Environmental Engineering Reference
In-Depth Information
outlined message characteristics that have been
validated repeatedly in the research literature as
enhancing response to hazard warnings. The
research shows that warnings are more likely to
be believed and acted on if they have these
characteristics:
.
Clarity (and lack of ambiguity).
5
.
Precision and detail as regards location, timing
and magnitude in message content (Gruntfest
1977; Mileti and Fitzpatrick 1992).
6
.
Consistency in message content and across me-
dia (e.g. Mileti and Fitzpatrick 1992).
.
Multiple messages
dissemination is very variable by event and area.
For example, in Carlisle and the Northwest of
England in 2005, 46% (Environment
Agency 2005b) recalled receiving a warning; in
later events in January 2007 (Environment
Agency 2007a) the proportion was 64%; in events
in 2003 and 2004 (Environment Agency 2005c) the
proportion was 33% overall. Variable success in
warning dissemination may in part be accounted
for by the nature of the flood events and the
difficulty in predicting and providing advance
warning of some extreme rainfall events. Such
events include the Boscastle event in Cornwall
(Environment Agency 2004b), extreme rainfall
events in rapid-response catchments, extreme
events occurring outside areas with a flood warn-
ing service, and events mainly involving surface
water runoff and drainage systems overwhelmed
by heavy and prolonged rain, as in Hull in 2007
(Pitt 2008).
The social characteristics, experience and atti-
tudes of recipients may also be factors in warning
dissemination. Most obviously, in England and
Wales, being registered to receive an automatic
telephone warning has been shown to be a key
factor, and registration in itself has, not surpris-
ingly, been found to be associated with flood ex-
perience (Tunstall et al. 2005). In some studies, in
Carlisle, for example, those with certain social
characteristics were found to be more likely to
recall receiving a warning: those aged over 35;
households with a resident aged 65 and over; those
owning rather than renting property; those aware
that their property was at risk from flooding; and
those who had undertaken preparatory action of
some kind (Environment Agency 2005a). Past
experience and awareness of flooding and long-term
residence are other factors that have been associated
with warning receipt (Tunstall et al. 2005).
The factors that constrain and enhance
responses to flood warnings have attracted sub-
stantial research interest in the UK and elsewhere
(e.g. Drabek 1986, 2000; Environment Agency
2007b; Parker et al.2009). The message itself is
important, here involving the translation of flood
risk science into a message suitable for a non-
professional audience. Drabek (1986, 2000) has
and repetitions
(Baron
et al. 1988).
.
Allow ameans of confirmation (in the sense that
it is recognized widely that initial behaviour on
receipt of a warning is to seek confirmation).
.
A credible, known and trusted source of warning
is perhaps the most important characteristic. This
may be an official source, but hazard warning
messages delivered by family and friends have also
been found to be effective (Dow and Cutter 1998).
In England and Wales, the EA has sought to
provide for all these characteristics in its warning
communications. In particular, the EAhas sought to
make itself known to the public as a warning agency
and to build up public trust. It has had some success
over the years in increasing spontaneous awareness
of its responsibility for issuing flood warnings in
flood risk areas (Environment Agency 2005a). How-
ever, in the FRMRC case studies a wide range of
organizations were cited as preferred warning agen-
cies. In only one area, Chertsey was the official
agency (the EA) the agency most commonly chosen
as preferred and most trusted to give a reliable
warning. Some even mentioned the 'National
Rivers Authority', which the EA superseded in
1996, indicating how long it takes for an agency to
become known and recognized by the public.
Although technology is a key driver facilitating
change in FRM, other contextual issues in turn
5
Ambiguity creates a tension, despite the ethical appropriate-
ness of communicating uncertainty in messages when it exists
(see Faulkner et al. 2007).
6
Again this aspiration is at tension with the scientists' aware-
ness that the message is always to some degree uncertain.
