The study of large-scale phenomena in the global climate system (such as ocean-atmosphere-land interactions, oscillations and teleconnections, etc.) is enabled through novel and innovative computational approaches. In particular, graph-based data representations and analysis methods from graph theory and network science are adapted for application to a variety of large climate datasets, both observed and model-generated. Broad research questions in this area include, among others, the identification and explanation of statistical relationships, potentially lagged in space (i.e., teleconnections) and/or time; characterization of dynamics and collective behavior from multiple interacting processes or systems; exploration of phenomena at multiple spatial and/or temporal scales and relationships or dependencies between them. Collectively, these efforts help develop an improved understanding of the global climate system with its many complex, dynamical processes and their interactions.