熱浪的發生往往造成了極為嚴重的災害，尤其是會造成人體極不舒適的感覺，甚至會造成死亡，例如：1995年芝加哥熱浪、2003年巴黎熱浪及2010年俄羅斯熱浪…等，隨著全球溫度持續暖化的趨勢也使得熱浪發生的頻率、強度和持續時間變得更強、更長。而在前人研究中，他們發現在歐美地區發生熱浪時，主要是由於阻塞高壓所造成，而在高壓的下沉區造成高穩定度及較低濕度的環境分布，導致在此區域會有異常高溫持續出現的情況。然而，對於在台灣地區夏季也常有熱浪發生，但對於熱浪發生時大尺度環境與中高緯度區域未必相同，因此冀望能研究分析台灣熱浪發生期間的大尺度環境的分布特徵。
前人所做的熱浪研究中對於熱浪的定義並沒有統一的標準，本論文在熱浪分析時所使用的熱浪定義，參考過去研究，是以最高溫的兩個極端百分位值加以定義，可以允許持續熱浪中短暫的降溫。我們除了針對熱浪發生的頻率、持續時間加以探討，也討論結果對選取百分位值的敏感度。而資料的使用有氣象局測站資料及美國環境預報中心的氣候預報系統再分析資料(NCEP Climate Forecast System Reanalysis, CFSR)，分別從每個測站及再分析資料網格點裡定義出熱浪時段，由測站及再分析資料的分析結果，發現如果採取較高的極端百分位值，使熱浪發生的頻率為平均每年或甚至是平均1~2年才發生一次的極端事件，而熱浪平均的持續時間則是介於10~15天左右，在平均每年最早起始時間約在7月中旬開始，而平均每年最晚結束時間則接近於7月底結束，不過年際之間個別熱浪事件還是有相當大的差異。而從大尺度環境角度我們藉由分析500hpa重力位高度場、風場、垂直速度場、相對溼度場，發現在熱浪發生期間的共同特徵是太平洋高壓西伸至台灣以西，並在台灣附近形成下沉氣流與乾燥的環境。
除此之外，我們同樣選取高解析度氣候模式GFDL-HiRAM所模擬的資料分析，比較現今氣候與未來21世紀末氣候變遷下熱浪特性型態的改變。以氣候預報系統再分析結果驗證GFDL-HiRAM模式所模擬的熱浪特性，兩者在熱浪發生的頻率、平均持續時間、平均每年最早起始時間及平均每年最晚結束時間的結果相似，如果熱浪定義的兩個極端百分位值的選擇一樣，HiRAM所模擬的平均熱浪持續時間比氣候預報系統再分析的結果略短。而從模式21世紀末的未來推估結果，可以發現在未來氣候變遷下，溫度的上升會使得幾乎每年都會發生一次持續整個夏季的熱浪事件，平均熱浪持續時間的增加幅度則會是現今氣候下的10倍以上，平均每年最早起始時間同樣會提早70~80天開始，而平均每年最晚結束時間則會延後50~60天結束，由熱浪特性型態的改變，顯示在未來氣候暖化下會使得現今氣候下所定義每一、兩年才發生一次的熱浪事件變成夏季的常態。

Heat wave is one of serious disasters. Their occurrences and lasting effects could cause extreme discomfort to human body, and even lead to death, such as: 1995 Chicago heatwave, 2003 Paris heatwave and 2010 Russian heatwave. As the global temperature continued the warming trend, the heat waves become more frequent, the intensity become stronger and the duration expected to be longer than the present climate. In previous studies, the heatwave in Europe and America are mainly caused by blocking events. The high stability, decending motion and low humidity in the persistent high-pressure region, often create sustained high temperatures anomalies over the area. However, for the summer heatwave occurred in Taiwan, large-scale environment settings may not be the same in extratropical regimes. Our study aimed to analyze the large-scale environment characteristics near Taiwan during the heatwaves.
There was no uniform or standard definition of heatwave in the previous literature. In our study, we also try to highlight the sensitivity of the result to the definition of heat waves through changes in the parameters used for heatwave threshold. Following past studies, we selected a definition of heat wave that use both high and extreme temperature percentiles to quantify the heatwave occurrences. The use of second thresholds allows temperature to drop for a brief period of time during a heatwave event. We use both CWB surface station and US NOAA NCEP Climate Forecast System Reanalysis (CFSR) data for heatwave analysis. We first examine the sensitivity of heatwave characteristics to the two percentile values in our heatwave definition for each individual station and reanalysis grid points. There is a strong dependence of heatwave occurrence to the most extreme percentile threshold used. If we chose a more strict heatwave definition by using a higher extreme percentile value, the frequency of heat waves occurrence would reduce to once or twice per year on average, with mean duration between 10 to 15 days, and the earliest start time began about mid-July and the latest end time close to the end of July, Note that, the individual heatwave events can still have considerable interannual variability. For these relatively rare heatwave events over Taiwan we analyzed the large-scale environment with 500hpa geopotential height, wind, vertical velocity, and relative humidity fields, and found common features during the heatwave events were the extension of Northwest Pacific subtropical high pressure system to the west of Taiwan, and formed a dry environment with dominant descending motion in vicinity of Taiwan.
In addition to the heatwave analysis using observation and reanalysis data, we also use a high-resolution climate models GFDL-HiRAM simulation and projection data, compare the heatwave events in the present climate and near the end of 21st century. Comparing to the present-day heatwaves and their sensitivity to the thresholds used in heatwave definition in the CFSR data. GFDL-HiRAM model simulated heatwave characteristics are reasonably well. When the same two high and extreme percentile values, applied, similar heatwave characteristics are found except the average duration in the HiRAM simulation was slightly shorter than in reanalysis data. The results from the future model projection near the end of the 21st century, show that with present-day heatwave definition, the warming temperatures in the future climate will lead to a single continued summer heatwave event almost every year. The mean increasing of heat waves duration will be more than 10 time of those in the present climate. The mean annual starting time for heatwave will advanced by 70 to 80 days, while the mean annual heatwave ending time will delay by 50 to 60 days, with such changes in the heatwave pattern, we can expect the once or twice per year heatwave events in the present climate will occurrence everyday through the summer.

