In the Warnings Survey, it was possible to examine whether the availability of help from outside the home made any difference to the protective actions taken at the time of the flood (Table 5.6). In that survey, although the amount of help available within the household was not significant, being given help from outside the home in protecting property was a significant factor. In the Warnings survey, there were some significant differences in the actions taken to prepare for flooding of those who did and did not receive outside help. Such help did make some difference
More of those who received such help tried to find out more about the possibility of flooding in various ways. They were more active in moving themselves and others to a safe place, in moving valuables and in moving pets and stock. More of those who were helped tried to stop the flood waters entering their property with sand bags. Thus neighbourliness and community cohesion played a part in generating action to prepare for flooding and to protect property.
5.4 Evacuation and disruption 5.4.1 Introduction
Evacuation is one possible behavioural response to flooding. The function of evacuation is to save lives and reduce the danger to people and animals during a flood event. Evacuation measures are only normally taken during serious flood events when it would not be safe or practicable for people to remain in their properties, or for those living in ground floor flats, bungalows or mobile homes. Green and Parker (1993: 2) define evacuation as: “movement, normally using the individuals concerned own resources, towards a place of safety where that safety is created by separation by distance or topography from the hazard”.
Table 5.6: Actions taken to prepare for flooding and to protect property according to help
Evacuation is a process and not a short-term response and it is not complete until those who have had to leave their homes have returned (FHRC, 1996). Although many people will spontaneously evacuate to relatives and friends before being asked to do so officially, there is evidence that the evacuation process itself is extremely distressing and worrying for people, particularly where family or social structures are disrupted. Evacuation can largely increase the overall disruption resulting from flooding. Ketteridge and Fordham (1995) discuss the trauma of a badly co-ordinated and managed evacuation, the effects of which can be long-lasting and potentially devastating, particularly amongst the most vulnerable members of society. This highlights the importance of the temporal dimension of a disaster outlined in Section 1.2.1.
Post-event evacuation is necessary when flooding lasts for a long period of time or when there can be serious health and safety risks. The loss of services such as electricity or heating can also warrant leaving the home (Ketteridge and Fordham, 1995). Data from 1,712 FHRC interviews with flood victims from 11 different surveys comprising the Full Flood Impacts study were combined into one composite data file. The data did not distinguish between pre-event evacuation and the household leaving the house because it was inhabitable, but inspection of the data suggests that it was mainly the second type. 28 % of households reported that at least one member left. A discriminant analysis showed that the likelihood of evacuation depended on the number of infirm adults in the household, depth of water, duration of loss of telephone service and the time of year that water entered the property (e.g. households were more likely to evacuate in winter due to the cold and absence of heating). The duration of the evacuation was a function of the damage to the house. The severity of the evacuation was a function of the duration; however those that stayed with friends or relatives reported a greater severity than those who stayed elsewhere (Ketteridge and Green, 1994).
However, mass flood evacuations have been shown to be effective. Prolonged rainfall in 1995 led to extensive flooding in the Netherlands, with a total of 250,000 people being evacuated to safety. Overall, the evacuation operation was deemed successful. The slow onset of the flood and long warning lead time allowed time to prepare (van Duin and Bezuyen, 2000). The level of public cooperation surprised the authorities and operational services, the public's behaviour and discipline during evacuation were praised and said to be a contributing factor to its success. Almost all evacuees departed and returned to their homes without any support by the authorities. However, in Limburg where the flooding was not life threatening, people were more reluctant to leave their homes. This reluctance to evacuate is not uncommon and many flood victims from the autumn 2000 floods commented that although they evacuated on that occasion, they would not do so in the event of future flooding.
For this report, the Intangibles Survey is the only one of the three datasets to contain information on evacuation. It has been further analysed to throw new light on this form of behavioural response to flooding and to examine whether the drivers of this behaviour have changed since the earlier data was collected over ten years ago.
