A Comparative Study on Appraisal of Lithological Boundaries and Litho-Models Using Well Log Data at Bhogpara Oil Field in Assam of NE India

Abstract

Accurate assessment of bed boundaries at a drill site is a crucial step for reservoir characterization. Generally, the interpreters in oil industries carry out such analysis through expensive softwares based on their experiences and geological knowledge in the study area. However, identifying bed boundaries using traditional interpretation is tedious. In this study, we adopted a synergistic approach of Walsh transform and a bed boundary demarcation technique to discriminate the lithological boundaries in Bhogpara oil field of Assam in NE India. This bed boundary detection technique was tested over self-potential (SP), gamma ray (GR) and laterolog deep resistivity (LLD) logs. Initially, we detected the possible lithological boundaries from GR log variations and hence proposed a lithological model as per the knowledge of traditional log interpretation techniques. Subsequently, Walsh low pass filter was applied to the self-potential (SP), GR and LLD logs to obtain their step function. Afterward, the step functions of logs were processed through the boundary detection algorithm to find out the possible thick and thin beds. Further, bed correlation was made among the Walsh-detected boundaries for the studied wells of the Bhogpara oil field. Finally, a bed boundary model was generated using the boundaries picked from the Walsh-based approach. Our analysis exhibit that the model generated from Walsh boundaries more explicitly envisaged the possible heterogeneity. Traditionally we discriminated the minimum bed thickness of the order of 5.15 m, whereas, from the Walsh-based approach we obtained thin bed of the order of 3.2 m. This proposed bed boundary assessment technique is also helpful in understanding the position of subsurface rock layers as well as possible thick and thin beds when we don’t have any pre-received core information.