Güneydoğu Türkiye tarımsal alanlarında, Radarsat-2 uydu görüntülerinin işlenmesi ile yarı-deneysel, üç toprak nemi tahmin modelinin geliştirilmesi

Soil moisture or soil water content is an important factor and parameter for agricultural, hydrologicaland meteorological applications. Usually the soil moisture definition is, divided into two branches: deep or root zonesoil moisture, and surface soil moisture.Surface soil moisture is refers to the water that is in the upper 10 cm of soil and only constitutes 0.0012% of all water available on Earth(Verhoest et al., 2008). Whereas root zone soil moisture is the water that is available to plants, which is generally considered to be in the upper 200 cm of soil. The water stored in the soil has different roles in the global water cycle. For instance, the growth of plants, irrigation scheduling management, good product, crisis management during drought, etc. Or, it controls the partitioning of rainfall into runoff and infiltration. Runoff commonly means both exporting fresh water to other areas and degradation of topsoil through leaching and erosion(surface moisture),infiltration mean filling underground aquifers (deep soil moisture) (Verhoest et al., 2008). It has been widely observed that soil moisture is also a key variable in flood forecasting.However, the measurement of soil moisture is very important in other branches of science. Methods for measuring the mass of soil water have been in use since the 15th Century. Today, the most common method is with regard to the mass, volume or saturation of the soil. There are various methods available to measure the soil moisture content both directly and indirectly (Stachder 1996, Prietzsch 1998, Marshall 1999). In principle, direct and indirect methods of measurement can be distinguished from one another. Direct methods include all measured processes in which the soil water is removed via evaporation, extraction, or chemical reactions and gravimetric method such as the FDR or TDR methods. On the other hand, research in to the remote sensing of soil moisture began in the mid 1970's thanks to a surge in the development of satellite technology. Soil moisture measurement and estimation from remotely sensed data has come a long way due to its unique capability of monitoring large areas with long term repetitive coverage. Remote sensing is data acquisition with out direct contact with the object of interest by using of particular wavelength of electromagnetic spectrum. The omitted or reflected signals in remote sensing methods far from the Earth's surface contain information about soil properties and surface details in both optical and microwave remote sensing and have been put to use for soil moisture study. Studies on backscattering coefficiencyhave beenpublished regarding many different models for surface soil moisture estimation; (i) the empirical model (EM) of Oh et al., 1994, (ii) the theoretical integral equation model (IEM) of Fung et al., 1992,X-Bragg model (2008), and (iii) the semi-empirical models of Oh et al., 1992, Oh et al., 2004 and the model of Dubois et al., 1995. In this study by using performance of the Oh et al., 1992, Oh, 2004 and Dubois et al., 1995a, surface soil moisture estimation models was evaluated with two RADARSAT-2 scenes (FQ1, FQ19) on agricultural area over the Harran ,Sanliurfa of east south of Turkey. The results was researched in soil moisture models that was used in thesis,analyzed by Lee, Enh_Lee, Forst and Kuan filters to reduce speckle noise and improve the models performance and the accuracy of RADARSAT-2 soil moisture maps. In addetion, the optimal filter sizeby study about of statistical calculated indices for different kernel size was estimated . Highest agreement for optimal filter kind is Kuan,15×15, 5×5 kernel size for OH92, OH04 for 07.September.2012 RADARSAT-2 's data and 5×5,5×5,15×15 kernel size for OH92,OH04,DU95 models for 08.September.2012 ,RADARSAT-2 's data . By using Pauli RGB image was estimated surface coverage. The surface coverage information was used to find estimated soil moisture values in study area. Finally, the local and estimated values werecompared. This research supports the idea of incidence angle effect on the performance of models (Baghdadi et al. 2008 and Mo et al. 1984) and the idea of surface roughness as having animportant effect on the estimation ofsoil moisture values (Baghdadi et al.,2002; Verhoest et al., 2000; S.Khabazan et al., 2013; Zribi and Dechambre, 2003).