FLOOD05639 - Laserfiche WebLink

Home Browse Search

15 serve both as warni ng and as confi rmation of a warning (Mi 1 eti, 1974: 134). The lack of environmental cues also may have lead to a less adaptive action or to failure to adopt any action at all. Two people who received official warning in the lower end of the canyon where it was not raining were among the victims because they did not believe a flood was possible. At the lower end of the canyon, as mentioned earl'ier, the first warning message consisted of word of landslides and flood'ing. One waitress at work at the time of the warning sa'id no one moved from the restaurant until the second warni ng which stated that the darn at Estes hi~d broken. No environmental cues would precede a dam break but prior to ii flood there would have to be rain in the upper reaches of the basin. The relative impact of environmental cues on the actions taken in the 8ig Thompson flood could not be measured in any standard form and were ex- cluded from the findings. The warning cha racteri sti cs, SOUl'ce, mode, content and number were not significant when initially cOlTelated to action or to survival and hence were not included in subsequent analysis.. The reason for their lack of significance may be due to the limited sample size, especially the unavailability of data on tourists who received warnings. An additional reason to disqualify the warning characteristics from the analysis is that the data collected for non-surv'ivors concentrated on determining actions taken and not upon warning characteristics. However, the influence a warning had on choice of behavior and on survival is analyzed in general terms. The variable "kind of warning" was included in the discriminant analysis. This information was available for most of the survivors and the non-survivors. In cases where it was not known whether a warning was received 'it was assumed that environmental