Assessment of extremes hazards during thunderstorm in France 2000-2019

The estimation of natural hazard levels must be considered for the protection of industrial sites for important safety issues. Our study estimates the occurrence, the spatial extension and the concomitance of natural hazards (rain, wind, hail and lighting) during thunderstorm over France during the period 2000-2019.

A thunderstorm is an atmospheric convective instability. In general, a thunderstorm needs a warm and humid air mass to rise in an unstable atmosphere under the effect of some kind of forcing initiating a more or less deep humid convection leading to the formation of cumulonimbus. Depending on the characteristics of the air mass and the wind regimes at different altitude levels, thunderstorms will be of different natures and will produce different types of hazards with a certain intensity: heavy rain, lightning discharges, hail, strong wind gusts and sometimes tornadoes. A severe thunderstorm is defined by the occurrence of at least one of the following conditions: heavy rain, hail with a diameter of more than 2 cm, wind gusts above 25 m/s and tornadoes.

Hence, the study considers all the available data and project them on the same grid (Meteorage lightning grid with 25 km resolution). A thunderstorm event is defined when more than ten lightning are observed on the same day. Then, the probabilities and conditional probabilities knowing the occurrence of a thunderstorm were derived for each hazard over the period and by seasons. It emerges that the regions particularly affected by thunderstorms and their hazards are in a diagonal going from south-west to north-east France, the mountains areas during the warm seasons and the Mediterranean region, the Cevennes during hot and cold seasons especially in autumn with the Mediterranean event which brings heavy rainfall due to the influence of the sea and the proximity with the mountain. However, although it seems that thunderstorms are more frequent over the decade 2010-2019 than in the years 2000-2009, it is difficult to draw a conclusion on the evolution and intensity of these events over such a short period with the inter-annual variability.

To go further, the study explores the potential of the so-called "self-organising" map to describe the different synoptic scenarios leading to these extreme events and their combinations.