Principal Investigator

At a Glance

Beyond assessing effects of greenhouse gas emissions on trends in global temperature increases, research efforts led by Pacala and Shevliakova have advanced analysis of extreme precipitation from observations and climate model simulations, as well as improved representation of processes that affect climate extremes on regional scales, such as urbanization and dust emissions.


Research Highlight

During the last few decades, the main focus of numerous climate studies has been on changes in the Earth’s mean climate and how anthropogenic emissions of greenhouse gases will affect future climate projections. However, a growing body of research has acknowledged a need to understand past and future changes in climate variability and climate extremes.

In 2015, CMI postdoctoral fellow Monika Barcikowska, in collaboration with Geophysical Fluid Dynamics Laboratory (GFDL) scientists, refined statistical tools to analyze changes in extreme precipitation.1 The new analysis identified two dominant patterns of multi-decadal scale internal climate variability over the North Atlantic. The study assessed the impact of these patterns on Mediterranean winter precipitation using long (1,000 to 4,000 years) GFDL CM2.1 and CM2.5 preindustrial simulations. The first pattern resembles the North Atlantic Oscillation, which could explain over 30% of decadal winter precipitation variability observed in the regions of Spain, Morocco, Italy and the Balkans. The second pattern has a longer period, which varies from approximately 55 to 62 years.

The joint Atmospheric and Ocean Sciences Program (AOS)-CMI postdoctoral fellow Dan Li has completed implementation and evaluation of the urban component in the GFDL Earth System Model (ESM) framework. This analysis involved historical (1850 to 2005) simulations with the GFDL urban tile, forced by the high-frequency ESM output, for historical and future climates. This enabled characterization of how, over given historical periods, the magnitude of urban heat island effect has interacted with climate variability and change in the continental United States.2,3 A new set of additional simulations has explored implications of urbanization for current regional climate and climate variability using a novel stretch-grid (~10 km over North America, ~25 km over the rest of the world) implemented in the GFDL atmospheric general circulation model.

In a third project, CMI postdoctoral fellow Stuart Evans developed a capability to interactively simulate land dust emissions in the GFDL climate models.4 Previously, land dust source emissions were prescribed based on 1990s observations and did not change throughout historical or future simulations. This study involved performing a set of 500-year simulations for preindustrial climate conditions with the new land dust emission configuration. Analysis of the new simulations with interactive dust emissions has shown that land surface is indeed important to accurately model dust variability and its implications for climate variability. By accounting for soil moisture and vegetation, the new runs produce inter-annual variability that closely matches satellite observations and recreate the relationships seen in observations between specific dust sources and major climate indices such as El Niño. The new capability is enabling a novel exploration of the interplay between enhanced incidences of dust and droughts in different regions of the world, particularly Australia (Figure 1.5) and Africa.

El Niño and El Niña composite map
Figure 1.5. (Left) Schematic diagram of dust-ecosystem-climate interactions, showing how the “Dusty” CM3 model connects Australian dust emissions with the El Niño–Southern Oscillation (ENSO). (Right) DJF composite anomalies of Australia in La Niña and El Niño phases of ENSO, showing relationships between precipitation and dust optical depth. ENSO leads to precipitation anomalies, which in turn lead to dust anomalies. Contours of optical depth are superimposed on a color scale for deviation from mean precipitation.


  1. Barcikowska, M., and S. Kapnick, 2016. Impact of Large-Scale Circulation in the North Atlantic Sector on Mediterranean Winter Hydroclimate. Manuscript in preparation, March, 2016.
  2. Li, D., S. Malyshev, and E. Shevliakova, 2016. Exploring Historical and Future Urban Climate in the Earth System Modeling Framework. Part I: Model Development and Evaluation. J. Adv. Model Earth Sy., in review.
  3. Li, D, S. Malyshev, and E. Shevliakova, 2016. Exploring Historical and Future Urban Climate in the Earth System Modeling Framework. Part II: Interactions Between Urban Heat Islands and Climate Change Over the Continental United States. J. Adv. Model Earth Sy., in review.
  4. Evans, S.M., P.A. Ginoux, S. Malyshev, and E. Shevliakova, 2016. The Importance of the Land Surface to Australian Dust Variability in Models and Observations. Manuscript in preparation, March, 2016.