Mitigation/Stimulation:
Determining if it is Matrix or Fractured Injection

Key Issues

Most injectors are believed to be fractured during their injection life. Injection into fractured wells is characterized by a derived injection pressure at the formation/sand face which is higher than a value that is representative of an in situ stress equivalent to a fracture propagation of closure pressures. However, without fracture closure stress information it is often difficult to identify when the injector is under matrix injection and when the injector is fractured. This is particularly true when thermal stresses and poroelastic effects on fracture propagation pressure are considered. The key issues and considerations for identifying the injection regime (matrix or fractured) include:

1. Compare the fracture propagation pressure with the formation/sand face pressure. Be certain to account for poroelastic and thermoelastic effects in your considerations.
   
2. Compare the formation/sand face pressure to any known closure pressure or far-field stress. Be certain to account for poroelastic and thermoelastic effects.
   
3. If there is no stress information (and even if there is) plot the formation/sand face pressure (do not use the wellhead pressure) against the surface injection rate and search for a significant break in the curve.
   
4. Where possible, always consider poroelastic and thermoelastic effects in calculating the far field stresses.
   

Relevant references are cited.

If the Fracture Propagation Pressure is Known?

Case 1: When you know the fracture propagation pressure, compare it with the formation face pressure. Be certain to consider poroelastic and thermoelastic effects.

When the relevant propagation pressure is known, for example, from a step-rate test, identifying whether the well is experiencing matrix or fracture injection requires calculation of the formation face injection pressure and comparing this pressure with the fracture closure stress (or equivalent). Remember that the pressures required to reopen, initiate, maintain open and/or propagate the fracture can change with changes in stress near the wellbore (or around a pre-existing fracture) due to poroelastic effects and due to temperature changes associated with injection of fluid at a temperature that is different from the static reservoir temperature.

If the formation face pressure (i.e., downstream of the completion) is greater than the pressure required to extend the fracture (or initiate a fracture if one does not already exist), the injection is occurring under fracturing conditions. If the fracture is stagnant, it is injecting under matrix conditions (as is also the trivial case where a fracture has not been initiated).

The Wellbore Hydraulics tool and the radial damage model, PWRAD, should be used in analyzing injection data to identify the relevant pressure (formation face - not bottomhole and NOT wellhead).

 Back to Key Issues

If the Closure Stress is Known?

Case 2: If closure pressure or far-field stress is known, compare the formation face pressure with the closure pressure. Be certain to consider poroelastic and thermoelastic effects..

When the fracture closure pressure or far-field stress is known, you may need to estimate fracture propagation pressure from far-field stress and fracture size. For sizable fractures in produced water injection applications, hydraulic fracturing simulation often shows that the net pressure may be small, and, for most formations, you can assume that the propagation pressure is fairly close to the in-situ stress. However, thermal stresses and poroelastic stress effects on fracture propagation pressure must be incorporated in estimating fracture propagation pressure.

Some very rough guidelines (based on Sponsor experience) are available. For every foot of additional fracture length the injection pressure will increase by approximately 2 psi for a 100 md formation and 3 psi for a 1 darcy formation. Use the Multilateral/Multilayer Injection tool to estimate these effects.

 Back to Key Issues

If there is No Stress Information?

Case 3: If you have no reliable stress information, with which to compare the formation face pressure, either measure it specifically (refer to Step Rate Testing in Monitoring) or, at the very least, attempt to infer performance from historical rate-pressure data.

• Plot formation face pressure (wellhead pressure plus the hydrostatic head minus tubing AND completion friction) versus the surface injection rate, using all available data. There is a Toolbox routine for doing this, and for estimating formation face pressure from wellhead pressure and rate (Figure 1).
• DO NOT USE WELLHEAD PRESSURE.

Figure 1: This is a plot of the formation face pressure with rate for Well NAM-1. In this plot, it is important not to draw too many conclusions on early time data because there was a non-return valve in line and friction may not have been adequately represented.

