The development of 3-D seismic is the single most important technological breakthrough in many years1 . Some guidelines to separate reserves into the traditional classification of proved, probable, and possible using seismic methods have been developed and refinements to the current definitions proposed. These guidelines supplement the SPE reserve definitions and the SPEE (Society of Petroleum Evaluation Engineers) guidelines, which were published in the 80's and do not detail the use of seismic data in classification of reserves. Since the oil and gas industry is increasing the use of 3-D seismic, it is lime for the SPE to recognize the usefulness of 3-D seismic data in classifying reserves. Use of 3-D seismic data is especially important in classifying reserves in complex structural and stratigraphic traps with limited well and production data. The seismic methodology and thought process and typical interpretation steps are illustrated in two simulated case studies. Smarter, faster, cheaper, more profitable oil and gas development and reserve evaluation result from the use of 3-D seismic.

Recent advances in seismic technology, computing power, and software provide engineers and geophysicisls with the tools to make quantitative measurements of reservoir parameters. These measurements can be made with very dense spatial sampling over an entire reservoir. The oil and gas industry frequently relies on the use of seismic data to determine reservoir parameters in areas which are a significant distance from the wellbore. Seismic imaging methods are emerging as promising new measurement tools which can be used to bridge the information gap between wells and help constrain stochiastic models2 . The seismic method can be used to define the limits of a reservoir and to characterize the rock properties within the reservoir. These properties include thickness, continuity, lithology, net/gross ratio, porosity, permeability, fracturing, and water saturation.

The SPE reserve definitions and the SPEE guidelines are generally designed to classify reserves based on well data and production history. The area of a reservoir considered proved is generally limited to:

  1. Producing wells.

  2. Their immediate offsets.

  3. Locations where interpretations of well data indicate lateral continuity of the producing formation.

  4. The area above the lowest known hydrocarbons.

In 1981 the SPE permitted assignment of proved reserves on the basis of core or log data by using the analogous reservoir concept (the data must indicate the reservoir is hydrocarbon bearing and analogous to other productive reservoirs in the same area). This paper extends the analogous reservoir concept to include application of modem seismic techniques. Once the seismic attributes of the productive formations have been recognized and verified by drilling, how should the remaining undrilled areas identified by 3-D seismic interpretation be classified? If there are proved reserves associated with any wells that lie within a 3-D seismic survey, and if the reservoir characterization process is performed with due diligence by a team of geologists, geophysicists, and engineers, then a larger spatial area of the reservoir than that defined by well control should be classifiable as proved.

In the "gap" areas between wells, and in areas extending beyond well control, current SPE/SPEE classification methods do not consider the accuracy of parameter determination that is available through the use of 3-D seismic data. Table 1,3  and Figure 1 4  illustrate the success of predictions from 3-D seismic interpretations.

Table 1

Comparison of well log measured values of net porosity feet versus those predicted by the seismic-cross plot-mapping technique

WellLog measuredSeismic predicted
7.6 7.0 
5.4 6.2 
10.3 10 4 
6.2 6.8 
7.6 8 1 
7.1 7.0 
WellLog measuredSeismic predicted
7.6 7.0 
5.4 6.2 
10.3 10 4 
6.2 6.8 
7.6 8 1 
7.1 7.0 

One of the most important reasons for determining reserve classification is to justify investment, which is especially important early in the life of the field. The analogous reservoir concept is especially applicable to reserve classification in areas where large capital investments must be justified before production can start (i.e. offshore). In the modem offshore industry, detailed facilities are designed, and multi-million dollar expenditures are customarily made, on the basis of data from a small number of wells integrated with 3-D seismic data. Onshore fields may also need to show sufficient reserves to justify building pipelines and surface facilities. The difficulties encountered without 3-D seismic data are illustrated by comparing Figures 2 and 3 5 . The 3-D structure is more complex and less continuous than the simplified 2-D structure. The methods discussed here may better reflect the current thought process in reserve classification when integrating 3-D seismic data with limited well and production data than the current SPE/SPEE classification methods.

The authors recognize the tremendous difficulties in classifying reserves, especially at early stages of development. The intent of this paper is to promote greater discussion on the use of the seismic method in classifying reserves. As technology advances the existing SPE/SPEE classification methods can run into difficulties. When a horizontal well is drilled the seismic method may be the only way the reservoir thickness and hydrocarbon-water contact can be determined.

