The analysis of precipitation variability and characteristics in Taiwan associated with climate changes.

• Main Authors: Bao-chun Tang, 湯寶君
• Other Authors: 隋中興
• Format: Others
• Language: zh-TW
• Published: 2010
• Online Access: http://ndltd.ncl.edu.tw/handle/05296562886926871426

碩士 === 國立中央大學 === 大氣物理研究所 === 98 === The aim of this study is discussing the precipitation variability and characteristics in Taiwan associated with inter-decadal and long-term trend climate changes. We analyze the 58 years (1951 to 2008) rainfall data temporally and spatially at 20 stations. When analyzing the precipitation characteristics and variability, we use the adaptive method-HHT (Hilbert-Huang transform) to decompose the time series of rainfall variability to many sub-series and trend, focusing on the low frequency component and trend to analyze and discuss. Furthermore, we intend to clarify the decadal and inter-decadal variability and characteristics of rainfall in Taiwan, ex: time series variability and spatial distribution…etc. The results show that the low frequency component of yearly cumulative rainfall, warm season rainfall and yearly extreme rainfall, their eigenvectors of the first EOF display an island-wide synchronizing mode. However, the eigenvectors of the second EOF display the northeast-southwest dipole rainfall pattern. Reconstructed rainfall variability based on the two leading EOF modes is consistent with that of individual station analysis. Both EOF analysis and individual station analysis show that after 2000, warm season rainfall increase much more significantly than yearly cumulative rainfall. The increasing trend after 2000 is more evident in extreme rainfall, such that the more extreme, the more significantly rainfall increase and the more difference between east and west in Taiwan. Both EOF and individual station analyses show strong rainfall before 1955 and after 2005, weak rainfall during 1960-1965, 1980 and 1995. We further perform a regression analysis of rainfall against the warming trend,MDV and PDO these three climate variability and trend signals that temperature signals display. Based on the results, we calculate the rainfall change rate influenced by different temperature signals. Overall, warming trend together with MDV can largely explain the low-frequency in yearly rainfall, warm season rainfall and the extreme rainfall before 1955 and after 2005.During the middle time, the offsets between warming trend and MDV make the little change in rainfall. Approximately 1985-1990, warming trend, MDV and PDO, all of them are positive contributions to Taiwan rainfall, therefore rainfall began to increase dramatically, the more extreme rainfall is, the more influenced by these temperature signals.