Range Queries for ESM Data Using Quadtree and Trapezoidal Map in Visualization System

Abstract

Climate modeling simulation plays a more and more important role in earth scientific research. For example, Earth System Model (ESM) can help scientists understand climate dynamics. As the output data of ESM is very large, scientists need visualization system to help them discover truth which is hidden in the massive datasets. When scientists find an interesting location on the visualization result, they need analytical tools to help them analysis the result precisely. Current visualization systems only have some simple analytical tools, but scientists need more useful and efficient analytical tools to satisfy their different analytical needs. We implement some analytical tools to solve this problem. These tools include rectangle range query and region range query which can be used to analysis large dataset of spatio-temporal datasets. To let rectangle range query be efficient to support real-time query, we use Quad tree to create a multi-resolution index file. Our experiments show that Quad tree is very efficient for rectangle range query, and that is because Quad tree can create multi-resolution index file which can reduce much I/O time. In addition to rectangle range query, we also provide tools to query an administrative region by using point location which can query the data of an administrative region like country or province.