Multivariate statistical “unmixing” of Indian and Pacific Ocean sediment provenance

• DOI: 10.1002/lom3.10645
• Publication Date: 2024-09-07T13:23:58Z
• Authors: Ann G. Dunlea; Kazutaka Yasukawa; Erika Tanaka; Ingrid L. Hendy

Abstract The geochemistry of marine sediment is a massive archive (paleo)oceanographic information. Accessing that information requires “unmixing” the various influences on to understand individual sources and geochemical processes. Q‐mode factor analysis (QFA) independent component (ICA) are multivariate statistical techniques have successfully been applied large datasets element concentrations identify number composition or end‐members. In this study, we apply both two geochemistry, compare output, discuss advantages each approach. datasets, ICA identified mixing trend between carbonates dust, whereas QFA represented end‐members as separate factors. Pacific Indian Oceans dataset, produced three factors components involving rare earth elements, but explained small, almost negligible, amount variability dataset. Also, more aluminosilicate (dust volcanic ash) than ICA. Ocean Sites 738 752 processes affecting Sr Ba components, while created representing covariation over intervals site's paleoceanographic history. results study exemplify identifies covariances finds discrete contributing bulk mass sediment. works best with non‐Gaussian distributions signals trends constitute characteristic structure multielemental data.