Scaling in 2D Rayleigh-Bénard convection

The last two decades it has been lively debated whether there exist an “ultimate regime” of heat transfer in Rayleigh-Bénard convection, in which the Nusselt number scales with Rayleigh number as Ra^{1/2} at very high Ra. In my talk last year, I argued that no such regime exists in the 3D case. This time, I consider the 2D case, and argue that there is even less reason to believe that such a regime exists in 2D. In the first part of the talk, I present the analysis of the 2D Rayleigh-Bénard system recently published in JFM ( doi.org/10.1017/jfm.2023.750 ), showing that, among other things, the computational cost of reaching stationarity is as high for the 2D as for the 3D system in the limit of high Ra. I also show that recent claims that such a regime has been simulated in 2D are unfounded. In the second part of my talk, I discuss the research ethical aspects of publishing unfounded claims of big research breakthroughs in leading scientific journals.