SIMILAR-TO-3710 AND GREATER-THAN-OR-EQUAL-TO-3790 MA VOLCANIC SEQUENCES IN THE ISUA (GREENLAND) SUPRACRUSTAL BELT - STRUCTURAL AND ND ISOTOPE IMPLICATIONS

The Isua supracrustal belt of southern West Greenland is the largest b ody of early Archaean supracrustal rocks, and as such has been the foc us of numerous studies to characterise early Earth processes. Based on new geochronological constraints we re-interpret the Isua supracrusta l belt stratigraphy and re-evaluate Nd isotopic constraints for the ea rly Earth from Isua supracrustal belt rocks. Part of the belt (similar to 50%) consists of a package of mafic chloritic schists (the garbens chiefer unit), layered amphibolites and felsic rocks, mica(+/-kyanite) schists and banded iron formation (BIF), and is interpreted as a sequ ence of mafic to felsic volcanic/volcanosedimentary rocks, turbidites and pelites with minor gabbro. Graded felsic volcanics in this package have yielded a U-Pb zircon age of similar to 3710 Ma, Along the south ern side of the belt, a package of amphibolites and ultramafic schists with BIF layers, interpreted as a mafic to ultramafic volcanic sequen ce, is cut by tonalite sheets with U-Pb zircon ages of 3790-3800 Ma. T hus the belt contains at least two unrelated supracrustal packages of different ages (greater than or equal to 3790 Ma and similar to 3710 M a), that were probably juxtaposed in the early Archaean, Felsic volcan ic rocks from a single similar to 3710 Ma unit have consistent initial epsilon(Nd) values of +0.8 to +1.8, suggesting minimal fractionation of Nd from Sm during later metamorphism. They may have formed from a d epleted (but not ultra-depleted) mantle source only a short time befor e 3710 Ma, with or without a contribution from LREE-enriched crust of 3750-3870 Ma with epsilon(Nd(3800)) of +2 to +4, There is debate as to whether the 3806 +/- 2 Ma 'A6' felsic unit with epsilon(Nd(3800)) of +1.0 to +2.9 is derived from volcanic rocks or from granitoid sheets, Thus in the published Isua supracrustal belt Nd database there are no samples which are definitely > 3710 Ma supracrustal or clearly mantle- derived rocks, Given the problems of locally intense alteration, the p resence of supracrustal rocks of two or more ages and of older inherit ed crustal components in some samples, the Isua supracrustal belt is n ot the ideal locality to find definitive answers on early Archaean man tle evolution. (C) 1997 Elsevier Science B.V.