Rock medium is not a virgin and tectonic events lose the homogeneity of intact rock. In order to model the rock masses, discontinues methods have been developed which is considered the deformability of rock blocks and joints. Using field mapping techniques on outcrops, trenches and/or subsurface measuring technique, statistical geometrical properties of discontinuity features such as length; aperture, location and direction (dip and dip direction) are approximated. The Monte Carlo method and statistical density function of geometrical parameters of fractures are used for generating a number of Discrete Fracture Networks (DFN) models. DFN models make a geometrical basis for numerical hydraulic and hydro-mechanical analyses in fractured rocks. However in the previous studies reported in the literature about fluid flow simulation and reservoir production in the hydrocarbon fields, either DFN models have not been consider or generated fracture pattern were based on the field mapping data from outcrops. In this research work, based on the recorded discontinuities information in a deep well in South Pars Gas field by Formation Micro Imager (FMI) logging tool, firstly natural and induced fractures are recognized. Using stereographic network analysis, number of natural fracture set is identified. The most fitted distribution function for all fracture properties of four fracture zones (K1 to K4) are evaluated and a large number of stochastic DFN models are generated. A William-Watson statistical test is run to find the most compatible generated fracture networks with the mapped fracture pattern in well for more efficient fluid flow modeling and quantifying the reservoir production in this field. The stress dependency of permeability and reservoir production (depth effect) are based on the dominant induced fracture orientations which give the direction of major and manor principle stresses around the well. Universal Distinct Element Code (UDEC) is facilitated for hydraulic and hydro-mechanical analyses in this thesis and statistical distribution of reservoir production for more favorable zone, K4 are calculated. The results show that the average values of hydrocarbon flow rates in K4 unit taken by hydraulic and hydro-mechanical numerical simulation are 3.424×10 -2 and 7.710×10 -4 m 3 /s respectively. Calculated average flow rate of hydro-mechanical analyses in this research work is close to the reality and compared with the determined flow rate in one of the adjacent wells in this field, only 15% error is approximated. Key Words Natural Fractures, Discrete Fracture Networks, hydraulic and hydro-mechanical analyses, South Pars Gas field, FMI