Understanding the Warning Process Through the Lens of Practice: Emancipation as a Condition of Action–Some Lessons from France - Reducing Disaster: Early Warning Systems for Climate Change - page 224

condensed into a single place (a forecasting centre, a crisis unit) visible to a single
person, which is what makes them so powerful (Latour 1985 ). One exemplary case is
tsunami warning: planetary scale monitoring allows the identifi cation of remote
phenomena whose impact is local (e.g. Soulé 2011 ).
However, this technology which, for example, allows fl ood forecasting data to be
published on the web can also be used to justify the suspension of direct calls by
forecasters to municipalities, which can now access the information on the website.
But many small town and village offi cials are at a loss with data of this kind.
The availability of online weather forecasts, which at fi rst sight looks like genuine
progress, can, from the point of view of the actors, further complicate their task. 4
More broadly, climate change creates a fl aw in our forecasting systems, which
are based on historical patterns, the laws arising from them and an assumption of
stationarity (Chateauraynaud and Torny 1999 ). This raises doubts about what we
might face in the future and potentially generates an across-the-board state of alert
that multiplies the number of entities to be considered. For instance, the use of the
'historic' fl ood as a reference point for risk management decisions under “Plans de
Prévention des Risques d'Inondation” (PPRI) process is currently at stake. Indeed,
the assumption is made that the historic fl ood represents the worst possible case.
With climate change, this assumption (i.e. the set of past events represents the full
range of actually possible outcomes from which it is then possible to infer return
periods and event frequencies) no longer holds 5 .
Finally, all these factors involve, or are linked with, issues of overlapping and
embedded scales, of growing relevance in the context of globalisation (Lussault 2013 )
and extreme weather phenomena (e.g. Ruin et al. 2012 ). In fact, local issues and
situations can no longer be assessed without reference to wider scales that add new
entities to the warning agenda.
This non-exhaustive inventory highlights two signifi cant fi ndings:
1. The proliferation of identities arises from contemporary, developing processes,
which are unlikely to run out of steam in the coming decades. In other words, the
question of the proliferation of entities that warning actors need to take into
account in order to take action - or the expansion of their AEs - needs to be taken
very seriously, because it constitutes one of the major challenges to action deci-
sions in the future.
2. This proliferation is often a side-effect of processes that otherwise offer signifi -
cant benefi ts, in particular scientifi c and technological progress. This means,
fi rstly, that we should 'not throw the baby out with the bathwater', but approach
this new challenge as a factor inherent to the operation of our globalised societies.
4 The need for a human face to help interpret and mediate this proliferation of information (like a
'decoder', see next section) is also found by researches in the UK that highlights the value added
by Public Weather Service Advisors, who serve a mediating role, putting a human face on Met
Offi ce forecasts. They are hugely popular amongst UK emergency responders (Demeritt 2012 ).
5 For further discussion of this issue and its implications for modelling and risk assessment, see
Demeritt and Wainwright ( 2005) .