There are as many variables associated with reservoir parameters as industry experts who are attempting to provide a satisfactory resultant reserve classification. It is the authors’ intent to present reservoir classification techniques that will increase the contribution offered by seismic data.

A useful way to emphasize the contribution of 3-D seismic data to reserve classification is to simulate a case history of field development. The first step in a development case history is to find the field, and often 2-D seismic data must be acquired and interpreted to find a promising area to drill. A bright spot (seismic anomaly) indicating a potential reservoir is shown on two crossing seismic lines (Figures 4 and 5). Since both seismic lines demonstrate some direct hydrocarbon indicators the decision to drill a wildcat well was made. The typical Gulf Coast wildcat probability of success (15%) was assigned to the well. Figure 6 shows the well location and the lateral extent of the amplitude anomaly in map view.

Possible causes of amplitude anomalies are:

  1. Hydrocarbon reservoir

  2. Lithologic/facies change

  3. Water reservoir with associated minor gas.

  4. Tuning of thin beds.

  5. Geometric focusing in migration.

  6. Acquisition or processing artifact.

Characteristics of direct hydrocarbon indicators (DHCIs) are:

  1. Flat spot

  2. Phase change

  3. Appropriate peak/trough relationship

  4. Dimming of the reflectors below the anomaly.

  5. Loss of high frequency data below the anomaly.

  6. Velocity sag (push down).

  7. Conformation of the amplitude outline to structure.

In this simulated case history the exploratory well was successful and had reserves of 10 BCF. Additional anomalies (bright spots on Figure 4) indicate that there may be more reservoirs adjacent to the new well which may contain additional reserves. This information coupled with the well data (log/core parameters, well test, etc.) justified the acquisition of a 3-D seismic survey for further development of the field.

A chart/timeline for field development is shown in Figure 7. Note the risk factors involved, the initial reserves (10 BCF) and the probability of success given to the prospect at stage 1 (pre 3-D seismic). Stage 2 includes the acquisition, processing, and preliminary interpretation of the 3-D seismic volume.

A time slice extracted from the seismic data during the initial stages of the interpretation (Fig 8) shows the presence of a birdfoot delta with a feeder channel. This is a discreet, isolated reservoir with several channels superimposed on top of the fan delta.

The seismic event amplitude map of the new reservoir (Figure 9) shows the spatial distribution and quantitative values (contours) of the amplitude, and shows the reservoir is an isolated discontinuous sand sequence. Since the reservoir is discontinuous the only reserves that could be classified as proved under the existing SPE/SPEE classification methods are determined by well control. It will probably take 4 or 5 wells to effectively drain this reservoir.

At the completion of stage 2 (preliminary interpretation of 3-D seismic data) a discrete discontinuous reservoir has been defined. There are at least 25 BCF of gas reserves with a probability of success of 30%.

Figure 7 shows the progress of field development. As the technical effort increases through the reservoir characterization steps, the probability of success, the reserve estimate and the rate of return all increase. For an economic project, it is still necessary to increase reserves and the probability of success. Any attempt to accomplish this should include the application of the steps included in the Basic Geophysical Guidelines ( Appendix I) which provide for determination of reservoir parameters and calculation of hydrocarbons in place.

The field development chart shows that the geophysical analysis increased the reserves to 40 BCF and improved the probability of success to 55%. As the reservoir characterization continues through specialized processes such as analysis of direct hydrocarbon indicators and inversion to acoustic impedance, more and more detail is known about the reservoir. Quantifiable attributes of the seismic data can be cross-plotted with reservoir parameters extracted from the log data. When good regression fits are accomplished, the seismic attributes can be directly transformed into reservoir parameters and reserve calculations can be made. In order to take the reserve calculation to a higher level of certainty a second well must be drilled.

When the second well was successful and confirmed the reservoir characterization numbers, the probability of success jumped to 80%. The reserves in place are not necessarily changed. Is this the point where remaining reserves as calculated using seismic data become classified as proved?

This field development plan may require 3 or 4 successful wells to classify all 40 BCF as proved reserves. After each well is completed the reservoir characterization is revised and improved. At some point in this development plan all the reserves will be classified as proved (after 4 or 5 wells).

To illustrate the importance of the contribution of 3-D seismic in the development of this field. Fig 10 shows the field development chart without the use of 3-D seismic data. In this simulated example, it took 8 wells drilled over a two-year period to achieve 40 BCF proved reserves. Wells 7 and 8 did not add reserves and were probably not necessary to drain the reservoir, but were needed to delineate the reservoir without seismic data. Well 5 was a dry hole. The rate of return on the field starts to decline when dry holes or excessive wells are drilled.

