In the last decade, the technological changes in environmental data acquisition, management, analysis, and dissemination have been astounding. However, extracting patterns and knowledge for decision-making is still tedious essentially because the human capacity to comprehend such large amount of heterogonous data. Visualization tools are required to allow identifying relationships and patterns that are not evident from raw data. Several techniques have been proposed; most often they are borrowed from other fields such as human computer interaction (HCI) and recently bio-informatics and software visualization. This paper discusses some of the limitations of the existing visualization techniques while introducing the concept of visual mining. An illustrative exemplar from immersive visualization is given.