climate variability on time scales of seasons to a year
for those areas affected by the El Niño-Southern
Oscillation (ENSO) events. Second, the field phase
of WOCE has produced, for the first time, a global
three-dimensional view of ocean structure as well
as of the trace substance distribution of the global
ocean (excluding the Arctic Basin and – due to
operational constraints – parts of the Southern
Ocean). At the same time there has been a significant improvement in the quality of ocean models
and the WOCE data have become the basis for
testing the ocean modules of climate (coupled
ocean–atmosphere–land surface) models.
This book is devoted to evaluation of ocean circulation as determined from the WOCE (and other)
data, to the development of ocean models and to
the understanding of the ocean’s circulation as a
component of the climate system. Therefore I will
concentrate on the role of the ocean as a component of the climate system.
1.1.2 The scientific approach to the
complex climate system
Complex systems are characterized by their ability
to develop transient organized structures when
reacting to internal or external forcing as a result of
their internal non-linear dynamics. Examples are
galaxies, stars, planets, the atmosphere, the ocean,
ecosystems, our body and a single cell. The preferred scientific approach to complex systems is via
experiments under controlled conditions for simpler
subsystems, leading to models that approach reality
and thus can be used for practical applications.
In the geosciences, however, where deliberate
experiments are impossible, the approach is via
long-term, nearly global observations that may lead
to a degree of understanding that allows numerical
prediction for restricted time scales. The most successful application of this method is deterministic
weather forecasting. In principle, weather forecasting can only be successful up to about 2 weeks and,
at present, mostly does not incorporate more from
the ocean than surface temperature as a starting
field for deterministic predictions up to 1 week
ahead. The climate system shows some statistical
stability in its long-term behaviour despite reacting
with deterministic instability to changed initial
fields. Thus, climate predictions are probabilistic
predictions that go beyond the predictability barrier for deterministic weather forecasts.
Since there were, until recently, no routine
observations of large parts of the oceans’ interior,
the barrier to climate variability predictions on
monthly to seasonal time scales could not be surmounted. An exception was the application of very
simple empirical rules to certain regions where persistent sea surface temperature anomalies allowed
some extrapolation in time. Long-term observations on nearly global scale are needed. The highest priority for seasonal forecasts was for upper
ocean observations in areas with especially strong
seasonal to interannual variability. This, together
with developing coupled ocean–atmosphere models, was exactly the strategy pursued by TOGA
scientists. They were able to implement the
TOGA observing system, including the Tropical
Atmosphere/Ocean (TAO) Array, a set of up to
64 moored buoys across the tropical Pacific,
measuring surface meteorological variables and
upper ocean structure and expendable bathythermograph measurements from Voluntary Observing Ships (VOS).
This observing system brought the breakthrough
to physically based climate anomaly predictions
for ENSO-affected areas by using coupled ocean–
atmosphere models assimilating near-real-time
observations. The success is due to the intrinsic
time scales of ocean–atmosphere interaction of up
to a year in the Pacific caused by the travel time of
equatorial ocean Kelvin and Rossby waves across
the Pacific basin.
To understand and predict the full spectrum of
climate variability, long-term global ocean observations are required (for a fuller discussion see
Wunsch, Chapter 2.1 and Smith, Chapter 7.4).
However, prior to WOCE no such system was
considered feasible. Thus a major challenge for
WOCE was to demonstrate that feasibility. Given
the complexity of the oceans and the limited
resources available, this required a range of different techniques (observations from research ships
and merchant vessels, surface drifters, subsurface
floats, moored instruments and, of course, satellites). All of these observing systems required some
degree of development.
WOCE’s main task was to observe for the first
time (largely on a basin-by-basin manner) the
three-dimensional structure of the global ocean
as the basis for ocean model improvement needed
for more reliable climate models. This required a
commitment by the research community to the
SECTION 1 THE OCEAN AND CLIMATE
4
for those areas affected by the El Niño-Southern
Oscillation (ENSO) events. Second, the field phase
of WOCE has produced, for the first time, a global
three-dimensional view of ocean structure as well
as of the trace substance distribution of the global
ocean (excluding the Arctic Basin and – due to
operational constraints – parts of the Southern
Ocean). At the same time there has been a significant improvement in the quality of ocean models
and the WOCE data have become the basis for
testing the ocean modules of climate (coupled
ocean–atmosphere–land surface) models.
This book is devoted to evaluation of ocean circulation as determined from the WOCE (and other)
data, to the development of ocean models and to
the understanding of the ocean’s circulation as a
component of the climate system. Therefore I will
concentrate on the role of the ocean as a component of the climate system.
1.1.2 The scientific approach to the
complex climate system
Complex systems are characterized by their ability
to develop transient organized structures when
reacting to internal or external forcing as a result of
their internal non-linear dynamics. Examples are
galaxies, stars, planets, the atmosphere, the ocean,
ecosystems, our body and a single cell. The preferred scientific approach to complex systems is via
experiments under controlled conditions for simpler
subsystems, leading to models that approach reality
and thus can be used for practical applications.
In the geosciences, however, where deliberate
experiments are impossible, the approach is via
long-term, nearly global observations that may lead
to a degree of understanding that allows numerical
prediction for restricted time scales. The most successful application of this method is deterministic
weather forecasting. In principle, weather forecasting can only be successful up to about 2 weeks and,
at present, mostly does not incorporate more from
the ocean than surface temperature as a starting
field for deterministic predictions up to 1 week
ahead. The climate system shows some statistical
stability in its long-term behaviour despite reacting
with deterministic instability to changed initial
fields. Thus, climate predictions are probabilistic
predictions that go beyond the predictability barrier for deterministic weather forecasts.
Since there were, until recently, no routine
observations of large parts of the oceans’ interior,
the barrier to climate variability predictions on
monthly to seasonal time scales could not be surmounted. An exception was the application of very
simple empirical rules to certain regions where persistent sea surface temperature anomalies allowed
some extrapolation in time. Long-term observations on nearly global scale are needed. The highest priority for seasonal forecasts was for upper
ocean observations in areas with especially strong
seasonal to interannual variability. This, together
with developing coupled ocean–atmosphere models, was exactly the strategy pursued by TOGA
scientists. They were able to implement the
TOGA observing system, including the Tropical
Atmosphere/Ocean (TAO) Array, a set of up to
64 moored buoys across the tropical Pacific,
measuring surface meteorological variables and
upper ocean structure and expendable bathythermograph measurements from Voluntary Observing Ships (VOS).
This observing system brought the breakthrough
to physically based climate anomaly predictions
for ENSO-affected areas by using coupled ocean–
atmosphere models assimilating near-real-time
observations. The success is due to the intrinsic
time scales of ocean–atmosphere interaction of up
to a year in the Pacific caused by the travel time of
equatorial ocean Kelvin and Rossby waves across
the Pacific basin.
To understand and predict the full spectrum of
climate variability, long-term global ocean observations are required (for a fuller discussion see
Wunsch, Chapter 2.1 and Smith, Chapter 7.4).
However, prior to WOCE no such system was
considered feasible. Thus a major challenge for
WOCE was to demonstrate that feasibility. Given
the complexity of the oceans and the limited
resources available, this required a range of different techniques (observations from research ships
and merchant vessels, surface drifters, subsurface
floats, moored instruments and, of course, satellites). All of these observing systems required some
degree of development.
WOCE’s main task was to observe for the first
time (largely on a basin-by-basin manner) the
three-dimensional structure of the global ocean
as the basis for ocean model improvement needed
for more reliable climate models. This required a
commitment by the research community to the
SECTION 1 THE OCEAN AND CLIMATE
4