This simulated development case history shows that some of the advantages of using 3-D seismic data are:

  1. Total reserves will be determined earlier in the life of the field.

  2. The field development plan will be more precise with improved economics (cheaper, faster, better).

This is a simulated reservoir characterization in which 3-D seismic has been acquired to enhance the performance of a mature field that is being waterflooded.

A geologic cross-section from the Graneros shale through the "J" sandstone from the D-J Basin is shown in Figure 11. The reservoir of interest is the "D" sandstone which was incised into the Huntsman shale. The sand presence is indicated by a thick isopach interval from the top of the D-sand to the base of the Huntsman shale. The overall reservoir is a complex of individual fluvial and estuarine sand bodies which complicates the estimation of porosity and permeability.

One of the most important steps in the geophysical interpretation process is correlating the borehole information and the seismic data. Lithological units must be accurately identified on the time-referenced seismic section for reservoir characterization to have any meaning. Seismic check shot surveys, vertical seismic profiles (VSP), and synthetic seismograms (seismic modeling) are all tools used to accomplish this task. These processes provide a time (seismic data) to depth (log data) relationship. In cases of poor or missing data it may be necessary to create pseudo-logs to generate the synthetic seismogram6 .

This kind of calibration is essential for accurate reserve calculations. A small error in the log data, the seismic data processing, the log interpretation, or the borehole calibration to the seismic data can negate a very diligent "state of the art" analysis of reservoir structure and stratigraphy.

Figure 12 is a synthetic seismogram/stratigraphic model that was made using the correct spatial position of a well within the 3-D seismic volume. The mathematical cross-correlation value .722 (1.000 is perfect) indicates the quality of the tie between the well log data and the surface seismic data. Values above .600 are excellent quality. This process is the first modeling step in seismic event identification and stratigraphic correlation.

The proper correlation can be confirmed by inserting the synthetic seismogram/well models into their proper spatial location within a seismic line. Figure 13 shows two synthetic seismograms from two different wells inserted into the seismic line. Both wells demonstrate an excellent correlation.

The fully corrected VSP from a well within the Sooner Field is shown in Figure 14. This is another method of seismic event identification and time-to-depth calibration. The VSP is an actual seismogram. directly recorded: whereas the synthetic seismogram is calculated from the sonic and density logs.

Once the seismic events have been correlated to the downhole log picks for all the wells within the 3-D seismic survey, the seismic events can be tracked throughout the survey and the seismic interpretation can begin. Figure 15 shows a seismic line where the events have been identified, interpreted, and tied to existing well control.

A simplified geological cross-section for geophysical modeling through the Sooner Field is shown in Figure 16. The vertical dimension is in time. The overall cross-section was created using seven individual well models similar to that shown in Figure 12.

The model was then converted to a synthetic seismic section (Figure 17). The upper synthetic seismic section shows the seismic response of a D-sand reservoir of constant thickness (15 feet). The lower synthetic seismic section shows the seismic response of the model in Figure 16. Note the increased amplitude of the peak associated with the D-sand event at the point where the sand is the thickest in the Sooner Field model (bottom).

A surface seismic line extracted from the 3-D volume is shown in Figure 18 (top). The line goes across a producing portion of the D-sand channel. The increased amplitude coincides with the thick part of the channel, as confirmed by well control and the seismic response of the Sooner Field model (Figure 18 bottom).

Figure 19 is an event (D-sand) amplitude map of the entire 3-D survey, in which the areas of highest amplitude are indicated by darker grays and black. This pattern represents the spatial distribution of the thickest part of the D-sand interval.

The black contours overlaying the grid in Figure 20 represent the isopach of the D-sand reservoir as mapped using well control. There is a good correlation between the two, but the grid from the seismic data shows more detail and an unexpected change in the lineament direction in the northern portion of the channel system. This anomalous lineation is probably the result of faulting that is present in a shallower interval.

Other attributes besides seismic event amplitude can be extracted from the seismic data to improve the quality of seismic interpretation and reservoir characterization. The attributes most commonly studied include acoustic impedance, pseudo-log velocity, instantaneous frequency and phase. These attributes can be quantitatively related to reservoir parameters by crossplot regression analysis, such as shown in Figure 1.

