Geoscience Reference
In-Depth Information
24.4
Climate change and dunes
respond to climate change and, in turn, influence the
system. Scenarios of potential response are therefore
the most likely product of this approach (Goudie, 2006;
Tooth, 2008).
Intermediate methods see some post-processing of the
climate change projections, typically by downscaling the
data to suitable spatial resolution. Empirical equations
calibrated to represent the current behaviour of the system
in question (e.g. dune mobility) are then used as the basis
to compute the new behaviour of the system given the
quantified and downscaled changed climate. Intermediate
methods in the case of arid zone geomorphology are likely
to be the most fruitful and have indeed made an impact
on the literature (e.g. Thomas, Knight and Wiggs, 2005;
De Wit and Stankiewicz, 2006).
The most sophisticated and complex approach involves
driving numerical models of the system of interest with
the climate change projections as the input to these mod-
els. There are few systems in geomorphology for which
numerical models have been developed although the value
of such models is enormous. Global climate models are
one example of numerical models. Typically these mod-
els represent space in terms of grid boxes within which
equations that govern the key processes in the system
are solved. With the equations in prognostic form, the
system can be integrated through time steps with new in-
puts to some of the controlling variables. Without climate
models, the issue of global warming and climate change
would most certainly only be at the stage of vague de-
bate and conjecture. The models have allowed both the
attribution of observed climate change to anthropogenic
emissions and the projection of future climate, resulting
in climate change becoming an established research prior-
ity. The background science evolved over many decades
through the economic imperative of weather forecast-
ing and was facilitated by (and indeed stimulated) the
fastest available supercomputers. Geomorphology has
lacked this imperative and therefore a commensurate in-
put of resources, although hydrological models are an
exception.
Of the Earth Systems for which numerical models have
been developed, the cycle of mineral dust is a leading
example. The need to construct models for the global
simulation of dust came about largely because of the im-
portant feedbacks that dust exerts on the climate system
itself. As a result, many of the leading climate mod-
els include components that simulate the deflation, ad-
vection and deposition of dust as well as its interaction
with the Earth System. While there has been progress
with the development numerical modelling in geomor-
phology, much of the capability is geared to timescales
Demonstration of the dynamism and sensitivity of dune
systems to changes in climate through concerted appli-
cation of dating techniques has been one of the major
achievements of aeolian geomorphology in recent decades
(see Chapter 3). Building on work in the Kalahari, which
points to episodes of punctuated aridity and dune mo-
bilisation during the Quaternary (Stokes, Thomas and
Washington, 1997), Thomas, Knight and Wiggs (2005)
have demonstrated the likelihood of twenty-first century
mobility of these linear dunes. Extending from northern
southern Africa to Angola and Zambia, these dunes are
currently stable with observed sand transport confined by
low erosivity and well-developed vegetation cover. The
investigation used a modified empirical index of dune mo-
bility (an example of the intermediate approach discussed
in the previous section), such that
U
3
/
A
p
,
GCM
=
(
P
lag
/
E
p
,
lag
+
P
rainy
/
E
p
,
rainy
)
where:
U
3
=
the cube of the mean wind speed
P
lag
/
E
p, lag
=
residual effect of recent rainfall and potential
evaporation, such that
P
lag
=
P
0
)/2, where
P
−
1
is precipitation in the previous month and
P
0
is rainfall
in the current month
(
P
−
1
+
E
p, lag
=
E
p, 0
)/2, where
E
p,
−
1
is potential evapo-
transpiration in the previous month and
E
p,0
is potential
evapotranspiration in the current month
(
E
p,
−
1
+
P
rainy
/
E
p, rainy
effect of rainy season precipitation and
potential evaporation on soil moisture, such that
P
rainy
=
=
(
P
N
+
P
D
+
P
J
···
)/
m
and
E
p, rainy
=
(
E
p, N
+
E
p, D
+
E
p, J
···
N (November), D (December),
J (January), and so on is the month under consideration
within the rainy season
)/
m
, where
m
=
This model was therefore specifically adapted for the
seasonal climate of southern Africa, and was driven with
data on moisture availability and erosivity derived from
the output of three global climate models, each forced by
several emission scenarios. Thomas, Knight and Wiggs
(2005) used model outputs, run for the twenty-first century
on a monthly basis, to demonstrate that dunefields across
this broad region are potentially reactivated by the end of