• If a change in slope is clearly shown on this plot, the portion before the slope change may be regarded as probable matrix injection while the portion after the change may be regarded as probable fractured injection.
• If there is a large scatter of data, moving average processing can be helpful in differentiating "characteristic" slopes. Plotting of the moving average of the data is also helpful in determining the inflection point (or the fracture opening/reopening stress).
• The plot should be demarcated with different symbols, depending on the date, to discriminate if there are events with chronological significance. This is especially true when different types of water (source, quality, temperature …) are injected during the life of the injector. Figure 2 is an example.

Figure 2: This is a plot of the formation face pressure with rate for Well NAM-1, showing all of the data after the NRV had been removed.

• Figure 2 shows that injection pressure is not strongly dependent on injection rate (often a good indication of fractured injection!) and in fact shows slight reductions in injection pressure as the rate was increased. What could be the cause of this? The possibilities include erroneous calculation of BHP, partial fracture reopening with greater fracture width at higher rates due to reduced near-wellbore entrance losses, thermal effects, upwards fracture growth, or opening of fractures in other zones, or pressure-dependent boundary leakage.... The analysis indicates that drawing conclusions without in-situ stress information can be particularly difficult. This is because pre-existing fractures are conductive at injection pressures below "closure" stress.

 Back to Key Issues

Stress Changes Due to Pressure and Temperature Changes in the Reservoir

Recommendation 1: Wherever possible account for stress changes due to poroelastic and thermoelastic effects.

If the far- or near-field pore pressure changes (either due to voidage effects) or simply due to the fluid leaving the wellbore or fracture, the in-situ stresses acting on the fracture will change. An increase in pressure leads to an increase in stress and vice versa. Similar effects arise due to temperature changes in the reservoir caused by injecting a fluid that is at a temperature that is different from the reservoir itself. Figure 3 is an excellent example, from a produced water injection pilot.

Figure 3: An example of the variation in stresses near a wellbore for injection of different temperature fluids - from a North Sea PWRI pilot.

• For radial injection, precise determination of the influence of poro- and thermoelastic stress changes is possible using the WellStress program in the PWRI Toolbox.
• For radial injection, precise determination of the influence of poro- and thermoelastic stress changes is possible using the WellStress program in the PWRI Toolbox.
• For radial and fractured injection, a good analytical estimation of the influence of poro- and thermoelastic stress changes is possible using the Multilateral/Multilayered program in the PWRI Toolbox.
• To learn more about poro- and thermoelastic effects, click here.

 Back to Key Issues

References

1. McLennan, J.D. and Roegiers, J-C.: “Do Instantaneous Shut-in Pressures Accurately Represent the Minimum Principal Stress,” Workshop on Hydraulic Fracturing Stress Measurement, Monterey, CA, December 1981.
2. McLennan, J.D. and Roegiers, J-C.: “How Instantaneous are Instantaneous Shut-in Pressures,” SPE 11064, 57th Annual Fall Technical Conference and Exhibition, SPE/AIME, New Orleans, September 1982.
3. Detournay, E., McLennan, J. and Roegiers, J-C.: “Poroelastic Constants Explain Some of the Hydraulic Fracturing Mechanisms,” Proc. Unconventional Gas Technology Symposium, SPE 15262, Louisville, KY, May 1986.
4. Detournay, E., Cheng, A.H.-D., Roegiers, J-C. and McLennan, J.D.: “Poroelastic Considerations in In-Situ Stress Determination by Hydraulic Fracturing,” 2nd International Workshop on Hydraulic Fracturing Stress Measurement, Minneapolis, MN, June 1988.
5. Detournay, E., Cheng. A.H.-D. and McLennan, J.D.: “A Poroelastic PKN Hydraulic Fracture Model Based on an Explicit Moving Mesh Algorithm,” J. En. Res. Tech.

 Back to Key Issues