first awareness of spatial and temporal sampling
problems, many years would pass before the ramifications of aliasing and poor spatial resolution
were appreciated, and it was not until WOCE that
a systematic attempt was made to resolve such
uncertainties.
World War II marked the start of a third era,
which was characterized by significant technological advances, such as the bathythermograph for
continuously recording temperature as a function
of depth, and by highly organized, intensive
oceanographic surveys that sought quasisynoptic
sampling of large regions. This era also marked
the introduction of non-ship instrumentation such
as drifting and moored buoys. One of the more
imaginative innovations of this period was the
neutrally buoyant float (Swallow, 1955). The principles of this technology provide the basis for
modern autonomous float instrumentation. Like
the previous era, significant advances in theory,
not the least being the first theoretical explanations of the gyres and intense western boundary
currents (for example, Stommel, 1948), also
marked this period.
The modern era has introduced a variety of
autonomous in-situ measurement methods, such as
profiling floats (Davis and Zenk, Chapter 3.2) and
moorings (Hogg, Chapter 4.5), remote sensing
from satellites (Fu, Chapter 3.3; Liu and Katsaros,
Chapter 3.4), and advanced numerical ocean models, made possible by high-performance computers
(Böning and Semtner, Chapter 2.2; Willebrand
and Haidvogel, Chapter 7.2). The growth in computing power has been phenomenal over the last
50 years, as has our ability to sense remotely the
state of the ocean and climate. The previous chapters provide many of the technical and scientific
details behind the contribution from WOCE to
these advances.
These simple reflections are important when we
try to examine the legacy of WOCE for current and
future research programmes and, in particular, for
the efforts to establish a sustained global ocean
observing system. Good research would have taken
place irrespective of WOCE and we would likely
have seen enhanced contributions to the ocean
observing system simply as a first-order response to
growing societal demand. However, as will be
shown in this chapter, WOCE has made distinctive
contributions to scientific and technical development in ways that would not have happened
through chance, and it is these aspects I would like
to draw out here. The views are eclectic in that I
will develop the focus around observations and
concentrate largely on those aspects of the WOCE
observational programme that offer greatest
promise in a follow-on, sustained effort within the
framework of the Global Ocean Observing System
(GOOS), and those aspects which seem destined to
enhance our ability to predict ocean and climate
variability. I leave discussion of the many other
contributions of WOCE to earlier chapters in this
volume.
A convenient starting point is the second goal of
WOCE (WCRP, 1988a):
To determine the representativeness of the
specific WOCE data sets for the long-term
behaviour of the ocean, and to find methods
for determining long-term changes in the ocean
circulation.
Within this goal there were several subobjectives
including:
1 To determine the representativeness of specific
WOCE data sets.
2 To identify those oceanographic parameters,
indices and fields that are essential for continuing measurements in a climate observing system
on decadal time scales.
3 To develop cost-effective techniques suitable for
deploying in a climate observing system.
It seems natural, perhaps even obligatory, that
research programmes such as WOCE include as
part of their objectives a ‘legacy goal’ that aims to
ensure that the practical benefits of the finitelifetime research programme are passed forward to
future generations. For observations, the selection
of elements that are sustained beyond the observing period of the research programme must be
founded on robust scientific understanding and
proven usefulness (Nowlin, 1999). There are of
course many observational methods that have
proven extremely effective for WOCE, but which
may not be strong candidates for sustained implementation, either because the methods were highly
technical and specialized or because they have
limited practical application at this time.
The issues of representativeness and long-term
climate are perhaps the clearest links through to
new research programmes and, in particular, to the
SECTION 7 INSIGHTS FOR THE FUTURE
586
problems, many years would pass before the ramifications of aliasing and poor spatial resolution
were appreciated, and it was not until WOCE that
a systematic attempt was made to resolve such
uncertainties.
World War II marked the start of a third era,
which was characterized by significant technological advances, such as the bathythermograph for
continuously recording temperature as a function
of depth, and by highly organized, intensive
oceanographic surveys that sought quasisynoptic
sampling of large regions. This era also marked
the introduction of non-ship instrumentation such
as drifting and moored buoys. One of the more
imaginative innovations of this period was the
neutrally buoyant float (Swallow, 1955). The principles of this technology provide the basis for
modern autonomous float instrumentation. Like
the previous era, significant advances in theory,
not the least being the first theoretical explanations of the gyres and intense western boundary
currents (for example, Stommel, 1948), also
marked this period.
The modern era has introduced a variety of
autonomous in-situ measurement methods, such as
profiling floats (Davis and Zenk, Chapter 3.2) and
moorings (Hogg, Chapter 4.5), remote sensing
from satellites (Fu, Chapter 3.3; Liu and Katsaros,
Chapter 3.4), and advanced numerical ocean models, made possible by high-performance computers
(Böning and Semtner, Chapter 2.2; Willebrand
and Haidvogel, Chapter 7.2). The growth in computing power has been phenomenal over the last
50 years, as has our ability to sense remotely the
state of the ocean and climate. The previous chapters provide many of the technical and scientific
details behind the contribution from WOCE to
these advances.
These simple reflections are important when we
try to examine the legacy of WOCE for current and
future research programmes and, in particular, for
the efforts to establish a sustained global ocean
observing system. Good research would have taken
place irrespective of WOCE and we would likely
have seen enhanced contributions to the ocean
observing system simply as a first-order response to
growing societal demand. However, as will be
shown in this chapter, WOCE has made distinctive
contributions to scientific and technical development in ways that would not have happened
through chance, and it is these aspects I would like
to draw out here. The views are eclectic in that I
will develop the focus around observations and
concentrate largely on those aspects of the WOCE
observational programme that offer greatest
promise in a follow-on, sustained effort within the
framework of the Global Ocean Observing System
(GOOS), and those aspects which seem destined to
enhance our ability to predict ocean and climate
variability. I leave discussion of the many other
contributions of WOCE to earlier chapters in this
volume.
A convenient starting point is the second goal of
WOCE (WCRP, 1988a):
To determine the representativeness of the
specific WOCE data sets for the long-term
behaviour of the ocean, and to find methods
for determining long-term changes in the ocean
circulation.
Within this goal there were several subobjectives
including:
1 To determine the representativeness of specific
WOCE data sets.
2 To identify those oceanographic parameters,
indices and fields that are essential for continuing measurements in a climate observing system
on decadal time scales.
3 To develop cost-effective techniques suitable for
deploying in a climate observing system.
It seems natural, perhaps even obligatory, that
research programmes such as WOCE include as
part of their objectives a ‘legacy goal’ that aims to
ensure that the practical benefits of the finitelifetime research programme are passed forward to
future generations. For observations, the selection
of elements that are sustained beyond the observing period of the research programme must be
founded on robust scientific understanding and
proven usefulness (Nowlin, 1999). There are of
course many observational methods that have
proven extremely effective for WOCE, but which
may not be strong candidates for sustained implementation, either because the methods were highly
technical and specialized or because they have
limited practical application at this time.
The issues of representativeness and long-term
climate are perhaps the clearest links through to
new research programmes and, in particular, to the
SECTION 7 INSIGHTS FOR THE FUTURE
586