SEISMIC HYDROCARBON INDICATORS LOWER RISKS

Otto Welper
Paramount Petroleum Co. Inc.
Houston

James L. Allen
Vulcan Exploration Inc.
Houston

Gus Fiongos
Discovery Oil & Gas
Houston

Recent advances in seismic hydrocarbon indicator (HCI) technology have been responsible for lowering exploration risk and opening new trends and stratigraphic plays in mature areas. The use of new HCI technology has led to a wildcat success rate greater than 50% in some trends.

For example, since 1987 over 300 bcf of gas has been discovered in stratigraphic traps in the Yegua/Cook Mountain (Eocene) sands along the Texas and Louisiana onshore Gulf Coast by using HCI analyses such as amplitude vs. offset (AVO) evaluation. A number of new fields have been found onshore by using other HCI techniques in the shallow Miocene sands of Texas and Alabama and the shallow Frio sands of Texas and Southwest Mississippi.

HCIs were originally developed offshore with great success due to the better quality seismic data available in that environment. The technology is becoming more commonly used onshore due to several factors. Seismic acquisitions and processing techniques have advanced rapidly in the last few years, dramatically improving onshore data quality.

The principal onshore data problems are caused primarily by increased weathered zone effects and cultural problems (houses, roads, rivers, etc.) interfering with ideal seismic acquisition and processing techniques. Also, an increasing number of HCI-related software products are now available for industry in-house computers, providing low-cost access to this technology.

Effective use of the advanced HCI techniques is resulting in lower exploration risk and new field discoveries, thereby lowering the finding costs for those companies that use them.

HCI technology is a method for directly interpreting gas-filled reservoirs in the subsurface by integrating borehole (well) and surface (seismic) data. The advanced techniques now being utilized involve detailed analyses of the four basic aspects of the seismic wavelet: amplitude, velocity, frequency, and phase. This paper is a discussion of the physical nature and proper application of these aspects in the search for hydrocarbons.

The high degree of success the technology has encountered in some trends is due to its ability to qualify with reasonable accuracy the three key elements of a field. The techniques can determine the existence and extent of (1) the hydrocarbon trap (the limit of the HCI should be the limit of the trap), (2) reservoir quality (since in many trends a thick, high porosity reservoir will cause an HCI anomaly), and (3) hydrocarbon accumulation (since the presence of gas is the usual cause of an HCI).

AMPLITUDE

Gas reservoirs can often be seen as high amplitude events (bright spots) on good quality seismic data. Bright spots become even more apparent relative to other seismic events when viewed on a relative amplitude processing (RAP) display.

This RAP display exhibits the true amplitudes of the seismic events throughout the section relative to each other. The traditional automatic gain control (AGC) processing attempts to equalize amplitudes of events to provide consistent reflectors for structural mapping but sacrifices some ability to distinguish relative amplitudes between separate events.

Geophysicists have been using limited bright spot technology onshore for at least 10 years, looking mostly for shallow Miocene and Oligocene gas reservoirs. Success rates exceed 60%. Recent advances in acquisition and processing technology allow geophysicists to see bright spots at greater depths and to qualify them by evaluating seismic attributes such as AVO.

Used properly, this technique often allows interpreters to distinguish gas sands from other possible bright spot causes, such as highly porous wet sands, hard streaks, or lignites. Therefore, the AVO technique allows explorationists to further reduce risk beyond drilling a well based solely on the existence of a simple bright spot.

An example of the effectiveness of this AVO technique is the wildcat success rate near 50% achieved by some exploration companies prospecting the Yegua and Cook Mountain (Eocene) sandstones along the onshore Texas and Louisiana Gulf Coast.

Fig. 1 illustrates the concept of a common reflecting point, often referred to as a common depth point (CDP) gather. The gather is the collection of all the traces generated by surface energy sources aimed at a common point in the subsurface.

The term "12-fold seismic data," for example, would consist of 12 different energy sources aimed at a CDP and then received by 12 corresponding recorders. The angle at which the energy arrives from the nearest source and is reflected to the nearest receiver is the same and is very small. The angle of incidence and equal angle of reflection increase as the distance increases.

The diagram shows how multifold data are acquired and demonstrates the relationship of different offset distances (offsets) in an individual CDP gather.

When seismic waves with various offsets reflect off a gas sand, each offset has a different amplitude response (reflection coefficient), depending on its angle of incidence. This response can increase, decrease, or remain the same across the range of offsets, depending on the lithology and fluid content of the sediments. This gather displayed by increasing offset distance (between seismic source and receiver) is what the geophysicists use in evaluating an AVO response.

An example can be seen in Fig. 2, which compares the model response to an actual seismic gather over a producing well. The event just below 2.1 sec shows good correlation from the model to the actual data. One would expect similar AVO responses in similar lithologies to indicate the presence of gas.

The physical parameters of the model can be varied to simulate different lithologies and make reasonable estimates of size and thickness of the gas column within the sand, estimates which are critical to evaluating the economics of a wildcat.

