## Midwest Mathematics and Climate Conference - Day 2 Afternoon Session

University of North Carolina at Chapel Hill Mathematics

#### Abstract

• , University of Texas, Austin
Dual-state Behavior of the Community Climate System Model

• Ice age climate between 80,000 and 11,000 years ago experienced abrupt (< 10 year) global-scale climate transitions every few thousand years. Evidence and theory suggests such behavior may be explained by transitions among multiple circulation states in the Atlantic Ocean’s meridional overturning circulation (AMOC) in response to variations in polar ice sheet mass wasting. However tests of more advanced coupled climate models do not exhibit the distinct circulation states that exist within more idealized ocean circulation models nor do they reproduce the scope of observed changes in tropical climate, particularly its monsoons. Here we report on the existence of dual-state behavior of the Community Climate System Model (CCSM3) tropical atmosphere under modern boundary conditions in response to changes in the AMOC brought about by an injection of Greenland ice sheet melt that sustains a 0.4 to 1.0 $Sv$  $(Sv =10^6m^3s^{-1})$ freshening of the north atlantic. the modeled transition involves a 5% increase in large-scale convection that reduces the elevation of convection detrainment that, in combination with a reduced hadely cell circulation, makes the tropical atmosphere less efficient in the export of moist static energy. changes in the distribution of clouds further exacerbate this altered state by increasing the amount of long-wave energy downwelling to the surface and the need for the model to make use of its large-scale, rather than sub-grid scale parameterized convection. the impacts of the transition are significant, causing a 30% to 70% reduction in precipitation across south america, africa, and south and east asia, similar in scale to those documented within ice-age climate proxy records of heinrich event 1 and earlier events. while certain aspects of this dual-state solution have been anticipated from theory and a hierarchy of modeling experiments, the emergence of these transitions may be an artifact of the model numerics and/or a consequence of subjective choices in model construction. nevertheless, the physics of the transitions seem plausible and finding such extreme sensitivity in models used to project future climate is relevant to our understanding of the potential risks associated with a changing climate. : ;

• , University of Michigan
Exploring Tipping Points in Cloud System-Environment Interactions

• The properties of dynamic circulation systems are strongly influenced by the environment in which they form. They subsequently modify their environment both through their dynamics, and via the diabatic influence of clouds on the thermodynamic state. Determining the response of precipitating cloud systems to changes in the Earth's climate is made complicated by the fact that there are multiple controls on the outcome of the system (e.g. precipitation distribution, radiative fluxes and heating rates). Interactions between controlling factors makes it difficult to attribute a specific change in a cloud system to any single factor. In addition, for many (perhaps most) precipitating cloud systems, the response to a perturbation may differ in magnitude, and even sign, depending on the environment in which it is embedded. Systematic sensitivity analysis reveals that this behavior is due to tipping points associated with the development of dynamic and moist convective instabilities. This presentation uses data assimilation and machine learning to explore cloud-environment interactions in three types of dynamic precipitation systems: tropical cyclones, extratropical cyclones, and orographic rainfall. In each case, distinct changes in the behavior of the system are caused by the presence of a dynamic or static instability.