This is one of 16 Rapid Response (RAPID) projects funded as the result of a Dear Colleague Letter (NSF 11-006) encouraging diagnostic analyses of climate model simulations prepared for the Intergovernmental Panel on Climate Change Fifth Assessment Report (IPCC AR5). Research conducted in these projects is expected to lead to more detailed model intercomparisons, better understanding of robust model behaviors, and better understanding and quantification of uncertainty in future climate simulations.

The project considers the drying of the Mediterranean region over the second half of the 20th century as portrayed in the ensemble of 20th century climate simulations prepared for the AR5. The central question motivating the project is whether this drying occurred due to natural variability or anthropogenic greenhouse warming. The PIs note that continued drying of the Mediterranean is a robust result across the ensemble of climate model projections of greenhouse gas-induced warming over the 21st century, so there is reason to believe that the region is susceptible to drying as a consequence of global warming. On the other hand, much of the observed drying of the Mediterranean can be linked to the prolonged period in the late 20th century during which the North Atlantic Oscillation (NAO) was in its high phase, as the high phase is associated with a northward shift of the Atlantic stormtrack that provides the Mediterranean with much of its wintertime precipitation. The extent to which the NAO is influenced by global warming is not known, and several mechanisms of influence have been proposed. Work performed in this project uses statistical analysis, primarily in the form of signal-to-noise maximizing empirical orthogonal functions (EOFs), to determine the relative importance of natural variability and greenhouse gas-induced global warming in producing Mediterranean drying in 20th century climate simulations. Additional analysis using a decomposition of the ensemble-mean moisture budget is used to discriminate between possible mechanisms for global warming-induced drying.

The broader impact of the project lies in its support of the IPCC AR5, which is intended to provide information on climate change and its consequences to decision makers worldwide. In particular, the work will provide guidance for decision makers concerned with Mediterranean drying, as the drying of recent decades may or may not reverse depending on whether it is part of a natural fluctuation or a consequence of secular global warming. The mechanisms of natural variability and forced change explored in this project may also be consequential for climate change occuring in other parts of the world including the semiarid subtropical regions of the United States.

Project Report

This grant supported research on the causes of the strong reduction in Mediterranean region rainfall that occurred in the second half of the 20th century. Using observed datasets derived from weather stations throughout the region, and suites of global climate models used in the most recent and upcoming Intergovernmental Panel on Climate Change Assessment reports, analysis was performed with several key goals in mind: first, to determine whether the previous generation of models was capable of simulating winter drying trends thirty years in length that were as robust as the observed decline in rainfall from the 1960s to the 1990s; second, to estimate using these same models how much of the observed winter trend was forced by increasing greenhouse gas concentrations in the global atmosphere; and third, to evaluate the improvement in the newest generation of global climate models with respect to simulation of climatology and trend, for the six-month "winter" and "summer" seasons. Water is a vital human resource, and continued declines in Mediterranean region rainfall during the 21st Century as predicted by the models would have important consequences for the people that live there. Therefore it is necessary to evaluate the uncertainty associated with the global climate model predictions and improve our understanding of the mechanisms that drive precipitation variability and change and how these mechanisms differ from winter to summer. A Mediterranean climate is one characterized by wet winters and dry summers. It is well known that precipitation variability over the region and over Europe is strongly influenced by the large scale dynamics of the northern hemisphere midlatitudes, particularly the atmospheric pressure patterns over the North Atlantic. These can be thought of as two distinct phases, a negative and a positive, that govern whether storms are more likely to occur equatorward bringing more rain to the Mediterranean, or shift poleward reducing the precipitation there. This natural oscillation of atmospheric pressure occurs not only from year to year, but also from several decades to the next several decades, based on the limited observational record. In the presence of such large natural variability it is important to know how much global warming may be influencing the drying of the Mediterranean region. In particular, how well can the models simulate the precipitation variability of the 20th Century in space and time, particularly the multi-decadal variability, and if so, how much has the precipitation response to the global warming signal already begun to emerge and what is projected for the 21st Century? This research includes several key outcomes. First, that the previous generation of models was able to simulate thirty-year trends in winter precipitation that were as large as the observed drying during the late 20th Century, but only as rare events. Second, based on model-derived estimates of the global warming signal, it is concluded that the observed winter drying trend was dominated by the natural variability which trended toward the positive phase of North Atlantic atmospheric variability during these several decades. It is expected however based on this same signal estimate that the forced global warming contribution to future drying will grow relative to the natural variability. Next it is concluded that the newest generation of global climate models, through increases in spatial resolution and other changes, collectively show modest improvement in their characterization of Mediterranean climate, although the uncertainty is much larger during summer. This work comprises a portion of Colin Kelley’s doctoral thesis.

National Science Foundation (NSF)
Division of Atmospheric and Geospace Sciences (AGS)
Standard Grant (Standard)
Application #
Program Officer
Eric T. DeWeaver
Project Start
Project End
Budget Start
Budget End
Support Year
Fiscal Year
Total Cost
Indirect Cost
Columbia University
New York
United States
Zip Code