The goal is to improve the industry's ability to quantitatively estimate reservoir parameters from geophysical data, thereby reducing the cost and improving the efficiency of oil and gas exploration efforts.

Objectives: The objectives are to: (A) investigate the feasibility of simultaneously inverting different types of geophysical data linked through a rock physics model to produce a single self-consistent earth model parameterized by hydrologic parameters (porosity, saturations, etc.) rather than geophysical measurements, and if such an inversion is feasible, to (B) develop specific algorithms to use land and marine geophysical data to arrive at these parameters.

Lawrence Berkeley National Laboratory (LBNL) – project management and all research products

• Investigated the interdependences between reservoir and geophysical parameters and determined the accuracy required to estimate reservoir properties from geophysical parameters,
• Developed a rock physics model that links velocity, density and electrical conductivity to reservoir parameters (porosity, pressure and fluid saturations),
• Developed acoustic and elastic impedance and amplitude versus offset (AVO) inversion algorithms for seismic data,
• Determined system sensitivity and requirements for optimal electromagnetic (EM) imaging of reservoir fluid saturations using 3-D model studies of controlled source marine system,
• Tested different EM source-receiver configurations for optimal reservoir delineations using
   new 3-D EM inversion code,
• Developed and tested a stochastic joint seismic-EM inverse to investigate benefits of stochastic inversion versus deterministic inversion,
• Produced a joint EM-Seismic inversion code for use with marine and land surface seismic and controlled source EM data, and
• Prepared papers for publication in Geophysics and for presentation at the 2004 Society of Exploration Geophysicists (SEG) annual meeting.

Detailed knowledge of fluid properties in oil and gas reservoirs can increase production and extend the life of proven reserves. Geophysical data sets (seismic, electromagnetic, gravity, etc.) do not individually have sufficient resolving power to provide the level of information required on fluid properties. However, previous studies indicate that when datasets are combined with a petrophysical model of the reservoir, detailed fluid properties can be extracted.

In the past, application of complementary data sets for identification of fluid properties has centered on an iterative, interpreter-driven process, where results from one data set were used to guide and constrain the interpretation of another. While this technique produces useful results, it is limited by a high degree of subjectivity and lacks the statistical measures needed to quantify uncertainties in the fluid properties of the reservoir. In this project, researchers will interpret complementary data sets by joint inversion/imaging, where fluid properties are calculated directly from the data by the mathematical process of inversion.

Researchers will consider various forms of geophysical data while designing the inversion codes, however, the focus of the project is on seismic, gravity and EM data acquired in the marine environment. The objective of the algorithm development is to create general codes capable of incorporating an appropriate petrophysical model to estimate reservoir fluid properties.

Thus far in the project, researchers have investigated using only seismic travel-time and event move-out data (eliminating the need to estimate the source waveform and source and receiver coupling). Full waveform seismic data will be considered as the project continues.

and Remaining Tasks:  Remaining tasks include evaluation of the merits and costs of using full waveform seismic inversion as compared to industry standard AVO inversion as the seismic component in an advanced joint seismic-EM inversion code. LBNL will develop a prototype 2-D cell-based full waveform seismic inversion code for this comparison.

$358,000

$0

NETL – John D. Rogers (john.rogers@netl.doe.gov or 304-285-4880)
LBNL – Gregory A. Newman (gnewman@lbl.gov 510-486-6887)