Application of Fuzzy Classification to the Structure of the Water Mass Distribution in the Easter Waters of Taiwan

Abstract

為探究台灣東部海域之溫鹽水團、營養鹽水團及葉綠素水團分布特性之相關性，本研究於2008年分4個季節採樣涵蓋22個測點，使用CTD的水文調查資料及水樣實驗，獲得水樣之溫度、鹽度、營養鹽及葉綠素數據，應用模糊分析以及交叉相關性分析法解析出台灣東部海域之溫鹽水團（T-S）、營養鹽水團（NP）及葉綠素水團（Chl a）型態之間的相關性及其分布結構。使用溫-鹽曲線圖對照水文型態分類圖後，不同季節的水文資料被劃分成不同類別的水團型態，另外使用群聚分析方法可以將水團大致分為3類，分別為黑潮水團、混合水水團及湧升水團其結果和溫鹽曲線圖對照之結果一致，且不同季節會有不同之水團特性。此外，應用模糊類別分析法將北緯122度的測線斷面圖所分析之水文資料，發現在溫鹽水團、營養鹽水團及葉綠素水團間有顯著相關性存在。當颱風過境時，相關性較正常時候更來的高。颱風經過海上時會形成冷渦流並形成湧升，帶來大量的營養鹽，大量的營養鹽會使得浮游植物大量生成。導致颱風路徑上，低溫、低鹽、高營養鹽的特殊情形出現。Four seasonal sampling cruises, covering a grid of 22 sampling stations, were carried out to collect physical and chemical oceanographic data in the eastern waters east of Taiwan in 2008. The aim of this study was to document the spatial and seasonal variabilities of water masses, as defined by temperature-salinity (T-S), nutrients, and chlorophyll (Chl a) using fuzzy classification. In order to determine relationships among T-S water mass types, nitrate and phosphate (NP) water mass types, and Chl a water mass types, this study used fuzzy classification methods and applied a cross-classification analysis. Hydrographic data were classified into three different water mass types in various seasons in a T-S diagram. The same results from a cluster analysis of each water mass type defined Kuroshio Current, mixed, and upwelling water types, which had different characteristics in the various season. Results of the fuzzy classification of the vertical transect hydrographic data of T-S, nutrients, and Chl a indicated that there were high correlations among T-S, nutrients, and Chl a. When a typhoon struck, the correlation was significantly higher. The typhoon caused a cold vortex to well up bringing with it abundant nutrients in the waters east of Taiwan, which caused the phytoplankton to increase.