PETROLOGY AND PROVENANCE OF THE NEOGENE USIBELLI GROUP AND NENANA GRAVEL: IMPLICATIONS FOR THE DENUDATION HISTORY OF THE CENTRAL ALASKA RANGE
KENNETH D. RIDGWAY, JEFFREY M. TROP, and DIANE E. JONES
Department of Earth and Atmospheric Sciences, Purdue University, West Lafayette, Indiana 47907-1397, U.S.A.
ABSTRACT: Neogene basins exposed along the north-central flank of the Alaska Range record the geologic development of the highest mountain range in North America. Strata from these basins consist of the Miocene Usibelli Group (~ 600 m thick) and the Pliocene Nenana Gravel (~ 1000 m thick). Petrographic analyses of 54 sandstone thin sections and 36 conglomerate clast counts from one of these basins, the Healy Creek basin, delineate two distinct petrofacies. A stratigraphically older, quartz-rich petrofacies defines the Healy Creek, Suntrana, and Lignite Creek Formations of the Usibelli Group. A stratigraphically younger, lithic-rich petrofacies defines the Nenana Gravel. Sandstones of the lower petrofacies are characterized by a dominance of quartz (Q60F27L12; Qm33F28Lt39) and a high proportion of polycrystalline quartz (Qp68Lvm21Lsm11). Schist and quartz clasts are common in conglomerates from this petrofacies. Imbricated conglomerate clasts and planar cross-stratification indicate southward to southwestward paleoflow for this quartz-rich petrofacies. Relative to the lower petrofacies, sandstones of the upper petrofacies are defined by a dominance of lithic fragments (Q39F19L43; Qm16F21Lt63), a lower proportion of polycrystalline quartz, and a higher proportion of volcanic lithic fragments (Qp37Lvm55Lsm8). Conglomerate clast compositions in this petrofacies consist mainly of sedimentary and igneous clasts. This lithic-rich petrofacies is dominated by northward paleocurrent indicators. The compositional and paleocurrent data of the Usibelli Group and Nenana Gravel suggest a regional paleodrainage reversal from southward-flowing fluvial systems transporting sediment from the quartz-rich, metamorphic Yukon--Tanana terrane, to northward-flowing fluvial systems transporting sedimentary and igneous detritus from the growing Alaska Range.
Superimposed on the first-order tectonic signal are second-order compositional differences. These differences may be related to a combination of factors but appear to be consistent with well documented changes in paleoclimate. For example, sandstones deposited during the warmest and wettest paleoclimates are characterized by a relative enrichment in quartz, especially polycrystalline quartz (Healy Creek Formation, Q81F16L3, Qp93Lvm3Lsm4; Suntrana Formation, Q61F25L14, Qp66Lvm25Lsm9). Sandstones deposited during cooler and drier climates, in contrast, have lower proportions of quartz and higher proportions of feldspar and lithic fragments (Lignite Creek Formation, Q45F33L14, Qp56Lvm28Lsm17). Our analysis cannot account for all the possible influences on sandstone compositions during deposition of the Usibelli Group, but we contend that the apparent correlation between sandstone composition and changes in paleoclimate suggest that climatic modification of sandstone composition may have been significant.
Data table 1 - Excel spreadsheet file
Data table 2 - Excel spreadsheet file
We analyzed 54 medium- to coarse-grained sandstone samples from a continuous measured stratigraphic section through the Usibelli Group and the Nenana Gravel of the Healy Creek basin (Fig. 5). The measured section is located along the northern bank of Healy Creek at the confluence with Suntrana Creek (Fig. 2). Sandstone samples were selected from the measured stratigraphic section in order to characterize the entire measured thickness of the Usibelli Group and Nenana Gravel. Stratigraphic locations of samples are shown in Figure 5 of the article. Epoxy grain mounts were made from the poorly consolidated sandstone samples; thin sections were cut from the grain mounts. Each thin section was etched and stained for calcium and potassium feldspar using the method of Houghton (1980). Prior to final analysis, a preliminary point count was taken to determine the proper counting grid and to locate any questionable grain types. Following the preliminary point count, a minimum of 400 framework grains was counted per thin section. Matrix and cement were not counted. Separate classification and tabulations of grain types were maintained using both the Gazzi-Dickinson and "traditional" methods (Gazzi 1966; Dickinson 1970, 1985; Ingersoll et al. 1984). Definitions of raw and recalculated parameters used in the analysis are listed in Table 1 of the article. Recalculated framework modes are tabulated in Table 1 in terms of the Gazzi-Dickinson method. Recalculated point-count data and raw petrographic data are archived and available in digital form at the World Data Center for Marine Geology and Geophysics, Boulder.