以衛星測高資料探討台灣海域之外海與近岸波高之統計特性

Abstract

本文使用ERS-2、ENVISAT-1及JASON-1等三顆衛星的測高資料，將全台灣周圍海域依海底地形及距岸遠近分區，加以分析2002年至2008年間衛星測高資料在不同地區的波高分布結果，綜合三顆衛星的測高資料之每月波高進行分布套配以了解波高之最適分布，並以K-Means群集分析方法明確指出夏冬季節分類的月份。本文另外使用定點港口實測資料，推定各港口波高資料在空間上的分布情形，及夏冬季節分類之月份。 最後比較衛星測高資料與定點港口實測波高後發現，衛星測高資料約為定點實測波高之1~6倍，其中臺北港更有12倍差異。整體來說台灣周圍海域的衛星測高資料與定點實測波高並無明顯關係，且外海波浪資料比近岸較無一致性。本文藉此希望能建立台灣周圍外海各區域與其近岸之關係，若是能從廣域範圍之衛星測高資料得知近岸波浪分布之情形，則必能夠在未來對於海洋工程或是海象預測有一定程度之幫助。

The thesis is to investigate Wave-height statistics at offshore and nearshore waters around Taiwan using sea surface heights (SSH) observation of satellite altimeters. SSH data collected by ERS-2, ENVISAT-1 and JASON-1 during 2002 to 2008 were separated into monthly groups to fit possible distributions. K-Means method, which is a kind of cluster analysis, is used to group seasonal months of summer time and winter time. Probability fitting and K-means method are also applied to measured wave data at some observations established by the Harbor and Marine Technology Center. Comparison between SSHs and observed wave heights in separated zones show that SSHs are higher than observed wave heights in common. In Taipei Harbor, SSHs are 12 times than observed wave heights. SSHs at offshore and nearshore waters around Taiwan have low correlation coefficients with corresponding observed wave heights. The paper is a prior study to connect SSHs to observed wave heights at some particular points. In further researches more data can be carried out to develop solid connection between SSHs and observed wave heights for application of ocean engineering and wave forecasting.