Improving the translation of 4D seismic attributes into reservoir properties

The optimization of the quantitative seismic data interpretation process is a key factor for reservoir<br>characterization and monitoring. In this paper, we show the results of a methodological investigation where Neural Network and Discriminant Analysis, two possible tools to incorporate the 4D seismic attributes in the quantification of saturation changes during the production of a reservoir, where analyzed. A special focus was applied to the evaluation of the impact of the quality and the amount of the calibration data to the final result, showing the<br>pseudo-well technology as a good alternative to improve the process.