with 10-day and 3° resolution. These data will
allow us to establish accurately the transports and
seasonal cycle of the ocean’s heat engine and
hydrological cycle and to track climatic variations
on time scales from years to decades. Because
Argo data will be made publicly available within
hours of collection, it will be possible to watch the
El Niño-Southern Oscillation (ENSO) develop,
see the upper ocean changes produced by tropical storms, and know almost immediately how
strongly convection is ventilating the ocean in any
winter. These data will not only support new areas
of research but, through data assimilating modelling, will make possible now-casts and forecasts of
the ocean on a regular basis. Details on Argo can
be found on the web at http://www.argo.ucsd.edu.
The fast evolution of observational technology
over the 1990s makes it foolhardy to guess what
the next decade will bring, but a hint of what is
ahead can be obtained by extrapolation of activities
already underway. For observation of subsurface
velocity, it is likely that both continuously tracked
neutrally buoyant RAFOS floats and autonomous
floats will be needed. For acoustic floats a major
limitation to economical sampling can be overcome
by widespread deployment of high-energy sound
sources. It is, for example, entirely feasible today to
install enough sound sources that a float could be
continuously tracked anywhere in the tropical or
North Atlantic. Since sound sources, like radio stations, can serve different users, the presently rather
simply structured network of moored sound
sources will require greater international coordination in the future. The benefits of an organized
RAFOS network will be linked with the responsibility of contributing parties to maintain such
arrays of sound sources over an extended period of
time on a basin-wide scale. Miniaturization of
receiver electronics and production in great numbers could result in a significant decline in float
prices. This, and development of new sensors,
could open other fields of research and even commercial opportunities such as fish tracking with
high spatial resolution (Rossby, personal communication). Biogeochemical sensors now under development may be suitable for Lagrangian instruments
and an operational oxygen sensor on RAFOS floats
has been successfully demonstrated by Rossby et al.
(2000) in the North Atlantic.
For autonomous floats the availability of Global
Positioning System (GPS) navigation and polar
orbiting satellites supporting two-way communication will reduce the length of on-surface periods,
making trajectories more nearly Lagrangian. This
will also allow float missions to be adjusted after
deployment by, for example, allowing cycle times
to be reduced as a float approaches concentrated
boundary currents or regions where small-scale
structure is of interest. The new satellite communication will increase the amount of data that is
feasibly relayed, allowing floats to report measurements of more variables, at higher precision and
higher depth resolution. The same satellite developments make feasible autonomous gliders that can
be used to gather time series of profiles or sections
at predetermined locations. Use of the ocean’s thermal gradient to supply power to both profiling
floats and gliders might greatly extend float life in
the well-stratified parts of the ocean.
Techniques for deploying floats are also expanding. Repeated deployments from one location are
made possible by the ‘float park’ concept (Zenk
et al., 2000) in which floats are inexpensively
moored until they are deployed on a schedule.
Seeding autonomous floats from commercial shipping vessels and their air deployment will likely
increase as the use of floats moves from specific deployments to the maintenance of observing
arrays.
Success of the Argo programme depends on
stable salinity sensors and several different sensors
have been used on profiling floats. Long-term stability has been a problem with some so that early
autonomous profiling has been most useful in the
upper ocean where salinity changes are large
enough that salinity uncertainties do not cloud the
picture. Progress is being made and recent results
suggest that long-term stability to O(0.01) in salinity is feasible. The T versus S plot in Fig. 3.2.12
(see Plate 3.2.12, p. 172), from an 18-month record
from a P-ALACE deployed in the subtropical
North Atlantic by S. Riser, shows that recent longterm salinity profiling has reached the point where
relatively subtle climatic variations of salinity can
be reliably observed autonomously. This figure
shows that observed salinity variations near 700 m
depth are O(0.01), substantially smaller than
the O(0.1) climatic variations already seen in the
historical record.
