| 摘要: | 本研究使用剖風儀與二維雨滴譜儀針對2008年西南氣流觀測實驗(SoWMEX/ TiMREX)期間發生在台灣南部的降水系統進行觀測研究，首先使用三種不同降水分類方法進行比較，基本上皆可以獲得不同降水型態的特徵；使用地面雨滴譜儀進行降水分類，可以發現2008年6月在西南部的觀測，層狀降水所占的時間比率為九成，對流降水則占一成左右；而使用Williams et al. (1995)提出的方法，則可以將降水系統的型態分類得更為仔細，其中混合降水所占的時間比率最大，值得注意的是當我們把混合降水與層狀降水合在一起時，其所占的比率也約為九成，對流降水占一成，與地面雨滴譜儀分類有一致的情況。 針對2008西南氣流實驗期間不同降水系統，去區分混合降水、層狀降水與對流降水，所做的垂直速度與回波統計分析，可以發現，層狀降水低層垂直落速，梅雨鋒系統略大於中尺度對流系統1m/s，而對流降水低層垂直落速則是中尺度對流系統略大1m/s；混合降水與層狀降水在高度4~5公里，垂直速度有明顯的速度梯度，回波有顯著躍升，此為融解層所造成的強回波；對流降水回波則隨高度的遞減而增加，並無亮帶的特徵出現。頻譜寬隨高度統計圖分析顯示，層狀降水在融解層上方的分布較為狹隘；混合降水則較為寬廣，顯示融解層上方可能有亂流的活動存在；對流降水的頻譜寬隨高度的曲線相較其他降水型態，則呈現底層窄而上層寬廣的形狀。 同一時間二維雨滴譜儀觀測結果，在梅雨鋒以及中尺度對流系統，降雨率小於50mm/hr情況下，中值體積直徑 似乎隨著降雨率的增大而增大，但當降雨率持續變大時， 似乎維持在2.1值左右，顯現雨滴的增長似乎有一定的限制。而在log10( )與降雨率R的關係中，降雨率較小時，其log10( )分布較廣，隨降雨率增大，其值也會隨之緩慢變大，一致性佳。在μ、Λ與降雨率R的關係中，降雨率小於10mm/hr時，μ、Λ值分布相當廣，但隨降雨率增大，μ、Λ值皆有減小的趨勢。而從μ與Λ的散布圖中，可知μ與Λ似乎有一正相關的關係存在。在午後對流方面，由於資料比數較少，且雨滴譜儀所觀測到的降雨率不大，所以各gamma參數與降雨率的關係不甚明顯，在降雨率小於10mm/hr時，μ、Λ值 分布相當廣，不過依然可看到μ與Λ似乎有一正相關的關係存在。 在混合及對流降水型態下，中尺度對流系統個案的平均雨滴粒徑分布比梅雨鋒個案大且多，而在層狀降水型態下，兩者則無太大差異，僅在大雨滴部分，梅雨鋒個案較多些。午後對流個案的平均雨滴粒徑分布最窄且雨滴濃度小，可能午後對流系統到達剖風儀與二維雨滴譜儀上空時，系統皆已開始消散，因此雨滴無法發展太大。 In this study, we used the wind profiler and 2D-Video disdrometer (2DVD) to observed different precipitating system during the Southwest Monsoon Experiment (SoWMEX/TiMREX). At first, we use three classification methods to classify precipitation type and basically we can get the same characteristics of different type of precipitation, no matter what method is used. When we used the classification algorithms from 2DVD, we could find that, the ratio of straitiform precipitation time is about 90%. However, the ratio of convective precipitation time is about 10%. And then we used the wind profiler to classify precipitating clouds into either stratiform, mixed stratiform/convective, deep convective, or shallow convective clouds by analyzing the vertical structure of Doppler vertical velocity and spectral width. If the mixed stratiform/convective classification and the convective classification are combined into one type, then we can also get similar result, the ratio of precipitation time is about 90%, and the convective precipitation time is about 90%. Using wind profiler to classify different precipitating system into either strati- form, mixed stratiform/convective, deep convective, or shallow convective clouds during the SoWMEX/TiMREX experiment, we can find that, the vertical velocity of stratiform precipitation of Mei-Yu front is greater than MCS about 1 m/s. But the vertical velocity of convective precipitation of Mei-Yu front is smaller than MCSs about 1 m/s. The CFADS (Contoured Frequency by Altitude Diagrams) of reflectivity and vertical velocity of mixed and stratiform types, it shows a distinct bright band and associated large vertical gradient in Doppler velocity at the altitude of 4~5 km. But for the convective rain, no bright band was found. The spectral width frequency distribution for the stratiform precipitating system shaws narrow distribution above melting layer. However, mixed type precipitating system had broad distribution of spectral widths above melting layer, this means turbulent motions are possibly occurring within the cloud. However the spectral width frequency distribution for the convective precipitating system, the shape is different from above. It appears narrow in the lower layer, but broad in the higher layer of atmosphere. At the same time, the observational results of 2DVD of Mei-Yu and MCS, when rainfall rate smaller than 50 mm/hr, will greater with rainfall rate. But when rainfall rate greater than 50 mm/hr, seems stop growing at the value 2.1, it shows drop size has some limitation. When rainfall rate is small, log10( ) is scattered. When rainfall rate greater, log10( ) will also larger. It seems the log10( ) and rainfall rate is much closed related in larger rainfall rate. The relationship between μ、Λ and rainfall rate, when rainfall rate smaller than 10 mm/hr, μ、Λ is scattered and broad. But when rainfall rate is greater, μ、Λ will also become smaller. From the μ、Λ scatter plot, we can see that μ、Λ has positive correlation. For the afternoon thundershowers, we also can see that when rainfall rate smaller than 10 mm/hr, μ、Λ are scattered and broad, and μ、Λ has positive correlation. For the rainfall of mixed and convective type, the average of drop size distribution of MCS is larger and greater than Mei-Yu front case. However there is no big DSD difference for stratiform type between Mei-Yu front and MCS, but Mei-yu front case has more big rain drop. For the afternoon thundershowers, the average of drop size distribution is the most narrow and N(D) is the smallest, the result may be due to the afternoon thundershowers systems already dissipated when they moved over the wind profiler and 2DVD site. |