Conclusions of the DJ Hasin Simulated Case History

  1. The reservoir characterization using 3-D seismic data confirmed estimates of OOIP from well control and would have provided this information earlier in the life of the field.

  2. The 3-D seismic interpretation showed the reservoir was more complex than expected: therefore recoverable reserves may decrease.

  3. It would have been advantageous to have obtained the 3-D interpretation prior to waterflooding to help design the injection pattern. Some changes in injectors have been made as a result of the 3-D seismic data.

  4. The 3-D seismic interpretation improved the geological model used for reservoir management.

  5. The northeast-southwest lineaments seen in the northern part of the Sooner Field are probably fault-related, and will interrupt the porosities and permeabilities across these boundaries.

Geophysical reserve classification is very dependent on geographical area, data quality, the amount and quality of processing and interpretation, and the judgment of the interpreter. Therefore it is not simple to develop hard rules for geophysical reserve classification. General principals of geophysical reserve classification which were not discussed in the case studies are illustrated in the following discussion.

2-D Seismic Data

The limitations of 2-D seismic are well known. Quantitative use of 2-D seismic data can lead to large errors in reservoir volumes7 . Generally speaking, reserves which are based on 2-D data should use the existing SPE/SPEE classification methods.

Migration

Migration is the most important step towards focusing the seismic data into a clear picture of the subsurface geology. Current SPE/SPEE classification methods already allow for the use of seismic data for determining structure and closure. The reservoir structure is often defined using seismic data.

Time migration corrects for structural dip and positions the subsurface correctly in the x and y directions. Depth migration positions the subsurface correctly in the z direction by converting time measurements to depth through the velocity model, and becomes significant when there are severe lateral variations in velocity.

Traditionally, seismic traces have been stacked to remove noise and improve data quality, and then migration performed (post stack migration). Advances in computing and software have made 3-D prestack migration easier to perform, and prestack depth migration improves results in areas of complex geology where lateral velocity variations exist.

Figure 21 8 , which is separated into three panels, illustrates the importance of migration. The top panel shows the raw data post-CDP-stack. The center panel shows the results of 2-D migration, and the third panel shows the results of 3-D migration. Since this set of data was collected using a known structural model, the third panel is an accurate representation of the focused image of the input model. The 2-D migration gives a false image of a second anticline in the middle of the panel.

Correct migration is necessary to determine reservoir volume and geometry. Uncertainty in the migrated positions of the seismic events will be translated into errors in calculated reserves and should be reflected in reserve classification.

AVO

AVO (amplitude vs. offset) analysis was introduced in the early 70's and has now been successfully used in many major oil and gas areas of the world. Reservoirs identified by AVO will generally be put in the probable category, as long as the area has a history of successful use of AVO. This technique is used in conjunction with seismic "bright spots" or amplitude anomalies and the "direct hydrocarbon indicators previously discussed.

A recently completed Gas Research Institute study9  conducted on drilling 96 AVO anomalies in the Gulf Coast found a 50% success rate. This is an excellent study which illustrates uses and pitfalls of AVO in the Eocene Yegua formation of onshore Texas.

Higher success rates (90%) using AVO have been attributed to the use of high quality 3-D data.10 . Probability of success was based on the AVO analysis, but reserve calculations and well location were based on the 3-D seismic interpretation.

The following additions to the existing SPE definitions, which are written in bold type, would include the contribution offered by the seismic method. Use of 3-D seismic data will generally lead to more accurate volumetric calculations and understanding of fluid flow.

  1. In certain instances, proved reserves may be assigned on the basis of electrical and other type logs and/or core analysis and/or 3-D seismic data that indicate subject reservoir is hydrocarbon-bearing and is analogous to reservoirs in the same area that are producing, or have demonstrated the ability to produce on a formation test.

  2. In the absence of data on fluid contacts, the lowest known structural occurrence of hydrocarbons controls the proved limit unless otherwise indicated by definitive engineering, performance, or 3-D seismic interpretation.

  3. Reserves for other undrilled locations are classified as proved undeveloped only in those cases where interpretations from wells indicate that [he objective formation is laterally continuous and contains commercially recoverable hydrocarbons at locations beyond direct offsets unless otherwise indicated by definitive interpretations of 3-D seismic data.

  4. In general, probable reserves may include (8) reserves identified by interpretations of 3-D seismic data that are near producible hydrocarbon reservoirs.

Some of the basic elements of geophysical due diligence for reservoir classification are summarized in  Appendix I.

