中文摘要：

利用地面和梯度塔的风连续观测数据进行的频谱分析已经开展很多,而受资料限制,高空风的频谱分析仍较欠缺。本文使用风廓线雷达获取的长时间序列连续测风数据,运用傅里叶变换的方法,计算了风的脉动谱密度。脉动谱能够反映不同频率的风速涨落对风动能的贡献。使用2012年4月江西宜春前汛期期间的高空风连续数据,结合地面降水资料进行了1 000~3 000 m高度区间的频谱分析,发现地处前汛期雨带上的宜春地区降水存在着两种不同时间周期的天气系统影响,脉动谱的分布表现出时间周期为5~7 d和2~3 d的峰值区。分别对两种不同时间周期的天气系统频谱进行了分析,并与平稳天气时的频谱进行比较。5~7 d周期峰区的脉动谱密度数值为2~3 d的4~5倍,脉动谱峰区在2 000~3 000 m高度上较强,峰值强度向下迅速降低;2~3 d周期的脉动谱峰区在低层比较明显,峰值强度较弱。风的脉动谱分布与地面降水的时间周期较为吻合。

英文摘要：

The first rainy season in southern China（ from April to June） accounts for approximately 40% to 50% of the total yearly precipitation.Therefore ,it is significant to conduct some research regarding the rainfall during the first rainy season.The precipitation is closely related to the changes in the winds, and wind spectrum distribution is considerably connected with the period of the weather system.Many previous studies regarding spectrum analysis using meteorological tower measurements have been carried out. However, due to the absence of wind measure- ments, there have been found to be several limitations for the spectrum analysis of the boundary layer.By applying Wind Profile Radar in atmospheric soundings, it has been possible to obtain high-altitude continuous wind data. Therefore, research studies regarding the spectrum based on high-reliability wind profile radar measurements at al- titudes from I00 to 3 000 m,have been completed.The turbulence spectrum density was calculated by the wind profiling radar data at the 1 000 to 3 000 m level during rainy season（April,2012） at Yichun based on the FFT. This spectral analysis provided a new method for weather forecasting. The Yichun boundary layer wind profile was detected by five wave beams,with sounding levels from 100 to 5 980 m,and the sampling interval was set at six minutes.The wind profile radar provided the horizontal windspeed data and signal-to-noise ratio （SNR） , echo intensity, spectral width, and so on.The data were obtained from the period ranging from April 1st to May 5th,2012,and a quality control was applied. The average wind is usually treated as the mean wind of the data collected during a short period.However, the average wind of data collected during a long period tends to change with time, which is called a trend. The fluctuation of the wind is obtained by using a least-squares curve fitting to calculate the trend, and then this is deducted from the wind data. In this study, a spectral analysis was carried out with the vertical w