Reservoir simulation is established as a good practice to make the best decision for a petroleum reservoir. The reservoir is characterized in terms of reservoir elements such as structural model, well data, rock and fluid properties. Then the reservoir model is enhanced through history matching and finally different prediction scenarios are tried to find the best plan for the reservoir understudy. The more accurate the reservoir is characterized, the faster and the more precisely the history match is finished and the more reliable predictions are obtained. The most important part of reservoir characterization is the rock typing, where the quality of CCAL (conventional core analysis) and SCAL (special core analysis) properties are evaluated and estimated for any simulation grid. The resulting oil in place must be confirmed by the OOIP (original oil in place) calculated based on average petro-physical parameters for any layer. To allocate different rock types to simulation grid, rock types should be assigned according to different ranges of rock differentiation parameter which has to be determined in any specific study. Based on our experience in Iranian carbonate reservoirs, most frequently irreducible water saturation is the rock differentiation parameter. In the oil zone, water saturation from log data is assumed to be the irreducible water saturation. Thus, the rock type is identified with no trouble. The problem arises in transition zone, where water saturation from log data is not equal to the irreducible water saturation of that rock. This study includes the observed variations in terms of water saturation data versus depth and how to assign rock types to the transition zone grids. The objective of the capillarybased method is to produce a water saturation map which honors laboratory data as well as the well log data and considers the depth so that it can handle the transition zone in a proper manner. In fact, novelty of this work is to explain how it is possible to consider log and capillary pressure data together so that the most accurate rock type is assigned to reservoir grids of the transition zone. Moreover, this method is consistent with equilibration method for initializing reservoir simulations. A procedure is presented for how to implement the capillary-based method in a stepwise manner. Once the proposed method is carried out, the initialized simulation model is consistent with all sources of data (core analysis and petro-physical data). In this procedure original oil in place calculated after the initialization for simulation is more accurate and can be crosschecked with volumetric calculation based on interpreted log data. Therefore, it is considered to facilitate subsequent stages of reservoir study, namely history match and prediction.

No abstract