考慮氣壓及水溫變化的潮汐調和分析

Abstract

台灣夏天及秋天常遭受颱風侵襲，且颱風帶來的強風及氣壓變化會導致海水面的異常變化，造成的水位異常的抬升，及海水溫度的改變亦會引起水位高低的變化，導致水位的上升或下降。傳統的調和分析方法只能分離出實測潮汐的天文潮，無法分離壓力及水溫造成潮汐的偏差量。因此，本文提出考慮壓力及溫度因素的調和分析，能分離實測潮位的天文潮、壓力及溫度影響成份的方法，並探討壓力及溫度對於潮汐的影響程度。本文選擇台中、高雄、花蓮及澎湖共四個測站，分析2005年至2007年的實測潮位、氣壓及溫度資料，以傳統調和分析法(HA)、加入氣壓因素的調和分析法(HAP)及加入氣壓與溫度因素的調和分析法(HAPT)進行後續的分析。 本文利用超量壓力差及超量溫度差提升與殘餘水位的相關係數，與壓力差及溫度差的相關係數相比，分別提升了0.157及0.087。接著經HA、HAP及HAPT取得殘餘水位，且隨著考慮條件增加，實測潮汐與天文潮的差值便依序下降。接著量化潮汐各成份，顯示天文潮佔潮汐約95%、壓力差水頭佔潮汐約1-2%及溫度影響水頭約1%，清楚瞭解潮汐各成份佔的比例。但殘餘水位仍佔潮汐約1-2%。經HA的殘餘水位、HAP及HAPT的壓力差水頭的比較，顯示HAP及HAPT的暴潮偏差趨勢相近，且較HA的暴潮偏差平滑。HAP及HAPT的最大暴潮偏差附近的水位上下振盪較HA的明顯。

Averaged three or four typhoons usually pass through or by Taiwan during summer and autumn. Strong winds and atmospheric pressure changes of typhoons cause abnormal variations of sea levels and the seawater temperature. The traditional harmonic analysis can only extract astronomical tides but cannot separate atmospheric tides and thermal tides from tidal data. Considering the variations of atmospheric pressure and seawater temperature the modified harmonic analysis method is proposed to improve the function of accurately computing storm surges and thermal tides. Data of tidal, atmospheric pressure and seawater temperature at four stations, which are Taichung, Kaohsiung, Hualien and Punghu, from 2005 to 2007, were collected to examine the validity of three models, that are HA, the traditional harmonic analysis, HAP considering atmospheric pressure only, and HAPT considering both atmospheric pressure and seawater temperature. The correlation coefficients between residual tides and threshold pressure differences or temperature differences are higher than those between residual tides and pressure differences or temperature differences by 0.157 and 0.087. Residual tides, defined the difference between measured tides and extracted tides, obtained by HA are higher than HAP and HAPT. Ratios of the potential of extracted astronomical, atmospheric and thermal tides are about 95%, 1-2% and 1%, respectively. The storm surges during the periods of typhoons by HAP and HAPT are close. Both are smoother than that by HA. The peak of each storm surge by HAP and HAPT is more obvious than that by HA.