Origin of komatiite at high pressures

Claude T. Herzberg, Eiji Ohtani

Research output: Contribution to journalArticlepeer-review

39 Scopus citations


Experimental melting relations to 20 GPa demonstrate that the commonly accepted model of komatiite formation by non-invariant melting (olivine + liquid) is too restrictive. Komatiites could also have formed by pseudoinvariant melting (olivine + pyroxene + garnet + liquid; modified spinel + majorite + liquid) along the anhydrous solidus at high pressures. The MgO content of komatiites cannot be used to infer the degree of partial melting in the mantle. There would have been little change in MgO at a high-pressure pseudoinvariant point, even for degrees of partial melting that ranged from very small to over 80%. Instead, the MgO content of komatiites may be a better indication of the depth of partial melting, than it is the degree of partial melting. Late Archean komatiites containing 20-28% MgO could have been generated at depths ranging from about 130 to 260 km; early Archean komatiites from the Barberton Mountain Land could have formed at depths corresponding to the present-day transition zone, or at shallower levels from an unusually primitive mantle source. Komatiites which originated by partial melting in a lower mantle formed by majorite fractionation should have geochemical characteristics that have not yet been reported. This indicates that either melting did not extend into the lower mantle or, if it did, the upper and lower mantles were isolated by a chemical and thermal boundary layer in Archean times.

Original languageEnglish (US)
Pages (from-to)321-329
Number of pages9
JournalEarth and Planetary Science Letters
Issue number3-4
StatePublished - May 1988

ASJC Scopus subject areas

  • Geophysics
  • Geochemistry and Petrology
  • Earth and Planetary Sciences (miscellaneous)
  • Space and Planetary Science


Dive into the research topics of 'Origin of komatiite at high pressures'. Together they form a unique fingerprint.

Cite this