Funded Grants

Modeling the human impact on rainfall on the U.S. east coast

One reason that climate change is simultaneously unfathomable and inevitable is that the causes and consequences are separated in time and place, such that societies are unable to accurately understand and quantify the risks associated with their collective actions. But what if our fireplaces and air conditioners meant that we could no longer grow Jersey tomatoes or even a few grassy lawns in the diminishingly green "Garden" state? Alternatively, what if our hour-long weekday commutes mean increased odds of weekend rain spoiling Sunday picnic plans? Recent findings suggest both of these scenarios are alarmingly likely for the U.S. East Coast population, but the scientific evidence needed to solidify those links continues to elude us (Cerveny and Balling, 1998).

An important focus of the past two decades of climate research has been the predicted change in temperature trends globally, but these trends - while overwhelmingly supported by recent work - have failed to compel people to implement major behavioral changes in our society. The changes in lifestyle that may be necessitated in 100 years due to gradual and seemingly minor global mean temperature trends of tenths of a degree simply have not convinced us to change our behavior. After all, who would not want a few extra days each year of tank tops and sandals rather than ear muffs and galoshes? And while drastic changes are likely for some regions, key highly populated regions of the U.S. (such as most of the original 13 colonies) are predicted to undergo changes that on average (and aside from sea level issues in low-lying coastal regions) clearly do not amount to much more than that.

Only recently have groundbreaking new data established the catastrophic implications of pollution on rainfall. Compelling evidence from satellite observations has provided a convincing link between dense pollution in southern India and recent droughts (Ramanathan et al., 2001). The implications for agriculture and other human activities are widespread, demanding behavioral adaptations ranging from food production to water usage. And more importantly, the climate impacts are linked in space and time to the emissions and their human sources. As a consequence, understanding the impact of human activities on rainfall reduction is not only a key missing component in the climate system but it is potentially a far more compelling way to close the cycle by allowing humans to react to climate change by providing real-time feedback on their actions that could result in policy and societal changes.

Because of this tight link in time and space to human activities, understanding the science that controls the impacts of pollution on clouds is critical. In order to understand the impact of man-made emissions of black carbon and other pollution particles in perturbing clouds and interfering with the rainfall cycle, we must be able to accurately model the rainfall cycle of clouds.

Why have we failed to either recognize or resolve this problem to date? The reason is that the human impacts on rainfall bridge two independently complex problems and require a truly interdisciplinary understanding that links spatial scales of nanometers to kilometers in describing physical processes with temporal events of microseconds to months in quantifying chemical reactions. The successful approach must build a multiscale theoretical model that faithfully represents measurements despite similar limitations in scale. My work and my research group bridge those traditional gaps between the disciplines of meteorology and aerosol chemistry, as well as the institutional gaps between measurements and modeling, in order to allow us to pursue a unique strategy in tackling this problem.

I lead a research group of nine graduate students in studying some of the key processes that determine the links in these scales. Elizabeth Fetter studies the microphysics of clouds, Anita Adhitya models the radiative impact of pollution on clouds, Steve Maria measures the chemical components of clouds, and Monica Rivera identifies tracers of human activities in particle emissions. I have established strong collaborations on theoretical atmospheric models with Venkatachalam Ramaswamy and Leo Dormer at the premier geophysical modeling laboratory located at Princeton, and through this work I have combined our observational evidence with comprehensive models. I have designed, carried out, and interpreted ship- and aircraft-based measurements in eight major multinational field campaigns with leading observational experts. While we and others have made important strides in reducing uncertainties in a number of key problems, significant progress on understanding the role of pollution in clouds requires a unified modeling and measurement approach that does not fit in the current course of academic study or the current climate funding structure.

The combined expertise along with my own 10 years of research in the field have lead me to favor a novel approach to predicting rainfall theoretically. The idea is to incorporate aerosol chemical evolution in a cloud-resolving model that can then be initialized to measured emissions. The predictions are compared to a time series of measurements collected by following an evolving cloud in a Lagrangian domain, so that the time history of the evolution can be used to directly evaluate the model processes.

The project will start by developing the model for rainfall predictions along the US, eastern seaboard, where emissions are well characterized by government-sponsored projects. We will then use a research vessel and instrumented pod suspended from a ship-based helicopter to track the modeled cloud evolution northward and eastward. The current drought in this region has already begun to limit water usage and crop production, so the possibility that this reduced hydrological cycle is human induced has important ramifications for future behavior. In other words, it has been clear to date that reducing pollution emissions in the industrial and urban areas of Pennsylvania, New Jersey and New York has associated economic costs, but are those costs sufficiently high as to outweigh the water consumption needs of the millions that live here? The goal of this project is to put a price on that pollution by linking it to the reduction in rainfall it causes.

The significance of this approach is that it addresses the rainfall process at the microphysical (nanometer) scale to the cumulus cloud (kilometer) scale. The outcome provides a direct cause-and-effect link between pollution and rainfall in a politicallypowerful region. For historical, institutional, and political reasons, existing rainfall studies focus on global models that cannot accurately predict rainfall and weather models that do not account for the impacts of pollution. Conversely, current studies of pollution fail to provide mechanistic links to cloud formation and precipitation, so that the resulting gap in understanding will tend to persist unless a novel approach is taken. In order to make a significant step forward we will have to provide this link by a multiscale cloud model tied to well-constrained Lagrangian observations.

References

Cerveny, RS., and R.C. Balling, Nature, 394: 561-563, 1998.

Ramanathan, V., P.J. Crutzen, IT. Kiehl, and D. Rosenfeld, Science, 294: 2119-2124, 2001.