The response shown in Fig. 2 is the typical response (amplitude increase with increasing offset distance) that today's explorationist is using in the Eocene along the Texas Gulf Coast and the Sacramento basin of California.1

AVO response in a gas sand has been shown to be limited not only to an increase of amplitude with offset.2 Depending on the density and velocity contrast of the geologic section, the presence of gas may cause an amplitude decrease with offset3 a consistent high amplitude across all offsets (Paramount's experience in the Southeast Louisiana state waters area), or a more complex response. For example, a reflector on the stacked seismic in this situation may show up as a dim-out rather than a bright spot.

As AVO technology is refined further, these more complex responses should find increased use in prospecting not only for gas-filled sandstones but for hydrocarbons in other reservoir lithologies as well. For example, AVO technology has found applications in determining the presence of fracturing in carbonate rocks.

VELOCITY

The seismic velocity of a rock is the speed at which a seismic wave passes through it. Fig. 3 shows the effect that gas has on the acoustic wave as it travels through an Eocene sandstone in the Texas Gulf Coast.

The bright orange trough labeled "Tavener field" represents a decrease in the interval velocity of a gas sand within a sand-shale sequence.

For this analysis, the seismic traces have been converted to low frequency sonic logs. (The reverse of this process is to analyze a sonic log and create a synthetic seismic trace.)

The color presentation increases the dynamic range of the seismic response, making it easier for the eye to detect subtle changes. It highlights not only interval velocity decreases that might indicate gas but also changes in lithology and both interparticle and fracture porosity.

FREQUENCY

Frequency has been relatively ignored as an exploration tool, but in the last few years there has been increasing interest in its use.

Frequency is basically the pitch of a sound wave and is measured as number of times an event occurs in a second (cycles per second), expressed as hertz. Seismic frequency is measured from peak to peak along the seismic trace.

A 10 hz signal would peak 10 times in 1 sec; a 50 hz signal would peak or vibrate 50 times/sec. (Human ears are sensitive from approximately 40 to 17,000 hz.)

A significant difference between hydrocarbons and brine (salt water) is that as the acoustic wave propagates through a brine-filled reservoir, all frequencies will be transmitted, but if gas is present, it will tend to absorb the higher frequency portion of the seismic spectrum (approximately 20-60 hz).

With modern broadband seismic data, one is afforded the opportunity to widen the spectrum enough to have a chance to observe the frequency absorption caused by hydrocarbons. This tool can lower exploration risks by directly identifying gas-filled or combination gas/oil-filled porosity in both structural and stratigraphic traps.

A few seismic processing contractors in the last 2 years have begun to offer a frequency analysis of the stacked seismic traces or the gathers. An example of this is shown in Fig. 4. The two yellow anomalies in the middle of the section represent an absorption of high frequencies due to hydrocarbon accumulation.

Each contractor has a different approach, and there are still some theoretical problems in the statistical analysis of the seismic data, but the industry will be seeing more of this analysis in the future.

PHASE

Phase response is another attribute of the seismic wavelet affected by gas. Fig. 5 illustrates the polarity of a seismic wavelet.

In many cases, gas causes a lateral phase change between traces on the stacked seismic section.

Sometimes gas causes a complete polarity reversal phenomenon to occur, such as a negative wavelet (trough) abruptly changing laterally to a peak.

This effect can take various forms, many of which go unnoticed until a field is discovered and its rock properties are thoroughly studied.

Figs. 6 and 7 show an example of the phase response of an event under the influence of gas. Fig. 6 illustrates the geologic concept, and Fig. 7 represents a comparison of the modeled seismic expression vs. the actual stacked seismic.

There is an excellent correlation between the synthetic model and real seismic data. The model reveals that as the gas sand pinches out, the seismic response exhibits a change in phase that looks like a polarity reversal. Note that the modeled traces are generated by using a synthetic AVO gather then summing this gather to form one trace of the synthetic seismic line, as opposed to the conventional approach of creating a synthetic for the log on the right and having the computer generate the interpolated traces.

Since conventional modeling does not account for the AVO phenomena, it has become necessary to use this advanced modeling approach to duplicate the stacked response.

SUMMARY

This paper has touched upon the basic attributes that geophysicists use in the qualification and interpretation of seismic HCIs.

Companies using these technologies in combination with conventional geological/geophysical concepts are lowering their exploration risk, opening new plays in previously explored areas, and finding significant amounts of oil and gas.

REFERENCES

1. Weagant, Frank E., and Sterling, Robert H. Jr., "Seismic amplitude with offset discovery used successfully in Sacramento basin," OGJ, Vol. 87, No. 7, Feb. 13, 1989, p. 52.

2. Ostrander, W. J., 1984, "Planewave reflection coefficients for gas sands at non-normal angles of incidence," Geophysics, No. 49, 1984, pp. 1637-1648.

3. Rutherford, S. R., Williams, R.H., "Amplitude-vs.-offset variations in gas sands," Geophysics, No. 54, 1989, pp. 680-688.

Copyright 1991 Oil & Gas Journal. All Rights Reserved.