In the theory of tectonic plates, at a convergent boundary between a continental plate and an oceanic plate, the denser plate usually subducts underneath the less dense plate. It is well known that oceanic plates subduct under continental plates, and therefore oceanic plates are more dense than continental plates.

My question is why are the oceanic plates always denser than the continental plates. I'm aware that the difference in density can be attributed to the plates differing compositions, but what I'm interested in is why these plates have different composition in the first place giving rise to their relative difference in densities.

  • $\begingroup$ Oceanic crust is denser than continental crust because: it contains silica $\endgroup$
    – user5121
    Jan 8 '16 at 14:30

Ocean lithosphere (geophysical definition of crust + upper mantle that acts as a 'plate') is primarily of basaltic composition - the upper levels are basalt and the lower levels are gabbro. The top levels have been proven with boreholes, whilst the lower levels have been inferred from transform fault sampling and comparisons with ophiolites. This sequence is produced by partial melting of mantle peridotite at a fairly controlled rate. So much so that basalts formed in this way even have a specific composition "MORB" (Mid Ocean Ridge Basalt).

In contrast, continent lithosphere is more complex and tends to be of a 'granitic' composition. This includes granites but can also include a lot of metamorphic rocks (eg. gneiss) and sediments. Sediments are lower density anyway (high pore space), but so are quartz-rich rocks such as granites. The various processes that build continents tend to favour silica rich compositions, resulting in this bulk "granitic" composition. For example, limited partial melting will initially produce high silica, high alkali melts. Erosion will tend to break down most common minerals before quartz - leaving quartz-rich sediments (hence sandstone is primarily quartz). Metamorphism of pelites (rocks rich in Al and Si) will tend from the initial mudstones & basalts through to gneisses & migmatites (which have a lot of quartz and feldspar). Migmatites are partially melted - and the melted bits are essentially granite.

Basalt is denser than granite. On gravity surveys, basalts and gabbros will appear as positive anomalies, whilst granites and sedimentary basins will appear as negative anomalies.

  • 1
    $\begingroup$ @windwaed: Your answer is great. Because the question speaks about plates I recommend editing "oceanic crust" to "oceanic lithosphere" and following from that, talk about the combined density of "basaltic oceanic crust plus mantle lithosphere" and "continental crust + mantle lithosphere". $\endgroup$ Apr 22 '14 at 16:58
  • $\begingroup$ @Speiss, yes I guess as we're talking about density, lithosphere is the more correct term. I'll modify. $\endgroup$
    – winwaed
    Apr 22 '14 at 19:19

Not the answer you're looking for? Browse other questions tagged or ask your own question.