 311-324
Emery WJ, Meincke J (1986) Global water masses: summary and review. Oceanol Acta 9: 383-391
Erez J, Luz B (1983) Experimental paleotemperature equation for planktonic foraminifera. Geochim Cosmochim
Acta 47: 1025-1031
Fairbanks RG, Charles CD, Wright JD (1992) Origin of
global meltwater pulses. In: Taylor RE (ed) Radiocarbon after four decades. Springer, Berlin
Heidelberg New York, pp 473-500
Fanning AF, Weaver AJ (1996) An atmospheric energymoisture balance model: climatology, interpentadal
climate change, and coupling to an ocean general
circulation model. J Geophys Res 101 (D10):
15,111-15,128
Simulation of Oxygen Isotopes in a Global Ocean Model
Garzoli SL, Gordon AL (1996) Origins and variability of
the Benguela Current. J Geophys Res 101: 897-906
Gat JR (1996) Oxygen and Hydrogen isotopes in the
hydrologic cycle. Ann Rev Earth Planet Sci 24:
225-262
Gill AE (1982) Atmosphere-Ocean Dynamics. International Geophys Series 20, Academic Press, San
Diego
Gent PR, McWilliams JC (1990) Isopycnal mixing in
ocean circulation models. J Phys Oceanogr 20:
150-155
Gonfiantini R (1978) Standards for stable isotope measurements in natural compounds. Nature 271: 534-536
Gordon AL, Lutjeharms JRE, Gründlingh ML (1987)
Stratification and circulation at the Agulhas Retroflection. Deep-Sea Res 34: 565-599
Gould WJ, Loynes J, Backhaus J (1985) Seasonality in
slope current transports northwest of Shetland.
Technical Report 7, ICES, Hydrography Commitee,
pp 1-13
Hoffmann G (1995) Stabile Wasserisotope im allgemeinen
Zirkulationsmodell ECHAM. Dissertation, Univ
Hamburg, pp 1-99
Hoffmann G, Heimann M (1993) Water tracers in the
ECHAM general circulation model. In: International
Atomic Energy Agency (ed) Isotope techniques in
the study of past and current environmental
changes in the hydrosphere and the atmosphere
Jacobs SS, Fairbanks RG, Horibe Y (1985) Origin
and evolution of water masses near the antarctic
continental margin. In: Oceanology of the Antarctic
Continental Shelf, Ant Res Ser 43: 59-85
Joussaume S, Jouzel J (1993) Paleoclimatic tracers: an
investigaton using an Atmospheric General Circulation Model under Ice Age conditions. 2. Water
isotopes. J Geophys Res 98: 2807-2830
Joussaume S, Sadourny R, Jouzel J (1984) A general circulation model of water isotope cycles in the atmosphere. Nature 311: 24-29
Jouzel J (1986) Isotopes in cloud physics: Multiphase
and multistage condensation processes. In: Fritz P,
Fontes JC (eds) The Terrestrial Environmont, Handbook of Environmental Isotope Geochemistry.
Elsevier, New York, 2, pp 61-112
Jouzel J, Russel GL, Suozzo RJ, Koster RD, White JWC,
Broecker WS (1987) Simulations of the HDO and
H218O atmospheric cycles using the NASA GISS
General Circulation Model: the seasonal cycle
for present-day conditions. J Geophys Res 92:
14,739-14,760
685
Juillet-Leclerc A, Jouzel J, Labeyrie LD, Joussaume S
(1997) Modern and last glacial maximum sea surface
δ18O derived from an Atmospheric General Circulation Model. Earth Planet Sci Lett 146: 591- 605
Jung SJA (1996) Wassermassenaustausch zwischen
NE-Atlantik und dem Nordmeer während der letzten
300000/80000 Jahre im Abbild stabiler O- und C-Isotope. PhD Thesis, Univ Kiel, pp 1-104
Kalnay E, Kanamitsu M, Kistler R, Collins W, Deaven
D, Gandin L, Iredell M, Saha S, White G, Woolen J,
Zhu Y, Leetma A, Reynolds R, Chelliah M, Ebisuzaki
W, Higgins W, Janowiak J, Mo KC, Ropelewski C,
Wang J, Jenne R, Joseph D (1996) The NCEP/NCAR
40-year reanalysis project. Bull Amer Meteor Soc 77:
437-471
Knauss J (1969) A note on the transport of the Gulf
Stream. Deep-Sea Res 16: 117-124
Labeyrie LD, Duplessy JC, Duprat J, Juillet-Leclerc A,
Moyes J, Michel E, Kallel N, Shackleton NJ (1992)
Changes in the vertical structure of the North
Atlantic ocean between glacial and modern times.
