Fault shadow pertains to the distorted visualization of a foot wall fault block in seismic, due to lateral velocity gradients across a fault zone, or strong layer velocity contrasts across a fault. These strong lateral velocity changes cause false geometric constructions around the fault zone. The distortions are more apparent in time domain seismic which increases an interpreter's risk when first analyzing a seismic volume that has not been depth migrated. Interpretation mistakes can further amalgamate to produce poor results within a general exploration workflow.
Sometimes called a fault shadow zone, fault shadow is characterized by poor illumination and false structural representation of sub-fault seismic reflection data. Although this phenomenon is known to occur in many modes of faulting, most cases accompany normal and reverse faulting associated with extensional and compressional stress environments, respectively. While the foot wall block imaging is untrustworthy, the hanging wall block typically remains undistorted, so the fault location can be accurately assessed but the form of the underlying layers may be disingenuous.
Causes and Effects
The local velocity changes across a fault are due to fault-slip placing beds with different seismic velocities adjacent to each other. This velocity difference controls the incident and reflected ray paths, defined by Snell’s Law, affecting the travel-time values associated with the foot wall reflectors. As a result, the effects of fault shadow are most aptly evidenced in time domain data. The dominant visual seismic features caused by fault shadow are velocity sags and velocity pull-ups.
As can be seen in Figure 2, a velocity sag in an extensional regime occurs when the relatively fast-velocity layer in Figure 1 becomes thinner across the fault zone as it is separated and juxtaposed with slower velocities. The paucity of the fast velocity layer causes a travel time increase and therefore a sag, or push-down effect of the underlying reflectors. Oppositely, a thinning of a low velocity layer across a fault will cause a pull-up effect within the foot wall block. This happens because the lack of low-velocity material decreases the ray travel time, creating an antiformal feature. Figure 3 shows a seismic example of normal faulting causing velocity sags.
Figure 1: Layer cake velocity model with normal fault separating a shallow high velocity layer in the depth domain. 
Figure 2: Layer cake velocity model (same as Figure 1) showing fault shadow effects of sag and pull-up in the time domain. 
Figure 3: Compound fault shadow developing from successive normal faulting of an anisotropic layer.
In compressional stress regimes, these effects still occur but happen by different means. A reverse or thrust fault generally places deeper, faster velocity lithologies, in the hanging wall, adjacent to shallower slower velocity lithologies in the foot wall as seen in Figure 4. This can cause an accumulation of fast velocity rocks that lower the ray travel time and lead to a pull-up effect in the foot wall seen in Figure 5. Likewise, a push-down or sagging effect will occur where there is an increased thickness of slower velocity rocks causing increased travel time.
Figure 4: Layered velocity model of a fault propagation fold caused by reverse faulting in the depth domain. 
Figure 5: Fault shadow effects in a reverse fault model in the time domain. Large pull-up structure caused by a stacking of the fast velocity layers seen in Figure 4. 
Another effect surrounding fault zones is the poor illumination of the foot wall fault block. This anomaly, characterized by a low seismic impedance or amplitude response of reflectors in the foot wall block, can be caused by either physical low impedance layers or may be indicative of additional local faulting.
Time domain data is generally the initial seismic data produced and interpreted at the onset of exploration. This data is also the most prone or susceptible to the fault shadow effects mentioned. In Figure 7, fault shadow effects result near an area of prospective hydrocarbon development. Thus, there are implications for interpreters regarding this seismic manifestation. Firstly, the pull-up and sag effects may misguide an interpreter into incorrectly identifying geologic structures in the seismic. What may look like a rollover feature could be an artifact of velocity perturbations caused by the adjacent fault. In an oil and gas exploratory mindset, this anticlinal feature may appear promising for hydrocarbon accumulation, raise undue interest, and possibly drive further development that results in no production. Along with creating completely erroneous features, fault shadow effects can also more subtly enhance features that do exist. Issues that arise from this circumstance are exaggeration of a structure's aerial and vertical extent. When the overall geometry of key target features are affected and they contain hydrocarbons, over-estimation of volume accumulations can occur. Additionally, both pull-up and sag effects can be mis-interpreted as fault drag deformation. Qualms expressed about illumination issues can also be tied back to oil discovery. An interpreter may interpret the poor amplitude response as low impedance reflectors or as a highly faulted and fractured zone associated with the main fault. Both have implications for hydrocarbon exploitation. A physical low impedance contrast interpretation may cause an overlook of the questioned area due to the lack of a high amplitude anomaly commonly associated with the presence of hydrocarbons, and being related to direct hydrocarbon indicators. The interpretation of the shadow zone as being highly faulted would have implications for more downstream applications such as production development plans.
Solutions to solve the fault shadow issues have been investigated. Both pre-stack and post-stack time migration and depth migration processing steps have been studied, with, generally, the best results in seismic event geometry resulting from pre-stack depth migration (PSDM) procedures. The migration techniques may involve either Kirchhoff, Reverse Time Migration (RTM), or Wave Equation Migration (WEM) algorithms to more adequately adjust the seismic image. However, these more sophisticated methods of processing have a higher cost association.
- Tucker, P., Yorston, H. (1972, March). Pitfalls in Seismic Interpretation. Retrieved October 22, 2018, from http://ngds.egi.utah.edu/files/GL01406/GL01406_2.pdf
- Julien, P., Broyer, J. L., & Bernet-Rollande, O. (2000, April). Fault Shadow Correction Methodology. Retrieved October 22, 2018, from http://archives.datapages.com/data/ipa/data/027/027001/1_ipa027g-065.htm
- Fagin, S. (1996, September). The fault shadow problem: Its nature and elimination. Retrieved October 22, 2018, from https://library-seg-org.ezproxy.lib.ou.edu/doi/pdf/10.1190/1.1437403
- Trinchero, E. (2000, February). The fault shadow problem as an interpretation pitfall. Retrieved October 17, 2018, from https://library.seg.org/doi/10.1190/1.1438549
- Habit 5: Exploring the 10 Habits. (n.d.). Retrieved from https://scacompanies.com/resources/ten-habits/habit-5-exploring-the-10-habits/
- Sumit, V., Mutlu, O., & Marfurt, K. (2013, August). Seismic modeling evaluation of fault illumination in the Woodford Shale. Retrieved October 22, 2018, from https://library.seg.org/doi/pdf/10.1190/segam2013-0491.1
- Azuoko, G. B., Ehirim, C. N., Ebeniro, J. O., & Uraechu, D. (2017, January 02). Analysis of multiples in onshore Niger Delta: A prelude to fault shadow phenomenon. Retrieved from https://link.springer.com/article/10.1007/s13202-016-0309-8
- Michel, L., Hoo, C. H., & Flynn, M. (n.d.). IPTC 14817 Application of PSDM Imaging for Reservoir ... Retrieved from https://www.cgg.com/technicalDocuments/cggv_0000011691.pdf
- Grasemann, B., Martel, S., & Passchier, C. (2005, June 09). Reverse and normal drag along a fault. Retrieved October 23, 2018, from https://www.sciencedirect.com/science/article/pii/S0191814105000738
- Nie, Y., Wei, J., Zheng, J., & Quan, W. (2012, March 14). A Forward Simulation Research to Resolve the Fault Shadow Problem. Retrieved October 23, 2018, from https://www.sciencedirect.com/science/article/pii/S1876610212000288