Data assimilation of a legacy 4D seismic in a brown field

Abstract

The quantitative use of four-dimensional (4D) seismic data in an assisted history matching process is still a major challenge in the industry. The problem becomes even more challenging when the quality of the seismic is deficient. Legacy 4D seismic data, i.e. 4D data from seismic acquisitions that were not originally parameterized for time-lapse processing, are typically regarded as not suitable for quantitative approaches. On the other hand, this kind of data are very common in the petroleum industry. Therefore, there is a practical interest in determining whether these data can bring useful quantitative information to improve reservoir characterization through assisted history matching. In this work, we investigate the use of low-quality 4D seismic data (legacy surveys with low repeatability) in an ensemble data assimilation process. We use data from a carbonate reservoir in Campos Basis with a long production history (38 years). The seismic data are the acoustic impedance from a joint 4D inversion. To overcome the overall low quality of the data, we propose to assimilate the seismic data only within the interpreted anomalies regions. We compare the data assimilation results considering only production data and different configurations combining production and seismic data. The results show that we were able to simultaneously assimilate production and seismic data. In our best case, we obtained a production data match with the same quality of a case using only production data, while significantly improving the seismic data match.