Using GLIDER for Knowledge Discovery in Climate Science to Visualize, Analyze and Mine Satellite Imagery

Satellite remote sensing provides an essential source of information for monitoring the earth´s atmosphere and climate. Thematic information derived from satellite imagery is applicable to a wide variety of applications at local, regional and global scales. Extracting thematic information using satellite imagery is a non-trivial task that usually requires preprocessing data by applying operations for radiometric and geometric corrections. Analysts need the ability to interactively apply different image enhancement operations to digitally enhance imagery, in order to identify subtle information that might be otherwise missed. Advanced algorithms are needed to allow users to analyze imagery and extract thematic information without having to alter the original data through geographic projections. The Globally Leveraged Integrated Data Explorer for Research (GLIDER) is an easy to use tool for visualization, analysis and mining of satellite imagery. Users can visualize and analyze satellite imagery in its native sensor view, which is an important capability since geographic projections can result in loss of information. A full suite of image processing, pattern recognition and data mining algorithms is provided for image enhancement and analysis. Analysis results and imagery can then be displayed simultaneously and compared using a 3-D globe display that supports image overlays and Web map service (WMS) layers. Examples will be shown to illustrate how GLIDER provides views and tools to facilitate knowledge discovery from satellite imagery.