Editor's note: We have something special for you this month. Donald A. Herron gives his take on interpreting unconformities. We thought it would be fun if you could try to pick the line yourself. Before you proceed, fire up your browser and go to pickthis.io/brazil. Once you have made your interpretation, you can see everyone else's picks, including Herron's.
|This tutorial originally appeared as a featured article in The Leading Edge in February 2015 — see issue.|
An unconformity is an important geologic boundary that has correlation aspects of horizons and faults, in the sense that horizons are tracked as continuous or predictably changing patterns of reflections, whereas faults are recognized and tracked as discontinuities. The seismic response to an unconformity depends on the nature of the contact between preunconformity and postunconformity sections:
- In areas where pre- and postunconformity sections are in layer-parallel contact, the seismic response to an unconformity is a single continuous reflection, meaning it can be tracked accurately as a horizon. It can be autotracked wherever possible and would have to be tracked manually where signal-to-noise ratio and image quality preclude feature- or correlation-based autotracking.
- In areas where pre- and postunconformity sections are in angular contact, the seismic response to an unconformity appears as reflection terminations or discontinuities between groups of reflections, meaning it can be tracked manually only in the same manner that faults are picked. I make the distinction that in the great majority of cases, signal-to-noise ratio and image quality preclude effective operation of fault autopicking algorithms.
The seismic image shown in Figure 1 is a 3D prestack time-migrated line from offshore Brazil. The unconformity in this example is most striking in the center of the figure where it is sharply angular, but careful observation in the finest tradition of the one-line interpretation reveals that the unconformity probably extends over the full extent of the line. The primary objective of this tutorial is to illustrate how an interpreter proceeds in picking this unconformity on the entire line.
As shown in Figure 2, manual tracking of the unconformity as a seismic horizon is straightforward where the unconformity is most clearly angular. The suggested practice is for the interpreter to pick this area first, as shown by the yellow horizon in Figure 2, and then decide how to pick into the areas where the unconformity is layer-parallel and can be autotracked (again, data quality permitting). Referring to Figure 2, this entails determining whether the unconformity is a peak (red) or a trough (black), which indicates the sign of the acoustic-impedance contrast across the unconformity as positive or negative, respectively.
In this example, the unconformity can be picked reasonably as a peak between the blue arrows and as a trough between the green arrows in Figure 2; note that the limits of these picks are interpretive and would not necessarily be placed similarly by two competent interpreters. The interpreter should extend the manually picked portions of the unconformity several traces into the areas where it can be autotracked so that the subsequent autotracking can be merged smoothly back into the manually picked areas (this might require iterating between manual and automatic picking until acceptable smoothness is achieved).
To the left of the trough reflection delimited by the green arrows in Figure 2, it is not obvious whether the unconformity extends farther to the left as a shallower peak or continues as a deeper trough; it is also possible that there are two unconformities on this part of the line. To the right of the area marked by the yellow horizon, the unconformity is a continuous peak for a short distance, but to the right of that area, its exact location is difficult to determine, and again, several unconformities might be present.
From the preceding discussion, it is clear that as a single surface, this unconformity is composed of different segments which alternately are tracked manually and automatically. Fortunately, most interpretation applications allow this for a single horizon as long as the waveform onset type for autotracking, that is, peak or trough, does not vary spatially — it can be one or the other, but not both (zero crossings are omitted from this discussion). If the waveform onset type does vary, as in this example, the unconformity must be tracked on separate horizons as a peak or trough, and the composite horizon must be built by merging these component horizons.
- Herron, D. A., 2011, First steps in seismic interpretation: SEG Geophysical Monograph Series No. 16. http://dx.doi.org/10.1190/tle34020238.1