摘要
基于中国海洋大学卫星地面站接收到的NOAA系列卫星数据,利用1种神经网络反演方法获取大气温度、湿度及风矢量垂直廓线数据。以2004年第14号台风"云娜"为例,在中尺度气象模式MM5中运用松弛逼近法对这些反演数据进行了一系列同化试验。结果表明:仅同化温度与湿度的效果比同时同化风矢量、温度与湿度的效果好;台风的路径与强度有所改善,不采用人造涡旋方案时同化效果相对采用人造涡旋方案时更能体现出改进作用。
Based on the NOAA satellite data received by the satellite ground station of Ocean University of China, vertical profiles of wind, temperature and humidities are retrieved by a neural network ap- proach. Typhoon Rananim（No. 14 in 2004） is taken as an example to ingest these retrieval data, and then a series of assimilation experiments are carried out using the MM5 model with its built-in observation nud- ging method. The results show that the effect of assimilation of retrieved temperatures and humidities is better than that of assimilation of both temperatures, humidities and winds, and that the forecast of ty- phoon track and intensity is a bit improved but this improvement is relatively more obvious in the numeri- cal simulation without bogussing scheme than that with bogussing scheme.
出处
《中国海洋大学学报（自然科学版）》
CAS
CSCD
北大核心
2010年第2期99-104,共6页
Periodical of Ocean University of China
基金
国家自然科学基金项目(40706004)
国家高技术研究计划项目(2005CB422301)资助
关键词
ATOVS反演数据
垂直廓线
MM5
台风
同化试验
ATOVS retrieval data
vertical profiles of temperature and humidity
MM5
typhoon
assimilation experiments
