There is great need for novel statistical methods for the analysis of massive environmental data to address challenges that examine the interdependences between multiple processes that influence the biosphere. The statistical challenges within just climatology or atmospheric chemistry or oceanography have been a focus for intense activity; however, considering the atmospheric and oceanic sciences as a whole will allow us to examine the critical aspect of interdependencies between processes. For example, in order to understand climate fully, it is essential to understand weather, atmospheric chemistry, oceans, ice, and terrestrial processes, particularly with respect to the carbon and water cycles. Biological processes and human activities are both affected by and affect climate on local (urban heat islands), regional (land use patterns), and global scales (greenhouse gas emissions), such that a comprehensive study also requires hydrology, forestry, ecology, agriculture, economics and public policy. The development of appropriate statistical models for these processes remains a challenge because of the immense difficulty of accurately capturing dynamic aspects of spatio-temporal processes, especially nonlinear and non-stationary behavior. This session discusses common statistical themes across these major scientific problems and innovations developed to address these challenges.