Atmospheric Sciences Editors' Vox

Four Perspectives on Order From Chaos

What makes thunderstorms clump, even to the point of singularity, over uniform oceans? Three recent papers in JAMES address this question, and a new Commentary ties them together.


Here’s a problem that sounds simple enough: What happens to an Earth-like atmosphere over a uniform surface? Infrared radiation from water vapor and other greenhouse gases will cool and destabilize the atmosphere causing it to overturn. Heat released by condensing water breaks the symmetry between upward- and downward-moving air, resulting in narrow, intense updrafts and broad regions of gently descending air: the familiar puffy clouds or towering thunderstorms we call “convection.”

This problem of “radiative-convective equilibrium” (RCE) is a lovely idealization of the Earth’s tropics. Because the surface is uniform, it’s natural to expect that the convection, too, will be distributed homogeneously within the domain. But a funny thing happens when models with fine-enough scale to resolve convective-scale (~km-scale) motions simulate RCE: the convection tends to clump, often to the point of creating a single, massive region of convection surrounded by hundreds of kilometers of dry air.

This self-aggregation has been noticed for more than 20 years, but the problem has become more pressing in the last few years as computational capabilities have advanced and the potential importance of the problem has become clear. How does self-aggregation arise, and what leads to the massive up-gradient transport of energy and moisture? On what aspects of the atmosphere and surface does it depend? And how relevant is it for explaining the behavior of the real atmosphere—the genesis of tropical cyclones, say, or the Madden-Julian Oscillation? To go a bit further out on a limb—does the temperature dependence of self-aggregation imply a role in cloud feedbacks?

Self-aggregation in RCE was the theme of several papers published in the last year in the  Journal of Advances in Modeling Earth Systems (JAMES), one of AGU’s open access journals. This week, JAMES is pleased to publish a commentary by Brian Mapes of the University of Miami on three of these. These papers (Arnold and Randall, 2015; Bretherton and Khairoutdinov, 2015; and Holloway and Woolnough, 2016) take different but complementary approaches to the problem; Brian, in his personal and engaging manner, neatly identifies what can be learned by looking across these studies and what questions they provoke. Reading the commentary is a bit like watching self-aggregation itself, but with order emerging from ideas instead of clouds.

—Robert Pincus, Editor in Chief, Journal of Advances in Modeling Earth Systems; email: [email protected]