Exploratory 3D Geovisual Analytics

摘要

In this paper, we extend our generic "GeoAnalytics" visualization (GAV) component toolkit, based on the principles behind Visual Analytics (VA), to also support time-oriented, geographically referenced and multivariate attribute volumetric data. GAV includes components that support a mixture of technologies from the three data visualization fields: information visualization (InfoVis), geovisualization (GeoVis) and scientific visualization (SciVis). Our research concentrates on visual user interface (VUI) techniques through dynamic and direct data manipulation that permit the visual analytical process to become more interactive and focused. This paper encourages synergies between well-known information- and volume data visualization methods applied in a multiple-linked and coordinated views interface. We address challenges for improved data interaction techniques with volumetric data and the need for immediate response. Varieties of explorative data analysis (EDA) tasks and the possibility to view the information simultaneously from different perspectives and scenarios are discussed. The effectiveness of our geovisual analytics framework is demonstrated in a tailor-made volume data explorer (VDE) application that integrates InfoVis, GeoVis and SciVis visualization methods assembled from GAV components. VDE facilitates dynamic exploration and correlation of temporal ocean space temperature and salinity data supplied in a NetCDF format from NOAA. This real-world phenomenon that corresponds to a huge volumetric data set comprises more than 31 million values for a time period of 12 months in 1994.