Current warming and likely future impacts - Climate Change: Biological and Human Aspects - page 348

especially if climate change disrupted water-distribution events such as the monsoon.
Even if major individual water-distribution events remained unaltered, globally water
distribution about the planet via the atmosphere will almost certainly change. Indeed,
this is connected to the predictions of regional precipitation change and water avail-
ability noted earlier in the case studies of climate and natural systems of the USA and
UK. Water availability is not only important for ecosystem function and geophysical
processing but (as shall be explored in the next chapter) it also fundamentally affects
human ecology and therefore economic activity. Further to the above work on the
hydrological cycle as it affects Arctic rivers, a global assessment of stream flow and
water run-off suggests that there will be considerable change in many areas by the
mid 21st century compared with the 20th century.
In 2005 an ensemble of 12 climate models was used by Milly and colleagues from
the US Geological Survey to see how coherently they forecast changes in run-off
(run-off is defined as precipitation less evapotranspiration). The picture emerging is
that as the climate warms there is more ocean evaporation and hence precipitation,
but also that this precipitation is subject to more evaporation. The picture is therefore
somewhat complex, with some areas having more precipitation but paradoxically
less run-off due to this increased evapotranspiration. Overall, most areas are likely to
see change in run-off and this change is likely to continue as climate change (in the
thermal sense) progresses through to the middle of the 21st century (which is as far
as this analysis forecasted). Initial increases of run-off seen in the 20th century are
projected to reverse in some regions in this century, namely eastern equatorial South
America, southern Africa and the western plains of North America. The modelled
drying of the Mediterranean extends north from Spain and Greece (which we were
beginning to see by the end of the 20th century) to south-eastern Britain, France,
Austria, Hungary, Romania and Bulgaria as well as the sub-Russian states. In many
regions all the models in the ensemble agreed on the likely mid-21st-century change.
These agreements include increases of typically 10-40% in the high latitudes of
North America and north Eurasia (as in Wu et al., 2005), in the La Plata basin of
South America and in eastern equatorial Africa as well as some of the major islands
of the eastern equatorial Pacific. Agreements on decreasing run-off include southern
Europe, the Middle East, mid-latitude western North America and southern Africa
(Milly et al., 2005).
And then there is the weather factor, and specifically how individual weather events
(such as cyclone/hurricane strength) are altered by global climate change. In 2007
US climatologists Ryan Sriver and Matthew Huber calculated the effect of tropical
cyclones on depth of sea-water mixing. Their results indicated that tropical cyclones
are responsible for significant heat transport by mixing heat from the air into the sea.
They concluded that around 15% of peak ocean heat transport may be associated
with vertical mixing by tropical cyclones and that the magnitude of this mixing is
related to sea-surface temperature. This phenomenon was not at the time included
in computer models (and so is another example of new discovery driving ongoing
model improvement).
The possibility of changes in oceanic and atmospheric circulation is one current
focus of climate research. It appears that circulation systems interact with the global
climate system through a number of feedback cycles (see Chapter 1). These result