Interpretation with Seismic Attributes

(adsbygoogle = window.adsbygoogle || []).push({}); Learning Objectives After completing this topic "Interpretation with Seismic Attributes", you will be able to: Provide the definition of seismic attributes. Describe how and where seismic attributes are applied. Explain how seismic attributes are classified. Outline how AVO attributes are used to analyze changes in the amplitude of seismic waves. Discuss why the application of the coherence attribute is very useful to interpreters. List the well-known instantaneous attributes and define which geologic and geophysical features they can enhance. Introduction To Interpretation with Seismic Attributes https://youtu.be/pOIUKSaIhrc (adsbygoogle = window.adsbygoogle || []).push({}); As described by Taner (2001), “Seismic attributes are all the information obtained from seismic data, either by direct measurements or by logical or experience based reasoning.” Attributes extend the study of seismic data from stratigraphy and structural interpretation to the direct visualization of the rock properties, fluids, elasticity, fractures, etcetera. Attributes provide detailed and accurate information to the interpreter on structural, stratigraphic and lithological parameters of the seismic prospect, by facilitating the recognition of hydrocarbon anomalies, and providing validation of anomalies that reveal the relationship between subsurface rocks and seismic data. From the study of the seismic amplitudes, it is possible to derive some physical parameters of the rocks such as acoustic impedance, velocity, reflection coefficients, etcetera. Using the phase component, the geometric configuration of reflectors can be obtained. Attributes are extracted from seismic data and applied using the following approach: Define a model that explains the existing anomalies. Compute the forward response to the model. Compare the response of the model to the actual data. Adjust the model as many times as necessary or possible to match the model to the seismic data. Verify new areas or intervals. In terms of input data, seismic attributes can be classified into two categories pre-stack (before the stack) and post-stack (after data is stacked). In terms of computational calculations, seismic attributes can be classified into the following categories: Instantaneous attributes: Some calculated examples are sample-by-sample, such as the trace envelope and its derivatives frequency and phase. Wavelet attributes: These are computed at the peak of the trace envelope with the Fourier Transform; instantaneous frequency (the rate of change of phase as a function of time) is an example of this type of attribute. Physical attributes: These relate to physical the qualities of rocks frequencies relate to the bed thickness, and the magnitude of the trace envelope relates to the impedance contrast. Geometric attributes: These describe the spatial and temporal relationship of all other attributes. For example, lateral continuity measured by semblance is a good indicator of a discontinuity. We can further subdivide the attributes into two categories based on their origin in the wave phenomena: Reflective attributes: Attributes corresponding to the characteristics of interfaces All instantaneous and wavelet attributes Pre-stack attributes, such as amplitude versus offset (AVO) Transmissive attributes: Relate to the characteristics of a bed between two interfaces Interval, RMS and average velocities, Q, absorption and dispersion The trace envelope (Animation 1) shows that the seismic data can be viewed analytically as a combination of real and imaginary parts. It is a physical attribute which can be used as an effective discriminator of the following characteristics: Reflectivity it mainly represents the acoustic impedance contrast Bright spots indicates possible gas accumulation Sequence boundaries Thin-bed tuning effects Major changes in the depositional environment Spatial correlation to porosity and other lithologic variations Indicates the group, rather than the phase component, of the seismic wave propagation https://youtu.be/H23R6yM5qdo (adsbygoogle = window.adsbygoogle || []).push({}); The first derivative of the trace envelope the time rate of change of the envelope shows the variation of the energy of the reflected events. This attribute is also a physical attribute that can be used to detect possible fracturing and absorption effects. The second derivative of the trace envelope provides a measure of the sharpness of the envelope peak. It can be used to identify all reflecting interfaces within the seismic bandwidth. Following is a list of additional attributes that play a significant role in rock and lithology interpretation: Instantaneous amplitude (also known as reflection strength): The square root of the total energy of the seismic signal at an instant of time; the amplitude independent of phase, and an effective tool to identify bright and dim spots because it provides information about contrasts in acoustic impedance Instantaneous phase: A physical attribute that can be effectively used for interpreting geometric shapes Instantaneous frequency: A physical attribute that is used as a hydrocarbon indicator through the identification of low frequency areas; also assists with locating