Multilevel and multiscale drought reanalysis over France with the Safran-Isba-Modcou hydrometeorological suite
- 1CNRM/GAME, Météo-France and CNRS, URA 1357, 42 avenue Coriolis, 31057 Toulouse Cedex 1, France
- 2Cemagref, UR HHLY, 3 bis quai Chauveau – CP 220, 69336 Lyon, France
- 3Météo-France, Direction de la Climatologie, 42 avenue Coriolis, 31057 Toulouse Cedex 1, France
- 4UMR-SISYPHE, Mines-ParisTech, Centre de Géosciences, équipe SHR, 35 rue St Honoré, 77305 Fontainebleau, France
Abstract. Physically-based droughts can be defined as a water deficit in at least one component of the land surface hydrological cycle. The reliance of different activity domains (water supply, irrigation, hydropower, etc.) on specific components of this cycle requires drought monitoring to be based on indices related to meteorological, agricultural, and hydrological droughts. This paper describes a high-resolution retrospective analysis of such droughts in France over the last fifty years, based on the Safran-Isba-Modcou (SIM) hydrometeorological suite. The high-resolution 1958–2008 Safran atmospheric reanalysis was used to force the Isba land surface scheme and the hydrogeological model Modcou. Meteorological droughts are characterized with the Standardized Precipitation Index (SPI) at time scales varying from 1 to 24 months. Similar standardizing methods were applied to soil moisture and streamflow for identifying multiscale agricultural droughts – through the Standardized Soil Wetness Index (SSWI) – and multiscale hydrological droughts, through the Standardized Flow Index (SFI). Based on a common threshold level for all indices, drought event statistics over the 50-yr period – number of events, duration, severity and magnitude – have been derived locally in order to highlight regional differences at multiple time scales and at multiple levels of the hydrological cycle (precipitation, soil moisture, streamflow). Results show a substantial variety of temporal drought patterns over the country that are highly dependent on both the variable and time scale considered. Independent spatio-temporal drought events have then been identified and described by combining local characteristics with the evolution of area under drought. Summary statistics have finally been used to compare past severe drought events, from multi-year precipitation deficits (1989–1990) to short hot and dry periods (2003). Results show that the ranking of drought events depends highly on both the time scale and the variable considered. This multilevel and multiscale drought climatology will serve as a basis for assessing the impacts of climate change on droughts in France.