An innovative technique resolves apparent anomalies in detection of gas zones based on neutron—density logs

Abstract

Cross-over effect on neutron—density logs often indicates gas-bearing sands and/or carbonates. Identification of gas based solely on the neutron—density response can be misleading because some gas producing formations may not display a significant (or any) neutron—density cross-over, and some oil-bearing formations may falsely exhibit an appreciable neutron—density cross-over. False gas effects arise from either silt/shale in the reservoir, or from a significant difference between the reservoir matrix density and the standard sandstone/limestone matrix scale generally employed on the density log. True neutron—density gas cross-over effects occur in non-argillaceous formations when the matrix density of the reservoir closely matches the matrix density on the bulk density log. Examples of false gas effects are illustrated in two newly discovered fields, H-field and K-field, in Upper Assam, India. The reservoir in H-field produces gas from a silt—sand sequence, and when the percentage of silt in the reservoir approaches a certain upper limit (about 18%), no cross-over effect is seen in gas-bearing sands on the neutron—density log. An equation has been developed to correct for the presence of silt before making inferences about the presence of gas from neutron—density logs. The second field, K-field, produces oil from thin sands that are typically less than 3 m thick, and false gas effects can be seen on the neutron—density log. The false gas cross-over is caused by a reservoir matrix density that is considerably lower than the standard non-argillaceous sandstone matrix density scale on the bulk density logs. A combination of cross-plots and laboratory measurements is used to derive a modified matrix scale that corrects for false gas effects in the K-field reservoir. These two field case examples in Upper Assam, India, demonstrate that interpretations from wireline logs may be misleading and that correct log analysis requires integration of geology, subsurface test results and production data.