Downscaling Land Surface Temperature on Multi-Scale Image for Drought Monitoring

摘要

Land Surface Temperature (LST) is an important indicator of environmental changes, especially related droughtmonitoring. It is necessary to accurately detect drought events using advanced technology proved information regardingthe drought areas. Remote sensing images have proven to be efficient in detecting drought events. MODIS Terra andLandsat 7 ETM+ (Enhanced Thematic Mapper Plus) and Landsat 8 OLI/TIRS (The Operational Land Imager and theThermal Infrared Scanner) represent remote imaging images with different spatial resolutions that enable us to providedrought information. However, proper methods are needed to optimize these images for monitoring drought events. Thepurpose of this study is to find out the ability of multi-scale images provide information about drought monitoring usingLST methods. The method used in LST is a Temperature Condition Index (TCI), the Crop Water Stress Index (CWSI),and Principal Component Analysis (PCA). All three equations are selected because they represent a modification of themethod for LST input. The results suggest that the three equations used in multi-level imagery have a critical alignment ofinformation regarding the drought. The results show that the drought pattern identified by MODIS Terra image was similarto the one detected by Landsat ETM+ and OLI/TIRS images. However, we found a temperature difference in the dryseason (especially in October) between Landsat ETM+ and OLI/TIRS. The degree of LST estimation accuracy betweenMODIS Terra and Landsat (ETM+ and OLI/TIRS) is indicated by the average difference between the results of thoseimages, which was 1 degree Celsius (1°C). The use of these three equations for drought monitoring with multi-levelimagery suggests that there is a positive relationship. This relationship manifests the same pattern, shape, and associationthat are produced, thus using a common equation for drought monitoring is more focused.