Ph.D. Tezi Görüntüleme

Student: Sedat YILMAZ
Supervisor: Asst. Prof. Dr. Nart COSKUN
Department: Geophysical Engineering
Institution: Graduate School of Natural and Applied Sciences
University: Karadeniz Technical University, Turkey
Title of the Thesis: Two-Dimensional Finite Element Modeling And Inversion Of The Dipole-Dipole Resistivity Data Including The Effect Of Topography
Level: Ph.D.
Acceptance Date: 1/1/2005
Number of Pages: 115
Registration Number: di486
Summary:

       The effect of the rugged terrain causes erroneous results on the two-dimensional dipole-dipole apparent resistivity field data. To overcome this problem, current study investigates two approaches using the synthetical modeling technique for forward and inversion solutions. In the first approach, the data corrected using finite element method for the rugged terrain effects is inverted using the least squares method. In the second approach, the data uncorrected for the rugged terrain effects is inverted to incorporate topography into initial model.

Two dimensional mesh with 103x19 nodes used for the finite element method calculation of the apparent resistivity pseudosection of the models presented in this study gives error 2.1 percent for N=1 level while less than 0.5 percent for the other levels. The gradual ınversion approach which is decreased with 101/2 coefficent of the initial damping factor, λ = 10, gives better solution according to the traditional one which is used fixed damping value. Also, for the homogeneous initial resistivity guess, the arithmetic average of the field data or its multiples could be choosen.

       The topographical resistivity effects are significant in areas where slope angels are more than 10 degrees. The effects are excessive together with increasing of the slope of terrain. The normalized data can be inverted as if there were no terrain variation, or they were the response of a flat earth. Incorporation of the terrain topography into two dimensional resistivity inversion is difficult, and time consuming. Thus, on the interpretation of the data acquired on rough terrain, the first approach is much convenient than the other one.

In conclusion, reliability of the current study tested on lanslide area of Güzelyalı, Gürbulak, Trabzon (NE Turkey). The three-dimensional resistivity imaging of the lanslide is predicted using two-dimensioal inversion.

      

      Key Words: Resistivity, Dipole-Dipole, Finite Element Method, Topography, Normalization, Least Squares Method, Landslide