遙測衛星影像應用於地表乾燥指標及地表土壤含水量之研究目前使用光學衛星影像偵測地表土壤含水量的方法大部份為產生地表乾燥指標，然而因為此指標無法直接代表實際的土壤含水量(g/cm3)，因此只能視為間接式的描述地表土壤含水量。本計畫主要目的是應用光學衛星資料產生描述地表乾燥程度的TVDI指標 (Temperature-Vegetation Dryness Index)分佈圖，再與同地區由微波衛星所獲得的地表土壤含水量進行相關性分析，進而將相對的TVDI指標轉換為具有絕對物理量(g/cm3)的地表土壤含水量空間分佈圖，並加以驗證與分析優缺點。計畫預定執行三年，第一年的工作是利用MODIS(Moderate Resolution Imaging Spectroradiometer) 衛星所提供的1 公里解析度的NDVI (Normalized Difference Vegetation Index)資料與地表溫度資料計算出TVDI指標，以產生描述地表乾燥程度的1公里解析度TVDI空間分佈圖，然後再以雨量資料分析驗證TVDI 時間序列資料的特性。第二年的工作是分析MODIS衛星TVDI(1公里)與AMSR-E 衛星地表土壤含水量資料(25公里)的關聯性，然後進行相關性分析並建立轉換關係，以產生實際土壤含水量(g/cm3)的1公里地表土壤含水量空間分佈圖，亦即將1 公里TVDI透過與25公里地表土壤含水量的相對關係，以產生1公里的地表土壤含水量。第三年的工作是以前兩年TVDI的研究為基礎，使用空間解析度更高的 Landsat衛星的NDVI與熱紅外資料(60公尺)產生TVDI，再透過相對分析以產生60 公尺的地表土壤含水量空間分佈圖。本計畫主要是結合不同空間解析度之衛星遙測影像產生長時間的地表土壤含水量空間分佈資料，並研究提高地表土壤含水量資料的空間解析度，未來更可據此分析乾旱現象在時間序列上的空間分佈變化。 The Application of Remote Sensing Satellite Images to Land Surface Temperature-Vegetation Dryness Index (TVDI) and Land Surface Soil Moisture The surface soil moisture has been proven as one of the most important factors in environmental studies, including hydrology, meteorology, and climate change. The main objective of this project aims at investigating the surface soil moisture from optical remote sensing using the Temperature-Vegetation Dryness Index (TVDI) method. This project is designed for three years to address three specific objectives. In the first year, the objective is to create a series of TVDI maps (spatial resolution of 1 km) using the Moderate Resolution Imaging Spectroradiometer (MODIS) Normalized Difference Vegetation Index (NDVI) and Land Surface Temperature (LST). The TVDI results will be verified with the precipitation data. In the second year, the objective is to compare the TVDI results with the soil moisture data (spatial resolution of 25 km) derived from the Advanced Microwave Scanning Radiometer - Earth Observing System (AMSR-E) sensor on NASA's Aqua satellite. It is expected to produce results that indicate strong correlation between TVDI and AMSR-E datasets. Therefore, the TVDI method will accordingly be proposed to create surface soil moisture data with a better spatial resolution. In the third year, another attempt will be made to use Landsat data (thermal bands with spatial resolution of 60 m; and optical bands with spatial resolution of 30 m) to derive the higher resolution of surface soil moisture (i.e., 60 m) using the same method (TVDI). The derived results will also be compared with that produced from MODIS data. Since the Landsat imagery contain gaps due errors of the satellite sensors (since 2003), much efforts will also be made to develop a gap-filling algorithm for Landsat data during the third year. In order to obtain the Landsat LST used for estimating the surface soil moisture, the atmospheric calibration of thermal bands of Landsat data is another challenging matter which needs to be implemented. The goal of this project is to create the time-series soil moisture maps in quantitative measurement and improved spatial resolution. In the future, the soil moisture maps obtained in this project may provide a better data sources for analyzing the spatial distribution of drought event in different periods of time. 研究期間：10008 ~ 10107