Quat Sci Rev 11: 401-413
Landau LD, Lifschitz EM (1987) Fluid Mechanics (2nd
ed). Pergamon, Oxford, pp 1-539
Large WG., Danabasoglu G, Doney SC, McWilliams JC
(1997) Sensitivity to surface forcing and boundary
layer mixing in a global ocean model: Annual-mean
climatology. J Phys Oceanogr 27: 2418-2447
Larsen JC (1992) Transport and heat flux of the Florida
Current at 27° N derived from cross-stream voltages
and profiling data: Theory and observations. Philos
Trans R Soc London A338: 169-236
Lehman SJ, Wright DG, Stocker TF(1993) Transport of
freshwater into the deep ocean by the conveyor. In:
Peltier WR (ed) Ice in the Climate System. NATO ASI
Series I: Global Environmental Change, Vol 12,
Springer, Berlin Heidelberg New York, pp 187-209
Levitus S (1982) Climatological Atlas of the World
Ocean. NOAA Prof Pap 13, pp 1-173
Levitus S, Boyer TP (1994) World Ocean Atlas. Volume
4: Temperature. NOAA Atlas NESDIS 4, pp 1-117
Levitus S, Russell B, Boyer TP (1994) World Ocean
Atlas. Volume 3: Salinity. NOAA Atlas NESDIS 3,
pp 1-99
Mackensen A, Hubberten HW, Scheele N, Schlitzer R
(1996) Decoupling of δ13CΣCO2 and phosphate
in Recent Weddel Sea deep and bottom water:
Implications for glacial Southern Ocean
paleoceanography. Paleoceanography 11: 203-215
686
Paul et al.
Mathieu R (1996) GENESIS Σ. A Water Molecule Stable
Isotopes and Tracers Version of the GENESIS Global Climate Model Version 2.0. Model Description.
Interdisciplinary Climate System Section, CGD,
NCAR, Boulder, Colorado. Institute of Alpine and
Arctic Research, Univ Colorado, Boulder, pp 1-19
McCorkle DC, Keigwin LD (1994) Depth profiles of δ13C
in bottom water and core top C. wuellerstorfi on the
Ontong Java Plateau and Emperor Seamounts.
