Modeling the climate impact of volcanic eruptions

Volcanic eruptions can have global climate impacts lasting several years. Large explosive eruptions can inject sulfur gases into the stratosphere, which are converted to sulfate aerosols. These large masses of stratospheric aerosols decrease incoming shortwave solar radiation, resulting in the cooling of the Earth’s surface. Sulfate aerosols injected into the tropical stratosphere are transported poleward with a global e-folding lifetime of about one year, meaning climate impacts of large volcanic eruptions can last up to several years. Because of the lack of observations, climate models are heavily relied upon to analyze the climate impact of large, explosive volcanic eruptions. While current climate models can reasonably reproduce many of the typical climate responses to volcanic eruptions—suppressed precipitation and droughts and surface cooling lasting two to three years—there are other observed responses that are not as well reproduced in climate models. For example, the Northern Hemisphere (NH) winter warming response in the first 1–2 winters after tropical volcanic eruptions, which is well observed, is not captured in most model simulations. The surface winter warming response over NH landmasses is caused by a strengthened polar vortex due to the heating of volcanic aerosols in the tropical stratosphere. A strengthened polar vortex has been associated with a positive phase of the North Atlantic Oscillation and the Arctic Oscillation, both indices of the wintertime variability of NH sea level pressure. In this thesis, I explore the model response to volcanic eruptions, focusing in particular on the apparent lack of a winter warming response in current climate model simulations. My first step is to examine the winter warming response to tropical volcanic eruptions in the Coupled Model Intercomparison Project 5 (CMIP5) historical simulations. Previous studies have analyzed the response in the historical simulations, but looked at only 13 CMIP5 models and averaged the first two winters, finding little to no response. Here, I analyze all 24 CMIP5 models, include only the two largest eruptions (1883 Krakatau and 1991 Pinatubo), and look at only the first winter after the eruptions. The CMIP5 historical ensemble has the advantage of a large number of models and a large number of ensemble members for each model. On the other hand, the drawback of analyzing the historical ensemble is that there are only two very large eruptions over the 1850–2005 historical period. Therefore, as a second step, I analyze the winter warming response in the CMIP5/Paleoclimate Model Intercomparison Project 3 past1000 ensemble and the Community Earth System Model (CESM) Last Millennium Ensemble. These experiments, which span 850–1850, are longer than the historical experiment, and therefore have fewer participating models and fewer ensemble members for each model. However, there were many more large volcanic eruptions over the 850–1850 period than in the historical period, which will provide a better look at the winter warming response to large volcanic eruptions. In contrast to the general winter warming response to tropical volcanic eruptions, I also focus on a specific eruption to which the response has not been well resolved by climate models. The Laki eruption in Iceland, which began in June 1783, was followed by many of the typical climate responses to volcanic eruptions: suppressed precipitation and droughts, crop failure, and surface cooling lasting two to three years. In contrast to the observed cooling in 1784–1786, the summer of 1783 was anomalously warm in western Europe, with July temperatures reaching more than 3 K above the mean in some areas. While climate models can generally reproduce the surface cooling and decreased rainfall associated with volcanic eruptions, model studies have failed to reproduce the extreme warming in western Europe that followed the Laki eruption. As a result of the inability to reproduce the anomalous warming, the question remains as to whether this phenomenon was a response to the eruption, or merely an example of internal climate variability. Using CESM from the National Center for Atmospheric Research, I investigate the role of the aerosol indirect effect of the “Laki haze,” and propose a mechanism for its effect on Europe’s summer climate. Understanding the cause of this anomaly is important not only for historical purposes, but also for understanding and predicting possible climate responses to future high-latitude volcanic eruptions.Ph.D.Includes bibliographical referencesby Brian Zambr