Geoscience Reference
In-Depth Information
disciplines, largely through their careful assertion (and empirical verifi cation) of
mutual causality between environmental change, landscape change and human
societies in a non-deterministic fashion. This stands in sharp contrast to the work
of scholars from other disciplines on the relationship between human societies and
their environments, in which the environment is seen to play a simplistically deter-
ministic role, overly limiting human agency, but also, at times, insidiously displacing
blame for inequality within human societies to 'natural' environmental factors
(e.g. Wittfogel, 1957; Landes, 1998; Sachs, 2005).
Water and Risk: (De)Constructing Water Hazards
The mutual imbrication of nature and society has also been a core theme of geo-
graphical analyses of natural hazards, a focal point of which has been water-related
hazards: fl oods, droughts, landslides and hurricanes. In the view of the 'hazards
school' in environmental geography, research on physical processes must necessarily
be integrated with research on the social construction of hazards if risk and vulner-
ability are to be comprehensively analysed and mitigated. For example, environ-
mental geographers have evaluated fl ooding and landslide hazards by studying the
siting of residential housing developments in fl oodplains or on landslide-prone
slopes. Physical geographers have documented the relationship between geomor-
phological and hydrological processes and risk; for example, research on slope sta-
bility and landslide risk is used to support analyses of slope failures (see, e.g., Petley,
2004; Glade et al., 2005; Hufschmidt et al., 2005).
Hazards research integrates the work of physical geographers with assessments
of vulnerability and risk. The work of geographer Gilbert White, for example,
entailed the identifi cation and classifi cation of perceptions of, and responses to,
fl ood hazards, enabling more accurate analyses of responses to fl oods in the United
States (Hinshaw, 2006). White argued that an over-reliance on structural hydraulic
works (such as levees, dams and barrages) in the United States had increased
damages caused by fl ooding, rather than decreasing them, because the high degree
of public confi dence in structural works enticed development in fl ood-prone areas
(White et al., 1958). White and his students demonstrated a lack of willingness of
individuals living in fl ood-prone areas to adopt protective measures such as insur-
ance or fl ood-proof doors and windows, refl ecting a reduction in perception of risk
due to the presence of engineered fl ood control structures (Kates 1962; Kreutzwiser
et al., 1994; Shrubsole et al., 1997). White's work resulted in the development of
the National Flood Insurance Program in the United States and was infl uential in
the promotion of 'non-structural' solutions to fl ooding, such as regulatory restric-
tions on the use of fl oodplains (Kates and Burton, 1986; Tobin and Montz,
1997).
Hurricane Katrina serves as an illustration of the relevance of hazards research
on fl oodplain development and fl ood control. Sited in a region in which hurricanes
regularly make landfall, New Orleans' urban expansion into the surrounding fl ood-
plain has been enabled by the construction of levees and draining of wetlands
(Lewis, 2003; Colten, 2005). As the city sank, the coastal wetlands protecting it
from storm surges in the Gulf shrank. The extensive wetlands which used to absorb
annual fl oods and storm surges in southern Louisiana have dramatically subsided,
as upstream damming and channelisation of the river has reduced sediment loads
and diverted them out to the Gulf; Louisiana has lost an estimated 1 million acres
