IADC/SPE 87223 Bounds on In-Situ Stress Magnitudes Improve Wellbore Stability Analyses

Bernt S. Aadnoy/Stavanger University and Ann K. Hansen/Shell

Copyright 2004, IADC/SPE Drilling Conference This paper was prepared for presentation at the IADC/SPE Drilling Conference held in Dallas, Texas, U.S.A., 24 March 2004. This paper was selected for presentation by an IADC/SPE Program Committee following review of information contained in a proposal submitted by the author(s). Contents of the paper, as presented, have not been reviewed by the International Association of Drilling Contractors or Society of Petroleum Engineers and are subject to correction by the author(s). The material, as presented, does not necessarily reflect any position of the International Association of Drilling Contractors or Society of Petroleum Engineers, their officers, or members. Papers presented at IADC/SPE meetings are subject to publication review by Editorial Committees of the International Association of Drilling Contractors and Society of Petroleum Engineers. Electronic reproduction, distribution, or storage of any part of this paper for commercial purposes without the written consent of the International Association of Drilling Contractors and Society of Petroleum Engineers is prohibited. Permission to reproduce in print is restricted to a proposal of not more than 300 words; illustrations may not be copied. The proposal must contain conspicuous acknowledgment of where and by whom the paper was presented. Write Librarian, SPE, P.O. Box 833836, Richardson, TX 75083-3836, U.S.A., fax 01-972-952-9435.

A field case from the North Sea demonstrates the application of the stress bounds, and shows that prognosis are obtained that better corresponds to observed behavior of the wells. Introduction Bradley1 is acknowledged as the person that introduced geomechanics to the drilling industryat about 1980. At that time the oil industry moved into deep wells , high-angle wells and long-reach wells. It became clear that wellbore stability was required to be able to drill these wells efficiently. Two decades ago, one was not certain if high-angle wells could possibly be drilled. Bradley's paper therefore received high interest in the drilling community. Today one routinely drill wells at any inclination. Stable horizontal wells are common in many oil fields. The initial fear of high-angle wells is removed. However, due to the increased lengths of today's wells, borehole stability remains an important issue. Hydraulic fracturing is a phenomenon that occurs at high wellbore pressures. When a critical pressure is exceeded, the borehole ruptures along the axis of the hole, leading to fluid losses from the borehole. Conversely, at low borehole pressures, a high shear stress arises, sometimes leading to borehole wall failure. The consequence can be an enlarged hole, with collapsed rock causing fill problems or stuck drillstring. Shear failure may also occur at high borehole pressures, but these may not result in hole enlargement. Today borehole stability analysis is routinely performed for offshore wells. It requires in-situ stresses as governing input parameters. Usually these are obtained from so called LeakOff-Tests, which are conducted after each casing is set to verify the pressure integrity of the formation below the casing. Aadny2 presents various methods to deduce the in-situ stress state from such measurements. One may also obtain stress information from extended LOT tests and massive fracturing operations. Although a common practice is not yet established, both Raaen and Brudy3 and Addis et.al.4 discusses various interpretation issues. This paper will not further discuss interpretation, but focus on the maximum permissible stress states that give physically correct results. Figure 1 shows the problem addressed in this paper. It is sometimes seen that the collapse curve exceeds the fracturing curve for high wellbore inclinations. Many horizontal wells

Abstract Wellbore stability analysis is becoming a necessary tool to improve operational economy and to avoid wellbore problems. Essential elements here are estimation of critical fracturing and collapse pressures to obtain an optimal mud weight window. The input data for this analysis is mainly pore pressure predictions from many sources like logs and drilling exponents, overburden stresses from logs or drilled cuttings, leak-off tests at casing shoes and breakout analysis from caliper logs. From these and other data, estimates for in-situ stresses and directions are obtained, which again serves as input for wellbore stability modeling. It is evident that the input data comes from many different sources and can therefore not be considered consistent. During wellbore stability modeling it has often been observed that unrealistic results appear. Sometimes one observes a critical collapse pressure that exceeds the fracturing pressure, clearly a faulty result. These results are often ignored. However, there must be an inherent error in some of the input data. This paper addresses this problem and concludes that the relative magnitudes of the in-situ stresses are the main cause of this. Equations are given that defines the bounds on the insitu stresses for different stress regimes. Using the bounds on the horizontal in-situ stresses on field cases, realistic fracturing and collapse prognosis are always obtained . Since the models also define the minimum permissible anisotropic stress state, these can be used as default parameters when field data are missing.