Alexander, L. V., et al. (2006). "Global observed changes in daily climate extremes of temperature and precipitation." Journal of Geophysical Research 111(D5).

Chu, Y. C. (2007). "Heat waves of the 21st century."

D.Horel, J. (1981). "A rotated principal component analysis of the interannual variability of the northern hemisphere 500 mb height field.pdf." Monthly Weather Review 109.

Ding, T., et al. (2009). "Changes in hot days and heat waves in china during 1961-2007." International Journal of Climatology

Fei, S. T. and L. H. bang (2003). "Abrupt change of the west pacific subtropical high and its interannual variation during the later spring and early summer."

G.Steadman, R. (1984). "Steadman's model function.pdf." Journal of climate and Applied Meteorology 23.

Huth, R., J.kysely, and L.Pokorna (2000) <A gcm simulation of heatwaves, dry spells, and their.pdf>.

Jiang, X., et al. (2011). "Interannual and interdecadal variations of the south asian and western pacific subtropical highs and their relationships with asian-pacific summer climate." Meteorology and Atmospheric Physics 113(3-4): 171-180.

Li, W., et al. (2011). "Changes to the north Atlantic subtropical high and its role in the intensification of summer rainfall variability in the southeastern united states." Journal of Climate 24(5): 1499-1506.

MuQ iaozhen, et al. (2001). "Variations of the western pacific subtropical high in summer during the last hundred years."

Michael A. Palecki, S. A. C., and Kenneth E. Kunkel (2001). "<The Nature and Impacts of the July.pdf>." Bulletin of the American Meteorological Society 82(7).

Meehl, G. A. and C. Tebaldi (2004). "More intense, more frequent, and longer lasting heat waves in the 21st century." Science 305(5686): 994-997.

ROBINSON, P. J. (2000). "On the Definition of a Heat Wave." Journal of Applied Meteorology.

SCHOEN, C. (2006). "<A New Empirical Model of the Temperature–Humidity Index.pdf>."

Sui, C.-H., et al. (2007). "Interannual and interdecadal variability of the summertime western North Pacific subtropical high." Geophysical Research Letters 34(11).

Schalge, B., et al. (2011). "Blocking Detection Based on Synoptic Filters." Advances in Meteorology 2011: 1-11.

Schneidereit, A., et al. (2012). "Large-Scale Flow and the Long-Lasting Blocking High over Russia: Summer 2010." Monthly Weather Review 140(9): 2967-2981.

Smith, T. T., et al. (2013). "Heat waves in the United States: definitions, patterns and trends." Clim Change 118(3-4): 811-825.

Wu, Z., et al. (2012). "Heat wave frequency variability over North America: Two distinct leading modes." Journal of Geophysical Research 117(D2).

龔道溢,王.,楊義文(1998)."<90年代西太平洋副高异常的分析.pdf>." 氣象 24(8).

慕巧珍,王.,朱錦紅,龔道溢(2001)."<近百年夏季西太平洋副热带高压的变化.pdf>." 大氣科學 25(6).

慕巧珍,王.,龔道溢(2002)."<近百年四季西太平洋副熱帶高壓的變化.pdf>." 氣象學報 60(6).

陶詩言,衛.捷.(2006)."<再论夏季西太平洋副热带高压的西伸北跳.pdf>."應用氣象學報 17(5).