透過海氣象浮標觀測資料探討西北太平洋之紊流現象

Abstract

颱風的經過會造成上層海洋溫度下降，形成冷尾跡 (cold wake)。引發此現象的主要機制包括紊流混合、Ekman pumping與海氣間的熱通量交換。然而，在這種極端的環境中觀測上層海洋是具有挑戰性的。為了更好地了解混合情況與其中的物理過程，需要進行更進一步的現場觀測與分析，以提高數值模式的預測能力。此外，中尺度渦旋也對當地背景條件和混合情況產生影響。臺灣的東南海域是颱風頻繁出現的區域，並且常有中尺度渦旋經過。因此，此處是研究颱風和中尺度渦旋期間，紊流混合情形的理想區域。本研究透過2018年與2022年在臺灣東南海域布放的海氣象浮標與都卜勒流速剖面儀 (Acoustic Doppler Current Profiler, ADCP) 錨碇觀測資料，分析颱風和中尺度冷渦旋期間的紊流混合情形。 研究中，使用單支與雙支溫度計計算溫度分層，兩種方法在非颱風期間的計算結果相近。然而，在颱風期間，強烈的外力作用攪拌下，海水混合較均勻，分層結構微弱時，兩種方法的結果會產生較明顯的差異。 兩次實驗總共觀測到三個颱風的經過。觀測結果顯示，山竹颱風期間，水深30公尺處的紊流動能消散率 (turbulent kinetic energy dissipation rate, ϵ) 大約為10-7 m2 s-3，溫度渦旋擴散係數 (eddy diffusivity of temperature, Κ_T) 大約為10-3 m2 s-1 。水深59公尺處的紊流動能消散率大約為10-7-10-6 m2 s-3，溫度渦旋擴散係數大約為10-4 m2 s-1。在分層結構持續微弱的環境下，水深30公尺處，理查森數 (Richardson number) 在機率分布中的峰值小於臨界值0.25，顯示了強烈的混合現象。此外，山竹颱風引起的紊流混合，在此處侷限於59公尺以上。軒嵐諾颱風期間，同樣觀察到Richardson number在機率分布中的峰值小於臨界值0.25。然而，當地的全日潮、半日潮與1/3日潮在呈現相同相位的情形下，內潮引起的垂直運動較強，下層冷水顯著抬升，水深20公尺處能夠觀察到分層結構變得較穩定，混合現象被抑制了大約7小時。接著，強度相對弱一些的梅花颱風期間，背景環境可能受到前一個颱風的影響，水深20公尺在持續微弱的分層結構中，流切控制了Richardson number的變化，使它在機率分布中的峰值接近臨界值0.25。而在水深75公尺處，軒嵐諾颱風和梅花颱風期間的Richardson number與平常時期相比，較無明顯變化，因此推測颱風引起的紊流混合在此處可能侷限於75公尺以上。 此外，兩次觀測實驗皆觀測到了一次中尺度冷渦旋的經過。2018年的水深30公尺處與2022年的水深20公尺處，能觀察到冷渦旋經過期間，分層結構變得較穩定，Richardson number增加，抑制了紊流混合。然而，2022年的冷渦旋離開後，75公尺深的Richardson number有增加的趨勢。透過垂直溫度梯度的變化可以推論，當冷渦旋離開當地，溫躍層的加深使此處的海水更趨向穩定，減少了紊流混合的發生。

Strong winds from typhoons decrease sea surface temperature, and this cooling area is called the cold wake. It is well known that the primary mechanisms causing this phenomenon include turbulent mixing, Ekman pumping, and surface heat fluxes. However, observing the upper ocean in such an extreme environment is challenging. More observations of turbulent mixing are needed to understand the physical processes involved and improve numerical model predictions. In addition, the passage of an eddy also affects background conditions and turbulent mixing. The northwestern subtropical Pacific Ocean is an area where typhoons are prevalent, and eddies often pass through it. Therefore, it is a suitable area to study turbulent mixing when typhoons and eddies occur. In this study, data were obtained from the surface buoy and the ADCP subsurface moorings in 2018 and 2022. We estimated the stratification using single and dual thermistors. The two methods yielded similar results during non-typhoon periods. However, during typhoon periods, the strong winds caused the seawater to become uniformly mixed. The weak stratification resulted in some discrepancies between the two sets of results. During the two experiments, three typhoons were observed. The observations during Typhoon Mangkhut showed that the TKE dissipation rate at 30-m depth was approximately 10-7 m2 s-3, and the eddy diffusivity was around 10-3 m2 s-1. At 59-m depth, the TKE dissipation rate ranged from 10-7 to 10-6 m2 s-3, and the eddy diffusivity was approximately 10-4 m2 s-1. Under sustained weak stratification, the Richardson number at 30-m depth had its peak in the probability distribution below the critical value of 0.25, indicating strong mixing. Additionally, the turbulent mixing induced by Typhoon Mangkhut occurred at depths shallower than 59 meters. During Typhoon Hinnamnor, the peak of the Richardson number in the probability distribution was also below 0.25. However, when the diurnal tide, semi-diurnal tide, and terdiurnal tide had the same phase, the vertical motion of seawater caused by the internal tides was stronger. This resulted in a more stable stratification at 20-m depth, suppressing mixing for approximately 7 hours. Subsequently, during Typhoon Muifa, which was relatively weak, the background environment might have been influenced by the preceding typhoon. In the sustained weak stratification at 20-m depth, shear dominated the variation of the Richardson number, and its peak in the probability distribution was close to 0.25. At 75-m depth, the Richardson number did not significantly change when Typhoon Hinnamnor and Typhoon Muifa passed by, suggesting that the turbulent mixing induced by these two typhoons might occur above 75-m depth. Furthermore, both experiments observed the passage of a mesoscale cold eddy. During the cold eddy period at 30-m depth in 2018 and 20-m depth in 2022, the stratification became more stable, resulting in an increase in the Richardson number, which suppressed turbulent mixing. Additionally, after the cold eddy period in 2022, Richardson numbers increased at 75-m depth. Based on the variations in vertical temperature gradients, it can be inferred that when the cold eddy left the area, the deepening of the thermocline intensified the stratification, reducing the occurrence of turbulent mixing.