fracture zones and bed thicknesses Statistics of attributes: This involves geostatistics, such as correlation and sequence Linear Discrimination Analysis (LDA) and Principal Component Analysis (PCA) Ratio and phase differences The Coherence cube Instantaneous dip azimuth Semblance lateral event continuity 2D and 3D AVO Attributes (adsbygoogle = window.adsbygoogle || []).push({}); AVO analyzes changes in the amplitude of seismic waves that are dependent on source-receiver distance, and relates to pre-stack seismic attributes for hydrocarbon discrimination. It is an attribute capable of distinguishing if a bright spot is generated by the presence of fluid or just lithological changes. Figure 1 illustrates the concept of the AVO attribute. In fact, near and far offset-stacks show noticeable differences. Figure 1: Near and far offset-stacks; arrow A is the reflector of interest AVO analysis is based on the Zoeppritz (1919) equations. These equations define a relationship between the amplitude of reflection of a seismic wave and the angle of incidence. The change will depend on the density and velocity contrasts (compressional and shear) of the two contacting layers. In fact, in the case of normal incidence, the amplitude of the incident layer is equal to the amplitude of the reflected layer, which depends basically on the densities and velocities of the two media (as shown by the equation below). This is called the P-wave reflection coefficient for normal incidence (NI): [latex]\dfrac{\rho _2 \, V _{\textrm{p2}} - \rho _1 \, V _{\textrm{p1}}}{\rho _2 \, V _{\textrm{p2}} + \rho _1 \, V _{\textrm{p1}}}[/latex] These relationships become much more complicated when the angle of incidence is oblique. Fortunately, several approximations have been defined in the literature for Zoeppritz equations to allow for a practical application. Shuey's (1985) approximation is one of the most popular, as defined by Castagna (1994). The P-wave reflection coefficient, as a function of angle of incidence, Rpp(θ), may be expressed as: [latex]R_{\textrm{pp}}(\theta) = A + B\, \sin^2 \theta[/latex] Where: (adsbygoogle = window.adsbygoogle || []).push({}); A is the normal incidence P-wave reflection coefficient, Rp B or AVO gradient (or slope) is given by: [latex]B = -2\, \dfrac{V^2_\textrm{S}}{V^2_\textrm{P}}\dfrac{\Delta \rho}{\rho} + \dfrac{1}{2}\dfrac{\Delta V_\textrm{P}}{V_\textrm{P}} - 4\, \dfrac{V^2_\textrm{S}}{V^2_\textrm{P}}\dfrac{\Delta V_\textrm{S}}{V_\textrm{S}}[/latex] Where: [latex]V_{\textrm{P}} = \frac{V_{\textrm{P2}} + V_{\textrm{P1}}}{2}[/latex] [latex]V_{\textrm{S}} = \frac{V_{\textrm{S2}} + V_{\textrm{S1}}}{2}[/latex] [latex]\boldsymbol{\rho} = \frac{\boldsymbol{\rho} _2 + \boldsymbol{\rho} _1}{2}[/latex] [latex]\Delta V _{\textrm{P}} = V_{\textrm{P2}} - V_{\textrm{P1}}[/latex] [latex]\Delta V _{\textrm{S}} = V_{\textrm{S2}} - V_{\textrm{S1}}[/latex] [latex]\Delta \boldsymbol{\rho} = \boldsymbol{\rho}_2 - \boldsymbol{\rho}_1[/latex] VP2= P-wave velocity in underlying medium VP1= P-wave velocity in overlying medium VS2= S-wave velocity in underlying medium VS1= S-wave velocity in overlying medium ρ1= density in underlying medium ρ2= density in overlying medium Different authors have proposed different attributes to discriminate gas sands from water-bearing sands. Each of these have advantages and disadvantages. A gas-bearing sand with low impedance laying between shales will show larger negative A and B variables than reflectors that don’t have any associated gas. So the product A×B is a good indicator of gas-bearing sands. Among other indicators, Castagna and Smith (1993) have shown that the indicator Rp−Rs (the difference of the normal incidence P-wave and S-wave reflection coefficient) can work better with gas sands with all kinds of impedances. Verm and Hilterman (1995) made a different modification of the Shuey's equation: [latex]RC(\theta ) \approx NI\, \cos^2(\theta) + PR \, \sin^2 (\theta )[/latex] Where: NI stands for Normal Incidence reflectivity, as defined earlier PR= Poisson Reflectivity = [latex]\frac{\boldsymbol{\sigma} _2 - \boldsymbol{\sigma} _1}{(1 - \boldsymbol{\sigma} _{\textrm{avg}})^2}[/latex] Rutherford and Williams (1989) developed a classification for gas-bearing sands based on AVO parameters, which was augmented by Castagna and Swan (1997). Figure 2 illustrates the differences between the four kinds of gas-bearing sands. Gas-bearing Sands Classifications Figure 2: Classifications for gas-bearing sands Class 1: Sands with a greater impedance than the surrounding media where a dim spot is present. The top curve in Figure 2 is representative of a Class 1 sand, which is usually found onshore, is mature and has undergone moderate to high compaction. Class 2: Sands with a similar impedance as the surrounding media where a phase reversal is present. The middle two curves in Figure 2 represent a range of possible AVO responses in Class 2 sands. Class 3: Sands with a lower impedance than the surrounding media where a bright spot is present. These sands are usually undercompacted and unconsolidated. Class 4: The shear velocity (Vs) is less than the Vs from the overlying primarily shale intervals. The sandstone Vs is less than the Vs of the shale interval above it. Castagna and Swan note that Class 3 and 4 gas sands may