4D reservoir volumetrics: a case study over the Izaute gas storage facility
Dr Duncan Bate (ARK Geophysics) presents a recent oil company sponsored case study in France pointing to the value of 4D microgravity modelling as a cost-effective and unintrusive way of enhancing reservoir information. Measuring the Earth’s gravitational field as a means to mapping subsurface density distributions has been an exploration technique for decades. Gravitational anomalies may be related to subsurface geological structures and thus indicate features including faults, intrusives, and salt structures. High spatial resolution gravity data (as acquired with 3D seismic surveys) may be used in combination with 3D modelling to help define end member salt volumes in seismic imaging processes such as pre-stack depth migration (PSDM). The measurement of the time varying (4D) gravity field as a method of observing subsurface fluid flow in a hydrocarbon reservoir is a more recent application of gravity data. ARK Geophysics (ARK) conducted a time-lapse microgravity field trial over the Izaute gas storage reservoir in the Gascoigne district of South-Western France approximately 60 km north of Pau, between January 2003 and June 2004. The reservoir is operated by Total Infrastructure Gaz France (TIGF). The project was coordinated by ITF and funded by Statoil, Shell, and Total. The reservoir is a domestic gas storage buffer – filled throughout the summer months and depleted during the winter. The objective of this trial was to investigate whether time changes in the observed gravity field could practically be measured and if so, whether the observations correlated to known changes in gas stock levels. Twelve repeat gravity surveys were performed (each comprising over 200 stations) and two continuous gravity meters were employed. Overall the measured time-lapse gravity data correlate well with the predicted gravity changes from the reservoir model. Any observed differences between the predicted gravitational response of the reservoir model and the measured time varying gravity field requires further investigation. Sources of such observed discrepancies include system noise but also geological information that may not have been captured in the reservoir model such as reservoir compartmentalization, porosity, and permeability differentials as well as unmapped thief zones. The continuous data recorded over the reservoir show a good correlation with the reservoir gas pressure data. The aim of this study was to assess whether the time varying gravity field generated by a propagating density interface (e.g. oil-water or gas-water contact) can be measured by currently available cost-effective instrumentation. A two-fold approach was employed, utilizing time-lapse (or dynamic) and continuous gravity acquisition techniques. A dynamic gravity project comprises repeat static gravity surveys and is sometimes referred to as time-lapse or 4D. Workers in this field have successfully applied this technique to hydrothermal energy exploration (Hallinan et al., 1989; Allis et al., 1986) and hazard monitoring (Rymer & Brown, 1986; Yokoyama, 1989). A continuous gravity survey involves recording the variations in gravity continuously over a period of time. This technique has previously been used for long term studies of earth tides and volcano monitoring. There are several designs of gravity meter that can measure to the required accuracy for this project. The LaCoste and Romberg G and D meters with Aliod electrostatic feedback nulling system were used. This system has a range of approximately 100 mgal (+/-50) with a resolution of 0.01, or 0.001 mgal depending on which version is fitted. The Izaute, and its neighbouring Lussagnet, gas storage facilities have been created from porous submolassic sand reservoirs in the upper section of a vast multi-layer reservoir that contains the ‘inframolassic aquifer’ stretching from the Pyrenees to north of Bordeaux. The Izaute reservoir is formed by a gently plunging anticline, with an overburden of approximately 500 m. The reservoir thickness ranges from 50 to 80 m, has a porosity of between 25% and 35% and a permeability of between 6 and 20 Darcys. Gas is injected in the summer months (May to October) and recovered in the winter months (November to April). The total seasonal variation in the depth to the gas-water contact is 24 m.