Geoscience Reference
In-Depth Information
matter supplied to coastal waters mainly via rivers). These components complicate
the accurate determination of chlorophyll. Intensive efforts to use optical data from
several spectral bands to determine all three components have, as yet, been only
partially successful. A recent and exciting development in the use of satellite sensors
in biogeochemistry of the open ocean is the detection of natural fluorescence from
surface chlorophyll, which can provide information on phytoplankton physiology,
such as nutrient stress (Behrenfeld et al., 2009 ).
The measurements of the sea surface elevation by microwave sensors (altimeters)
on satellites have provided pictures of geostrophic currents and the mesoscale circu-
lation in the deep ocean but works less well in shallow seas because of the large
spatial scale of the 'footprint' of the microwave beam and interference from land at
the coastal boundary. A more useful microwave technique for shelf seas is Synthetic
Aperture Radar (SAR) which detects changes in surface roughness and thereby, in
favourable conditions, allows the detection of internal wave motions. We will use
some examples of this in Chapter 10 .
1.6
The role of models - a philosophy of modelling
......................................................................................................................
Numerical models are another important and powerful tool available to oceanog-
raphers. Increased computing power and improved analytical software such as
M atlab # have facilitated the construction of a great variety of numerical models
of ocean systems. Such models can be used for many purposes, but from the scientific
point of view, they play a crucial role in the testing of our conceptual models of how
the ocean works. Here we set out the approach to the use of models in our own
research which will be reflected in later chapters of the topic.
As in other branches of science, oceanography advances as new observations
stimulate hypotheses about how the system works. Predictions, based on one or
more hypotheses, are then tested against further observations, eventually leading to
improved understanding of the system. A major difficulty in oceanography is that we
are rarely in a position to be able to carry out manipulative experiments on the ocean.
Instead, we make observations of the real environment, and then have to try to tease
apart the relative influences of the possible forcing mechanisms and interactions.
In dealing with such complex systems in which many diverse processes are operating,
the testing of hypotheses can often be undertaken only through the use of numerical
models. We can think of the construction of the model as the assembling of a number
of hypothetical ideas about how the system works which together amount to a
'compound hypothesis'. Comparing the results of the model with observational data
then constitutes the required test of the compound hypothesis.
If the model compares well with observations, we conclude that the compound
hypothesis is provisionally verified (i.e. not falsified) and our understanding is
increased. We may then seek more exacting tests of the hypothesis in different
scenarios and against new observations. We can also use the model to switch different
forcing mechanisms on or off, which is a powerful way of gaining insight into how
Search WWH ::




Custom Search