Severe flooding occurred along the River Thames in January 2003. This was caused by heavy rain falling on a saturated catchment with already swollen rivers. The flooding took place relatively soon after the June 2002 completion of the £110 m, Maidenhead, Windsor and Eton Flood Alleviation Scheme (a diversion channel named the Jubilee River), immediately upstream from the Lower Thames area. The flood alleviation scheme saved approximately 1,000 properties in Maidenhead, Eton and Windsor from being flooded but there was internal flooding affecting about 500 properties elsewhere including the downstream Lower Thames area. There was also widespread disruption to traffic and businesses. The extent of the River Thames flooding was the worst since 1947. Following the flooding, there was great public concern that the operation of the Jubilee River had exacerbated the downstream flooding. An independent Chairman was appointed to lead an investigation and report on the causes of the flooding along the River Thames (Onions, 2004) and a Strategy Study was initiated to explore the options for the Lower Thames area and in particular for the reach of Lower Thames with a narrow, confined and highly urbanised flood plain where the Lower Thames Survey was mounted (Halcrow Group Limited et al., 2005).
4.2 Flood events and experience in the surveys
The ‘Intangibles’ and ‘Warnings’ Surveys cover a wide range of flood events occurring at different times since April 1998, affecting different local communities, and evoking varied responses from the responsible organisations. The Intangibles Survey involved only river flood events. However, the kinds of rivers and event varied greatly covering slowly developing events on long rivers such as the Thames and Severn, to flash flooding in steep catchments, and very extreme rainfall events. The Intangibles Survey focused heavily upon areas affected by two major national flood events: the Easter flooding of 1998 and the floods of autumn 2000. The Warnings Survey also covered some localities affected by the autumn 2000 floods but not included in the Intangibles Survey. In addition, it included a large variety of later small fluvial events and also covered some coastal flooding and flooding that was affecting very small ‘ordinary watercourses’ for which the Environment Agency was not, at the time, responsible. The Lower Thames Study is different since it focused on a particular locality affected mainly by one specific recent event: the flooding from the Thames in January 2003. The different populations and sampling approaches used in the surveys reported here mean that those included differed in the extent to which they had experienced recent and past flooding at their current address (Table 4.1).
Flood experience has been found to be a salient variable in many studies of flood perception and response from the early 1960s (e.g. Lowenthal, 1961; Kates, 1962; White, 1973, 1974; Burton, Kates and White, 1978; Azjen, 1988; Tunstall and Fordham, 1994). There is support for an “innoculation hypothesis” and other conceptualisations that emphasise the advantage of being familiar or experienced with a stressor that is at hand (Norris and Murrell, 1988). However, Tunstall and Bossman-Aggrey (1988) reported that previous experience of flooding did not leave residents with knowledge about how to cope with a future flood, but with a feeling that there is little they can do.
Of those interviewed in the Lower Thames survey, very few respondents (24) reported being flooded above floor level in that event, and only 44 respondents reported that they had been flooded at all.
4.3 Depth and duration of the survey flood events Flood event characteristics that may have an influence on flood impacts, and hence on people’s capacity to cope with these impacts, are depth and duration. Both the Intangibles Survey of the flooded and the Warnings Survey contained some information on the nature of the recent or worst flood event, where more than one recent event had been experienced by those surveyed. Flooding in England and Wales is generally mild by international standards and the survey data on flood depths confirm this. However Table 4.2 shows that those interviewed in the Intangibles Survey were more seriously affected in terms of flood depths than those in the Warnings Survey, with substantial minorities experiencing deep flood waters (60cms or more) in their homes.
Flooding in England and Wales is usually of relatively short duration. However, in both surveys, a majority had flood waters in their homes for at least a day. In the Warnings Survey, there were more cases in which the flood waters receded quickly, within less than 12 hours.
Table 4.2: Maximum depth of flooding and duration of flooding inside the home: Intangibles Survey and Warnings Survey
The Intangibles Survey of the flooded included further detailed questions because it was hypothesised that specific flood characteristics and the extent of flooding might have an influence on the health effects of flooding. These factors may be important for the analysis of vulnerability and resilience in Section 6. Thus, this survey included questions on the particular rooms affected.
Respondents were asked ‘How quickly did the flood waters rise?’ because it was thought that the speed of onset of flooding might have had more impact on health and well being. Most respondents (64%) reported that the waters rose so quickly that you could see them rising, 16% reported waters rising slowly over many hours, 10% thought the speed was somewhere in between, and 10% did not know. Flooded respondents were also asked whether the flood waters contained sewage or other pollution, because it was hypothesised that these might add to the adverse impact of flooding on health. Most (77%) thought that the flood waters were contaminated, 15% thought they were not, and the remainder 8% did not know.
4.4 Flood risk constructions and awareness
Flood risk constructions
How people construct the risk of flooding (how they perceive it) is both subjective to the individuals and groups concerned as well as influenced by the objective realities with which they live and function (see Section 1.2.3). One main hypothesis in this report is that people in low or infrequent flood risk areas will be willing to live with the risk in exchange for other benefits associated with living in an area, such as amenities, environment, social networks and so on. There is very little data on flood risk constructions across the three surveys reported here. Some perceptions of risk were a focus of the Lower Thames Survey and there is more scope for analysis there. In an earlier 1989 study of Thames residents (Tunstall and Fordham, 1994) respondents were asked what they thought was an acceptable (tolerable) level of flood risk that they were prepared to live with each and every year. A majority of the 488 sample stated that they would be prepared to live with a 1 in 200 or 100 risk of flooding and half were prepared to live with a 1 in 50 risk; the proportion prepared to accept the risk declined as the level of risk increased. However, over a fifth stated that they would be willing to accept a 1 in 5 risk each and every year; much of this was attributed to a trade-off being made between living in a highly valued environment and the flood risk. It would be interesting to know if these perceptions have since changed.
Awareness of flood risk may be a factor affecting response to flooding. Lack of awareness of the risk may be seen as making residents more vulnerable to the impacts of flooding. All three surveys contained a question on awareness of flood risk, although the form of the questions varied significantly across the studies (Table 4.3).