The full power of neutrally buoyant devices to
observe ocean properties will be realized only when
accurate, long-lived sensors become available to
SECTION 3 NEW WAYS OF OBSERVING THE OCEAN
138
allow us to establish accurately the transports and
seasonal cycle of the ocean’s heat engine and
hydrological cycle and to track climatic variations
on time scales from years to decades. Because
Argo data will be made publicly available within
hours of collection, it will be possible to watch the
El Niño-Southern Oscillation (ENSO) develop,
see the upper ocean changes produced by tropical storms, and know almost immediately how
strongly convection is ventilating the ocean in any
winter. These data will not only support new areas
of research but, through data assimilating modelling, will make possible now-casts and forecasts of
the ocean on a regular basis. Details on Argo can
be found on the web at http://www.argo.ucsd.edu.
The fast evolution of observational technology
over the 1990s makes it foolhardy to guess what
the next decade will bring, but a hint of what is
ahead can be obtained by extrapolation of activities
already underway. For observation of subsurface
velocity, it is likely that both continuously tracked
neutrally buoyant RAFOS floats and autonomous
floats will be needed. For acoustic floats a major
limitation to economical sampling can be overcome
by widespread deployment of high-energy sound
sources. It is, for example, entirely feasible today to
install enough sound sources that a float could be
continuously tracked anywhere in the tropical or
North Atlantic. Since sound sources, like radio stations, can serve different users, the presently rather
simply structured network of moored sound
sources will require greater international coordination in the future. The benefits of an organized
RAFOS network will be linked with the responsibility of contributing parties to maintain such
arrays of sound sources over an extended period of
time on a basin-wide scale. Miniaturization of
receiver electronics and production in great numbers could result in a significant decline in float
prices. This, and development of new sensors,
could open other fields of research and even commercial opportunities such as fish tracking with
high spatial resolution (Rossby, personal communication). Biogeochemical sensors now under development may be suitable for Lagrangian instruments
and an operational oxygen sensor on RAFOS floats
has been successfully demonstrated by Rossby et al.
(2000) in the North Atlantic.
For autonomous floats the availability of Global
Positioning System (GPS) navigation and polar
orbiting satellites supporting two-way communication will reduce the length of on-surface periods,
making trajectories more nearly Lagrangian. This
will also allow float missions to be adjusted after
deployment by, for example, allowing cycle times
to be reduced as a float approaches concentrated
boundary currents or regions where small-scale
structure is of interest. The new satellite communication will increase the amount of data that is
feasibly relayed, allowing floats to report measurements of more variables, at higher precision and
higher depth resolution. The same satellite developments make feasible autonomous gliders that can
be used to gather time series of profiles or sections
at predetermined locations. Use of the ocean’s thermal gradient to supply power to both profiling
floats and gliders might greatly extend float life in
the well-stratified parts of the ocean.
Techniques for deploying floats are also expanding. Repeated deployments from one location are
made possible by the ‘float park’ concept (Zenk
et al., 2000) in which floats are inexpensively
moored until they are deployed on a schedule.
Seeding autonomous floats from commercial shipping vessels and their air deployment will likely
increase as the use of floats moves from specific deployments to the maintenance of observing
arrays.
Success of the Argo programme depends on
stable salinity sensors and several different sensors
have been used on profiling floats. Long-term stability has been a problem with some so that early
autonomous profiling has been most useful in the
upper ocean where salinity changes are large
enough that salinity uncertainties do not cloud the
picture. Progress is being made and recent results
suggest that long-term stability to O(0.01) in salinity is feasible. The T versus S plot in Fig. 3.2.12
(see Plate 3.2.12, p. 172), from an 18-month record
from a P-ALACE deployed in the subtropical
North Atlantic by S. Riser, shows that recent longterm salinity profiling has reached the point where
relatively subtle climatic variations of salinity can
be reliably observed autonomously. This figure
shows that observed salinity variations near 700 m
depth are O(0.01), substantially smaller than
the O(0.1) climatic variations already seen in the
historical record.
The full power of neutrally buoyant devices to
observe ocean properties will be realized only when
accurate, long-lived sensors become available to
SECTION 3 NEW WAYS OF OBSERVING THE OCEAN
138