Its useful to consider these 2 broad categories of fractures when characterising fracture reservoirs (i) diffuse fractures and (ii) fracture swarms. Geologically, there are of course different type of fractures i.e., joints, veins, faults, stylolites which populate these 2 broad categories. Here we look at the key features of fracture swarms and diffuse fractures and how we can tell them apart when characterising fractured reservoirs.
DIFFUSE FRACTURES
These are usually joints which are spread throughout the rock body (Fig. 1, 2) - often referred to as background joints or regional joints. They are not related to faults or fracture swarms and may be bed bound (mechanical stratigraphy) i.e., thinner beds often contain more fractures than thicker beds. There are many ways in which joints can form from burial to unloading or exfoliation.
Figure 1. Diffuse fractures (joints) in limestone, Kilve Pill, Somerset, UK.
Fracture swarms
Also called fracture corridors, these are zones of concentrated fractures (joints and faults) across which there may be little or no displacement. These are shown by arrows in a dolomite cliff section in Fig. 2 There is a widespread diffuse fracture network in between the arrows in Fig 2. Another example in a carbonate (oolitic grainstone) given in Fig. 3 - the white arrow shows the extent of the small fault/swarm to the right hand side of the slip surface (dashed red line).
In general, fracture swarms are more through-going (of the stratigraphy) than diffuse fractures which can be bed-bound.
Figure 2. Fracture swarms in dolomite - on cliff sections, Marsden Bay, NE England.
Figure 3. Small fault or fracture swarm in (high porosity) oolitic grainstones, Mere Knolls Road, Sunderland, England. Highest fracture density is within the zone between the 2 dashed white lines. Extent of damage zone shown by blue arrow across the top.
In the Subsurface…
As mentioned above, fracture swarms are often below seismic resolution, therefore we need to look at other types of data. The obvious one is an image log where a clustering of fractures at a particular depth is a clear indicator. Its worth trying to reconcile this data with any observed penetration of a seismically resolved fault.
Ideally we should have a PLT (production log test) to lay along side the image log data to check out the percentage flow contribution of the swarm. A through going fracture swarm is illustrated on the left hand side of Fig. 4 - on the right hand side an expected PLT response is given (red arrow). Note how there is a larger % contribution to flow over the same interval from the swarm, than the flow contribution from interval with only diffuse fractures. In general, we’d expect large influxes of flow over short intervals - it may be that only a few large fractures control production.
Figure 4. (a) Block diagram showing the main characteristics of diffuse fractures (Wennberg et al. 2005) vs. fracture swarms (b) possible production response (from Wennberg et al. 2006)