|Abstract: ||隨著科技的發展，近年來科學家們嘗試使用高空間解析度的衛星資料探討都市熱島效應，本研究嘗試在大台北都會區中整合並擴展Landsat系列衛星的都市熱島研究方法。使用土地利用與地表型態變遷(Land Use and Land Cover, LULC)、地表溫度(Land Surface Temperature, LST)、都市熱島影響區域(Urban Heat Island area, UHI area)、都市熱島環境評估(Urban Thermal Field Variance Index, UTFVI)探討大台北都會區過去25年的都市變遷以及都市熱島影響，並測試三組遙測指數: 常態化差異建地指數(Normalized Difference Built-up Index, NDBI)、常態化差異潛熱指數 (Normalized Difference Latent heat Index, NDLI) 、常態化差異植生指數 (Normalized Difference Vegetation Index, NDVI)與地表溫度進行分析，其中NDLI為國際上新的遙測指數，本研究特別針對大台北都會區的不同土地利用型態適用範圍進行測試和整理。在都會區和市中心的比較上，本研究也進行了時序比較，尤其是在2016年的極高溫事件運用LST搭配地表觀測站計算兩者差異性測試，並整理大台北都會區及不同土地利用型態的都市熱島強度(Urban heat island intensity, ∆UHI)。|
研究顯示，市中心雖然經綠地規劃植被(Vegetation)面積近幾年上升約2 %，但是市中心仍有擴張的情況，以致整個都會區的建地(Built-up)面積比例上升約2.18%，整個都會區的植被面積比例有下降約0.5%；在大台北都會區發展衍生的都市熱島增強影響下，可以發現近幾年都會區溫度上升約3°C，且市中心的溫度上升程度更為嚴重約上升4°C；而都市熱島影響範圍的測試中，近幾年越往市中心集中且擴張14% (0.5*SD)面積、11.5%(1*SD)，且在UTFVI評估下，市中心的Excellent等級範圍面積下降約20%，Worst等級範圍面積上升約16%。在2016個案中大台北都會區整體都市熱島強度為4.6°C，其中建地的∆UHI最高為4.8°C，遠高於測站間所測得的∆UHI為3 °C。
在遙測指數的應用上，分別於夏季以及冬季進行測試，發現相關性在夏季時會優於冬季且市中心會優於都會區，其中夏季市中心的LST與NDBI有著高度的正相關性約0.87，NDLI有著高度的負相關性約-0.84；NDVI在冬季市中心呈現低度的負相關性約 -0.51，在夏季市中心才呈現較高度的負相關性 -0.75。在NDLI的運用中: 1. 夏季時NDLI在水體(Water Body)的負相關性均高於NDVI，適用於淡水出海口以及翡翠水庫區域的NDLI水體監測；2而台北市和新北市間的都市規劃差異及3.山區和都市地區間的環境保育發展評估上，也提供未來更為廣泛的應用和測試於台灣與國際都市熱島效應研究中。
;The use of high-resolution satellite data for analyzing the Urban Heat Island effects (UHI) is becoming typical in the recent years due to advancement in the satellite technology. The study integrates and expands the earlier used Landsat series of urban heat island research methods for the Taipei metropolitan area. The Land Use and Land Cover (LULC), Land Surface Temperature (LST), Urban Heat Island area (UHI area), and Urban Thermal Field Variance Index (UTFVI) are used to explore the urban changes and analyze the effects of urban heat islands in the Taipei metropolitan area for the past 25 years. It also tests three sets of telemetry indices: Normalized Difference Built-up Index (NDBI), Normalized Difference Latent Heat Index (NDLI), and Normalized Difference Vegetation Index (NDVI) along with analyzing surface temperature for the same region. However, among all indices, NDLI is a new international telemetry index. The study specifically aims to differentiate LULC of the Taipei metropolitan area. A time-series comparison is done between the metropolitan and the urban area of Taipei for the recorded highest temperature year of 2016. The LST and the surface station based observational data are compared for both UHI and non-UHI area of the Taipei Metropolitan.
Results show that the vegetation area of the urban area is increased by about 2% in the recent years along with the expanding urban area. As a result, the proportion of the Built-up area of the Taipei metropolitan is increased by about 2.18%. The proportion of the vegetation area is decreased by about 0.5% under the influence of the increased urban heat island caused by the recent developments in the Taipei Metropolitan. It is found that the temperature of the metropolitan is gradually increased by about 3°C in recent years, and the temperature rise in the urban area is also increased even more (about 4°C). The area of influence in the urban heat island is more concentrated in the urban and expands by 14% (0.5*SD)和11.5% (1*SD) . The UTFVI assessment reveals that the area of the Excellent level in the urban is decreased by about 20%, whereas the area of the Worst level is increased by about 16%. In the year 2016, the overall urban heat island intensity of the Taipei Metropolitan was 4.6°C. The highest urban heat island intensity (ΔUHI) is formed on the construction site which is 4.8°C, it is much higher than the ΔUHI measured between the stations (3°C).
The indices are tested for both summer and winter seasons, which reveals that their correlation is better in summer than in winter especially for the urban area. The LST and NDBI in the summer have a high positive correlation of about 0.87, whereas NDLI has a high negative correlation of about -0.84 in the urban area. The NDVI presents a low negative correlation of about -0.51 in the winter and a high negative correlation of -0.75 in the summer of urban area. The application of NDLI found in the present study are: 1. The NDLI is more suitable for monitoring the water bodies like Tamsui River and Feicui Reservoir area, because of its negative correlation for the Water Bodies in summer than NDVI. 2. The NDLI is capable to show the difference in the urban greenery between Taipei and New Taipei city, 3. The NDLI shows the difference of nature between the urban and mountains areas, which can be further analyzed in detail in the future researches.