The addition of seismic data to reserve classification can have an impact on property evaluation. The effect may be negative, since previously unrealized complexities may be revealed by 3-D seismic data. In many cases the use of 3-D seismic data will reveal additional reserves and increase property values. The impact 3-D seismic had on the cash flow for the High Island 24L field is shown in Figure 22 11 .

In the 80s and 90s a significant amount of research has been devoted to taking the geostatistical techniques first developed in the mining industry and applying them to the petroleum industry. Application of geostatistical techniques to reserve classification is not addressed by the current SPE definitions and SPEE guidelines. Geostatistical techniques are particularly useful for analyzing spatial variations in large data bases, such as a 3-D seismic survey with well data.

Geostatistical reserve analysis should usually take advantage of the increased spatial sampling a 3-D seismic survey provides. Chambers12  performed a study on a West Texas Permian Basin reservoir, which compared the error obtained by kriging porosity with and without calibrated 3-D seismic data. In the case of limited well data the error is greatly reduced when the seismic data is integrated into the reservoir description.

  1. Some guidelines which may better reflect the current thought process in reserve classification with seismic data have been developed.

  2. Classification of reserves using these geophysical guidelines may differ from those obtained by strictly following the existing SPE definitions and SPEE guidelines. Use of these geophysical guidelines for reserve classification can have a significant impact on property evaluation.

  3. I Indeveloped reserves for complex reservoirs (stratigraphie traps, sand lenses, isolated porosity developments) will not meet the SPE/SPEE continuity requirement for the proved classification, but if these reservoirs are seismically detectable and demonstrate similarity to seismic models of analogous reservoirs they could be upgraded to a higher classification.

  4. Case studies which compare reserves after full field development to earlier predictions are needed to increase confidence in reserve classification based on seismic data.

  5. Development of seismic guidelines and pitfalls for specific areas, similar to the work SPEE presents for conventional reservoir methods, would increase confidence in classifying reserves from seismic data.

  6. The SPE Oil and Gas Reserves Committee and SPEE should include geophysical experts that can monitor the impact of 3-D seismic on reserve classification and develop standards.

  7. Reservoir characterization using 3-D seismic data will provide critical reservoir parameters early in the life of the field. This will provide for better, more informed development planning which will ultimately improve the economics of the project.

  8. As with traditional methods, the application of seismic methods to reserve classification not only rely on the integrity, skill, and judgment of the reserve estimator, but also are effected by the amount and quality of data available and the geological complexity, stage of development, and degree of depletion of the reservoir. However, the use of 3-D data will generally reduce uncertainty and improve classification of reserves.

This paper was selected for presentation by an SPE Program Committee following review of information contained in an abstract submitted by the author(s). Contents of the paper, as presented, have not been reviewed by the Society of Petrolem Engineers and are subject to correction by the author(s). The material, as presented, does not necessarily reflect any position of the Society of Petroleum Engineers, its officers, or members. Papers presented at SPE meetings are subject to publication review by Editorial Committees of the Society of Petroleum Engineers. Permission to copy is restricted to an abstract of not more than 300 words. Illustrations may not be copied. The abstract should contain conspicuous acknowledgment of where and by whom the paper is presented. Write Librarian, SPE, P.O. Box 833836, Richardson, TX 75083-3836, U.S.A., Telex, 163245 SPEUT.

Special thanks to Jewel Wellborn for the discussion of Case Study One, Mark Sippel and 1ESC for the discussion of Case Study Two, and the SEG and AAPG for allowing reproduction of figures.

SOME BASIC ELEMENTS OF GEOPHYSICAL DUE DILIGENCE FOR RESERVOIR CLASSIFICATION

  1. Borehole logs should include sonic and density logs (sonic log should extend from surface to total depth).

  2. Edit logs/prepare pseudo-geophysical logs for poor or missing data.

  3. Wellbore locations need to be accurate-resurveying may be necessary.

  4. Check Shot/VSP used to calibrate downhole logs to the surface seismic data and determine velocity.

  5. 2D seismic data should be used to help design 3D acquisition parameters and determine the areal extent of the 3D survey.

  6. 3D Seismic Data Acquisition should be designed for maximum frequency bandwidth and smallest spatial sampling economically justifiable.

  7. 3D Seismic Data Processing

    • Process to preserve maximum frequency bandwidth(vertical sampling=resolution).