www.petroman.ir

IADC/SPE 87223

have been drilled the past decades, demonstrating that the previously discussed crossing of stability curves is not physically correct. Aadny and Chenevert5 showed that by transforming the borehole inclination from vertical to horizontal, incorrect results might appear. This was handled by introducing a correction factor. In Fig. 1a is a wellbore stability showing a crossing of the critical collapse and fracturing curves at an inclination of about 60 degrees. The conclusion that this well cannot be drilled for higher inclinations is obviously wrong, as it is not possible for a well to fracture and collapse at the same pressure. The problem is obviously the relative magnitude of the principal in-situ stresses. Figure 1b shows a correct wellbore stability plot, where the minimum distance between the critical fracturing and collapse curves are defined as . The problem of Fig. 1a is frequently observed also in recent borehole stability analysis. It is seen as a requirement to correct this problem to improve analysis. We will apply the following conditions to the analysis: 1) The pore pressure can never exceed the fracturing pressure or the horizontal stress, but it can exceed the critical collapse pressure. 2) The critical collapse pressure cannot exceed the critical fracturing pressure. The second condition is visualized in Fig. 1b. In the following will we determine bounds on the three principal in-situ stresses based on the conditions above.

seen that the only method that simultaneously estimate both maximum and minimum horizontal stress and direction is the inversion method6. A field study of the Snorre field offshore Norway shows very good results 7. Djurhuus and Aadnoy8 further extend the inversion method to include analysis of borehole fractures from image logs. Bounds on the in-situ stresses. Tan et. al.9 presented a method to establish bounds on the insitu stress state from wellbore data. For each stress state (normal, reverse, strike/slip), two bounds are established based on the critical fracturing and collapse pressures. A different approach is taken in this paper. We will analytically search for the minimum difference between the critical collapse and fracturing pressures, as derived in the following. The most common scenario is a normal fault stress state. The in-situ stresses are as follows:

v > H > h
We are searching a general expression, considering any orientation inside a 3-dimensional space. We define the most extreme conditions as found within the principal stress space. Investigation shows that these are found when the borehole is aligned along one of the principal in-situ stress axes. We will rotate the borehole along these three axes, one at a time, and compare the results. First we will consider a vertical hole, = 0. The transformed in-situ stresses, the borehole stresses and the principal stresses are obtained from Appendix A, and shown in Table 2. First consider the direction = 0. The following principal stresses results:

The In-Situ Stresses The in-situ stress state is usually assumed to coincide with the vertical and horizontal directions. One assumes that the principal in-situ stress tensor consists of a vertical principal stress, which is equal to the overburden stress, and two unequal horizontal stresses. In a relaxed depositional geological setting these are usually lower than the overburden stress. However, in strongly tectonic stress regimes the horizontal stresses may exceed the vertical stress. Based on the relative magnitude of the three principal in-situ stresses, the stress state is defined as either normal fault, or as reverse or strike slip fault stress states. Many indirect methods have been used to estimate the magnitude of the in-situ stress state. However, only fracturing measurements gives a direct measure for the in-situ stress level. There are several different interpretations of this pressure. Addis et.al.4 evaluate different ways of interpreting in-situ stresses from LOT tests. They conclude that the minimum horizontal stress can be estimated from the shut-in response of an extended LOT test, or from a conventional LOT test. The maximum horizontal stress is more difficult to estimate from single borehole measurements. Table 1 summarizes common methods used to assess the in-situ stress state. It is

r = Pw

= H + h 2 ( H h ) cos 2 Pw

(1)

z = v
Fracturing will arise in the direction of the maximum normal stress,that is at = 0, resulting in the following expression for the fracturing pressure:

3' = 'j = 3 h H Pwf Po = 0

Pwf = 3 h H Po

(2)

Collapse will arise in the direction of the minimum normal stress, = 90.

r = Pw = 3 H h Pw z = v z = 0

(3)

www.petroman.ir

IADC/SPE 87223

With reference to Eqn. A3, is the maximum principal stress, whereas r is the minimum principal stress. Inserting these into Eqn. B3, the collapse pressure becomes:
Pwc = 1 ( 3 H h )(1 sin ) + Po sin o cos 2 (4)