Paleoceanography 9: 197-208
McGuffie K, Henderson-Sellers A (1996) A Climate Modelling Primer (2nd ed). J Wiley & Sons, Chichester,
pp 1-253
Mikolajewicz U (1998) A meltwater induced collapse of
the ‘conveyor belt’ thermohaline circulation and its
influence on the distribution of 14C and δ18O in the
oceans. J Geophys Res, in press
NCAR Data Support Section (1986) NGDC ETOPO5 global ocean depth & land elevation, 5-min. Data Set
759.1, NCAR, Boulder, Colorado, pp
NCAR Oceanography Section (1996) The NCAR CSM
ocean model. NCAR Technical Note NCAR/TN423+STR, NCAR, Boulder, Colorado
Östlund HG, Craig H, Broecker WS, Spencer D (1987)
GEOSECS Atlantic, Pacific, and Indian ocean expeditions. Shorebased data and graphics. GEOSECS
Atlas Series, US Govt Printing Office, Washington
DC, 7, pp 1-200
Pacanowski RCE (1996) MOM 2. Documentation, User’s Guide and Reference Manual. Technical Report
3.2, GFDL Ocean Group, GFDL, Princeton, New
Jersey, pp 1-329
Peterson RG, Stramma L (1991) Upper-level circulation
in the South Atlantic Ocean. Progr Oceanogr 26:
1-73
Roemmich D, Wunsch C (1985) Two transatlantic sections: Meridional circulation and heat flux in the
subtropical North Atlantic Ocean. Deep-Sea Res 32:
619-664
Rohling EJ, Bigg GR (1998) Paleosalinity and δ18O: A
critical assessment. J Geophys Res 103: 1307-1318
Schmidt GA (1998) Oxygen-18 variations in a global
ocean model. Geophys Res Lett 25: 1201-1204
Schotterer U, Oldfield F, Frölich K (1996) Global network
for isotopes in precipitation. IAEA, Vienna, pp 1-48
Semtner A (1986) Finite-difference formulation of a
World Ocean model. In: O’Brien JJ (ed) Advanced
Physical Oceanographic Modelling. NATO ASI
Series C 186, Reidel, Dordrecht, pp 187-202
Shackleton NJ (1974) Attainment of isotopic equilibrium
between ocean water and the benthonic foraminifera
genus Uvigerina: Isotopic changes in the ocean
during the last glacial. Colloq Int Centr Nat Rech Sci
219: 203-209
Shea DJ, Trenberth KE, Reynolds RW (1990) A global
monthly sea surface temperature climatology. NCAR
Technical Note NCAR/TN- 345, NCAR, Boulder,
Colorado, pp 1-167
Stössel A, Oberhuber JM, Maier-Reimer E (1996) On the
representation of sea ice in global ocean general
circulation models. J Geophys Res 101 (C8):
18,193-18,212
Stramma L, Lutjeharms JRE (1996) The flow field of
the subtropical gyre of the South Indian Ocean.
J Geophys Res 102: 5513
Toggweiler JR, Samuels B (1995) Effect of sea ice on the
salinity of antarctic bottom waters. J Phys Oceanogr
25: 1980-1997
Toggweiler JR, Dixon K, Bryan K (1989). Simulations of
radiocarbon in a coarse-resolution World Ocean
model. 1. Steady state prebomb distributions.
J Geophys Res 94: 8217-8242
UNESCO (1981) Tenth report of the joint panel of oceanographic tables and standards. UNESCO Technical Papers in Marine Sciences 36, UNESCO, Paris
Veronis G (1973) Large scale ocean circulation. Adv Appl
Mech 13
Veshteyn VY, Malyuk GA, Rusanov VP (1974) Oxygen18 distribution in the Central Arctic Basin. Oceanology 14: 514-519
Wajsowicz RC (1993) A consistent formulation of the
anisotropic stress tensor for use in models of the
large scale ocean circulation. J Comp Phys 105:
333-338
Washington WM, CL Parkinson (1986) An Introduction
to Three- Dimensional Climate Modeling. University
Science Books, Mill Valley, California
Weaver AJ, Hughes TMC (1996) On the incompatibility
of ocean and atmosphere models and the need for
flux adjustments. Clim Dyn 12: 141-170
Weiss RF, Östlund HG, Craig H (1979) Geochemical studies of the Weddell Sea. Deep-Sea Res 26A: 1093-1120
Whitworth T III, Peterson RG (1985) Volume transport
of the Antarctic Circumpolar Current from bottom
pressure measurements. J Phys Oceanogr 15:
810-816
Zahn R, Mix AC (1991) Benthic foraminiferal δ18O in the
ocean’s temperature-salinity-density field: constraints on the ice age thermohaline circulation.
Paleoceanography 6: 1-20