Fault movement can drag rock layers up or down a fault plane, depending upon whether the fault movement is normal (Figure 1) or reverse (Figure 2). Such fault drag is common where there are ductile rocks in the faulted stratigraphy. Brittle rocks are less likely to drag against the fault plane.
Figure 1. normal drag of shales & limestones against a normal fault, kilve pill, somerset, UK
Figure 2. Reverse drag of shales and (hand pointing at) limestones against a reverse fault, Kilve Pill, Somerset, UK.
Interpreting Drag in the Subsurface
Figure 3. fault drag from bedding dips (interpreted on image logs).
Fault drag (Figures 1 and 2) of bedding can be detected on image logs. This is because image logs can be used to interpret bedding dip. The resulting tadpoles (where the blue dots on Figure 3 represents the amount of dip and their tail is the bedding azimuth) can be plotted on a track. An example from a PETREL project is given in Figure 3. You'd normally show the image itself next to this track in a well correlation panel.
The well was drilled through a seismically identifiable reverse fault and you can see where it approaches the reverse drag zone just below the dark red line (about 2 thirds of the way down); where the bedding dip starts to increase from about 20 degrees right up to about 85 degrees. I've interpreted the fault plane in the data gap area (F) - just below the green line to where the dips go back to 20 - 30 degrees.
Drag is commonly seen on seismic sections through faults. Rob Butler has interpreted the folding in a section through a normal fault in Figure 4a (point 2) as being caused by distributed shear. It could also be interpreted as drag (which we'll carry here; either way there are implications for communication) although I'm struggling with an explanation for the footwall folding.
Impact of Fault Drag
Fault drag changes the communication path across a fault (Hesthammer & Fossen, 2000). I've put some well pairs into 2 scenarios (in the red layer in Figure 4b, c) to illustrate this. They wouldn't communicate in a non-drag situation. In the most likely drag-situation (as the data suggests), they would. Bear in mind this is a single (seismic) line interpretation and we'd need to check other lines and integrate other data (e.g., Figure 3), when planning wells.
Concluding Remarks
Think about fault drag when building structural framework models and turning them into geo-cellular grids. This is because it can alter the communication pathways in your reservoir, impacting history matching efforts and the degree to which wells will "talk to each other". If your stratigraphy is like that at Kilve Pill in Somerset (shale/limestone - Figures 1, 2), then drag is more likely than if more brittle rocks such as dolomite are present.
Figure 4 (a) Seismic section from the Moray Firth Basin, with potential normal drag against the fault at point 1. Scene is approx 2 km wide by 1200 ms. (b) non-drag scenario, (c) drag - most likely as suggested by the data. Red and yellow sand units are separated by shales.
VIRTUAL SEISMIC ATLAS: http://www.seismicatlas.org/uploaded/image/201501/0c926fe1-13c5-45d4-84ad-3f0b03f17197.jpg. AUTHOR ROBERT BUTLER.
References
Hesthammer, J, Fossen, H. 2000. Uncertainties associated with fault sealing analysis. Petroleum Geoscience, 6, 2000, 37 – 45.