Ocean-atmosphere coupling for prediction of Mediterranean heavy precipitation events : better modeling and impacts of river runoff and sea state.

The Western Mediterranean Sea area is frequently affected in autumn by heavy precipitation events (HPEs). These episodes, characterized by strong offshore low-level winds and heavy rain in a short period of time, can lead to severe flooding and wave-submersion events. This thesis work aims to progress towards integrated short-range forecast system via coupled modeling for a better representation of the processes at the air-sea interface. The methodology consists in studying the impact of a realistic representation of freshwater flow into the ocean, then examining the impact of better taking into account the sea state, particularly through coupling.In a first part, we evaluated the sensitivity to different river flow representations in the NEMO ocean model during the HyMeX campaign (SOP1, fall 2012). For this purpose, two ocean configurations were used, WMED (1/36°) covering the Western Mediterranean Sea and a new configuration: NWMED (1/72°) covering the Northwestern Mediterranean Sea. Three river discharge forcings are used: a monthly climatology as well as daily and hourly observations. The results showed a significant local impact on the ocean stratification when river discharge observations are used compared to climatology. The surface salinity is modified as well as the mixing layer, becoming thinner, delimited by a well marked halocline. Secondly, we focused on a HPE that occurred between October 12 and 14, 2016 in the south of France. In order to study the role of sea state in air-sea exchanges, a set of numerical simulations was carried out with the Météo-France AROME kilometric atmospheric model - including the turbulent sea surface fluxes parameterization WASP - forced or coupled with the WaveWatchIII wave model. The results showed that taking sea state into account has a significant impact on the lower levels of the atmosphere, reducing the surface wind speed and modifying the precipitation forecast over sea, particularly the location. In the third part of the thesis, we evaluated the contribution of ocean-atmosphere-wavecoupling. In order to identify and quantify the coulping impacts, coupled ocean-atmospherewave simulations were performed using the coupled NEMO-AROME-WaveWatchIII system and notably compared to coupled atmosphere-wave and ocean-atmosphere simulations. The results showed, first of all, that the forecast is sensitive to coupling, and that the interactive coupling with the ocean leads to significant changes in the heat and moisture supply of HPE, while coupling with a wave model mainly leads to changes in the low-level dynamics. These results were finally compared with atmospheric sensitivity tests to the turbulent sea surface fluxes parameterization and to SST. It emerges that the choice of the parameterization can be of great importance for the HPE forecast as large as the coupling with an interactive ocean, whereas the dynamic effect is only produced by considering waves forcing/coupling.