On the possibility of using suspension-log data to estimate historic overburden in shallow unconsolidated sediments

Abstract

Suspension-log data are typically used for engineering and seismic studies in shallow unconsolidated sediments. At the USDOE Savannah River Site, suspension data were acquired in a series of deep boreholes (up to ∼400 m) to support high-resolution geological and geophysical investigations in the Upper Atlantic Coastal Plain. Although regional coastal plain stratigraphy is historically inferred from core, seismic, and log data, it became apparent that both major and minor unconformities and possibly formation contacts in the unconsolidated sediments of the Upper Atlantic Coastal Plain were strongly correlated to distinct variations in suspension data that translated from the deep to very shallow horizons. Overall, the suspension data contained more information (character) than typical wireline sonic logs, most likely because wireline logs are optimized for consolidated sediments (shales, sandstones, carbonates) while the suspension log is more optimized for unconsolidated sediments. Most coastal plain wireline-log stratigraphic correlations are based on gamma and resistivity data that often do not show much character in the freshwater-saturated sediments and correlations from paleontological and palynological analyses are often difficult because of the scarcity and oxidation of fossils. However, because of the correlation of suspension data with high-resolution near-surface engineering data from direct push, core samples, and shallow geophysics, it was noticed that it might be possible to correlate a deep regional engineering stratigraphy using the assumptions that coastal deposition proceeded to a zero stress level (i.e., subaqueous or subaerial exposure), that remnant burial stresses remain in deeper sediments, and that the Poisson's ratio determined from the P and SH data captured the remnant stress history. To explore this possibility, suspension-log data were correlated to stratigraphic horizons from geophysical, wireline, and direct push data. Correlated intervals were then analyzed using simple best-fit regression curves extrapolated to a value where the Poisson's ratio intersected a value for surface sediments (zero overburden) based on literature data. Initial results indicated that for some intervals an interpreted depth of historic overburden was reasonable. Overall results were mixed, but intriguing, suggesting that this technique might have value in estimating historic overburden and in providing better stratigraphic correlations. However, a better understanding of the suspension stratigraphic correlation and possibly a better model are needed.