極端降水事件：物理與預測

Abstract

台北都會區的強降雨變得越來越頻繁和強烈。本研究的目的包括：(1)利用逐10分鐘降雨觀測，探討台灣北部暴雨事件(>40 mm h-1)的特徵；(2)藉由偏極化雷達參數：差異反射率(differential reflectivity, ZDR)和比相位差(specific differential phase, KDP)柱狀結構的演化，解釋暴雨期間上升氣流增強的機制及相關微物理特徵；(3)利用水凝物識別(hydrometeor identification, HID)算法和地面雨滴尺寸分佈(raindrop size distribution, RSD)總結台北盆地暴雨的整體微物理過程；(4)建立台灣北部極端降水事件期間偏極化參數極值與劇烈降雨率之間的關係。 逐10分鐘的降雨觀測有利於細分暴雨發生時段並作進一步的應用。統計期間北台灣發生上百次最大時雨量超過40 mm的大雨事件，而40 mm h-1 約為觀測降雨率的第 99.5 百分位。 最極端降雨事件為2015年6月14日的午後雷暴 (afternoon thunderstorm, TSA)，並導致市區內淹水；其最大降雨率131.5 mm h-1超過第99.9百分位，且有較長的生命期。當日降雨分佈與長期觀測特徵相似，但相較TSA日的平均值，海風更有效輸送水汽、且降雨期間溫度下降得更快。 是次雷暴演變以胞合併特徵區分為兩階段：組織期(organizing stage, OG)和豪雨期(heavy rain stage, HR)。OG期間，較淺的孤立對流胞以大雨滴為特徵，隨後產生> 10 mm h-1但持續時間短的陣雨。在最大降雨率(20 分鐘內超過 60 mm)前發生對流胞合併：ZDR和KDP的柱狀特徵在合併胞中變得更加明顯，這與向上運動和混合相水凝物的廣泛分佈相關。低於環境 0 ºC高度的KDP 向地面增加，與融化霰有關並導致HR期間的強降雨率。由於碰撞-破碎(collision–breakup)過程，幾乎所有粒徑雨滴的最高濃度和較小的平均粒徑都出現在最大降雨階段。顯示冰相和暖雨過程對此極端降水都有貢獻：融化霰或雹為主要降雨源且在胞合併後顯著增加；粒徑淘選(size sorting)或碰撞過程則形塑RSD並影響降雨率。 分析3小時累積降雨量 > 150/200 mm的強降雨事件，指出近地面偏極化參數變化與降雨時序高度相關；但環境 0°C高度以上出現偏極化參數的極端值時，則可能是地面暴雨的早期指標。融解層高度以上的廣泛 ZDR >1.0 dB 分佈意味著更猛烈的向上運動以支持更大的降水顆粒；經由粒徑淘選過程而在近地面被觀測到的大雨滴暗示鄰近的上升運動與隨後增強的降雨率。但ZDR特徵變化在不同天氣系統則有顯著差異。當融解層高度以上的 KDP > 0.75° km-1的體積增加時，隨後在地面附近的分佈通常也會增加，這與冰粒子融化為雨滴的過程有關。若考慮融解層上方之KDP變化，對暴雨預警時間最佳能提前20分鐘。此方案適用於深對流系統，然而對於冰相過程不顯著的天氣系統，更有效的暴雨前導指標仍待研究。

Severe rainfall has become increasingly frequent and intense in the Taipei metropolitan area. The aims of this study include: (1) To explore the characteristics of heavy rain events (>40 mm h-1) in northern Taiwan using 10-minute rainfall observation; (2) To explain the mechanism of updraft enhancement and related microphysical features through the evolution of differential reflectivity (ZDR) and specific differential phase (KDP) columns during heavy rainfall; (3) To summarize the overall microphysical process of heavy rain in the Taipei Basin with the hydrometeor identification (HID) algorithm and ground raindrop size distribution (RSD); (4) To establish the relationship between the extreme values of polarimetric variables and the high rain rate during the extreme precipitation events in northern Taiwan. 10-minute rainfall observation benefits segment a period of heavy rain and further applications. Hundreds of heavy rain events with maximum hourly rainfall exceeding 40 mm were identified in north Taiwan, and 40 mm h-1 is about the 99.5th percentile of the observed rainfall rate. A complex afternoon thunderstorm (TSA) in the Taipei Basin on 14 June 2015 produced an extreme rain rate (131.5 mm h−1), leading to an urban flash flood. On this day, the spatial distribution of heavy rain was similar to the general long-term observations. However, the sea breeze transported the moisture more efficiently than average TSA days, and the temperature decreased more rapidly during the rainfall. The evolution of the convective storm on 14 June 2015 was divided into two stages based on the process of cell merger: the organizing stage (OG) and the heavy rain stage (HR). During the OG, shallow isolated convective cells were characterized by large raindrops, followed by >10 mm h-1 but short-duration showers. The storm's evolution highlighted the behavior of merged convective cells before the heaviest rainfall (exceeding 60 mm within 20 minutes). The columnar features of ZDR and KDP became more evident in merged cells, which correlated with the broad distribution of upward motion and mix-phased hydrometeor. The KDP below the environmental 0 ºC level increased toward the ground associated with the melted graupel, resulting in subsequent intense rain rates during the HR. Due to the collision–breakup process, the highest concentrations of almost all drop sizes and smaller mass-weighted mean diameter occurred during the maximum rainfall stage. Both the ice phase and warm rain process contributed to the extreme precipitation: melting graupel or hail was the main rainfall source and increased significantly after cell merger; size sorting or collision process shaped RSD and affected the rainfall rate. Severe rain events of 3-hour cumulative rainfall > 150/200 mm are analyzed and revealed that the variation of polarimetric variables near the ground highly correlates with the rainfall. However, the appearance of extreme polarimetric variables above the environmental 0 °C level might indicate heavy rain on the ground early. A large ZDR >1.0 dB volume above the melting level means a more violent upward motion to support larger precipitation particles; large raindrops observed near the surface via the size sorting imply the adjacent upward motion with subsequent increased rain rate. However, there are significant differences in the ZDR characteristics in different weather systems. When the volume of KDP > 0.75° km-1 above the melting level increases, soon after, its distribution near the ground usually also increases; which is related to the process of melting ice particles into raindrops. It causes more vital rainfall rates, and there is a chance to advance the lead time to 20 minutes before the surface rainfall. This scheme is suitable for deep convective systems, but the effective leading indicator for the weather system with an inactive ice process is still under investigation.