Global Model Prediction Optimization Using Three DimensionalTruncation

使用三维截断的全局模型预测优化

• 批准号: 9002835
• 负责人: Ferdinand Baer
• 金额: $ 15.86万
• 依托单位: University of Maryland, College Park
• 依托单位国家: 美国
• 项目类别: Continuing Grant
• 财政年份: 1990
• 资助国家: 美国
• 起止时间: 1990-08-15 至 1993-01-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=9002835&HistoricalAwards=false
• 关键词: Global; Model; Prediction; Optimization; Using

Solutions to the atmospheric (or oceanic) prediction equations require an understanding of the three dimensional spatial nature of the fluid, including dynamics and physics, conservation conditions, and forcing. Since the equations are exceptionally complex and highly nonlinear, it is necessary to convert those equations to some numerical form which will provide meaningful iterative solutions within the constraints of available computational capabilities. The challenge is to do so optimally. While the problem has been investigated, because of the nature of the system, horizontal and vertical representations have been considered independently in the past. It is the contention of the principal investigator that systematic truncation of the atmosphere/ocean prediction equations in all three dimensions simultaneously could improve predictive skill. Thus, the purpose of this research is to study the correlation of truncation in all dimensions in a numerical model, and to find a truncation which is systematic in all three dimensions. The principal investigator will select a truncation according to scaling arguments that are based on the results from his previous studies and define appropriate functions. To establish the optimal three dimensional truncation, the functions will be tested in the operational numerical prediction model of NOAA's National Meteorological Center (NMC). This project is part of the NSF-NMC Program Joint Program in Numerical Weather Prediction.

大气（或海洋）预测的解决方案 方程需要理解三维 流体的空间性质，包括动力学和物理学， 保护条件和强迫。 因为方程是 异常复杂和高度非线性，有必要 将这些方程转换成某种数值形式， 有意义的迭代解决方案的约束下， 可用的计算能力。 挑战是这样做 最佳的。 虽然问题已经调查，但由于 系统的性质，水平和垂直表示 过去一直被独立考虑。 是 主要研究者的论点是， 大气/海洋预报方程的截断 三个维度同时进行可以提高预测能力。 因此，本研究的目的是研究 数值模型中所有维度的截断相关性， 并找到一个截断，这是系统的所有三个 尺寸. 主要研究者将选择截断 根据基于以下结果的缩放参数， 他以前的研究，并确定适当的功能。 到 建立最优三维截断， 将在业务数值预报模式中进行测试， NOAA国家气象中心（National Meteorological Center，NMC） 该项目是NSF-NMC计划联合计划的一部分， 数值天气预报