Applying DEM data to improve performance of water extraction indices using landsat 8 OLI images in mountainous area

Water is one of the most important earth resources which is essential to human health, society and environment. Studies on water extraction and changes have been subjects of academic studies for many years. Remote sensing as an efficient and reliable tool has been used in recent years and Landsat satellite imagery were one of the most common data due to their advantages in spatial resolution and cost. Improvement of new Landsat 8, the Operational Land Imager (OLI) data attracted more attentions recently. This study uses the Landsat 8 OLI imagery data source for water information extraction based on the Normalized Difference Water Index (NDWI), Modified Normalized Water Index (MNDWI) and Automated Water Extraction Index (AWEI) to compare the effect of using The Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) Global Digital Elevation Model (GDEM) in mountainous area. The study area is Kermanshah, in west of Iran, a mountainous area which has difficulties for water extraction due to shadows and dark objects. Due to small area of water bodies in study area user's accuracy were used for evaluation of results. User accuracy for water class gives results of 23.68%, 24.34% and 22.57% for NDWI, MNDWI and AWEI, respectively. In other words, around 77% of pixels which classified as water are not water and are misclassified pixels. Applying DEM data improves results to 27.44%, 29.1% and 27.22% for NDWI, MNDWI and AWEI, respectively which shows slight increase of 3.76%, 4.88% and 4.65%.