Talk:Mantle convection

University links
Regarding Angelpeream's recent move and comment in the Edit Summary, I would very much support removing all these university links - most geology departments will, one way or another, have someone connected with studying mantle convection. My 2 cents. Cheers Geologyguy 19:42, 28 October 2007 (UTC)
 * Since this list could easily be one of all the major universities in the world, I removed it. Geologyguy 15:34, 4 December 2007 (UTC)

Mantle convection caused by cold Oceanic crust ??
Convection occurs when the temperature gradient (change of temperature through distance) is steep, specifically too steep for static conduction to carry the resulting heat flow. Thus, to speed heat flow, bulk motion begins, i.e. convection. Steep temperature gradients → convection. Now, cold Oceanic crust is the principle heat-sink of earth. Temperature profiles ("geotherms") are steep under Oceanic crust. Perhaps, then, Mantle convection is caused by cooling effects from cold Oceanic crust, and cold oceans? The lack of (cold) seas on Venus, Mars, coincides with their lack of Mantle convection. 66.235.38.214 (talk) 13:34, 20 October 2012 (UTC)
 * Convection involves hot light rising material, and cold dense sinking material. Material within a convecting layer cannot be characterized with a single temperature & density, at each depth.  Rather, the upwelling & downwelling streams must be considered separately.  The difference in temperature (and density) could put the upwelling material into a different phase regime, as compared to the downwelling material.  For example, the hot upwelling material might be a wet-melted slurry, whereas the cold downwelling material might be in solid state:
 * http://s9.postimage.org/ln2hqyppr/Convection_multiphase.png
 * Note that a "steep temperature gradient" (large change in temperature per change in depth) looks horizontal, on typical geology diagrams, whereupon depth is plotted on the vertical axis. 66.235.38.214 (talk) 08:44, 21 October 2012 (UTC)

Speed of convection: an inconsistency?
The speed-of-convection section gives a speed of 20 mm/yr near the surface and seems to be implying that this is faster than a 200-million-year "deep cycle". But if we take the depth of the mantle to be 3500 km in round numbers, the length of a cycle from the bottom of the mantle to the top and back would have to be on the order of 7000 km. A 200-million-year cycle would therefore have to travel at a rate of 7000/200 = 35 km/million years, or 35 mm/yr. This is nearly twice the claimed 20 mm/yr near the top.

Likewise, assuming the 50-million-year shallower cycle is slower than 20 mm/yr (= 20 km/million years), a similar calculation shows that it could not be deeper than 500 km or one-seventh the depth of the mantle.

Did someone slip a decimal point, or does mantle convection actually slow down near the top, or are the cycles much shorter than implied by the figure? Vaughan Pratt (talk) 16:44, 19 November 2015 (UTC)

External links modified
Hello fellow Wikipedians,

I have just modified one external link on Mantle convection. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:
 * Added archive https://web.archive.org/web/20110928001747/http://www.iris.edu/hq/gallery/photo/4344 to http://www.iris.edu/hq/gallery/photo/4344

When you have finished reviewing my changes, you may follow the instructions on the template below to fix any issues with the URLs.

Cheers.— InternetArchiveBot  (Report bug) 20:23, 15 January 2018 (UTC)

Better cartoon needed?
Fellow Wikipedians,

I am concerned about the main cartoon at the top of this page. While all cartoons are certainly abstractions, this one shows a number of demonstrably incorrect aspects including (1) double-sided subduction (subduction is one-sided), (2) slabs sinking straight to the CMB (slabs follow complex and variable paths), (3) mid-ocean ridges sourced from the CMB (MOR are sourced from the shallow mantle), (4) simple cellular convection patterns (the mantle Rayleigh number is far too high for this type of pattern). This cartoon and its variants have become ubiquitous on the web, possibly spreading these misconceptions. Does a better cartoon exist that can replace this one?

Regards, Geo jt (talk) 12:54, 17 December 2018 (UTC)
 * What do you mean by double-sided subduction? RockMagnetist(talk) 18:13, 17 December 2018 (UTC)
 * The problem with finding a replacement issue is that it's hard to find good ones without copyright restrictions - see, for example, the results of this search. If we want a better cartoon, someone may have to create it. RockMagnetist(talk) 18:13, 17 December 2018 (UTC)
 * I meant that the arrows made it look like material from both plates was subducting, rather than a single subducting and single overriding plate. Though I suppose one arrow could be the plate and the other the corner flow.