聖嬰現象對臺灣沿海水位波動的影響

Abstract

我們用Huang et al. (1998)所發明的新分析方法HHT (Hilbert-Huang Transform)，分析臺灣沿海的潮位資料，這方法可以用來分析非線性、非平穩性的資料。傳統上的分析方法只適用在線性系統，但實際的自然界資料通常為非線性的；潮位計大多設在岸邊淺海，所以量測的水位資料會受到地形和摩擦力的影響，其非線性的性質很明顯。HHT分析資料的第一個步驟，是用經驗模態分解法(Empirical Mode Decomposition method, 簡稱EMD)將資料分成有限個數的IMFs (Intrinsic Mode Functions) ，每個IMF代表不同時間尺度的訊號，但是因為有些尺度的訊號並不是連續的，所以會發生混模(mode mixing)的現象。Wu和Huang(2005)發表了EEMD的方法，改善原本EMD經常發生mode mixing的缺點。我們用EEMD分析臺灣水位，可以看到全日潮振幅都顯示約19年的默冬周期(Metonic cycle)變化，而富崗測站位於臺灣東岸，其年變化振幅和聖嬰現象有很大的關係，在聖嬰現象發生時年變化的振幅較大，其年變化的包絡線和聖嬰現象的指標(El Niño 3.4 index)的相關性可達到0.6。此外我們比較基隆、梗枋、富岡和高雄四測站的水位趨勢(trend)，發現從1991到2000年間水位都呈上升的趨勢。

We use the new Hilbert – Huang Transform (HHT) method, developed by Huang et al (1998), to analyze the coastal sea-level fluctuations in Taiwan. This new approach can be used to analyze non-linear and non-stationary data. Traditional analysis method like Fourier transform and Wavelet transform can only used in linear system, but in real world, the data are usually non-linear. Tide gauge usually located at near shore sites. So when we analyze the gauge data, we can not ignore its non-linear property. The key part of the HHT is the “ empirical mode decomposition “ method with which any complicated data set can be decomposed into a finite and often small number of “ intrinsic mode function “(IMF). Each IMF represents different driving mechanism. When the driving force appears intermittency, it will cause the decomposed IMF consisting of oscillations of dramatically disparate scales, hereafter called “mode mixing”. Wu and Huang (2005) presented a new Ensemble Empirical Mode Decomposition (EEMD). This new approach consists of sifting an ensemble of white noise-added signal and treats the mean as the final true result, thus prevents “mode mixing”. We analyze the sea level data in Fu-Kang Harbor, and find that the annual cycle oscillation in Fu-Kang is primarily affected by air pressure: When the air pressure is high in winter, the sea level is low. On the contrary, the air pressure drops in summer while the sea level rises. The amplitude of annual cycle increase during El Niño event. The cross-correlation between upper envelope of annual cycle and El Niño 3.4 index reaches 0.6 with -7 month lag time (envelope leads index). We also find that the sea level around Taiwan rises during the period 1991~2000.