have identical normal incidence reflection coefficients, but the magnitude of Class 4 sand reflection coefficients decreases with an increasing angle of incidence, while Class 3 reflection coefficient magnitudes increase. Plots of A versus B (or the gradient versus R0) allows for discrimination of gas-bearing sands from water-bearing sands or dry sands. The Coherence Attribute (adsbygoogle = window.adsbygoogle || []).push({}); Among the different kinds of attributes, the coherence attribute is one that does not require extensive interpretation work. Coherence coefficients for seismic trace similarity are calculated in a grid using a nine-trace operator. The coherence algorithm is used to enhance discontinuities (faults and/or stratigraphic details) in 2D and 3D data. Figure 1 shows an example of an application of the coherence algorithm, with a nominal seismic time slice (colored image) and a coherency time slice (black and white image). Faults are difficult to interpret in the conventional time slice, but these are clearly shown in the coherency time slice. Figure 1: Time slice (colored) and coherency time slice (black and white) Instantaneous Attributes (adsbygoogle = window.adsbygoogle || []).push({}); There are instantaneous attribute variations of different parameters—for example, the trace envelope and its derivatives, frequency and phase. These are calculated from complex traces sample-by-sample. Some of the well-known and often used instantaneous attributes are the instantaneous phase, instantaneous amplitude and instantaneous frequency. A 3D seismic data interpretation usually starts with the search for high amplitude anomalies that can be associated with bright spots, sequence boundaries, etcetera. The instantaneous amplitude attribute is often used to enhance amplitude anomalies. Figure 1 is an image of an amplitude time slice showing a linear feature running diagonally from top left to bottom right as the most remarkable feature in this time slice. Figure 1: Amplitude time slice Figure 2 shows the instantaneous frequency of the same time slice. Notice that a new meandering feature not shown in the previous image is clearly recognizable in this time slice. Figure 2: Instantaneous frequency time slice Attribute Applications (adsbygoogle = window.adsbygoogle || []).push({}); This subject contains examples of how the application of attribute analysis helps exploration efforts to become more focused and precise when determining exploration targets. The first example is from South Texas and shows how small contrasts in acoustic impedance values and significantly lower Poisson’s ratios, along with far-offset stacks, can indicate gas-bearing horizons. The second example provides an application of the techniques of AVO analysis based on linear fit to calculate attributes, which are used and interpreted through crossplot in prospecting for gas. Example 1: South Texas Gregg and Bukowski (2001) published an interesting paper on the application of AVO analysis for locating exploration wells in the clastic Oligocene Vicksburg Formation in South Texas. South Texas Example of AVO Application Figure 1 illustrates two different prospects with similar geological conditions. Only the well drilled in the top prospect resulted in a gas producing well. This result triggered the search for a better discriminating approach before drilling. Figure 1: Two similar geological prospects in South Texas Figure 2 shows small contrasts in acoustic impedance values (ρV) and significantly lower Poisson’s ratio (σ) for the gas-bearing horizon. Figure 2: Petrophysical parameters for the discovery wells (adsbygoogle = window.adsbygoogle || []).push({}); Figure 3 shows modelled CDP (common depth point) gathers to analyze the AVO responses using log data and Verm and Hilterman’s (1995) approximation of Shuey's equation. A Class 2 gas-bearing sands anomaly is very clear at the far-offset in the producing well, while the water-bearing well shows near-offset positive amplitudes that fade out at far-offsets. Figure 3: Modelled CDP gathers After reprocessing the seismic data, the AVO anomaly in the far-offset was clearly enhanced, as shown in Figure 4. Figure 4: Far-offset anomaly for gas-bearing sand The results obtained after reprocessing the 3D data encouraged the application of the AVO attribute to the full field, revealing a number of new anomalies to be drilled (Figure 5). Figure 5: Full field AVO far-offset anomalies Example 2: Coast of Veracruz, Mexico (adsbygoogle = window.adsbygoogle || []).push({}); In this section, we will highlight some of the results of Malcotti et al. (2004) on the analysis of AVO anomalies from the Kosni Gas Field. The paper discussed an application of the techniques of AVO analysis based on linear fit to calculate attributes (Shuey, 1985), which are used and interpreted through crossplot in prospecting for gas. It also discusses recent techniques for multidimensional classification of attributes in prospecting for gas. The Kosni Field is located in the waters off the coast of the State of Veracruz, Mexico. The shallower sands belong to the Upper Pliocene, and have average porosities of 22%, with bright spots observed in the stacked, migrated seismic image. The deeper targets belong to the Miocene sands. These are fine-grained and