中亚区域快速更新多尺度分析和预报系统研发进展

Advances in the Central Asia Rapid-Update Multi-scale Analysis and Forecast System

摘要:

数值天气预报是保障国计民生和国防安全的核心科技的重要组成部分。为了提升“一带一路”沿线地区的气象服务水平，中国气象局乌鲁木齐沙漠气象研究所自2017年起开始推进中亚区域快速更新多尺度分析和预报系统RMAPS-CA的研发工作，并在多源资料联合同化技术、参数化方案优化和动力框架调整方面取得了一定的研究成果。在回顾RMAPS-CA业务研发历程的基础上，重点对雷达反射率同化、高空次网格地形重力波拖曳与近地层阻塞拖曳分离的次网格物理过程、混合地形追随坐标动力框架以及基于贝叶斯模式平均（Bayesian model averaging, BMA）方法订正预报产品等方面所取得的研究成果进行了综述。针对中亚地区常规观测稀疏且沙漠下垫面范围广等特点，从卫星地面通道同化等方面提出了加强中亚区域数值天气预报业务研发及应用的建议。

Abstract:

Numerical weather prediction is an essential component of core technologies that safeguard national economy, people's livelihood, and national defense security. In order to enhance the meteorological service level along the "Belt and Road" regions, the Urumqi Desert Meteorological Research Institute of the China Meteorological Administration has been advancing the development of the Central Asia Regional Rapid Update Multi-scale Analysis and Forecast System (RMAPS-CA) since 2017. Significant research achievements have been made in the joint assimilation of multi-source data, optimization of parameterization schemes, and adjustment of the dynamic framework. Based on a review of the operational development history of RMAPS-CA, this paper focuses on summarizing the research achievements in the assimilation of radar reflectivity, the separation of sub-grid physical processes such as gravity wave drag due to high-altitude sub-grid terrain and near-surface blocking drag, the hybrid terrain-following coordinate dynamic framework, and the correction of forecast products based on the Bayesian model averaging (BMA) method. In response to the characteristics of sparse conventional observations and extensive desert surfaces in Central Asia, suggestions are proposed to strengthen the operational development and application of numerical weather prediction in the Central Asia region, especially on the assimilation of satellite ground channels.