| 应用3种方法对降水在各尺度空间不均匀性的分离 |
| |
| 引用本文： | 肖蓓,房佳蓓,杨修群,任雪娟.应用3种方法对降水在各尺度空间不均匀性的分离[J].气象科学,2013,33(6):602-609. |
| |
| 作者姓名： | 肖蓓 房佳蓓 杨修群 任雪娟 |
| |
| 作者单位： | 南京大学 大气科学学院, 南京 210093;南京大学 大气科学学院, 南京 210093;南京大学 大气科学学院, 南京 210093;南京大学 大气科学学院, 南京 210093 |
| |
| 基金项目： | 国家重点基础研究发展计划（973计划）项目（2010CB428504）；江苏高校优势学科建设工程资助项目 |
| |
| 摘 要： | 利用滑动平均、傅里叶分解和小波分析3种方法，对基于COMRPH高分辨卫星降水资料的中国区域2004—2009年夏季平均降水场进行尺度分离，分析了1 000 km以上、500~1 000 km，200~500 km、64~200 km及64 km以下的5个尺度上的降水空间不均匀性，比较了3种方法在降水空间尺度分离上的优缺点和适用范围，并着重分析了长三角地区降水在各尺度空间的不均匀性及其成因。结果表明：空间尺度分离可以得到具有物理意义的降水非均匀分布特征，可以反映出大尺度海陆分布和季风环流、地形、洋流以及城市和城市群等不同尺度因子对降水的影响。空间滑动平均方法简单、易操作，适用性广泛，但是精确度较低。傅里叶分解能进行解析的分离，但是依赖于信号的平稳性，对于小尺度分离存在较大误差。小波分解则可以对非平稳信号进行处理，且具有较好的细节分辨能力。
|
| 关 键 词： | 降水 尺度分离 空间滑动平均 傅里叶分解 小波分析 |
| 收稿时间： | 2012/12/27 0:00:00 |
| 修稿时间： | 4/6/2013 12:00:00 AM |
Scale separation of spatial rainfall heterogeneity based on three methods |
| |
| XIAO Bei,FANG Jiabei,YANG Xiuqun and REN Xuejuan.Scale separation of spatial rainfall heterogeneity based on three methods[J].Scientia Meteorologica Sinica,2013,33(6):602-609. |
| |
| Authors: | XIAO Bei FANG Jiabei YANG Xiuqun and REN Xuejuan |
| |
| Institution: | School of Atmospheric Sciences, Nanjing University, Nanjing 210093, China;School of Atmospheric Sciences, Nanjing University, Nanjing 210093, China;School of Atmospheric Sciences, Nanjing University, Nanjing 210093, China;School of Atmospheric Sciences, Nanjing University, Nanjing 210093, China |
| |
| Abstract: | Based on CMORPH high-resolution satellite precipitation data, three methods including spatial moving average, Fourier transform and wavelet analysis are used respectively to separate the 6-a mean (04-09) summer precipitation over China region on different scales. The characteristics and possible mechanisms of spatial precipitation distribution on five scales (more than 1 000 km, 500-1 000 km, 200-500 km, 64-200 km, and less than 64 km) are investigated, especially over Yangtze River Delta region. The advantages and disadvantages of the three methods on scale separation as well as their application range are also analyzed. The results show that scale separation can obtain distribution features of spatial rainfall heterogeneity, based on physical reasoning, which reflect the influences of large-scale land-sea distribution, monsoon circulation, topography, oceanic current, city and city groups. The spatial moving average method is easier to operate, but is relatively less precise. Fourier transform can support full spectrum decomposition, but it strongly depends on the stationarity of the signal and generates considerable deviation for small-scale disturbance. While wavelet analysis can deal with nonstationary signal and has the ability to distinguish details. |
| |
| Keywords: | Rainfall Scale separation Spatial moving average Fourier transform Wavelet analysis |
|
| 点击此处可从《气象科学》浏览原始摘要信息 |
| 点击此处可从《气象科学》下载免费的PDF全文 |
|
