In this thesis, 41 years of bright sunshine duration (SD) data of 192 meteorological stations in Turkey were analyzed. The main objective is to determine the trends of SD data and the importance of such analyses is the high correlation between SD data and solar irradiation reaching the surface of the earth. Because of the missing value problems, only the data set for 72 stations were examined. After imputing missing values of these stations by expectation maximization algorithm, to test the homogeneity Kruskal Wallis test (K-W) and Wald-Wolfowitz runs test for randomness were applied. Only 36 of the stations passed from these homogeneity tests therefore, trend analysis was carried out for these locations. To exclude the data sets which did not pass from the tests was important to reach more accurate trend analysis of the data in hand. Results of the trend analysis showed that the change of SD over the 41 years are in agreement with the globally identified surface solar radiation dimming and brightening time periods all over the world. The dimming period is mainly between the years 1970 and about 1990 while the brightening period is from about 1990 to 2010. The yearly averages of SD data sets of 27 locations out of 36, for the years in the dimming period, were in a good agreement with the global dimming trends. However, for the brightening period the agreement was not as clear as it was in the dimming period. Nevertheless, during the brightening period, the data set of most of the locations had zero trends or noticeably reduced rates of decrease of SD. The dimming might be attributed to the increase in air pollution and this might be an indication of human induced climate change. Larger amounts of negative trends during winter months supported this conclusion. However, to reach a concrete conclusion more accurate of different climatic parameters should be analyzed. Satellites images may be helpful for further clarifications of such conclusions on climate change issues.

Modeling of flow in porous and fractured media is a very important problem in reservoir engineering. As for numerical simulations conventional Navier-Stokes codes are applied to flow in both porous and fractured media. But they have long computation times, poor convergence and problems of numerical instabilities. Therefore, it is desired to develop another computational method that is more efficient and use simple rules to represent the flow in fractured media rather than partial differential equations. In this thesis Lattice Boltzmann Automaton Model will be used to represent the single phase fluid flow in two dimensional synthetic fractures and the simulation results obtained from this model are used to train Artificial Neural Networks. It has been found that as the mean aperture-fractal dimension ratio increases permeability increases. Moreover as the anisotropy factor increases permeability decreases with a second order polynomial relationship.