Evolutionary Reasoning and the Interpretation of Convective Organization

Convection has a job to do, and flow configurations that do it better are rewarded by gravity in the currency of energy. These successful configurations thrive and persist, and thus are far more likely to be observed deep in time than their complexity would naively suggest. Naively, being far from random, complex configurations might seem unlikely, but as the word "organization" implies (if used with care), they enjoy a fitness bonus.

This is evolutionary reasoning, different from biological evolution in several important ways which must be carefully framed, but nevertheless an inexorable consequence of pure logic and the longness of time. Unlike biological life with its detailed millennium-spanning genetic codes, the memory of convective configurations is merely inertia, allowing only hours or days for the process of selection-of-the-more-fit. But unlike life, convective clouds are necessary structural elements their sustaining energy flow, and are not constrained by second-law entropy considerations: While chemical life hinges on the higher emission temperature of solar photons vs. longwave photons, atmospheric convection would be quite on a planet whose surface was heated by microwave radiation, for instance. Still, the Shannon information = -entropy formalism can be utilized to characterize the "complexity" of flow and cloud configurations. That information-entropy simply need not follow a second-law monotonic inequality, and has some nonuniqueness in the definition of its "unorganized" maximum reference point.

These ideas imply a set of logical predictions about convective organization and its environmental relationships. Those predictions serve as a paradigm for re-interpreting observations and findings that have previously been catalogued as a matter of flat empiricism, or "understood" with flabby heuristics lacking this sense of a complex configuration's functional meaning. Some of these evolutionary-convection predictions could be tested incisively, for example via domain dependence experiments in a peculiar scale-restricted setting: Multiscale or "Super-Parameterization" models.