The atmospheric heat engine response to climate change

Moist convection is characterized by complex interactions between dynamics and thermodynamics. As air parcels within the atmosphere, they experience multiple thermodynamic transformations, such as compression and expansion, diabatic heating and cooling, condensation and mixing. These transformations correspond to those of a heat engine that produces kinetic energy while transporting energy from a warm source to a colder sink. This atmospheric heat engine is however directly affected by moist processes. First, falling precipitation acts as a break on the circulation by dissipating a significant amount of kinetic energy. Second, evaporation of unsaturated water and diffusion of water vapor are irrevesible processes that also reduce the amount of work that can be produced. An important challenge is to quantify the impacts that these two effects have on the generation of kinetic energy.

Here, I will introduce a new technique - the Mean Air Flow As Lagragian Dynamics Approximation (MAFALDA) - that can be used to systematically analyze the thermodynamic behavior of complex atmospheric flows. This approach relies on sorting the upward mass transport in terms of the equivalent potential temperature of the air parcels to obtain an isentropic streamfunction. This streamfunction is then used to determine the thermodynamic evolution of air parcels as they move through the atmosphere. This approach is applied to analyze how convective systems would behave in a warmer climate. It is shown that an increase in atmospheric temperature lead to a significant increase of the amount of kinetic energy that is produced per unit of mass of air transported. At the same time, the total generation of kinetic energy is only slightly affected. Taken together, these findings imply that, in a warming atmosphere, the number of intense convective events will be reduced, while their intensity should increase. I will also discuss the new possibility of systematically studying the thermodynamic transformation in atmospheric models.