The article can be found here: https://www.nature.com/articles/s41561-018-0157-x
One of the most iconic pictures of Venus of this decade is probably the image acquired by the Long Infrared camera (LIR) on-board the Japanese spacecraft Akatsuki, right after its orbital insertion in 2015. It shows a fixed bow-shaped feature at the cloud top, despite pervasive 130 m/s winds. Soon, it appeared that the most likely explanation for this structure was an atmospheric mountain wave, occurring when the airflow is perturbed by a mountain, producing vertical oscillations. On Earth, they are usually very common, gentle, local, meteorological phenomena. When it comes to Venus, Akatsuki revealed that mountain waves are astonishing, being as big as the planet itself, extending vertically to huge altitudes, and lasting 30 days, the duration of the Venusian afternoon.
As a modeler, I never cease to be fascinated by how disparate physics equations, put into hundreds of thousands of lines of code, result in multiple arrays of numbers that turn out to represent, sometimes very accurately, actual matters happening in our Universe. But in this case, no one had ever envisioned such a wave, even less produced it in a numerical simulation of Venus’s atmosphere. If we were able to reproduce this wave in a simulation, we could have a clue of what is going on in the atmosphere of Venus by deciphering the interactions of the various physical processes at stake in our virtual planet.
But first, what were we missing in our climate model? Everyone suspected that the wave assuredly emanates from the surface, so I had a closer look at what happens down below, where the hot surface can melt lead, and 90 bars of pressure make the air a supercritical fluid. For obvious reasons, climate models of extraterrestrial planets have a strong heritage from Earth. Among the jumble of terrestrial leftover chunks of unemployed subroutines (for instance moist convection or water clouds), I noticed that a cryptic switch was off, pertaining to surface effects on the atmospheric circulation. It controlled the atmospheric drag of mountains, which is now a standard for any self-respecting Earth climate model. It nevertheless took a few months of effort to understand and fully adapt this scheme for Venus, and voilà! The model was now able to reproduce these observed mountain waves, with revealing access to why they form the way they do.
Among the things we were now able to explore with our functioning, better-than-ever, virtual planet, one of my collaborators pointed out the possible atmospheric effects affecting the rotation rate of the solid body, that he correctly theorized many years ago. Inevitably, it turned out that mountain waves, computed from our simulations, are a predominant term in the total atmospheric torque on the solid body. So, if you were standing on the surface of Venus and somehow managed to survive a full day (116 days on Earth), you would want to constantly adjust your watch. Indeed, the rotation rate of Venus can change by a few minutes when the winds blow strongly enough against the mountain flanks, and thus create these massive mountain waves. Admittedly, there is a similar effect on our planet, but of a few milliseconds only, that is to say a relative change of the day three orders of magnitude less than on Venus. Once again, Venus proves to be quite an intriguing planet.