台北地區不同海拔長期氣溫趨勢比較分析

COMPARISON OF LONG-TERM TEMPERATURE TRENDS AT DIFFERENT ALTITUDE REGIONS IN THE TAIPEI METROPOLITAN AREA

邱祈榮(Chyi-Rong Chiou) ； 鄧翔耀(Xiang-Yao Deng)

《中華林學季刊》 53卷4期 (2020/12) Pp. 169-186

摘要

近年來全球持續暖化，影響到不同地區的氣溫變化。本文研究目的在於透過台北周邊地區不同海拔的測站氣溫資料，比較不同海拔的長期變化趨勢是否一致，探討海拔及都市熱島效應對於氣溫趨勢所產生的影響。彙整台北、竹子湖及鞍部測站1970－2019年共50年的歷史氣象資料，利用趨勢檢定法、斜率推估法及改變檢點法建立各測站的長期氣溫趨勢。接著將資料分成十個時期，分析測站間各時期的溫差的變化，再以雙因子ANOVA變異數分析（Two-way ANOVA）並利用事後檢定分析時期差異。結果顯示各測站的溫度皆有顯著的上升趨勢，鞍部測站年平均最高溫上升最快、台北測站次之、竹子湖測站最緩；台北測站的年平均最低溫上升最快、鞍部測站次之、竹子湖測站最緩，冬夏季節各測站的上升趨勢不同，顯示季節對於溫度趨勢有一定的影響。各測站的測站位置及時期具有顯著差異，兩者也存在顯著的交互作用。台北測站與其他兩個測站之間的年平均最低溫溫差逐漸增大，明顯受到熱島效應的影響。鞍部與竹子湖測站的年平均最高溫溫差及最低溫溫差逐漸縮小，可能為海拔及地理環境的影響。如果每增溫1℃，對越高海拔地區的溫度上升影響程度越大，顯示高海拔受到暖化的影響可能更加劇烈，未來必須適當地解決目前的暖化問題。

關鍵字

熱島效應 ； Mann-Kendall趨勢分析 ； Theil-Sen斜率推估法 ； Pettitt-Mann-Whitney改變點檢定法 ； 鄧肯新多重差距檢定

並列摘要

Global warming has continued to affect the climate in many regions and the influence is different. Speeds of global warming impacted by many factors, including altitude, degree of urbanization, and geographic location. This study used the historical temperature records from 1970 to 2019 of Taipei, Zhizihu and Anbu meteorological stations in the Taipei metropolitan areas. We analysed the long-term temperature change tendencies and the influences of urban heat effect and altitude with these data. Statistic methods of the Mann-Kendall test, Theil-Sen estimator and Pettitt-Mann-Whitney test were adopted to calculate statistic values for calculating the long-term change trends, time of change point. The fifty years records of each station were divided into ten periods to compare temperature difference among stations and time periods by the Duncan's new multiple range test and two-way analysis of variance. Though the geographical conditions are similar, the altitude and degree of urbanization may affect their microclimate and resulting in different trends. The results showed that the temperature rates of each station were increased. The annually maximum temperature ascendant trend in the Anbu station was the highest, followed by Taipei and Zhizihu stations. The ascendant trend of each station was different in winter and summer, which indicated that temperature trends will be impacted by season. The ascendant trends of temperatures were divided into 3- 4 stages. The results of the two-way ANOVA analysis showed that locations and periods have significant impacts on temperature trends and both have interactive effects. The difference of the annually maximum temperature and the annually minimum temperature between the Zhizihu and Anbu were gradually shrinking and this phenomenon is conjectured to be affected by altitude and geographic conditions. The ascendant trend of the annually maximum temperature at Anbu station was greater than that of the Taipei station, indicating that the global warming may affect the high altitude areas more seriously.

並列關鍵字

Temperature trend ； Mann-Kendall Test ； Theil-Sen estimator ； Pettitt-Mann-Whitney test ； Duncan's new multiple range test