    • Improve seismic event continuity utilizing 3D migration to focus the seismic image, (prestack orpoststack)

    • Attribute Analysis-Hubert Transforms

  8. Seismic Modeling (forward and reverse)

    • 2D modeling to determine the synthetic seismic response of the reservoir.

    • Inverse Borehole Modeling to derive synthetic log responses from AVO analysis of surface seismic data.

    • Seismic inversion of the surface seismic data to acoustic impedance or pseudo-log velocity.

  9. 3D Seismic Interpretation

    • Seismic Event Mapping (peak, trough, zero crossing).

    • Structure Maps (Time/Depth)

    • Fault Plane Maps (Time/Depth)

    • Isochron/Isopach Maps

    • Amplitude Maps

    • Amplitude vs. offset (AVO) analysis and direct detection of hydrocarbons.

    • Crossplot regression analysis to relate seismic event properties to rock properties.

      1. Seismic amplitude vs. reservoir thickness.

      2. Seismic Impedance vs. reservoir porosity.

    • Reservoir Characterization Maps

      1. Gross reservoir thickness

      2. Net reservoir thickness

      3. Reservoir porosity

      4. Net Pore Volume

      5. Hydrocarbons in Place

Myers
,
R.C.
and
Rijks
,
E.J.H.
: "
Optimization of Field Development Through Early Acquisition of 3D Seismic
",
paper SPE 22343, presented at the SPE International Meeting on Petroleum Engineering
held in
Beijing, China
24–27 March 1992
.
Wang
,
Z..
Hirsche
,
W.K..
and
Sedwic
k.G.
: "
Seismic monitoring of waterfloods? A petrophysical study
".
Geophysics
, Vol.
56
, No.
10
, pp.
1614
1623
,
October
,
1991
.
Neff
,
D.B.
: "
Estimated pay mapping using 3D seismic data and incremental pay thickness modeling
", in
Reservoir Geophysics
, edited by
R.E.
Sheriff
,
SEG
,
Tulsa
,
1992
, p.
214
.
Stanulonis
,
S.F.
and
Tran
,
H.V.
: "
Lisburne Porosity-Thickness Determination and Reservoir Management from 3-D Seismic Data
", in
Interpretation of 3-Dimensional Seismic Data
. Third Edition, edited by
Alistair R.
Brown
,
AAPG
,
Tulsa
,
1991
, p.
315
.
Brown
,
Alistair R.
,
Interpretation of 3-Dimensional Seismic Data
. Third Edition,
AAPG
,
Tulsa
,
1991
, p.
58
.
Loren
,
J.D.
,
Kulha
LT.
,
Renbarger
,
K.S.
, and
Sneider
.,
R.M.
: "
Reservoir Properties Inferred from Seismic Response in Areas with Minimal Log Control
",
Proceedings of the 1st Archie Conference
, held
October 22–25, 1990
, in
Houston, Texas, U.S.A.
, p.
271
.
Van Sickle
,
K.H.. Soroka. W. L.
, and
Valusek
,
J.E.
"
AVO analysis of 3-D seismic data identifies untested reservoirs in old gas field, California
", in
Reservior Geophysics
, edited by
Robert E.
Sheriff
.
SEG
,
Tulsa
,
1992
. p.
215
.
Bulling
,
Thomas P.
and
Olsen
.
Rebecca S.
: "
Extending Field Life in Offshore Gulf of Mexico Using 3-D Seismic Survey
", in
Interpretation of 3 -Dimensional Seismic Data
. Third Edition, edited by
Alistair R.
Brown
.
AAPG
,
Tulsa
,
1991
. p.
283
.
Chambers
,
R. L.
"
Geostatistical Application for Exploration and Development: Porosity Estimation From 3-D Seismic Data Calibrated to Well Data
",
paper SPE 25655, presented at the SPE Middle East Oil Technical Conference and Exhibition
in
Bahrain
3–6 April 1993
.
Denham
,
L.R.
"
Why that Great Discovery Wasn't So Great
",
paper OTC 7076. presented at the 25th Annual OTC
in
Houston, Texas
,
3–6 May 1993
.
Brown.
Alistair
R.
,
Interpretation of 3-Dimensional Seismic Data
. Third Edition.
AAPG
.
Tulsa
,
1991
, p.
6
.
Allen,
James
L.
, and Peddy,
Carolyn
P.
;
Amplitude Variation with Offset: Gulf Coast Case Studies
.
SEG
.
Tulsa
.
1993
.