The Vertical Structure of Convective Mass-Flux Derived From Modern Radar Systems: Data Analysis in Support of Cumulus Parametrization

Principal Investigator(s):
Christian Jakob, Monash University

Convection is a fundamental process in the earth atmosphere. Despite a sizable research effort, atmospheric convection remains one of the least understood phenomena in atmospheric science. Consequently it is also one of the least well simulated features in climate models. This not only significantly affects the quality of the simulation of important quantities, such as rainfall, but also directly contributes to the significant spread in our estimates of Earth's sensitivity to anthropogenic influences on the climate. Despite progress in recent years, all contemporary climate models exhibit severe limitations in predicting many of the observed phenomena associated with tropical convection, such as the inability to predict the correct tropical mean rainfall distribution as well as the inability to represent the major modes of tropical variability, ranging from El Niño to the diurnal cycle of rainfall over land. Many of the shortcomings have been shown to be directly related to the models’ representation of convection. This is largely due to the fact that the processes in convective systems act on spatial scales that cannot be resolved using the horizontal grid-spacing currently permissible in global atmospheric models. Individual thunderstorm cells have spatial scales of a few kilometers, complexes of thunderstorms scales of a few tens of kilometers, while grid sizes of climate models are usually on the order of a hundred kilometers or more. Hence, convection in climate models needs to be represented by means of a technique widely known as parametrization.

In this research we will exploit observations taken at the tropical sites of the Atmospheric Radiation Measurement (ARM) Climate Research Facility (CRF) to support and inspire a fundamental redesign of the treatment of convection for climate models. Specifically, we will make use of observations made by radar systems - the atmospheric scientists equivalent to the X-Ray machine - to probe the properties of tropical thunderstorms in new ways. We will use the three-dimenisonal view that these systems provide to derive key properties of convective cloud fields in the tropics, such as the distribution of cloud size, cloud depth and cloud number and their association with the background state of the tropical atmosphere. Importantly, we will also derive profiles of vertical motion inside thunderstorms, a quantity of great importance to determining the intensity of the storms. Combining the information gleaned from the observations will allow us to evaluate current representations of convection in global models and, perhaps more importantly, will inspire and support the development of new approaches to solving what is a long-standing puzzle.

The proposed research will have a large impact in at least three major areas of atmospheric research. First, it will enhance our ability to retrieve key properties of convective systems from radar observations. The algorithms we will develop to do so are highly innovative and our goal of deriving vertical motion over long periods of time and an entire radar domain is groundbreaking. Second, by relating the small-scale convective properties to the larger scales the research will not only support the design of revolutionary new treatments of convection in models, but it will also discover key physical mechanisms responsible for the behavior of a convective cloud ensemble in an area the size of a GCM grid box. This will significantly enhance the body of knowledge on tropical convection. Finally, the research will directly support the construction of a fundamentally revised treatment on convection in climate models, which is under development in the PI’s research group. Successfully implementing this approach in world-leading weather and climate models will enhance our ability to predict weather phenomena associated with convection, such as heavy rainfall, as well as contribute to reducing the uncertainty in predictions of climate and climate change.