Online Journal for E&P Geoscientists

Banerjee, Tapan K.^*1; Haider, Abdul A.¹; Al-Khalifa, Nasser F.¹
(1) Kuwait Oil Company, Kuwait, Kuwait.

Initial water Saturation (Swi) is one of the key parameters that strongly influence the estimation of original oil in place. Correct interpretation and modeling of Swi is thus critical and very challenging particularly in case of carbonate reservoirs. In Umm Gudair field of Kuwait, historically several methods of Swi calculation has been adopted for Minagish Oolite reservoir. This paper briefly summarizes the different approaches adopted for Swi modeling and their related limitations.

A simple log derived Swi model was interpreted initially which was severely constrained by limited “m” & “n” data. Subsequently, a sensitivity study has been carried out with additional core petrophysical data. An approach to calculate log derived saturation height function (SHF) based on different porosity bins as well as zones were attempted mainly for 3D modeling. The next generation study included HPMI capillary pressure data from core. It was used to generate saturation height function (J function) from capillary data for five empirical rock types. It was also not perfect due to low confidence of interpreted permeability along the well section. Subsequently, in another attempt, core derived hydraulic flow units were directly combined with core capillary curves in order to estimate Swi. Though method was accurate, it was restricted by limited core data points and eventually failed to capture reservoir heterogeneity in true sense.

In the present study, linear regression technique is applied to log derived SHF in order to generate suitable regression coefficients for individual zones. The function is well constrained with representative core derived J functions. Swi is subsequently interpolated for each cell within the 3D grid using its relative position with respect to cell height and porosity. As the real data control on effective porosity for MO reservoir is quite satisfactory, the coefficients based on actual curve matching technique it is by far the most reliable method of Swi modeling applied in UG field.