5.4.2 Evacuation behaviour
Flooded respondents in the Intangibles Survey were asked four questions:
whether or not the respondent had to leave the home (60% did);
whether or not another family member had to leave the home (these two are not mutually exclusive) (56%);
whether no-one left the house (35%), so in 65% of homes at least one person had to leave the home due to the flood); and
how long it was before the whole household could live in the property again.
The research shows that usually people evacuate as whole families (Fisher et al., 1995; Drabek, 2000, Heath et al., 2001). Our data shows that there were some instances where someone other than the respondent left home but the respondent did not, and vice versa. But our data do not enable us to calculate whether all household members left home.
The survey did not include questions to establish when people left their homes. In England and Wales the police take the final decision on whether to initiate any official structured evacuation, and firefighters and local authorities would also assist in the process. However, although the police may strongly suggest that people evacuate they have no powers to enforce evacuation for flooding. Should an extreme flood event occur, as in the 1953 coastal floods, such a large-scale evacuation would prove extremely difficult and would need substantial prior-preparedness planning and resources, including military assistance.
No mass evacuations were organised during the Easter 1998 or autumn 2000 floods in England and Wales (which are the main events covered by the Intangibles Survey). During the autumn 2000 floods around 11,000 people were requested to leave their homes by the police, however, not everyone complied with the request (EA, 2001). It is likely, therefore, that most of the evacuations that did take place were the result of family decisions and that most people left very shortly before, during or after the flood when living conditions in the home became intolerable. The last choice is the most common response in England and Wales (Ketteridge and Green, 1994).
Green and Parker (1993: 2) differentiate between types of evacuation according to when during the hazard the evacuation takes place:
A precautionary or pre-event evacuation takes place before the hazard has occurred. In some cases, precautionary evacuations can exist as a form of land use control, e.g. property on a floodplain can be purchased by the government and its residents relocated somewhere else. However, most precautionary evacuations are undertaken because there is a forecast of a hazardous event, e.g. a dam break, toxic material release, etc. Pre-event evacuation is the main protective action against hurricanes (Sorensen, et al., 1987).
Aftermath or post-event evacuations take place in the aftermath of the event because of deterioration of living conditions in the area. This type of evacuation should be avoided or reduced to the shortest possible duration as it disrupts the social support network of the victims and makes it more difficult to put their lives back together. Consequently, disruption is particularly important if the relocation is permanent (see also Bland et al., 1997).
It is striking that so many households in the Intangibles Survey had at least one member leave home because of the flooding. This proportion (65%) is more than twice the proportion (28%) of people doing so in events that occurred at least ten years earlier, as reported in the full flood impacts study (Ketteridge and Green, 1994). It may be that the explanation is simply that the earlier events were less severe with fewer properties flooded and a lower depth of flooding. All the flooded in the Intangibles Survey had flood waters inside their homes. However, it is possible that current households have higher standards and expectations of comfort and convenience in their homes and are less prepared to live with the discomforts of a flooded home than households interviewed even a decade earlier.
The evacuations reported in the Intangibles Survey were long (Figure 5.10) and lasted from 12 weeks to up to 6 months for 30% (192) of households and between 6 months to 9 months for a further 30% (191). A few, 4% (23) of households could not live in the property again until over a year after the flood. The mean duration of evacuation of those households where someone left home was 23 weeks.
Figure 5.10: Duration of evacuation: Intangibles Survey
5.4.3 Flood event characteristics as drivers of evacuation behaviour
We would expect the nature of the flood event, particularly the depth and extent of the flooding in the home to be a key determinant of evacuation and its duration. In the Full Flood Impacts Study (Ketteridge and Green, 1994), the depth of flooding, the £ value of damages incurred, the time flood waters entered the property, the duration of loss of telephone services and the number of infirm adults in the home were key determinants of the likelihood of evacuation. In the Intangibles Survey there was also a strong association between whether someone had to leave home and the length of time before all the household could live together in the home and the depth of flooding inside the home. However, a minority of those who only had flooding in other parts of their property evacuated possibly because of loss of utility services and of access to the home (Table 5.7).
Table 5.7: Evacuation and depth of flooding: Intangibles Survey