Impacts of atmospheric radiative forcings on the tropical climate.
Directeurs.rices de thèses : Bony S.
Composition du jury
Dr Hervé Douville, CNRS-CNRM (Toulouse), Rapporteur
Dr Mark Ringer, MetOffice (Exeter), Rapporteur
Dr Jean-Louis Dufresne, LMD (Paris), Examinateur
Dr Romain Roehrig, Meteo-France-CNRM (Toulouse), Examinateur
Prof Bjorn Stevens, MPI (Hamburg), Examinateur
Prof Laurence Picon, LMD (Paris), Examinatrice
Dr Sandrine Bony, LMD (Paris), Directrice de thèse
The radiative effects of tropical low-clouds have long been recognized as critical for climate sensitivity estimates due to their role in modulating the Earth’s radiative budget at the top of the atmosphere.
Within the atmosphere, they exert a strong radiative cooling, and have thus the potential to affect also the large scale atmospheric circulation and the global hydrological cycle. The motivation of this thesis is to better understand the role of atmospheric cloud radiative effects (ACRE) of low-clouds in tropical circulation and precipitation.
To investigate this issue, planetary boundary layer (PBL) clouds are made transparent to radiation in a climate model. It is found that PBL ACRE enhance tropical precipitation and large-scale circulation through their coupling with surface turbulent fluxes. These results are shown to be robust across models and model configurations. The time-scale of the climate response to PBL ACRE is investigated by running the climate model in a numerical weather prediction mode. This approach is shown to be useful for studying the fast climate response to radiative perturbations. Short-term forecasts show that the coupling between PBL ACRE and surface heat fluxes and its impacts on precipitation and atmospheric circulation occur within days. Such a fast response allows for the use of short-term forecasts to investigate the impacts that changes in model parameterizations have on this coupling and their implications for some model biases.
Finally, the role of PBL ACRE on climate change is investigated in a multi-model framework. It is shown that in all models PBL ACRE decrease the hydrological sensitivity (defined as the response of global mean precipitation to an increase in surface temperature) and reduce the spread estimates among models. On the other hand, regional patterns of precipitation changes do not depend much on PBL ACRE, suggesting that PBL clouds affect the patterns of precipitation changes more through their impact on climate sensitivity than circulation.