more consolidated, rich in clay (40-50%), calcite, and illite. They also contain potassium, a low porosity (18%), and facies sands of a platform edge (from the low-stand system track). The 3D seismic data from this project was processed preserving amplitudes that were migrated before the stack through a zero-offset algorithm after dip moveout (DMO). Kosni Gas Field Example of AVO Application Figure 6 shows the same seismic that has been modeled using three different modeling techniques to highlight a water or gas response at Sand 13. Place your mouse over or tap the letters below each image for more information. Figure 6: Seismic modeling and fluid substitution made in the well Kosni-1, Sand 13 A: Modeling with the Aki-Richard equation and 100% water B: Full wave equation 1D modeling with multiples C: Modeling with the Aki-Richard equation and 100% gas Once the synthetic seismograms are calculated, and combined with plotted attributes (intercept [R0] and Poisson’s reflectivity [PR] for the AVO interpretation) on real CDPs, the calculation of AVO attributes can be done by analyzing crossplots (Figure 7). The results shown are as reported from the literature by Hilterman, 2001. The figure shows the rotation and classification of lithology, and the presence of gas for two wells. The colors represent the time of analysis around the sands of interest: 8, 10 and 13. The top of the gas sands are enclosed by the ellipse. The AVO analysis and classification was conducted in wells Kosni-1 and Kosni-101. The analysis mainly focused on two targets: one named Shallow Sand 27 (sands with an average porosity of 22%) and a deeper target named Sand 8 (Miocene sands of a platform edge with an average porosity of 18%). These sands have different gas saturations, with anomalies of Class 2 observed in CDPs, and bright spots observed in the stacked, migrated volume. Figure 7: Crossplot of R0 versus Poisson reflectivity for wells Kosni-1 and Kosni-101 Figures 7, 8 and 9 show maps obtained for sands 8 and 27. These maps show the sands that are more likely to contain gas (in red) after being calibrated with the wells, through the seismic modeling (Figure 6), to reproduce the response observed in the real seismic data. Figure 8 shows the map obtained with the product of the R0 and PR attributes rotated 45° (counterclockwise) so that the AVO anomalies are interpreted as Class 3 gas-bearing sands. Figure 8: Classified map corresponding to Sand interpreting the crossplot R0 versus PR of Figure 6 Sand with a red plot color is more likely to have gas while orange and pink coloring are most likely to be a sand with water (Figure 9). Figure 9: AVO attribute map rotated 45° for R0 PR for the top of the Sand 8 (Miocene) Figure 10 shows red plotted sands with gas, and yellow plotted sands with water. Trajectories of wells Kosni-1 and Kosni-101 and the path of the new location in the southwest of the well Kosni-1 are shown. Figure 10: Maps obtained for attribute [latex]R_0^* \ PR[/latex] rotated 45° for the horizons of sands Application Summary We have shown from the first example how small contrasts in acoustic impedance values and significantly lower Poisson’s ratios, along with far-offset stacks, can indicate gas-bearing horizons. The second example provides an application of the techniques of AVO analysis based on linear fit to calculate attributes, which are used and interpreted through crossplot in prospecting for gas. These are two real field examples of how attributes can be used to help locate economic, fluid-bearing reservoirs by visually locating these zones on seismic, and by helping develop targeted drilling plans. Assessment Interpretation with Seismic Attributes (adsbygoogle = window.adsbygoogle || []).push({}); 1. AVO analyzes changes in the amplitude of seismic waves that are dependent on __________ distance, and relates to __________ seismic attributes for hydrocarbon discrimination. 2. The coherence algorithm is used to enhance ___________ in 2D and 3D data. 3. In terms of computational calculations, seismic attributes can be classified into which of the following categories? (Select all that apply.) A .Physical attributesB .Geometric attributesC .Instantaneous attributesD .Wavelet attributesE .Chemical attributes 4. The magnitude of Class 4 sand reflection coefficients __________ with ___________ angle of incidence. 5. What type of coefficients for seismic trace similarity are calculated in a grid using a nine-trace operator? A .CoherenceB .DerivativeC .SemblanceD .Instantaneous 6. Which attributes play a significant role in rock and lithology interpretation? (Select all that apply.) A .Instantaneous phaseB .Transmissive attributesC .Reflection strengthD .Instantaneous frequency 7. What is the instantaneous amplitude attribute often used to identify? A .PinchoutsB .Magnetic anomaliesC .FaultsD .Amplitude anomalies 8. AVO analysis based equations define a relationship between the _________ of reflection of a seismic wave and the angle of incidence. 9. Which derivative of the trace envelope shows the variation of the energy of the reflected events and can be used to detect possible fracturing and absorption effects? A .FourthB .FirstC .SecondD .Third Recommended for You Seismic Interpretation Methods Seismic Stratigraphy Structural Interpretation