5.1.1 The problem
Ocean water mass characteristics are set by direct
contact with the atmosphere and subsequent transport and mixing in the ocean interior. Section 5
deals with these interior processes, including the
conservative mixing of heat and salt to produce
water masses of intermediate densities. Lazier
et al. (Chapter 5.5) discuss changes in surface
water mass properties that influence the deeper
layers, which become an essential part of the
global thermohaline overturning. In the absence of
significant mixing, surface water mass properties
can be transported to the ocean interior, where
they influence the structure of the ocean interior
and thus further influence climate variability
and change. (For example, in Chapter 4.4 Liu and
Philander demonstrate how water mass properties set in the extratropics affect the structure of
the tropical thermocline, thus affecting tropical
ocean–atmosphere interactions, and hence climate
fluctuations and change at decadal and longer time
scales.)
The focus of this chapter is air–sea interaction,
which creates new density classes at the surface,
and in particular extrema of both light and dense
water. Such surface water mass transformation
depends both on the surface density and the surface flux of density (Speer and Tziperman, 1992;
Walin, 1982). As such, it is not easily measured or
modelled. A direct determination requires better
observations of the ocean’s near-surface temperature and salinity than are commonly available.
Even more problematic is the measurement of the
surface heat and freshwater fluxes that make up
the density flux. This complicated subject is comprehensively covered by the WCRP/SCOR Working Group on Air-Sea Fluxes (WGASF, 2000).
In ocean modelling a common practice has been
to restore surface temperature and salinity to
observed values (Cox and Bryan, 1984), which
strongly limits surface density errors. However,
to do so often requires very erroneous fluxes
(Killworth et al., 2000), especially of fresh water,
and hence water mass transformation (Large et al.,
1997). Conversely, forcing with observed fluxes
has resulted in poor representation of surface temperature and salinity (Rosati and Miyakoda, 1988)
and hence water mass transformation. The two
approaches can be combined so that relaxation
terms are added to climatological heat and salt
fluxes (Barnier et al., 1995). However, Nurser et al.
(1999) find that the relaxation substantially modifies the surface density flux of a North Atlantic
model. Large et al. (1997) use more physical surface boundary conditions, where in particular, the
freshwater and heat fluxes are coupled through an
explicit evaporation/latent heat flux. The global
result is an improved equilibrium water mass
distribution, because the combined effect of surface
density and surface flux errors on water mass
transformation is reduced (Doney et al., 1998b).
In this chapter, we first set out Walin’s theory
of water mass transformation, focusing on how
surface fluxes drive diapycnal flow in the surface
mixed layer. We briefly discuss how lateral mixing within the mixed layer and entrainment at the
5.1
Ocean Surface Water Mass Transformation
William G. Large and A. J. George Nurser
317
OCEAN CIRCULATION AND CLIMATE
Copyright © 2001 Academic Press
ISBN 0-12-641351-7
All rights of reproduction in any form reserved
CHAPTER
Ocean water mass characteristics are set by direct
contact with the atmosphere and subsequent transport and mixing in the ocean interior. Section 5
deals with these interior processes, including the
conservative mixing of heat and salt to produce
water masses of intermediate densities. Lazier
et al. (Chapter 5.5) discuss changes in surface
water mass properties that influence the deeper
layers, which become an essential part of the
global thermohaline overturning. In the absence of
significant mixing, surface water mass properties
can be transported to the ocean interior, where
they influence the structure of the ocean interior
and thus further influence climate variability
and change. (For example, in Chapter 4.4 Liu and
Philander demonstrate how water mass properties set in the extratropics affect the structure of
the tropical thermocline, thus affecting tropical
ocean–atmosphere interactions, and hence climate
fluctuations and change at decadal and longer time
scales.)
The focus of this chapter is air–sea interaction,
which creates new density classes at the surface,
and in particular extrema of both light and dense
water. Such surface water mass transformation
depends both on the surface density and the surface flux of density (Speer and Tziperman, 1992;
Walin, 1982). As such, it is not easily measured or
modelled. A direct determination requires better
observations of the ocean’s near-surface temperature and salinity than are commonly available.
Even more problematic is the measurement of the
surface heat and freshwater fluxes that make up
the density flux. This complicated subject is comprehensively covered by the WCRP/SCOR Working Group on Air-Sea Fluxes (WGASF, 2000).
In ocean modelling a common practice has been
to restore surface temperature and salinity to
observed values (Cox and Bryan, 1984), which
strongly limits surface density errors. However,
to do so often requires very erroneous fluxes
(Killworth et al., 2000), especially of fresh water,
and hence water mass transformation (Large et al.,
1997). Conversely, forcing with observed fluxes
has resulted in poor representation of surface temperature and salinity (Rosati and Miyakoda, 1988)
and hence water mass transformation. The two
approaches can be combined so that relaxation
terms are added to climatological heat and salt
fluxes (Barnier et al., 1995). However, Nurser et al.
(1999) find that the relaxation substantially modifies the surface density flux of a North Atlantic
model. Large et al. (1997) use more physical surface boundary conditions, where in particular, the
freshwater and heat fluxes are coupled through an
explicit evaporation/latent heat flux. The global
result is an improved equilibrium water mass
distribution, because the combined effect of surface
density and surface flux errors on water mass
transformation is reduced (Doney et al., 1998b).
In this chapter, we first set out Walin’s theory
of water mass transformation, focusing on how
surface fluxes drive diapycnal flow in the surface
mixed layer. We briefly discuss how lateral mixing within the mixed layer and entrainment at the
5.1
Ocean Surface Water Mass Transformation
William G. Large and A. J. George Nurser
317
OCEAN CIRCULATION AND CLIMATE
Copyright © 2001 Academic Press
ISBN 0-12-641351-7
All rights of reproduction in any form reserved
CHAPTER