The determination of lithology from core physical properties measurements

I performed statistical analysis of shipboard physical properties data from the Ocean Drilling Program to investigate relationships between the physical properties data and the lithology of deep ocean cores. The use of non-invasive experiments on deep-sea core samples offers a near real-time view of sediments and requires little user interaction or interpretation. The speed, density, and accuracy of these experiments make efficient use of limited space and expensive ship time. The fact that these experiments are noninvasive also allows for further post-cruise studies. For the study I chose Leg 162 (July-September 1995 in the North Atlantic) for the density of data, the experiments performed, the quantity and quality of post-cruise publications and the influence of different, yet dominant, environments. Combining similar lithologies across a Leg increased sample size and offered a more statistically normalized sample. Interpolation of the physical properties data matched the intervals used for the lithological determinations. Statistical methods included univariate and multivariate correlation matrices, mean and standard deviations, the significance of the correlations, and a model equation for each lithology and the Leg as a whole. By looking at the physical properties, one can estimate the lithology. This research is important because sedimentological and geophysical approaches can be merged to offer a more accurate, more detailed view of the depositional history of oceanic cores. Through statistical analysis of geophysical data, the findings duplicate the findings of the sedimentologists without the painstaking examinations typical of this type of research. Performing experiments and analysis quickly and accurately with minimal operator error allows for immediate discussions and results. Use of this research as a data verification tool provides the ability to distinguish data acquisition problems and misidentifications. This application has proven invaluable for allowing a non-sedimentologist quantitative insight into the lithology.