气象信息科学视算环境及其若干问题

多维动态气象数据的科学视算环境通过多视角的图形、图像表达方式 ,能够直接、准确地展示复杂多样的天气过程及其时空演变 ,辅助解释人员和预报人员形成天气系统的四维空间 /时间结构 ,已成为天气系统分析的重要工具。文中首先讨论了描述天气系统和天气过程的常用图形、图像表达方式 ,然后通过若干典型天气系统数值模拟结果的示例分析 ,介绍了目前气象信息科学视算环境的主要功能 ,最后结合当前的应用状况和气象科学发展的需求 ,讨论了气象信息科学视算环境中的若干问题及其解决方案 ,并展望了未来气象信息科学视算环境的发展前景。

METEOROLOGICAL DATA VISUALIZATION SYSTEM

Meteorological data presents a subst antial challenge for scientific visualization application, because of the very nature of the data itself. The visualization of meteorological data is the last element in the pipeline of meteorological data processing. It enables the forecaster or the researcher to understand the complex circumstances by representing the spatial distributions and temporal evolution of weather situation, and allows scientists to extract knowledge efficiently from rapidly expanding multidimensional datasets. With the amount of data increasing exponentially, a fundamental question is "What is the most efficient and most easily underst andable way to examine these datasets?" It is difficult to use traditional methods designed for two dimensional data sets to extractscientific information from these multidimensional data sets. In this paper, a summarization of graphical representations of multidimensional data sets in meteorological area was presented, a powerful meteorological datavisualization system running on PC-Windows operating system was introduced, several typical weather (cyclone, typhoon, etc.) analysis of mult idimensional data sets generated by numerical model were used to demonstrate some of the capability of this system. Although the visualization systems have played an important role in the understanding of complex weather situations, but there are still many limitations in current visualization system. For example, too many complex parameter adjusting are discom modious for the forecaster, can't support the experience or knowledge of the forecaster which formed in past analysis of weather situations (viz.intelligent analysis), there is a lack of ability to sumup the conceptive models of special weather situation which are very important in practice, and can't have access to the fullrange of meteorological data, etc. At the end of this paper, these limitations, and expectations in improving were indicated.