2005年梅雨季6月12號台灣西南部豪雨之分析

蘇炯瑞(Jyong-Ruei Su)

指導教授 ： 陳景森

國立中央大學/地球科學學院/大氣物理研究所/碩士(2007年)

摘要

台灣受東亞季風影響，在梅雨季(5月中～6月中)適當的綜觀條件下，大雨(時雨率>15mm而且日雨量>50mm )受西南季風及地形的影響，大部分發生於台灣西南側中央山脈的迎風面，而背風處的東部及東南部降雨較少。顯示台灣梅雨季地形效應對於較大雨量之分佈扮演很重要的角色(Chen C.-S. et al., 2007b)。
根據氣象局傳統測站過去30年雨量資料顯示，有17年梅雨季雨量大於30年平均雨量，其中有6年5月下旬的雨量大於6月上旬之雨量(81、83、84、86、88、01)，有11年(77、79、85、90、93、95、97、98、03、05、06) 6月上旬比5月下旬大。因此梅雨季在較多雨之年，6月上旬雨量較大。97、98、05、06年6月上旬雨量較多，但05年雨量在台灣西南部特別大，在雪山山脈附近的雨量相對較少，因此檢討這4年6月上旬的大氣平均狀態的差異性。97、98和06年在鄂霍次克海有明顯的阻塞高壓，05年則無。98及06年在貝加爾湖附近有滯留性的低壓中心，但05年則無。97、98和06年由滯留性低壓的低壓中心向南延伸的東亞槽延伸到30 以南靠近台灣北部，有利斜壓系統影響台灣，但05年東亞槽在30 以北，並且偏向東海東部，斜壓系統較不易影響台灣。97、98和06年太平洋高壓經菲律賓延伸到中南半島，較不利熱帶系統發展，但05年太平洋高壓在台灣及菲律賓以東。而05年南海北部的短槽相當明顯，其他3年則無，顯示05年受南海天氣系統影響很大。2005年台灣在西南部的主要豪雨日(台灣西南部至少有5個雨量測站大於130mm/day，而且至少有1個雨量測站大於200mm/day)發生在12～15號之間。這四天在西南部的累積雨量佔05年西南部梅雨季總雨量的50％以上。12號西南部之降雨特性為降雨極值發生在早上8點和下午3點，和其他三天(13、14、15日)不同，6月13號在西南部的降雨日變化早上有一個極值發生在08LST，但午後沒有明顯的極值；14號06LST有一個極值，另一個較大極值在傍晚21LST；15號的極值與12號相似，極值在05LST及15LST，但午後極值比清晨大。又12號超大豪雨的測站數是12~15日這四天中最大，因此本篇論文著重於研究6月12號豪雨發生降雨的機制。
分析EC(European Center for Medium-Range Weather Forecasts)資料、JMA (Japan Meteorological Agency)日本天氣圖、衛星雲圖及OLR(Interpolated Outgoing Longwave Radiation)外逸長波幅射(Outgoing Longwave Radiation)得知，12號的鋒面在台灣北部近海。8號在南海西部有熱帶擾動生成，隨時間向東移動，在11號18LST移至南海東北部，在12號接觸台灣西南部，影響台灣西南部降雨。豪雨主要受南海對流系統的影響。
分析雷達回波得知，清晨台灣西南部沿海地區有強回波生成，推論沿海地面有東南風和海上盛行西南風輻合。在925百帕有受地形阻擋轉向的南風和盛行風輻合，使得降雨發生在沿海和平地地區。另外由南海東北部及近海降雨系統的移入台灣西南部，使得西南部降雨極值發生時間由沿海向內陸延遲。為了研究西南部降雨雙峰的特性以及更進一步瞭解沿岸及斜坡地區發生豪雨之機制，使用天氣預報模式(Weather Research and Forecasting Model, WRF)，來研究南海降雨系統進入陸地之後受地形影響如何造成沿岸和山坡地區之豪雨。
模擬結果顯示在清晨斜坡及平地的降雨使虛位溫降低，造成離岸風，和盛行西南風輻合，使沿岸輻合增強。另外平地地區由於西南氣流受地形影響，減速轉向形成南風和西南風在平地輻合。這兩種因素皆增強平地與沿岸的降雨。但在午後，沿海輻合較不明顯。由近海移入的對流系統加強西南部的降雨，但雨量比清晨小。模擬結果驗證觀測沿海及平地降雨的推論。但日雨量分佈較大的降雨區屏東平地到斜坡地區，雖和觀測的雨區一致，但模擬日雨量(約350mm)比觀測值(508mm)小。由於模擬槽的南海東北部的短槽移速較觀測分析的短槽快且強度較弱，沿海地區的近海模擬降雨移入系統較弱，模擬雨量比觀測要小，但影響台灣的時間比觀測早。模擬雨量極值發生時間在05LST及12LST比觀測提早2～3個小時。另外，無地形測試顯示地形造成盛行風減速輻合的效應對於平地之降雨作用相當顯著。

