Cross correlation of paleoecological and geochemical proxies: A holistic approach to the study of past global change

As our understanding of ancient greenhouse intervals has grown, the difficulties in applying uniformitarian models for paleoenvironmental reconstruction of these intervals has become more apparent. In order to meet this challenge, sedimentary geologists conduct multiparameter studies of these critical intervals of Earth history. Various parameters are now commonly employed as proxies for primary depositional processes. These parameters include paleoecologic data, such as planktic and benthic biofacies, as well as geochemical data, including abundance and/or isotopic ratios of carbon, sulfur, nitrogen, iron, and other elements. As proxies for changes in primary production, nutrient dynamics, benthic and water column oxygenation, and paleo-pCO₂ levels, these data have the potential to greatly enhance our understanding of the dynamics of climateocean interactions in the geologic past. Analysis of stratigraphic covariance between proxies offers a means to test hypotheses for the relationships between paleoenvironmental processes and to evaluate the significance of temporal variations in those relationships. In this paper we present a method of moving window cross-correlation for analysis of stratigraphic covariance between paleoecological and geochemical proxies in ancient strata. The method improves upon past versions by employing multiple sizes of windows to assess correlation at different levels of temporal resolution, and by integrating biotic and geochemical data. It is illustrated with examples from a critical interval of mid-Cretaceous (Cenomanian-Turonian) strata in the Western Interior Basin. Application of the method reveals decoupling among processes (detrital dilution, organic production, carbonate production, benthic oxygen deficiency, and OC preservation) that have been interpreted previously to be coordinated due to climatic cyclicity. Moving window cross-correlation significantly improves the information gained from application of statistical methods to stratigraphic data and should be useful in studies of other critical intervals of Earth history.