3-D Moisture Budget of Changes to Precipitation Extremes in a Warming Climate

As the climate warms, the most extreme precipitation events are projected to increase over most of the globe but may decrease in some areas. The interplay between dynamical and thermodynamical processes during these extreme events and how these processes will change for future extremes remains poorly understood. In this study, the Community Earth System Model (CESM) Large Ensemble (LENS) is analyzed for the present and a projected future (2071-2080) climate with 40 ensembles in each group. Percentiles of precipitation are identified using 6-hourly data and the mean 3-D structures of water-vapor mixing ratio and horizontal mass convergence are calculated for a given percentile, both in the present and future climates. Conditioning on a given percentile allows a separation into thermodynamic and dynamic tendencies for changes to that percentile of precipitation from the current to future climate at each grid point and vertical level. This analysis allows an examination of how the 3-D moisture budget is projected to change for extreme events in a future climate. Consistent with previous studies, precipitation extremes increase over most of the globe in winter and summer, but decrease over some subtropical ocean regions. The sign of these changes to precipitation extremes is determined by the relative contributions of the dynamical and thermodynamical components of the moisture budget. The thermodynamical tendency is positive everywhere (i.e., it has a tendency to increase precipitation extremes) and relatively uniform with height, whereas the dynamical tendency varies between region, season, percentile of precipitation, and between different levels of the atmosphere. In the tropics, the greatest increases to precipitation extremes are seen because of a positive dynamical tendency, even at low percentiles of precipitation. In parts of the subtropics, the dynamical tendency is sufficiently negative (i.e., it has a tendency to decrease precipitation extremes) to overwhelm the positive thermodynamic tendency, so that precipitation extremes become weaker in the future. In the midlatitudes, in winter, the dynamical tendency is negative for most percentiles of precipitation, indicating a tendency to reduce precipitation associated with atmospheric rivers, but is overwhelmed by the positive thermodynamic tendency, i.e., the increased moisture contained within atmospheric rivers in the future. With increasing percentile of precipitation, the dynamical tendency becomes closer to zero and above the 99th percentile becomes positive over most of the midlatitudes in winter, indicating that some atmospheric-river events will benefit from both more moisture and greater convergence, and hence be doubly enhanced in the future climate. In summer, the dynamical tendency is negative over most of the midlatitudes (as well as subtropics) even at the highest percentiles of precipitation, indicating the importance of atmospheric rivers for distinguishing summer from winter trends. Finally, the vertical structure of the dynamical tendency is examined for various regions and percentiles of precipitation, and compared between winter and summer.