ALLUVIAL STRATIGRAPHY OF THE CENTRAL CONNECTICUT RIVER VALLEY: TOWARDS AN ARCHAEOOGICAL MODEL

The alluvial history of the central Connecticut River Valley was reconstructed using: (1) an analysis of the historic flood and sediment regimes; (2) geospatial reconstructions of channel changes; and (3) reconstruction of the alluvial stratigraphy based on split-spoon samples from 16 conventional geotechnical boreholes and continuous cores from 27 hydraulic push sampling devices (geoprobes) extending from 7m above to 3 meters below modern sea level. Core analysis involved macroscopic logging and description, magnetic susceptibility, granulometry, pollen analysis, and radiocarbon dating (14 dates ranging from 990 +/- 40 to 8920 +/-40 14C yr BP (0.1 to 8.9 ka). From the bottom up, the cores included: diamict and clay associated with withdrawal of the Laurentide Ice Sheet prior to 12 ka; relict terrace deposits associated with reworking of Glacial Lake Hitchcock by braided streams; a transitional sediment regime between 7-9 ka dominated by channeling, rapid vertical accretion and stabilization of the floodplain by vegetation; a prolonged stable phase during most of the mid- to late-Holocene of thoroughly bioturbated, uniformly silt-clay, overbank deposits; a return to a more rapid sedimentation during a more volatile overbank flood regime between 1.5 and 2.5 ka; formation of a well developed paleosol approximately 1 ka; a return to flooding during or before European settlement; and the deposition of fill and other disturbances during urbanization. The abrupt onset of the Holocene floodplain between 7-9 ka coincides with the inception of the world's major deltas, suggesting that bottom-up stabilization of global sea level was the main forcing agent. The roles of other forcing agents including isostatic recovery, local sea level, watershed changes in vegetation and climate influencing flood flows and human activity are being explored. The Middle-Late Archaic archaeological transition from small, sparse sites to more extensive, abundant and diverse sites coincides with floodplain stabilization. The Late Woodland phase coincided with a reduction in overbank sedimentation and soil formation. Beyond these compelling correlations, we are developing a GIS model for future archaeological research.