High-Resolution 3-D and 4-D Imaging of the Near Surface II: Innovative Wave-Based Techniques
Presiding: A Pidlisecky, Stanford University; T Barber, Schlumberger
NS34A-01 INVITED 15:30h
3D and 4D Seismic Imaging in the Oilfield; the state of the art
Seismic imaging in the oilfield context has seen enormous changes over the last 20 years driven by a combination of improved subsurface illumination (2D to 3D), increased computational power and improved physical understanding. Today Kirchhoff Pre-stack migration (in time or depth) is the norm with anisotropic parameterisation and finite difference methods being increasingly employed. In the production context Time-Lapse (4D) Seismic is of growing importance as a tool for monitoring reservoir changes to facilitate increased productivity and recovery. In this paper we present an overview of state of the art technology in 3D and 4D seismic and look at future trends. Pre-stack Kirchhoff migration in time or depth is the imaging tool of choice for the majority of contemporary 3D datasets. Recent developments in 3D pre-stack imaging have been focussed around finite difference solutions to the acoustic wave equation, the so-called Wave Equation Migration methods (WEM). Application of finite difference solutions to imaging is certainly not new, however 3D pre-stack migration using these schemes is a relatively recent development driven by the need for imaging complex geologic structures such as sub salt, and facilitated by increased computational resources. Finally there are a class of imaging methods referred to as beam migration. These methods may be based on either the wave equation or rays, but all operate on a localised (in space and direction) part of the wavefield. These methods offer a bridge between the computational efficiency of Kirchhoff schemes and the improved image quality of WEM methods. Just as 3D seismic has had a radical impact on the quality of the static model of the reservoir, 4D seismic is having a dramatic impact on the dynamic model. Repeat shooting of seismic surveys after a period of production (typically one to several years) reveals changes in pressure and saturation through changes in the seismic response. The growth in interest in 4D seismic has been driven by the need for improved recovery on declining fields, and for newer fields the requirement for optimised production. The vast majority of 4D activity is currently offshore based on the economic imperative in those environments. Much effort is being employed in the acquisition and processing domains to improve the resolution of the method, and to move from a qualitative to quantitative assessment of reservoir changes. Key developments in this field have been the improvements in data quality attained by new acquisition systems, growing maturity in processing of repeat datasets and the integration of seismic information into the reservoir management workflow. For the future there is much interest in the emplacement of semi-permanent seismic arrays to monitor reservoir changes on a very frequent basis (months rather than years). These arrays would also enable passive monitoring of reservoir changes between repeat seismic surveys. Despite the maturity and success of the 3D seismic method in the oilfield there are remaining challenges. Improvements to imaging schemes are required to address the need for ever more accurate subsurface images and the desire for improved reservoir performance is driving 4D technology forward. For the future we can expect even greater data densities than are currently acquired and developments in imaging based around increased computational resources.
NS34A-02 15:50h
Integration of GPR and Laser Position Sensors for Real-Time 3D Data Fusion
Non-invasive 3D imaging visualizes anatomy and contents inside objects. Such tools are a commodity for medical doctors diagnosing a patient's health without scalpel and airport security staff inspecting the contents of baggage without opening. For geologists, hydrologists, archeologists and engineers wanting to see inside the shallow subsurface, such 3D tools are still a rarity. Theory and practice show that full-resolution 3D Ground Penetrating Radar (GPR) imaging requires unaliased recording of dipping reflections and diffractions. For a heterogeneous subsurface, minimum grid spacing of GPR measurements should be at least quarter wavelength or less in all directions. Consequently, positioning precision needs to be better than eighth wavelength for correct grid point assignment. Until now 3D GPR imaging has not been practical: data acquisition and processing took weeks to months, data analysis required geophysical training with no versatile 3D systems commercially available. We have integrated novel rotary laser positioning technology with GPR into a highly efficient and simple to use 3D imaging system. The laser positioning enables acquisition of centimeter accurate x, y, and z coordinates from multiple small detectors attached to moving GPR antennae. Positions streaming with 20 updates/second from each detector are fused in real-time with the GPR data. We developed software for automated data acquisition and real-time 3D GPR data quality control on slices at selected depths. Standard formatted (SEGY) data cubes and animations are generated within an hour after the last trace has been acquired. Examples can be seen at www.3dgpr.info. Such instant 3D GPR can be used as an on-site imaging tool supporting field work, hypothesis testing, and optimal sample collection. Rotary laser positioning has the flexibility to be integrated with multiple moving GPR antennae and other geophysical sensors enabling simple and efficient high resolution 3D data acquisition at different GPR frequencies and polarizations. Existing GPR systems can be upgraded with a simple start-of-trace trigger modification and an investment similar to the cost of a standard GPR unit. This opens new opportunities for wider application of fully resolved 3D and 4D imaging in the near surface environment.
