Geology Reference
In-Depth Information
climate change studies, but existing scenarios
can be taken into account by models, as shown in
a few examples in this chapter. Feedbacks between
soil erosion, agricultural productivity and land-
use changes (e.g. Bakker
et al
., 2004, 2005; Avni,
2005) are less well known and more difficult to
take into account; for example, Nunes (2007) sug-
gested that in regions of degraded soils, long-term
soil erosion might have as much impact on vege-
tation productivity as changes in climate, with a
possible feedback relation with soil erosion.
In particular, one poorly understood subject is the
relationship between desertification and climatic
or land degradation thresholds; more data and
research on this issue would allow its inclusion
in soil erosion studies. This will be particularly
important when the focus is on semi-arid 'thresh-
old regions', with dry climates tending to become
drier. One possible approach to consider these
interactions is to link physical erosion models
with socio-economic models; however, interdis-
ciplinary models of this nature are still rare, and
existing approaches (e.g. Wu
et al
., 2007) are still
not developed enough to be combined with rela-
tively complex erosion models.
Furthermore, most of the studies do not con-
sider adaptation to climate change. The studies
that did tested different adaptation methods (e.g.
conservation and no tillage) to assess whether they
are efficient tools to counterbalance the negative
impacts of climate change. A further issue that
needs to be addressed is the interaction between
measures to adapt to different climate change
impacts. For example, the study by O'Neal
et al
.
(2005), reported above, shows that agricultural
adaptations to future climate aimed at increasing
productivity might also lead to increased erosion.
On the other hand, Lacombe
et al
. (2008) showed
how the adoption of extensive soil and water con-
servation measures in the past has led to a decrease
in available water downstream. These issues need
to be addressed, particularly at the watershed scale
and in light of proposing integrated watershed
management methods to adapt to multiple impacts
of climate change on water resources, floods, agri-
cultural productivity, soil erosion, nutrient exports,
and so on.
These and other limitations must be addressed
before the impacts of climate change can be evalu-
ated with some measure of confidence. Some ongoing
research projects are proposing modelling strategies
that take these issues into account; one such exam-
ple is the ongoing MESOEROS21 project, aiming
to study the impact of global climate change on
soil erosion in the Mediterranean (MESOEROS21,
2006). The modelling approach takes into account
soil erosion drivers such as land-use changes,
intensification of irrigation and desertification, as
impacted by climate change, as well as direct
impacts on soil erosion; in some cases, more com-
plex vegetation models are used to generate vegeta-
tion cover scenarios. Changes to erosion processes
are explored with complex models in intensively
monitored small catchments, but results are also
upscaled to the regional scale using simpler models
validated with long-term soil erosion databases for
the region. Finally, this project also studies the vul-
nerability to soil erosion - the expected impact of
changes to soil erosion rates on soil water storage,
crop productivity, and so on. MESOEROS21 shows
an example of how more complex, and hopefully
more robust, approaches can tackle a number of
the research gaps outlined above.
In summary, this chapter has hopefully shown
how climate change can impact soil erosion
through a number of processes, many of which
are not linearly dependent upon changes to rain-
fall patterns. Current soil erosion modelling
approaches have been developed and applied to
test the impacts of climate change for different
case studies involving different climate scenarios
and locations, providing an insight into the proc-
esses linking climate and soil erosion. However,
a significant number of research gaps are still
present, including the upscaling of results for
larger spatial scales; uncertainties in climate
change scenarios and their impacts, particularly
on soil erosion drivers not fully taken into account
by current models; and the links between climate
change, soil erosion and land-use changes involv-
ing socio-economic as well as biophysical proc-
esses. More complex modelling approaches can be
developed to address these limitations; however,
some effort in other areas such as data collection
