The Atmosphere and Climate Modeling - Introduction to Carbon Capture and Sequestration - page 91

Figure 2.5.11 The velocity of climate change
The fi gure shows the average speed at which an ecosystem needs to move in km per
year to ensure a constant temperature given the climate change predictions following
the A1B scenario. Figure by Loarie et al. [2.14 ], reproduced with permission from
Macmillan Publishers.
There are two types of answers to this question. Philosophers will tell
you that the idea of a True theory is problematic to start with. Scientists
will point to the various sorts of uncertainty in their models [2.15]. Both
answers can be read as a version of “ We're not sure .” This might sound
like grist to the mill of the climate skeptics, but it is not. To see why not,
it is important to understand why “ we're not sure ” is not the same as “ our
models may well be wrong, and until we're absolutely certain there's no
reason to act on them .”
For philosophers, the main reason to doubt the idea of a single, True
theory is this: given a set of data, there are always many different theories
that are mutually incompatible, but all equally consistent with the data. If
the data exhaust the available evidence, we have no evidential reason for
believing one theory over another: our theory is underdetermined by the
data. To complicate matters, if we fi nd that our theory confl icts with
experiment, there is no single right thing to do to make our theory more
True. Because all the parts are interconnected, the experiment does not
simply prove that one particular hypothesis is wrong; and because of
underdetermination, there are no hypotheses that we must hold true
come what may. From this point of view, making a climate model is not
like building a house on the solid foundations of the laws of physics, but
rather like a ship that we have to rebuild at open sea, replacing the parts
with new scientifi c insights, and keeping it afl oat as best as we can [2.16].