國立臺灣大學地理環境資源學系; 國立臺灣大學地理環境資源學系賴栗葦姜善鑫2006-12-192018-06-282006-12-192018-06-282004http://ntur.lib.ntu.edu.tw//handle/246246/2006121215551260The diurnal temperature range (DTR) has significantly decreased worldwide since 1950. In this study, the trends of monthly mean of DTR in Taiwan were analyzed. The methods of time series regression model and principal component analysis were used to analyze the daily minimum and maximum temperature data of Taiwan from 1950 to 2000. The trends of monthly mean of DTR in most stations indicated a statistically significant decrease. This was mainly due to faster increasing trends in monthly mean of daily minimum temperatures than those in monthly mean of daily maximum temperatures. The decreasing trends in the monthly mean of DTR in urban areas were larger than those in other areas and much more obvious in autumn. It was found that the trends in the monthly mean of DTR were correlated with human activities and geographical locations.1950年以來全球大部分地區氣溫日較差(diurnal temperature range﹔DTR) 呈減少的趨勢,顯示出全球暖化下氣候變遷的現象。本研究目的在了解臺灣地區月平均DTR的趨勢,並找出可能影響月平均DTR的因子。本研究以選取臺灣地區氣象資料較長的18個測站為基礎,選取10項逐日氣象資料,應用時間序列迴歸模型及主成分分析法進行分析。結果顯示,近五十年來臺灣地區大部分測站月平均DTR均呈顯著減小趨勢,這主要是月平均最低溫上升趨勢較月平均最高溫顯著所致,其中又以大都市地區秋季最為明顯,月平均DTR的趨勢還能反應出受到人為活動及地理位置不同的影響。application/pdf737576 bytesapplication/pdfzh-TWdiurnal temperature range (DTR)daily minimum temperaturedaily maximum temperaturetime series regression modelprincipal component analysismonthly mean氣溫日較差(DTR)日最低氣溫日最高氣溫時間序列迴歸模型主成分分析月平均臺灣地區月平均氣溫日較差趨勢分析journal articlehttp://ntur.lib.ntu.edu.tw/bitstream/246246/2006121215551260/1/11600184730672064711255.pdf