Improved Understanding of the Response of Mean and Extreme Precipitation to Climate Change

The goal of this work is to understand the basic mechanisms which determine how increases in global temperature affect precipitation, including both changes in the spatial distribution of precipitation and changes in the intensity of the most extreme precipitation events. Model simulations of the response of the climate system to greenhouse gas increases show substantial changes in precipitation as a consequence of global warming, but models disagree on the details of these changes and the mechanisms for them are not well understood. One mechanism commonly invoked to explain the changes is predicated on the fact that the moisture content of air typically increases with warming, so moisture convergence and subsequent precipitation increase in regions where moisture is already converging and causing precipitation in the present-day climate (drying is likewise expected in regions which are already dry). But previous work by the PI shows that this "wet get wetter" argument does not adequately account for precipitation changes over land, where they have the greatest societal impacts. Work pursued here seeks to better understand the precipitation response over land by considering the roles played by spatial differences in land surface warming and relative humidity change, through experiments with a simplified atmospheric general circulation model and analysis of simulations from the Climate Model Intercomparison Project version 5 (CMIP5) . Another argument holds that precipitation will increase over the tropical oceans where local sea surface temperature (SST) exceeds the overall warming of tropical SSTs, as the warmer SSTs cause the overlying atmosphere to be less stable than in neighboring regions. But this "warmer gets wetter" argument neglects potential contributions from near-surface wind convergence, the radiative effects of water vapor and clouds, and changes in dry static stability. These effects will be examined together using a diagnostic model in which precipitation is related to a shallow vertical mode which responds to low-level convergence, and a second mode which captures the dependence of deep convection on relative SST change.

Research on changes in the intensity of extreme precipitation events uses a cloud-system resolving model (CRM, specifically the System for Atmospheric Modeling) in idealized configurations to make up for the limitations of climate models in representing extreme precipitation. Some simulations are performed using hypohydrostatic scaling, in which the vertical momentum equation is artificially modified to reduce the scale gap between the small scales on which convective precipitation occurs and the much larger scales of typical of the weather systems and high and low pressure centers found on weather maps. This approach enables experiments incorporating both scales which would otherwise be too computationally expensive. A further topic to be addressed is the effect of warming on extreme snowfall events. The PI's previous work posits an optimal temperature for snowfall extremes which occurs because precipitation extremes increase with temperature whereas the fraction of precipitation that falls as snow decreases sharply in a range near the freezing point. Work conducted here uses observed snowfall data and model outputs to test this theory and explore its implications for a warming climate.

The work has broader impacts due to the potential impacts of changes in mean precipitation and the severity of extreme precipitation events. Mean precipitation is important for agriculture and for water resources and their management, while extreme precipitation is often disruptive to society, and extreme snowfall events are associated with a number of costs in urban environments. The project also supports and train a graduate student, thus contributing to workforce development in this research area. The project also provides summer support for an undergraduate student.