PROVENANCE AND U-PB GEOCHRONOLOGY OF THE MESOPROTEROZOIC BELT SUPERGROUP (NORTHWESTERN UNITED-STATES) - IMPLICATIONS FOR AGE OF DEPOSITION AND PRE-PANTHALASSA PLATE RECONSTRUCTIONS

The Mesoproterozoic Belt-Purcell Supergroup is a thick succession of siliciclastic and carbonate sedimentary rocks in the northwestern United States and adjacent Canada. Recent stratigraphic studies provide strong evidence for an enclosed basin setting with a tectonically active western margin. The age of the western (present coordinates) landmass is unknown, but has been inferred to be different from North American basement that borders the Belt basin on the basis of whole rock Sm-Nd studies. We report U-Pb isotopic analyses of individual detrital zircon, monazite, and xenotime grains separated from westerly derived clastic units in the Belt Supergroup as well as Sm-Nd isotopic data for the same grains where appropriate. These data provide new constraints on the age and isotopic character of the western landmass. Most U-Pb data are concordant or slightly discordant; exceptions are noted below. The Revett Formation (Ravalli Group, lower Belt Supergroup) contains grains with concordant ages of 1590-1600 Ma and a grain with a discordant Pb-207/Pb-206 age of 1780 Ma: the ca. 1600 Ma age is very unusual in North America and effectively dates "Belt Island", long postulated to be the Revett source in what are now terranes of southeast Washington and northeast Oregon that were accreted in the Mesozoic. Detrital zircons from the Missoula Group (Bonner and Mount Shields formations; upper Belt Supergroup) form a distinct group at 1670-1859 Ma with two ca. 2.6 Ga grains. U-Pb ages from monazite range from 1642 to 1786 Ma, and together with Nd isotopic data (T(DM) model ages of 2.05-3.36 Ga) suggest variable interaction with previously differentiated crust. The Buffalo Hump Formation in northeast Washington may correlate with the Bonner Formation or may lie unconformably on the Belt Supergroup; it contains detrital zircons of ca. 1840 Ma and, surprisingly young, 1070-1244 Ma (several grains). Monazites have U-Pb ages of 1753-1774 Ma and Nd signatures similar to the Bonner Formation data. Several major conclusions can be drawn from these data. The ages of 1590-1600 Ma have no known possible source within western North America (although 1576 Ma augen gneiss is present in the Priest River Complex northwest of the Belt basin), and the 1642-1786 Ma ages are uncommon to rare in Precambrian rocks to the immediate south and east; together these data support previous Sm-Nd whole rock and stratigraphic conclusions of a dominantly western source for the Belt basin. The ages in the Buffalo Hump Formation can be interpreted in two ways. Either the Buffalo Hump correlates with the Bonner Formation and therefore the top of the Belt Supergroup is substantially younger (< 1070 Ma) than previously inferred (1250 Ma) and may overlap with Grenville-age tectonic activity. Alternatively, the Buffalo Hump Formation may represent a new unit that is younger than the Belt. Irrespective of this issue, these data help to constrain proposed pre-Paleozoic plate reconstructions; from the 1070-1244, 1590-1600 and 1642-1786 Ma source ages and related Nd data, we suggest that the basement blocks of south-central Australia (Gawler, Musgrave, Willyama, Arunta and possibly Mount Isa) were joined to the western side of Laurentia, adjacent to the Belt basin, prior to Neoproterozoic formation of the Panthalassa Ocean.