應用衛星資料探討地表溫度

Abstract

隨著台灣的經濟與科技的發展，大台北地區的建設不斷地擴增，因而改變了周遭的環境，造成地表溫度的差異日益明顯，此即為典型都市發展之現象，原地表天然的植被漸漸的被人工鋪面所取代，地表溫度的分布亦隨之改變。為探討近地表熱環境的異常，本研究使用2000年至2003年Landsat的紅外頻道影像，首先針對北台灣地區所反演地表溫度分布的特性進行研究，其次為結合常態化差異植生指數(Normalised Difference Vegetation Index, NDVI )、常態化差異含水量指數(Normalised Difference Water Index, NDWI )、常態化差異建物指數(Normalised Difference Building Index, NDBI )等指數，分別以定性和定量的方式分析土地利用之空間分布情形，並探討不同覆蓋物與地表溫度分布間之相關性。另一方面，由於大屯火山群位於台灣北部地區，其所潛含的地熱能量亦將影響地表溫度之呈現。因此，除土地利用類型外，本研究亦將針對衛星紅外頻道資料在火山區地熱異常勘查的潛能及可行性進行探討，亦即藉由分析其在時間序列上的變化，針對不同季節間溫度差異，來找尋可能的地熱分布區。整體研究結果顯示，地表溫度與地物的含水量呈反比，與地物裸露地程度呈正比；同時，透過分析衛星反演地表溫度之相對變化，亦可呈現火山區域可能的分布位置。

With the expanding and growth in both economy and technology of the greater metropolitan Taipei in the recent decades, the change of its surrounding environment, such as the land surface temperature contrast between day and night time, is getting obvious. The coverage of natural vegetation area is replaced by artificial pace, which is often seen in a developing city. It is expected that the land surface temperature distribution is no longer in similar patterns from past to current. To investigate the near-surface thermal phenomenon in association with environmental change, this study is proposed to utilize the infrared imagery from Landsat during 2000 to 2003. The characteristics of the retrieved land surface temperature (LST) from Landsat observation in the northern part of Taiwan are analysed. Those indices from NDVI (Normalized Difference Vegetation Index), NDWI (Normalized Difference Water Index), and NDBI (Normalized Difference Building Index) are examined for the correlation with land use qualitatively and quantitatively. The correlation between different land cover and LST is investigated as well. On the other hand, the potential geothermal energy in conjunction with LST is discussed in this study due to Datun volcano is in the vicinity of metropolitan Taipei. Through the change in LST from time series analysis in four seasons, this study would like to identify the volcanic area from geothermal anomalies in Landsat infrared observation. The result suggests that LST is in a negative correlation with soil moisture content, while LST has a positive correlation with the portion of bare soil. The spatial distribution of possible volcano location could be identified through the analysis of temporal change in Landsat LST.