Natural fractures can improve the permeability of tight rock and are therefore often targeted by oil wells. But where do they occur in the reservoir and how do we find them ?
Seismic is limited by resolution but techniques such as interval velocities have some success. Here we look at some geological factors that can explain their occurrence. These are often known as fracture drivers. We should consider these during fracture modelling.
Rock type
In the outcrop in Figure 1, the dolomite on the right hand side is pervasively fractured (by joints), whereas the limestone on the left hand side is unfractured.
Figure 1. Cupido Platform Carbonates, Las Palmas Canyon, Mexico; unfractured limestones on the left, fractured dolomites on the right.
In another example from the same locality (Figure 2), dolomite infills burrows (darker grey patches) and is fractured, then cemented (small white streaks). The background limestone (lighter grey) is unfractured. See Fig 3-5 of Nelson (2001) which shows higher fracture density in dolomite than limestone. This is because dolomite is brittle.
Figure 2. Cupido Platform Carbonates, Las Palmas Canyon, Mexico. Only the burrows infilled by dolomite are fractured.
Layer thickness
Often, thicker layers will have less fractures (greater spacing) than thinner layers (e.g., the observations by McQuillan (1973) on some Zagros anticlines in Iran). A nice example of this from the Khaviz Anticline in Iran is given in Figure 3. We can explain this by the assumption that the width of the stress shadow (release of energy during fracturing) is a function of the height of the fracture. Taller fractures cast a wider stress shadow, forcing fractures to form at a wider spacing in thick beds.
Fracture swarms are usually more through going, and not confined to individual beds.
Figure 3. Fractures in the plunge area of the Khaviz Anticline, Iran, showing different fracture spacing in layers with different mechanical layer thickness. From Ole Petter Wennberg of Statoil.
Structural position
We have to be careful about using curvature/dip attributes to locate fractures on folds. This is because present day curvature will not always show all the fractures in a rock which has experienced several tectonic events. It will however, tell us which areas have experienced the most strain. In many fault-propagation folds, the steep forelimb will be more fractured than the backlimb as it has experienced more strain. However, in the backlimb of the fold in Figure 4, the higher permeabilities are controlled by structural location (curvature/dip).
Other common structural locations for fractures are close to seismically resolvable faults.
Figure 4. Fracture permeability model (x-direction, where red is highest permeability and pink/purple lowest) on the backlimb of an anticline (From Wennberg et al. 2006).
Concluding remarks
There is no golden bullet (seismic) attribute to predict fractures so it very useful to consider the above geological drivers when building a fracture model. Statistical techniques like discriminant analyses (Gauthier et al. 2000) work quite well which consider a combination of attributes and geological factors to explain the distribution of fractures.
References
Gauthier, B, Zellou, A, Toublanc, A, Garcia, M, Daniel, J-M. 2000. Fractured Reservoir Characterization: a Case Study in a North Africa Field. SPE 65118.
McQuillan, H. 1973. Small-scale fracture density in Asmari Formation of Southwest Iran and its relation to bed thickness and Structural setting. AAPG Bulletin, 57, 2367–2385.
Nelson, R.A. 2001. Geological analysis of naturally fractured reservoirs 2nd edn. Gulf Professional Publishing, Houston.
Wennberg, O.P, Svånå, T,M. Azizzadeh, M, Aqrawi, A.M.M, Brockbank, P, Lyslo, K.B and S. Ogilvie, S. 2006. Fracture Intensity vs. mechanical stratigraphy in platform top carbonates: the Aquitanian of the Asmari Formation, Khaviz Anticline, Zagros, SW Iran. Petroleum Geoscience. Vol. 12, 235–245.