A Solid Step Toward Accurate Interpretations

Contributors: Ritesh Sharma

This month’s column deals with quantifying facies variation.

In last month’s Geophysical Corner we discussed the unsupervised seismic waveform classification method, which provides qualitative information on lithology in terms of facies variation in a given subsurface target zone.

While this information is useful, more work needs to be done for characterizing such formations of interest in terms of porosity and fluid content. For this purpose, impedance inversion of seismic data could be used, which essentially means transforming seismic amplitudes into impedance values.

This month we discuss here such impedance characterization of the formations of interest.

Seismic signals have a narrow frequency bandwidth, say 8-80 Hz. Frequencies below 8 or 10 Hz are lost either by bandpass filtering circuits in the recording instruments or while processing of seismic data to eliminate the low frequency noise usually present in the data.

Sonic logs, on the other hand, have a very broad frequency bandwidth, extending from 0 to well over many kilohertz.

While the high frequencies show the high-resolution information on the log, the low frequencies exhibit the basic velocity structure or the subsurface compaction trend. As these low frequencies are not present in the seismic data, the impedance traces derived from the seismic do not exhibit the compaction trend that the impedance logs do.

When seismic data are transformed into impedance values, the low frequency trend usually is added from outside – say, from the available impedance logs. By doing so, the derived impedance values will have a range of values similar to the value range exhibited by the impedance log curves.

Such a method is termed as absolute acoustic impedance inversion.

In figure 1, the addition of the low-frequency trend is demonstrated to the inverted impedance trace from seismic.

Sometimes it is not possible to determine the low frequency trend accurately, which possibly could be due to the unavailability of the log data. In such cases, the derived impedance range will be devoid of the compaction trend, but would still exhibit the relative variation of the impedance values.

Such a method is termed as relative acoustic impedance inversion.

In the last Geophysical Corner we discussed the seismic waveform classification method for describing the facies variation of Middle Jurassic Doig sandstones in Western Canada. We pick up the same example to determine the changes of impedance in the Doig sandstones in a lateral sense, which would be reflective of the porosity changes.

In figure 2a we show a horizon slice from the relative acoustic impedance volume derived from the input data volume. Within the sand boundary (in black), there are variations in impedance.

We also computed the absolute acoustic impedance on the same volume, which is shown in figure 2b. Though the overall impedance pattern still appears more-or-less the same, there is a significant variation in the impedance (red color) within the sand boundary.

The May 2008 Geophysical Corner described a thin-bed reflectivity inversion method that outputs a reflectivity series –and demonstrates that the apparent resolution of the inversion output is superior to the resolution of the input seismic data used to generate the reflectivity response. This aspect makes the method ideal for detailed delineation and characterization of thin reservoirs.

We derived the reflectivity from the seismic data used for generating the impedance displays in figure 2, and then generated relative acoustic impedance from it. The result is shown in figure 3, where the impedance variation exhibits a pattern that is more focused than the impedance variation in figure 2a and similar to that in 2b – but superior.

The main reason for this is probably the absence of the seismic wavelet in the data used to derive the relative acoustic impedance in figure 3.

Impedance inversion on highly resolved seismic data retrieved in the form of reflectivity is useful for making accurate interpretations, and so proves advantageous.

We thank Arcis Seismic Solutions for permission to present this work.

Comments (0)


Geophysical Corner

Geophysical Corner - Satinder Chopra
Satinder Chopra, award-winning chief geophysicist (reservoir), at Arcis Seismic Solutions, Calgary, Canada, and a past AAPG-SEG Joint Distinguished Lecturer began serving as the editor of the Geophysical Corner column in 2012.

Geophysical Corner - Ritesh Kumar Sharma

Ritesh Kumar Sharma is with Arcis Seismic Solutions, Calgary, Canada.

Geophysical Corner

The Geophysical Corner is a regular column in the EXPLORER that features geophysical case studies, techniques and application to the petroleum industry.


Image Gallery

See Also: CD DVD

Desktop /Portals/0/images/_site/AAPG-newlogo-vertical-morepadding.jpg?width=50&h=50&mode=crop&anchor=middlecenter&quality=90amp;encoder=freeimage&progressive=true 5489 CD-DVD

See Also: DL Abstract

This lecture presents a history of sea-level changes focusing on the last 100 Myr. Prior to the Oligocene (ca. 33.5 Ma), the Earth had been a warm, high CO2 Greenhouse world that was largely ice-free back to 260 Ma, though recent evidence suggests that 15-25 m sea-level changes observed may have been caused by growth and decay of small, ephemeral ice sheets. The growth and decay of a continental scale ice sheet in Antarctica caused 50-60 m variations on the 106 yr scale beginning ~33.5 million years ago (Ma).

Desktop /Portals/0/images/_site/AAPG-newlogo-vertical-morepadding.jpg?width=50&h=50&mode=crop&anchor=middlecenter&quality=90amp;encoder=freeimage&progressive=true 11668 DL Abstract

See Also: Environmental Geosciences Article

We use sediment ages and mercury (Hg) concentrations to estimate past and future concentrations in the South River, Virginia, where Hg was released between 1930 and 1950 from a manufacturing process related to nylon production. In a previous study, along a 40 km (25 mi) reach, samples were collected from 26 of 54 fine-grained deposits that formed in the lee of large wood obstructions in the channel and analyzed for grain size, Hg concentration, and organic content. We also obtained radiometric dates from six deposits. To create a history that reflects the full concentration distribution (which contains concentrations as high as 900 mg/kg [900 ppm]), here, we treat the deposits as a single reservoir exchanging contaminated sediments with the overlying water column, and assume that the total sediment mass in storage and the distribution of sediment ages are time invariant. We use reservoir theory to reconstruct the annual history of Hg concentration on suspended sediment using data from our previous study and new results presented here. Many different reconstructed histories fit our data. To constrain results, we use information from a well-preserved core (and our estimate of the total mass of Hg stored in 2007) to specify the years associated with the peak concentration of 900 mg/kg. Our results indicate that around 850 kg (1874 lb) of Hg was stored in the deposits between 1955 and 1961, compared to only 80 kg (176 lb) today. Simulations of future Hg remediation suggest that 100-yr timescales will be needed for the South River to remove Hg-contaminated sediments from the channel perimeter through natural processes.
Desktop /Portals/0/PackFlashItemImages/WebReady/hero-EG-Journal-21-1-2014.jpg?width=50&h=50&mode=crop&anchor=middlecenter&quality=90amp;encoder=freeimage&progressive=true 5780 Environmental Geosciences Article

See Also: Online Certificate Course

Unconventional Resources is an online course that enables participants to learn about shale gas, shale oil and coalbed methane.

Desktop /Portals/0/PackFlashItemImages/WebReady/oc-cc-unconventional-resources.jpg?width=50&h=50&mode=crop&anchor=middlecenter&quality=90amp;encoder=freeimage&progressive=true 1473 Online Certificate Course

See Also: Online e Symposium

Desktop /Portals/0/PackFlashItemImages/WebReady/oc-es-eagle-ford-shale-prospecting-with-3d-seismic-data.jpg?width=50&h=50&mode=crop&anchor=middlecenter&quality=90amp;encoder=freeimage&progressive=true 1482 Online e-Symposium