FROM THIN SECTIONS TO BASINS; USING WELL CUTTINGS TO GENERATE SEQUENCE STRATIGRAPHIC FRAMEWORKS

Detailed examination of plastic-impregnated, thin sectioned well cuttings has been used to generate regional high resolution, lithology-based sequence stratigraphic frameworks in areas with limited outcrop and subsurface core control. This dataset, when incorporated with wireline logs, seismic data, and chronostratigraphic control is a highly effective, yet low-cost method of identifying lithologic, diagenetic, and pore system information at a scale that can be easily tied to hydrocarbon reservoirs or aquifer systems. Subsurface sediments from the Albemarle basin of eastern North Carolina were selected for this study, because they form a significantly thicker and more complete stratigraphic package than coeval highly-thinned and eroded updip outcrops. This basin, like many Cenozoic basins, has a handful of shallow cored wells concentrated in updip facies; however, more than 500 exploratory oil and water wells have been drilled and sampled by means of well cuttings across the basin. The deeper Jurassic-Lower Cretaceous interval has no outcrop equivalent and only one highly discontinuous core, yet more than 20 wells (with cuttings and wireline logs) penetrate the interval across the basin. Integration of petrographic analyses from thin sectioned carbonate-prone cuttings and binocular analysis of poorly-lithified, siliciclastic-prone cuttings allowed calibration of wireline log responses and seismic reflectors with appropriate lithologies, revealed a variety of exceptions to conventional wireline log and seismic interpretations. In addition to regional lithologic information, petrographic analyses of cuttings from this passive margin succession have been used to document and map the regional cementation and dissolution fabrics preserved in carbonate-prone facies. While these analytical techniques are well established, they have been underutilized in broad-scale stratigraphic studies, largely due to the general bias that the cuttings dataset is too prone to mixing and hence, unreliable. Results from this study indicate that while some downhole mixing does occur, it generally can be readily identified and amended when properly integrated with additional datasets, yielding valuable information from readily available datasets that have long been overlooked.