论文部分内容阅读
由一元线性回归分析的正则方程计算表明，在验潮记录较短的客观条件下，高频扰动对确定海平面变化趋势的影响是显著的，其中影响最显著的高频带是ｆ≥０．０２０８ｃｐｍ（ｃｉｒｃｌｅｐｅｒｍｏｎｔｈ），即对应周期Ｔ小于４年的高频扰动．要从短序列中确定海平面的变化趋势，必须消除周期小于４年的高频扰动的影响．本文探讨用傅氏变换方法来消除高频扰动的影响，由低通序列一元线性回归分析就能从短序列中准确计算海平面的变化趋势．
The regular equation calculated by the univariate linear regression analysis shows that under the objective conditions with short tide records, the influence of high-frequency disturbance on the determination of sea-level change trend is significant. The most significant high-frequency band is f≥0. 0208cpm (circlepermonth), that is, the corresponding period T is less than 4 years of high-frequency disturbance. To determine the trend of sea level changes from short sequences, the effect of high-frequency disturbances less than 4 years in duration must be eliminated. In this paper, the influence of high-frequency disturbance is eliminated by Fourier transform, and the change trend of sea level can be accurately calculated from the short sequence by the univariate linear regression analysis of low-pass sequence.