Mapping thin targets such as carbonate stringers is a major challenge in all fields of Kuwait. The target reservoir is three Upper Jurassic carbonate units impeded in anhydrite sequences of the Gotnia Formation at approximately 10,000 feet below the surface. Available data consists of a newly re-processed 3D high density single sensor seismic volume and borehole measurements from eight wells existing within the study area. The zone of interest is an upper Jurassic that has Hith Formation underlies Makhul and overlies Gotnia Formations.  
A conventional method that relies on extending the limits of seismic resolution such as inverted acoustic impedance won’t properly estimate the thicknesses. We present a new approach to the joint inversion based on facies classification. The study started by conducting joint impedance and facies post stack inversion where we rely on depth trends with no variograms and demonstrate the procedure using a 3D seismic survey acquired over a one-thousand square kilometers. This method inverts for facies and for Impedances-per-facies has been applied to by solving for absolute property models which are comparable to impedances from well log data.  
The quality of the inversion was evaluated by displaying inversion results at the well location overlaid on the corresponding real log curves. It gave a significant increase in quality over modelbased Inversion and gave a presentation of the data in a form that looks geologically plausible. In addition to, facies-based inversion results show higher resolution and support more accurate interpretation which in turn facilitates better thickness maps of the thin carbonates within the three anhydrite layers.  
This paper shows how the stringers impeded in anhydrite layers were delineated by facies-based joint impedance technique and reasonably matching with the facies generated from wells. The promise of a facies-based inversion is not only it resolves the thin bed layers, but also it honors the elastic properties of each layer. We found that structure-oriented filter better handles noise and helped in reducing them, especially on Burgan structural high. Moreover, the utilized joint inversion algorithm proves to be capable of producing both acoustic impedance and facies volumes with a great efficiency and a high resolution, which will make a significant impact on thin beds mapping of this field and similar ones. Finally, we would recommend before applying such advanced technique to have all the interpreted horizons re-assessed, the microlayers of the solid models to be built with accuracy, the essential geophysical logs reviewed and inversion steps iterated toward the best result. Such approach can be used as a diagnostic tool in the exploration for hydrocarbons in thin bed