Is it possible for an earth analog planet* to have more than 3 convection cells per hemisphere?

In our system we only have Jupiter and Saturn with 20 atmospheric bands, even Uranus and Neptune have only 3 bands. So why should a fourth cell appear instead of simply having a larger mid latitude cell? And if it is possible what condition does it needs?

For what I know a faster rotation rate should be the key factor, as described here: https://www.lpl.arizona.edu/~showman/publications/showman-etal-exoplanets-review-revised.pdf (pag 14 and 34) But how much precisely is needed for a new cell to appear?

Also a smaller axial inclination should reduce the equatorial and the polars cells (Milankovitch cycles), but i don’t know if it can be of any relevance. Can also others characteristic of the planet like surface area, gravity and atmospheric density influence the likeability (or less likeability) of having a 4th cell?

*Rocky planet in an habitable zone with a volcanic activity and magnetosphere compatible with terrestrial pluricellular carbon-water life, also with a surface gravity close to 1G and with water oceans (anyway the swamp-planet model is perfect of course), other characteristics, like atmosphere density/thickness and surface area of the planet etc... can be adapted to the needs.

  • $\begingroup$ A key term here is "coriolis effect." If you are not already familiar with it, you need to read up. If you are, then perhaps you could put that in your question, and tell us specifically what additional data you need? $\endgroup$
    – cobaltduck
    Feb 13, 2017 at 20:43
  • 2
    $\begingroup$ Hi. Short answer - No. Land / Ocean ratio affects cells but 'smooth' models show the same three bands under earthlike conditions. If you assume similar size / gravity / atmospheric density, voila - three bands. You may be able to achieve a fourth if you increase radius, decrease density to maintain 1g $\endgroup$
    – pHred
    Feb 13, 2017 at 23:24
  • 1
    $\begingroup$ Sorry, bumped POST. ... continued. @cobaltduck is on the money. If you increase spin remember you will have much greatet tectonic activity. How non-earthlike is acceptable? Atmospheric composition is less important than mean density. Cf the oceans with greater density have 1 cell (sort of) but here composition and heat transfer dominate. See thermohaline circulation. $\endgroup$
    – pHred
    Feb 13, 2017 at 23:31
  • $\begingroup$ I have read that increasing a planet radius without changing the other characteris may lead to the formation of more jet streams (which are linked with convection cells) but I haven't found any clear reference or even a rough explanation. For that i haven't included it with the rotation rate. $\endgroup$ Feb 14, 2017 at 13:14
  • $\begingroup$ Actually I’m interested in such scenario of a less dense but with larger surface area earth (with same surface gravity), since all this came out from a discussion about how many geographical and climatic barriers an inhabited earth analog could have, and of course a larger surface may mean more lands, so larger distances and more diversification. I assume too less density should mean absence of a sufficiently large iron core needed for a sufficiently strong magnetosphere, so is the 4 cells scenario plausible in the range of density needed for a sufficient magnetosphere? $\endgroup$ Feb 14, 2017 at 13:16

1 Answer 1


There may be a specialised set of geography under which this could occur but ordinarily no. The Hadley and Polar cells are heat engines directly driven by the latitudinal temperature differences. The Ferrel Cell (mid-lat circulation) is indirectly driven by these and, bingo, three cells. Threre is a relation between the molecular weight of atmospheric gases, surface gravity (thus thickness) heat flux and spin. Thus Neptune / Uranus, higher surface gravity but low molecular weight atmosphere and low heat flux have three bands. Look at the Wikipedia entry on atmospheric circulation.

  • $\begingroup$ Also, for a weather system based on insolation there must be an odd number of bands. The answer cannot be four. 3 or 5 work. The easterlies at the 'top' (higher lat) of the Hadley cell and the Westerlies at the 'bottom' (lower lat) of the Polar cell are counter rotating and are part of the driving mechanics of the Ferrel cell. The atmospheric dynamics of a system sufficiently energitic to drive three cells between Hadley and Polar cells would be diabolical and probably not conducive to life as we know it. $\endgroup$
    – pHred
    Feb 16, 2017 at 1:23

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .