Geography Reference
In-Depth Information
with events occurring independently of each
other. In the United States, using hail data for a
twenty-one-year period at twenty locations and
tornado occurrence data for 1916-80, they
demonstrated that both hazards were more likely
to occur in a particular year if they had been
experienced in the previous year. They argued that
if the same holds true for a hazard such as early or
late frost, then rotating crops to plant a frost-
resistant crop in any year following early or late
frost damage would save millions of dollars of crop
damage.
In drought studies, in addition to the
considerable research on drought impacts and
human response, particularly in the Sahel (e.g.
Trilsbach and Hulme 1984; Hulme 1986;
Trilsbach 1987; Walsh
et al
. 1988), one of the most
useful contributions of geography has been in
developing definitions of droughts and dry
periods that are relevant to particular societies or
specific physical or human issues. Thus Hulme
(1987), in defining wet season length and
character, attempted to incorporate farmers'
perception and decision making on crop planting
dates into his analysis of the impact of extreme
drought in semi-arid central Sudan and the
response of the rural water supply system. In rain
forest environments, where much shorter and less
intense periods of low rainfall constitute
'drought', Walsh (1996a) used archival rainfall
records and definitions of drought specifically
geared to tropical rain forest transpirational
demand in identifying changes in drought
magnitude frequency in Sabah (Borneo) over the
past 120 years and exploring their implications
for rain forest dynamics.
Relatively few studies have investigated changes
in heavy rainstorm frequency. An early study by
Howe
et al
. (1967) linked an increase in flood
frequency in the Rivers Wye and Severn since the
1920s to an increase in heavy rainstorm frequency
in central Wales. Later work demonstrated an
increase in heavy rainstorm frequency since 1925
to be widespread in south Wales and linked to a
parallel increase in flood frequency of the Rivers
Tawe and Ebbw (Walsh
et al
. 1982). The United
Kingdom Climate Change Impacts Review
Group (1996), which includes geographers on its
panel, used the simulated daily rainfall data outputs
of the UKTR model to suggest that return
periods of heavy daily rainfalls in the twenty-first
century will shorten significantly in summer in the
north and in winter throughout the United
Kingdom and pointed to the consequences for
flooding and soil erosion. Boardman
et al
. (1996)
found that soil erosion resulting from a (currently)
1000-year return period rainstorm in which up
to 128.7 mm fell over parts of Berkshire and
Oxfordshire in central England on 26 May 1993
was particularly severe on fields with spring-
planted crops, averaging 66 m
3
ha
-1
on a maize
field. They warned that the current trend towards
spring-planted (rather than winter-planted) crops
would increase the erosional risk of any increase
in large rainstorm frequency.
In the tropics a marked reduction in heavy
rainstorm frequency has accompanied the drought
epoch since 1965 in the Sudan and has been
linked to a decline in
wadi
flows, shallow
groundwater recharge and rural water supply
(Hulme 1986; Walsh
et al
. 1988), and a decline not
only in flood frequency but also in human
perception of flood hazards, resulting in large-scale
flooding of poorly located squatter settlements of
recent migrants in Khartoum and Omdurman in
the exceptional rainstorm of August 1988 (Walsh
et al
. 1994). In the eastern Caribbean, significant
increases in the return periods of large daily
rainstorms have marked the two dry epochs of
1899-1928 and since 1959 compared with the
wetter late nineteenth century and 1929-58
periods (Walsh 1998).
Finally, geographers involved in
multidisciplinary teams assessing global warming
and its impacts have been paying increased
attention to formulating scenarios regarding
changing frequencies and impacts of weather
extremes (IPCC 1996; United Kingdom Impacts
Review Group 1996; Hulme andViner 1998) and
Parry and Carter (1997) have incorporated
extreme event considerations in their manual on
climate impact and adaptation assessment.
