Investigation of potential thermal water areas by using remote sensing and geographical information system: Exploration of Kaz mountains and environment

Today, the importance of renewable energy sources and geothermal energies is increasing due to the ever increasing need for energy. In recent years, Remote Sensing methods and Geographic Information Systems (GIS) have been extensively used in geothermal energy researches. In this study, Kaz Mountain and surroundings located within the borders of Balıkesir Province was determined as a study area. Today, the need for renewable energy is increasing and the importance of geothermal energy, one of the renewable energy sources, is understood. In this thesis study, SPOT 4 multispectral satellite image with 20 meter geometric resolution and SPOT 4 panchromatic satellite image with 10 meter geometric resolution s s dated 19.07.2011 have been used along with multispectralLandsat 5 TM satellite image with30 meter geometric resolution dated 16.07.2011. Image fusion technique is used in various areas of remote sensing. The aim of image fusion is to obtain multispectral images having both high geometric and high spectral resolution. In this study, six different image fusion techniques have been applied to satellite images with high spectral and low spatial resolution by using high spatial and low spectral resolution images to obtain fused images with increased spatial resolution, while preserving spectral information as much as possible. Wavelet Transform, Multiplicative Transformation, Intensity-Hue-Saturation (IHS), Principal Component Analysis (PCA) and Brovey Transformation are used in image fusion methods. Image enhancement techniques were applied to the data both before and after application of the fusion techniques. IHS method transforms a color in RGB environment from IHS environment. IHS environment provides information and parameter that are very close to human color perception. The modified IHS transform also improves the resolution of the fused image. It changes the spectral contents of the image. Wavelet transform is used to decompose two-dimensional signals into different levels of resolution for multi-resolution analysis. This multi-resolution feature is used to combine images having different resolutions. The Principal Component Analysis is a linear transformation that transforms data with highly variable and highly correlated data into a new coordinate system without correlation. There is no correlation between the new components because they have perpendicular axes to each other. Multiplicative Transform applies a simple multiplicative algorithm which integrates the two raster images. The multiplicative method is the simplest of the PCA and Brovey methods. Brovey Transform is used to visually increase the difference between the lowest and highest edge values of the image. Because of this reason, this method is mainly used in studies where importance of perception of differences is prominent. In the first phase of this work, new satellite data with high geometric and spectral resolution is obtained by combining high geometric resolution satellite data with high spectral resolution multispectral satellite data. Results obtained by image fusion methods were classified in two stages as supervised and unsupervised. When supervised classification is made, the sample control areas selected for the thermal areas are the basis for working from the Hattusha thermal facilities and surrounding area in the study area. Accuracy analyzes of the resulting images were obtained after classification. By comparing the results of general accuracy analysis of the methods used in the study area, fusion techniques that can be obtained with high accuracy for the region have been determined. The percentages of general accuracy analysis of SPOT 4 MS and PAN fusion results are as follows: Waveleth Transform %92.09, Multicative Transform %90.43, IHS Transform %87.83, Brovey Transform %83.82 and Principal Component Analysis %83.75. The percentages for Landsat 5 TM MS and SPOT 4 PAN results are as follows: Dalgacık Transform %90.09, Multicative Transform %86.83, IHS Transform %83.83, Brovey Transform %75.98 and Principal Component Analysis % 78.75 accuracies were achieved. Among these two datasets, the Wavelet transform is more accurate. Accuracy assesment percentage of SPOT 4 MS – SPOT 4 PAN fusion has higher than Landsat 5 TM MS – SPOT 4 PAN fusion. In the second phase of the study, land surface temperature (LST) analysis of the basin was executed. Land Surface Temperature Analyzes (LST) of the region were also performed in the Erdas Imagine 2013 and Arc GIS 10.1 programs using the single window algorithm to establish the elevation-temperature relationship. For the NDVI (Normalized Difference Vegatation Index) which is a step of the single window algorithm, the 3rd and 4th channels of the Landsat 5 TM satellite data are used. For the single window algorithm, channel 6, the thermal channel of Landsat 5 TM satellite data, was used. Owing to the thermal band, a single window algorithm has been realized by obtaining the parameters of emissivity, atmospheric permeability and average atmospheric temperature.The mono window algorithm consists of six steps converts the digital number (DN) into spectral radiance, and the spectral radiance into a sensor brigtness temperature and then calculates land surface emissivity, atmospheric transmittance, mean atmospheric temperature, and finallized with LST retrieval. The results indicate that the basin varies between 18 and 40 °C approximately. Between elevations of 1400 – 1750 m, it was seen 18 – 22 °C, elevations of 1100 – 1400 m it was observed 20 – 25 °C, elevations of 750 – 1100 m it was observed 22-37 °C, elevations of 400 – 750 m it was seen 25-30°C and between elevations of 0 - 400 m it was observed that the temperature is always above 30 ° C in these areas. In the third phase of the study, in order to establish the height-warm relationship, two and three dimensional digital elevation models (DEMs) belonging to the region were created in Arc Map 10.1, Surfer 13 and Global Mapper 18 programs and the results were compared with each other. Finally, maps of fault lines belonging to the region have been downloaded from the site of the General Directorate of Mineral Research and Exploration (MTA), digitized and integrated into satellite images. As a result of all data obtained with above-mentioned techniques,the potential thermal water areas are estimated in the study area.