Laserfiche WebLink
<br /> <br />28 <br />Deaths <br />160 <br />140 <br />120 <br />100 <br />80 <br />60 <br />40 <br />20 <br /> <br />ROGER A. PIELKE. JR. <br /> <br /> <br />o ~ ~ g\~ ~ ~ ~ ~ ~ ~ ~ g g ~ ~ ~ ~ ~ ~ i m ~ ~ <br />~ - - - - - - - - - - - - - - - - - - - - - - <br /> <br />Figure 2. Flood deaths in the United States; 1903-1994; each point represents average annual deaths <br />of previous 25 years. <br /> <br />2.7. DATA ON FLOOD DAMAGES IS A PROXY FOR FLOOD RISK <br /> <br />As in the case of trends in people at risk to floods. analysts have sought to use <br />trends in flood damages as a proxy for trends in property at risk to floods. However, <br />it is alleast as difficult to form definitive conclusions about vulnerability from the <br />damage data as it is from the casualty data. Flood damages occur every yea~ in <br />various places around the United Stales. Such damages, per se. are not sufficl~nt <br />evidence of a policy problem. As the Task Force on Federal Flood Con.trot POI.ICY <br />noted in 1966 (p. 13), 'it may well be that the advantages of flood plam location <br />outweigh the intermittent costs of damages from floods. Further, there are some <br />kinds of activity which can only be conducted near a watercourse'. <br />Flood damages (or losses) have been defined as the 'destruction or impairment, <br />partial or complete, of the value of goods or services, or of health, resulting from <br />the action of flood waters and the silt and debris they carry. Easy to define, flood <br />losses are difficult to set down in dollar figures' (Hoyt and Langbein, 1955, p. 77). <br />Because of the methodological difficulties in assessing flood damages, as well as <br />the limited data available, 'taking all in all, it is evident that any evaluation of <br />flood damage is only a rough approximation' (Hoyt and Langbein, 1955, p. 79). <br />Nevertheless, the historical record of flood damages provides some insight as to <br />trends in flood impacts on society. <br />Figure 3 shows annual flood damages for the period 1903-19~4 a~ tabul~ted by <br />the National Weather Service (cf. F. Richards, personal communIcation). Figure 4 <br />shows the same data from the standpoint of a 2S.year moving average. The data <br /> <br />NINE FALLACIES OF FLOODS <br /> <br />429 <br /> <br /> 18 <br /> 15 <br />... <br />"' <br />Sl 12 <br />~ <br />@. 9 <br />t 6 <br />0 <br /> 3 <br /> <br /> <br />o <br />1900 19<;1 1920 1930 1940 1950 1960 1970 1980 1990 2000 <br /> <br />Hydrological Year <br />Figure J. U,S. floM damages. J903-1997. wlIh 5-yearrunning mean. <br /> <br />show that flood damages have been increasing steadily at this time scale (using <br />constant dollars). <br />A pattern of climate underlies the trends in flood casualties arid losses. For many <br />years hydrological analyses assumed that climate of floods - the distribution of <br />flood events around some central tendency - remained constant (FIFMTF. 1992). <br />In recent years, scientists and policymakers have come to realize that we live in a <br />climate: that is changing in ways that are difficult to assess and predict (e.g., Karl et <br />al., 1996). In other words, borh the distribution of events and the central tendency <br />may be changing in unpredictable ways. A consequence is that climate fluctuations <br />such as those associated with EI Nino events might be responsible for some of the: <br />variance in flood-related deaths and damages. This means that determination of <br />vulnerability CO floods muse consider explicitly both physical and societal factors. <br /> <br />2.8. KNOWLEDGE LEADS TO ACTION <br /> <br />Those wh.o study public response to natural disasters are well-versed in the per~ <br />sistence of this fallacy which really means: knowledge by itself is generally not <br />sufficient for policy action. As Sims and Baumann (1983, p. 167) note in the <br />context of public response to natural disaster, 'it doesn't necessarily follow that <br />because information is given it is received or because education is provided there <br />is learning'. Knowledge does not always lead to action, but only 'under highly <br />specified conditions. and if properly executed, with certain target public, informa~ <br />