Using data mining and visualization techniques for the reconstruction of ocean paleodynamics

Ocean dynamics modeling is essential for predicting the impact of climatic change on human activities. The need of large time series (e.g. paleoclimatic data) has been exposed as one of the challenges to embark on decadal climate predictability. The goal of this study was to reconstruct the surface water dynamics in the South China Sea during the last 130 kyears. In order to do that it was necessary to assemble integrated comprehensive datasets obtained from a gravity core recovered in the Sunda Slope using different techniques such as quantitative analyses from coccolithophores, stable isotopes and biomarkers. Thanks to data mining, through a variety of data analysis tools, from Silicon Graphics' Mineset, such as decision trees and clustering we have established the variations in the water column stratification (relative position of nutricline/thermocline). This study is a good example of combining data mining and paleo-oceanography to explain some general paleodynamics, including short-time events, showing the potential to monitor and predict in the context of decadal time-series.