Geoscience Reference
In-Depth Information
On July 5, 2005 the extreme southeast of France was ravaged by large forest
fires. For example, approximately 1,000 hectares of forest were burnt between the
towns of Puget sur Argens and Fréjus (which are located in the east of the Var
department), right next to the department of the Alpes-Maritimes. More than
100 hectares were burnt in the districts of Villeneuve-Loubet and Vallauris (Alpes-
Maritimes), despite the fact that these areas are partially urbanized and that they
have extremely efficient methods available for fighting such fires. The fire risk
index of these areas was in the range of 17 and 19 out of 20, irrespective of whether
the indirect or direct method was used to calculate the risk index, which highlights
the fact that the method used to work out the fire risk index was extremely effective.
8.4. Conclusion
Throughout this chapter we have dealt with the topic of forest fires, which can
be seen as being a risk that is partially associated with climate, and we can observe
that geographical information is omnipresent throughout the chapter. Throughout
this topic, geographical information has been applied to different factors that have
had an effect on the world of climatology, however, geographical information also
influences other factors that, in turn, influence forest fires. Geographical
information needs to be taken into consideration as far as other risk factors that are
not directly associated with climatology are considered, like susceptibility or
vulnerability. The aim of this is to be able to understand how the fires start and
develop long-term strategies for better prevention and protection. In other words
geographical information is used for operational purposes.
The aim of this chapter is not to use geographical information as a measurement
that can be used in calculations, but to use it in as many different approaches as
possible rather than only using it to recreate climatological data. This phase is not
present in the direct method, which is used to recreate the risk index that is,
nevertheless, a synthetic climate variable.
How valid is the map that represents information linked to a forest fire
meteorological risk index, whatever the method used to elaborate it? This question
needs to be asked at two levels: first, the index needs to be validated, yet not based
on real fires. As we have seen in this chapter, forest fires are not dependent only on
meteorology. An ideal index would not, therefore, be one that had the best
correlation with burnt surfaces, or would not be the one that generates the least
number of missed and false alerts. The best way to avoid missing any fire alerts is to
constantly increase the risk level that has been set, as this leads to an unbelievable
number of false alerts, and as a result of the high number of false alerts the
emergency services may question the severity of any future risk levels. In theoretical
terms, a good index should never lead to the creation of any missed alerts. However,
if a false alert does occur there are two possible explanations as to why it might
have occurred: first, the risk was exaggerated, and second, the index was not wrong,
it was just the fact that there was no initial spark present to start the potentially
