This study applies the stochastic seismic inversion algorithm to delineate the reservoir (Sand) from non-reservoir (Silt and Shale) of Upper Gharif which is poorly delineated by using deterministic seismic inversion in Western Flank. The challenge of Upper Gharif sand detection and delineation away from wells is due to thin bed nature and low contrast in acoustic impedance of sand and shale of Upper Gharif. Stochastic seismic inversion study was applied by using three partial stacks and four wells. One well was used as a blind test for the result. The rock physics feasibility study was conducted to find optimum petrophysical parameters to discriminate between sand and shale facies in Upper Gharif. At well log level the P-impudence vs. Vp/Vs ratio shows good discrimination of different facies. The well petrophysical data were QCed and conditioned to carry out the study. The QC includes logs correction for Vp, Vs and density where there is a washout at Upper Gharif level. A fluid substitution analysis was conducted to assess seismic response for fluid content by using Gassmann fluid substitution. The result shows that the seismic is less sensitive to fluid content. The deterministic seismic inversion poorly resolves the sand distribution. The stochastic seismic inversion was carried out by using wells and 3D seismic jointly where a facies lithological logs probability density function (PDF) were included in inversion calculation. It shows good results to detect the sand distribution. The study shows that the application of stochastic inversion in the area can contribute to other studies related to Upper Gharif reservoir such as reservoir architecture and distribution. The study suggests applying stochastic seismic inversion algorithm in other Upper Gharif field where there is low contrast in acoustic impedance of shale and sand.