Temporal Data Mining for the Discovery and Analysis of Ocean Climate Indices
Michael S. Steinbach, Pang‐Ning Tan, Vipin Kumar, Steven A. Klooster, Christopher S. Potter · 2002
To predict the effect of the oceans on land climate, Earth Scientists have developed ocean climate indices (OCIs), which are time series that summarize the behavior of selected areas of the Earth's oceans. For example, the Southern Oscillation Index (SOI) is an OCI that is associated with El Nino. In the past, Earth scientists have used observation and, more recently, eigenvalue analysis techniques, such as principal components analysis (PCA) and singular value decomposition (SVD), to discover ocean climate indices. However, these techniques are only useful for finding a few of the strongest signals and, furthermore, impose a condition that all discovered signals must be orthogonal to each other. We have developed an alternative methodology for the discovery of OCIs that overcomes these limitations and is based on clusters that represent ocean regions with relatively homogeneous behavior. The centroids of these clusters are time series that summarize the behavior of these ocean areas. We divide the cluster centroids into several categories: those that correspond to known OCIs, those that are variants of known OCIs, and those that represent potentially new OCIs. The centroids that correspond to known OCIs provide a validation of our methodology, while some variants of known OCIs may provide better predictive power for some land areas. Finally, we show that, in some sense, our current cluster centroids are relatively complete, i.e., capture most of the possible candidate OCIs.