Kansas Geological Survey, Current Research in Earth Sciences, Bulletin 241, part 3
Prev Page--Start || Next Page--GPR Data Acquisition and Processing

Introduction

Pennsylvanian, cyclothemic, carbonate strata that crop out in northeastern Kansas are direct analogs for stratal units that are petroleum reservoirs in central and western Kansas. Although the reservoirs have been successfully produced for many years, they still contain significant accumulations of hydrocarbons. New methods are being sought to identify thinner zones and better understand reservoir architecture at increasingly finer scales. Many of the reservoirs have multiple pay zones within the complex reservoir heterogeneity that are affected by poorly understood facies changes, stratal geometries, diagenesis, and interbedded shales and sandstones that provide seals to reservoir facies (Newell et al., 1987).

Outcrop studies of reservoir analogs can provide important insights into complex reservoir architecture. Increasingly, outcrop studies of Pennsylvanian strata in Kansas integrate modern concepts of sequence stratigraphy (e.g., Watney et al., 1989; Feldman and Franseen, 1991) with other methods, such as high-resolution seismic data (Miller et al., 1995; Franseen et al., 1995), in order to better understand geometries, facies patterns, and diagenetic trends, not only at the cyclothem scale, but also of strata within individual cyclothems. Such studies, focusing on higher resolution, can aid in better understanding reservoir architecture and controls, especially in smaller-scale reservoirs where general patterns of stratal geometries, thicknesses, and diagenetic trends, which do not correlate with structure are important in controlling reservoir development (e.g., Brown, 1963; Watney, 1980).

Seismic-reflection data provide general information about subsurface structures and usually image features greater than 10-15 m (33-49 ft) in thickness. Well data provide very detailed information concerning the area immediately around the well. Outcrop-analog studies are an important facet of reservoir studies because they bridge the gap in resolution between seismic and well information and allow detection of small-scale lateral variations of detailed stratigraphic architecture, which can affect oil recovery and may be missed by using only seismic or well data.

Ground-penetrating radar (GPR) is a near-surface, non-intrusive geophysical technique similar to seismic reflection that images the subsurface at a much higher resolution (Beres et al., 1995; Bridge et al., 1995; Gawthorpe et al., 1993; Jol et al., 1996; Liner and Liner, 1995; Martinez et al., 1996; Pratt and Miall, 1993). In addition to the high-resolution imaging, GPR is a potentially useful method in outcrop studies because it can provide stratal information in poorly exposed areas and three-dimensional stratal characteristics beyond the outcrop face (e.g., Beaty et al., 1997). Because GPR profiles are usually gathered as common offset data, minimal digital-signal processing is involved when compared to shallow seismic-reflection methods. Data collection is rapid and non-invasive, allowing profiles to be collected easily and quickly without altering a study site. The digital data collected are easily manageable on computer workstations with software developed for the petroleum industry to interpret seismic data. Collection of multiple profiles or three-dimensional grids of GPR data at study sites allows outcrop studies to be extended into the third-dimension, sometimes greatly enhancing understanding of stratigraphic architecture (Beaty et al., 1997). Grids of GPR data can also be used to supplement and connect discrete core information.

Typical GPR frequencies for stratigraphic studies range from 10 to 100 MHz, resulting in vertical imaging resolutions of 1.5-1.0 m (4.9-3.3 ft) (Beres et al., 1995; Bridge et al., 1995; Dominic et al., 1995). Although such resolution is sufficient for targets that are relatively large or laterally extensive, it may be insufficient for imaging detailed stratigraphy needed for some reservoir studies. Use of higher-frequency antennas (e.g., greater than 200 MHz) increases resolution enough to image fine-scale stratigraphic variations, but penetration depths decrease due to signal attenuation, which increases with frequency (Davis and Annan, 1989).

The purpose of our study was to determine the usefulness and limitations of GPR as an additional tool in characterizing Pennsylvanian, carbonate, reservoir-analog outcrops at a high resolution. To date, only a few studies have examined the usefulness of GPR as a stratigraphic tool in carbonate strata (Pratt and Miall, 1993; Liner and Liner, 1995). These studies focused on thick, relatively homogeneous carbonate packages. Our study differs in that we imaged interbedded carbonate and shale strata (0.2-5 m; 0.7-16.4 ft), and the carbonate beds variously contained thin (less than 0.01 m; 0.03 ft) shale layers.

The study outcrops were divided into several different units based upon GPR reflections and stratigraphy. The interpreted data show excellent correlation between stratigraphic surfaces seen on the outcrop face and those imaged via high-frequency GPR. In this study GPR imaged subsurface features as thin as 0.1-0.2 m (0.3-0.7 ft). However, data quality was severely diminished in some areas by significant signal attenuation, which was caused by either shales or clay-rich soils at the surface. The results of our study indicate that GPR is a useful method for imaging and adding to outcrop studies of carbonate-reservoir analogs. Our study also provides information on the limitations of GPR in the study of cyclic strata composed of interbedded carbonate and siliciclastic strata and gives direction for future studies.

Prev Page--Start || Next Page--GPR Data Acquisition and Processing

Kansas Geological Survey
Web version September 15 1998
http://www.kgs.ku.edu/Current/1998/martinez/martinez2.html
Comments to webadmin@kgs.ku.edu