關鍵字

none

並列摘要

During southeast monsoon, heavy rainfall ( hourly rainfall rate > 15mm and daily rainfall rate > 50mm ) occur frequently in Taiwan Mei-Yu season(mid-May to mid-June) the windward side of Central Mountain Range (CMR) in southwest of Taiwan. Over lee side of CMR in the eastern and southeast of Taiwan heavy rainfall are less than the windward side of CMR. The distribution of heavy rainfall was affected by orographic effect significantly (Chen C.-S. et. al 2007b).
The pass 30 (1977~2006) years rainfall data exhibited there were 17 years rainfall greater than the average of 30 years. Among these 17 years, there were 6 years the rainfall during late May (15-31 May) greater than early June (1-15 June) (81,83,84,86,88 and 01). However, there were 11 years the rainfall during early June greater than late May (77,79,85,90,93,95,97,98,03,05 and 06). In the years with above average rainfall, rainfall is large in early June. In early June of the last 10 years (97~06), the rainfall in 97, 98, 05 and 06 was significantly. But in 05, rainfall in southwest of Taiwan was particularly large and rainfall was relatively less in northern part of central Taiwan. This fact was differ from other 3 years (97,98 and 06). The difference of these 4 years is evident. There was a blocking high in Okhotsk in 97, 98 and 06, but the blocking high disappeared in 05. The stationary low near the Lake Baikal in 98 and 06, also disappeared in 05. The East Asia trough stretched southward from the center of stationary low to 30 N near north of Taiwan, favorably for the baroclinic systems influencing Taiwan. But in 05, the East Asia trough was to the north of 30 N and stretched south to east of the East China Sea. As a result the baroclinic systems hardly influenced Taiwan. In 97, 98 and 06, the west Pacific subtropical high stretched over Philippines to Indochina. The environment was unfavorable for the development of tropic systems in north South China Sea. In addition, the trough in the north South China Sea was significantly in 05, but disappeared in the other 3 years. It show that the rainfall in 05 influenced by the weather system in South China Sea. Heavy rainfall day (at the least 5 stations the daily rainfall greater then 130mm, and at the least 1 station the daily rainfall greater then 200mm) in southwestern Taiwan occurred during 12~15 June 2005. The total rainfall of these 4 days contributed more than a half of rainfall accumulation of Mei-Yu season in 2005. Characteristic of the precipitation had a pair of rainfall peaks happened in 08LST and 15LST in southwestern Taiwan on 12 June. This is differ from other three days(13~15). In 13 June, there was a rainfall peak in 08LST. Afternoon rainfall peak was no significantly. In 14 June, the two rainfall peaks were in 06LST and 21LST. In 15 June, the characteristic of rainfall was similar to 12 June. The rainfall peaked in 05LST and 15LST, but the second peak was higher than in early morning. The number of station with daily rainfall accumulation exceeding 350mm on the 12 June was higher than other three days. Therefore, in this study we investigate the heavy rainfall case on 12 June 2005. Through the analysis 05 the European Center for Medium-Range Weather Forecasts (EC) data, the analysis Chart of Japan Meteorological Agency (JMA), Satellite images and Outgoing Longwave Radiation data, we found a short-wave trough formed over the northeast of South China Sea on 8 June. Moving northeastward, and caused heavy rainfall in southwest Taiwan. During 12~15 June 2005, Mei-Yu front was located on the offshore of north Taiwan. The heavy rainfall event cause by mesoscale convective systems from the South China Sea.
In this study, there were intense radar echoes formed along the coast in the early morning. There may have convergence in the coast of southwestern Taiwan by southeast wind near surface of southwestern Taiwan and the prevailing southwest wind on the ocean. In 925hPa, because the prevailing southwest wind decelerated and became southerly wind cause by terrain blocking. The convergence between prevailing wind and southerly wind occurred over southwestern Taiwan. Rainfall fell in coast and plane. The peak rainfall moved from coast to slope in the southwest of Taiwan. In order to study the characteristic of double rainfall peak and advance our understanding of the mechanism of heavy rainfall in coast and slope, we use the Weather Research and Forecasting Model (WRF) to serve our objectives.
The model results show that intense coastal convergence was due to the prevailing wind and offshore flow which formed by the decrease of virtual potential temperature. This decrease was due to rainfall in slope and plane in early morning. In the plane, southwest wind converged with the south wind due to orographic effect. Both of the two factors enhanced the rainfall in coast and plane. In the afternoon the convergence in coast wasn’t evident. The rainfall become intense when the convective system moved into southwest of Taiwan. Although the mechanism for rainfall of model results can explain rainfall characteristics, but the magnitude(350mm/day) was less than the observation(508mm/day). In the model results, the convective system peaked in 05 and 12LST, 2~3 hours earlier than observation became of the rapid movement of short wave trough in the simulation results than observed. In addition, no Taiwan terrain test exhibit the orographic effect is important to influence coastal and plane’s rainfall due to the prevailing wind decrease become south wind convergence with prevailing wind.

並列關鍵字

none

參考文獻

彭文君，2006：梅雨季台灣西南部豪雨豪雨之初步研究（1997年～2004年）。國立中央大學，大氣物理研究所碩士論文，122PP。

Carbone, R. E., J. D. Tuttle, W. A. Cooper, V. Grubiši , and W.-C. Lee, 1998: Trade Wind Rainfall near the Windward Coast of Hawaii. Mon. Wea. Rev., 126, 2847-4863.

Chan, J.-C.-L., W. Ai, and J. Xu, 2002: Mechanisms responsible for the maintenance of the 1998 South China Sea summer monsoon. J. Meteor. Soc. Japan, 80, 1103–1113.

Chen, C.-S., and Y.-L. Chen, 2003：The rainfall characteristics of Taiwan. Mon. Wea.

Chen, C.-S., C.-H. Lu, and W.-C. Chen, 2007a：Numerical experiments investigating the mechanisms of a heavy rainfall event over northeastern Taiwan and a mesovortex during TAMEX. Meteorol Atmos Phys 95, 155-177.