Method for Quantifying Correlation in Oceanographic Profiles

This document is a research article that presents a method for quantifying correlation in oceanographic profile data using functional data analysis (FDA). The study focuses on vertical profiles, which are essential for understanding the physical and biogeochemical features of ocean water columns. The authors apply a recently developed technique to define a scalar correlation coefficient for functional data, allowing for the assessment of similarities in overall profile shape rather than just specific depth values. Two case studies illustrate the method's application: one examines the correlation between different oceanographic variables, while the other estimates the temporal autocorrelation function of a single variable. The findings indicate that this approach can effectively identify physical drivers of variability in biogeochemical profile structures and reveal regional differences in relationships between profiled variables. The technique has broad potential applications in oceanographic data analysis, including observing system optimization and the analysis of vertically structured ocean processes.