http://www.3dgpr.info
NS34A-03 16:05h
The Integration of GPR, GIS, and GPS for 3D Soil Morphologic Models
Ground-Penetrating Radar (GPR) has become a useful and efficient instrument for gathering information about subsurface diagnostic horizons in Florida soils. Geographic Information Systems (GIS) are a popular and valuable tool for spatial data analysis of real world features in a digital environment. Ground-Penetrating Radar can be linked to GIS by using Global Positioning Systems (GPS). By combining GPR, GPS, and GIS technologies, a more detailed geophysical survey can be completed for an area of interest by integratinghydrologic, pedologic, and geologic data. Thus, the objectives of this research were to identify subsurface soil layers using GPR and their geographic position with a highly accurate GPS; to develop a procedure to import GPR data into a popular software package, such as ArcGIS, and; to create 3D subsurface models based on the imported GPR data. The site for this study was the Plant Science Research and Education Center in Marion County, Florida. The soils are characterized by Recent-Pleistocene-age sand over the clayey, marine deposited Plio-Miocene-age Hawthorn Formation which drapes the Eocene-age Ocala Limestone. Consequently, soils in the research area vary from deep quartz sands (Typic Quartzipsamments) to shallow outcrops of the Hawthorn Formation (Arenic Hapludalfs). A GPR survey was performed on a 160 m x 320 m grid to gather data for processing. Four subsurface models estimating the depth to argillic horizon were created using a variety of specialized GPR data filters and geostatistical data analyses. The models were compared with ground-truth points that measured the depth to argillic horizon to validate each model and calculate error metrics. These models may assist research station personnel to determine best management practices (including experimental plot placement, irrigation management, fertilizer treatment, and pesticide applications). In addition, the developed methodology exploits the potential of combining GPR and GIS.
NS34A-04 INVITED 16:20h
New developments in high-resolution multicomponent georadar imaging
Scalar imaging algorithms originally developed for processing seismic reflection or remote sensing data are commonly used for imaging georadar data. Unfortunately, such algorithms do not account for the radiation characteristics of georadar antennas and the vectorial nature of radar waves. Consequently, georadar images obtained from such algorithms are often not optimally focused, exhibiting the effects of antenna radiation and wave propagation. Multicomponent imaging is a viable technique for taking into account the electromagnetic properties of georadar data, but it therefore requires measuring two components of the electric field. In a three-dimensional (3-D) high-resolution multicomponent survey, it is crucial to measure trace positions precisely while maintaining survey efficiency. We achieve this by linking a multichannel georadar system to a high-quality differential GPS unit. This setup allows us to record two components of the electric field and accurate position information simultaneously, without the need to adhere to a fixed grid during the survey. Using this setup, we have collected two pairs of co- and cross-polarized 3-D data sets on a 40x40~m area in a braided river environment. The relative trace positions obtained after post-processing the GPS data were mostly accurate to ±0.1~m. From these data, we derived two independent multicomponent 3-D images. Our original multicomponent imaging algorithm was based on far-field approximations of the radiation patterns for infinitesimal dipole antennas. To describe the effects of wave propagation more accurately, we have extended the multicomponent imaging algorithm to take into account the total antenna radiation characteristics and antennas of finite length. This should result in more accurate images, especially for the dipping, quasi-planar structures that dominate our data. Computational efficiency has previously been a major problem that inhibited using the total antenna radiation characteristics. Our new implementation is reasonably efficient, because it is based on wavenumber-frequency domain formulations of the Green's functions that describe directional antenna radiation and wave propagation. We thus avoid evaluating the corresponding space-frequency domain integral expressions of the Green's functions. The performance of the new total-field imaging operator has been tested on finite-difference time-domain modeled data containing variously oriented dipping reflectors. Compared to far-field images, the total-field images of these synthetic data show superior amplitude and phase (i.e., location) consistency of the imaged reflectors. An analysis of the total-field images obtained from the braided river field data and comparison with the corresponding far-field images revealed similar enhancements of the image quality. These results indicate that total-field multicomponent imaging of georadar data can increase our confidence in the interpretations of reflector locations and reflection amplitudes.
NS34A-05 16:40h
Investigation of the Near Subsurface Using Acoustic to Seismic Coupling
Agricultural, hydrological, and civil engineering applications have lead to a need for high resolution information of the near subsurface over large areas. In order to obtain this high resolution data over a large area, the measurement technique must be highly mobile with a short acquisition time. For this reason, ground penetrating radar (GPR) has seen much use for these applications. We propose to use acoustic to seismic (A/S) coupling to investigate the mechanical properties and stratification of the near subsurface down to a depth of a few meters. In this technique, seismic waves are created via the coupling of airborne energy into the ground surface. An airborne source has the benefit of being non-contact (mobile) and allowing for high frequency (high resolution) investigation of the subsurface. Sensors, either geophones or laser doppler vibrometers, record the resulting solid particle motion at the surface due to energy reflected from interfaces between dissimilar media at depth. Interference between waves produce a frequency response curve containing information about the distribution of mechanical properties of the near-surface soil. We will discuss the results of the A/S coupling measurements in relation to shallow seismic and cone penetrometer measurements obtained at an agricultural field site and how these measurements relate to a spatially variable genetic hard pan