EOARCHEAN OPHIOLITES? NEW EVIDENCE FOR THE DEBATE ON THE ISUA SUPRACRUSTAL BELT, SOUTHERN WEST GREENLAND

The Eoarchean Isua supracrustal belt (Greenland) comprises two distinct packages that have been dated at ca. 3800 Ma and 3700 Ma. Both packages consist of strongly deformed, pillow lavas and lesser amounts of gabbro (island arc tholeiite, picrite and boninite protoliths), felsic schists (andesite-dacite protoliths), chemical sedimentary rocks and depleted mantle dunites, most are at amphibolite facies. This paper focuses on the ca. 3700 Ma assemblage in the northwestern arm of the belt. In a rare low strain area layered gabbro preserves an igneous texture and contains high Th/U igneous zircons with an age of 3720 to 3710 Ma. Boninitic amphibolites with relict pillow structure are cut by a 3712 +/- 6 Ma tonalite sheet and an amphibolite-ultramafic schist tectonic contact is transgressed by a 3717 +/- 6 Ma mafic tonalite intrusion. Strongly deformed felsic schists of likely volcanic origin contain 3720 to 3700 Ma igneous zircons, showing they are broadly contemporaneous with the intercalated mafic volcanic rocks. The 3720 to 3710 Ma supracrustal rocks are also intruded by 3696 6 Ma tonalite. Hence Isua juvenile 3720 to 3710 Ma arc development was dominated by boninite, tholeiite and picrite eruption with mafic tonalite intrusion and evolved to maturity between 3710 to 3700 Ma, with formation of andesites, dacites, related volcano-sedimentary rocks and then intrusion of 3700 to 3690 Ma tonalites. The 3720 to 3710 Ma assemblage comprising mantle peridotite, gabbros, pillowed and non-pillowed volcanic rocks and chemical sedimentary rocks, is interpreted as an example of a ca. 3710 Ma ophiolite. However, it is not an intact section through an ophiolite for several reasons. First, this assemblage was repeatedly deformed and partitioned by Eoarchean shear zones and second, because a convincing "sheeted dike" complex has not yet been identified. Geochemical data from a candidate for sheeted dikes proposed by other workers does not match the geochemistry of nearby ca. 3700 Ma amphibolites with relict pillow structure, but instead better matches ca. 3510 Ma noritic Ameralik dykes. Nonetheless, the 3720 to 3690 Ma juvenile assemblage is juxtaposed to the south with a 3800 Ma complex, and hence it resembles Sierran-style ophiolites that are ensiniatic island arc terranes accreted against older crust. This evidence supports the concept that crust-formation processes at convergent plate boundaries were operating almost 4 billion years ago.