Geology Reference
In-Depth Information
soil erosion during the subsequent window of
vulnerability, while vegetation recovers (e.g.
Imeson & Lavee, 1998; Shakesby & Doerr, 2006).
Eventually, degraded regions subjected to these
pressures can be pushed beyond their resilience
threshold by extreme disturbances, leading to land
abandonment (Puigdefábregas, 1998; Martínez-
Fernández & Esteve, 2005). This can result in
the recovery of natural vegetation and gradual
increase in soil quality (e.g. Vicente-Serrano
et al
.,
2004; Martínez-Fernández & Esteve, 2005), but
severe land degradation and continuing climatic
instability can reinforce desertification processes
in abandoned lands, and as described above, the
process itself is self-reinforcing (Puigdefábregas,
1998; Bakker
et al
., 2005). This conclusion is sus-
tained by several observations of increased soil
erosion processes, particularly gully erosion and
vegetation patchiness in degraded landscapes (e.g.
Cammeraat & Imeson, 1999; Oostwoud Wijdenes
et al
., 1999; Seixas, 2000; Ries & Hirt, 2008). In
these cases, land degradation and desertification
can be irreversible without extensive human
intervention, and misapplied intervention prac-
tices (e.g. intensive irrigation, afforestation) can
even increase the problem (Thornes, 1998;
Puigdefábregas & Mendizabal, 1998; Martínez-
Fernández & Esteve, 2005).
In the past, drylands have experienced periods
of alternating human expansion and contraction
due to changes in climate; humid periods, leading
to increased pressure on natural resources, alter-
nated with dry periods, leading to irreversible deg-
radation if pressure is not released before resilience
thresholds are exceeded (Puigdefábregas, 1998). As
described above, climate change can be expected
to bring about a transition into a drier period for
these regions, leading to lower support for vegeta-
tion productivity and soil structure stability, com-
bined with constant or increasing rainfall intensity
and increased frequency of disturbances. The
interaction between these three factors is likely to
determine the impacts of climate change on soil
erosion in drylands (Imeson & Lavee, 1998).
Desertification is an example of how the com-
bined impacts of climate change on vegetation
cover and soil properties can interact and reinforce
each other, increasing soil erosion in a region
where lower rainfall rates might indicate other-
wise. This complexity should be taken into
account in modelling studies where the aim is to
assess accurately the multiple impacts of climate
change on soil erosion processes.
15.3 Erosion Modelling Approaches
and Climatic Change
The previous section described a number of
potential impacts of climate change on soil ero-
sion drivers; this section presents modelling
strategies used to assess these impacts for partic-
ular problems or locations. Model-based impact
assessment studies can evaluate the potential and
magnitude of these impacts. In the context of cli-
mate change vulnerability assessment frame-
works, such as the one proposed by Adger (2006),
modelling studies can be useful to:
●
assess the sensitivity of soil erosion processes
to climate shifts;
●
evaluate whether climate change will move
soil erosion beyond existing thresholds for deser-
tification, land degradation, or loss of ecosystem
services;
●
test eventual adaptation measures that can
mitigate the impacts of climate change.
Most modelling studies so far have focused on
sensitivity assessment, with a small number (e.g.
O'Neal
et al
., 2005) testing adaptation measures.
Threshold evaluation studies have been ham-
pered by the difficulty in their delineation, due
to the current lack of knowledge on the interac-
tions between soil erosion, soil quality and vege-
tation support (Herrmann & Hutchinson, 2005;
Boardman, 2006). Sensitivity assessment has usu-
ally focused on coupled hydrological and erosion
prediction, attesting to the fact that an accurate
estimate of runoff depth and velocity is at least as
important as the correct estimation of other
soil erosion parameters (Aksoy & Kavvas,
2005). Modelling exercises have differed in terms
of objectives, model used, spatial and temporal
extent of the study, and climate change sce-
nario strategy. The processes used to represent
