Logging Measurement of Residual Oil, Rangely Field, CO

Abstract

Summary This paper presents results of an effort to determine from well logs the amount of oil remaining in the Rangely field in Colorado after water flooding. Four logging methods for measurement of residual saturations were attempted: resistivity, pulsed neutron capture, high-frequency dielectric (electromagnetic propagation). and nuclear magnetism. Cores were analyzed to determine parameters needed for log interpretation and to establish the minimum possible oil saturation. The well was under reamed after the mud salinity was increased, and chemicals were added to the mud to cause the aqueous phase not to be detected by the Nuclear Magnetism Log (NML). Resistivity, pulsed neutron capture, and nuclear magnetism logs were run both before and after under reaming. Invasion of mud filtrate was not uniformly deep enough to enable accurate residual saturation determination by shallow-focusing resistivity or pulsed neutron capture logs. Because only the region very near the wellbore was invaded uniformly by mud filtrate, the shallow investigation of the nuclear magnetism and high-frequency dielectric tools made them the only logs that provided accurate measurement of the oil in place (OIP) after water flooding. Saturations were calculated from the Electromagnetic Propagation Tool (EPT) by using values determined from a zone below the oil/water contact for the dielectric properties of the rock matrix. Residual OIP calculated from the EPT averaged essentially the same as that from the NML. Comparison of errors identified for each of the measurement methods shows the NML is potentially the most accurate logging measurement of residual oil because it is the only method that measures oil directly. Introduction and Plan The Rangely field (CO) was discovered in the early 1930's and has been produced since the early 1940's. To assess the economic potential of proposed assisted-recovery efforts, it is necessary to know the quantity and distribution of the O[P after the water flooding that began in late 1957. The measurement methods described are based on analyses of routine cores (as distinguished from those taken with retained pressure) and of wireline logs recorded in open hole. Assessment of alternative methods for measurement of residual oil, such as pressure cores, partitioning tracers, and cased-hole log-inject-log, was deferred until the test established the degree of uncertainty from the less expensive measurements by openhole logs. The salinity of water in the formation at the time the well was drilled varied in an unknown way, from that of the freshest injected water (500 ppm NaCl) to that of the brine originally in the formation (115,000 ppm NaCl). To measure residual saturations, a plan was adopted to measure resistivities and thermal neutron-capture cross sections after the mixed-salinity brine in the region around the wellbore was replaced by mud filtrate. The well was thus logged with a Dual-Spacing Induction Log (DIL) and a Dual-Spacing Laterolog (DLL) both before and after under-reaming the hole, with mud having changed salinity. The well was cored through a 141-ft (43-m) interval and drilled to total depth using a relatively fresh (15,000 ppm NaCl) mud with a 30-minute fitter loss of 10 mL (10 cm 3). JPT P. 1735^