ABSTRACT This paper presents the results of an integrated reservoir management study of a large carbonate reservoir. Geostatistical and scale averaging tools were used to develop detailed geologic models and a simulation window model. The geostatistical simulation techniques generated realistic models of lithology, porosity, and permeability. The public-domain GSLIB software, with some custom programs for data manipulation and interactive variogram modeling, allowed for the timely and efficient construction of a multi-million cell geological model. A two-step approach to permeability modeling was successful; it allowed the straightforward integration of core and well test data and provided improved geologic models for use in history matching in flow simulations. Because of the large number of cells in the geologic model, the porosity and permeability models were scaled up for use in the simulation model. A calibrated power-law average approach to scale-up was found to work well. The geostatistical tools used in this study are applicable to both siliciclastic and carbonate reservoirs. The two-step approach to permeability prediction is applicable whenever a significant difference exists between core-based permeability measurements and production-scale permeability. The authors suggest that such an integrated approach to reservoir management is widely applicable and yields good results. This paper presents an overview of the methodology and techniques used to develop the geologic and simulation models for the area being studied.