USC.GeoDA: Effective Visualization and Decision-Support for On-the-fly Retrieval and Fusion of 5-D Earth Science Data
Currently, due to the high time-complexity of the data retrieval and fusion processes, the 5-D earth science data can only be fused in offline, with which a limited set of predefined combinations of the dataproducts are integrated in advance to prepare for an a priori known visualization scenarios; hence, offline visualization. In contrast, we use our fast and efficient wavelet-based pre-computation techniques to reduce the time-complexity of the data retrieval and fusion. Consequently, we can perform online visualization with which we can choose the combinations and the resolutions of the data products to be visualized on-the-fly. This capability significantly improves the flexibility and effectiveness of the visualization and decision-support earth science applications.
We are building USC.GeoDA (for Geospatial Data Analysis) by integrating and extending the mutually complementing systems, ProDA and GeoDec, into a unified online data retrieval and fusion framework for visualization of the 5-D earth science data (see Figure 1). ProDA and GeoDec are both built as 3-tier architectures. In each system, the functionality of a tier (i.e., the bottom, middle, or top tier) corresponds to that of the same tier at the other system. We integrate these systems by merging the corresponding tiers of the two systems. Since both systems benefit from a web-service infrastructure, the implementation is expected to be straightforward but certain overlap or missing functionalities need to be accounted for.
Figure 1. USC.GeoDA