Generalized stress bounds. In the following generalized stress bounds will be developed. Using similar assumptions as in the previous example, the stress bounds are shown as a function of pore pressure in Fig. 3. It is seen that the stress limits are clearly dependent on the pore pressure. It is also clear that all three bounds given in Table 3 are not governing, one principal direction is always more critical. Based on the results shown in Fig. 3, Table 4 defines the upper and lower bounds of the in-situ stresses. Figure 3 shows the physically permissible regions for the three in-situ stresses for all stress states. It is seen that an increase in pore pressure leads to an reduction of the permissible stress range. At the extreme pore pressure, only a hydrostatic stress state is possible. The connection and dependence between the in-situ stresses and the pore pressure is clearly seen.

Referring to Fig. 1b, and the conditions defined earlier, we require that the critical fracture pressure always exceeds the critical collapse pressure by an amount , called the stability margin. This can be stated as:
Pwf Pwc

(5)

Inserting Eqns. 2 and 4 into Eqn. 5, the following condition results.

h ( 7 sin )

Application of the model (6) The model can be applied in various ways to present realistic results. Below are a few applications described. a) Knowledge of instability problems in a well, or stable borehole sections. Some wells have known stability problems in given directions. Having such a case at hand, the critical frac and collapse pressure must be established. From these the stability margin can be established along one of the principal in-situ stress directions. If the well is hardly drillable, a very small stability margin may result. Also, trouble-free borehole sections can be used to assess stress level limits. b) Having no prior knowledge of stability problems. The evolution of modern drilling technology has led to the observation that most boreholes can be drilled in any direction, provided good planning and operational follow-up. However, the margins between success and failure are sometimes small. For this general scenario, a default model is proposed as follows. Letting the stability margin equal the cohesive strenght in Table 4, the last argument of constant C vanishes, and it becomes: C = 2Po(1+sin)/v Another advantage given by the model is the possibility to develop a continuous stress profile with depth, as opposed to current practise where one obtain discrete stress states at each casing shoe (LOT measurement). This will be demonstrated in the following field cases.

H ( 5 3sin ) + 2 Po (1 + sin ) 2 o cos + 2

Realistic results will be obtained if this condition is satisfied. However, we have only studied one stress state, with the wellbore pointing along one principal axis. For the present problem, a normal fault stress state, a horizontally oriented wellbore must also be considered, pointing along both horizontal principal stress axis. The analysis will be similar to that above, only that the normal stresses and the failure positions change. It can be shown that Eqn. 6 can be generalized as follows:

min ( 7 sin )

max ( 5 3sin ) + 2 Po (1 + sin ) 2 o cos + 2

(7)

where min is the least normal stress and max is the maximum normal stress on a borehole, assuming the hole points in one principal direction. Figure 2 shows the three possible principal stress orientations for a borehole. For the present case of a normal fault stress state, the following bounds are found:

h ( 7 sin ) H ( 5 3sin ) + 2 Po (1 + sin ) 2 o cos + 2 h ( 7 sin ) v ( 5 3sin ) + 2 Po (1 + sin ) 2 o cos + 2

(8) The same conditions can be derived for tectonic cases, using the following definitions for the in-situ stress state: Strike/slip fault stress state: Reverse fault stress state:

H ( 7 sin ) v ( 5 3sin ) + 2 Po (1 + sin ) 2 o cos + 2

H > v > h H > h > v

Table 3 summarizes the results of the analysis.

www.petroman.ir

IADC/SPE 87223

Field cases Field case A In this example we will investigate the in-situ stress bounds that gives realistic wellbore stability analysis results. The following data are used from a field in the North Sea: Depth: Overburden stress gradient: Pore pressure gradient: Cohesive rock strength: Friction angle of rock: 1700 m 1.8 s.g. 1.03 s.g. 0.2 s.g. 30 degrees

produce a stress state that covers the entire depth interval. Table 5 gives the bounds for the in-situ stresses. The minimum horizontal stress and the fracturing prognosis are shown in Fig. 4 The existing wells in the area have the 20 in casing strings placed at about 1100 m, and the 13-3/8 in. casings set at about 1800 m. Since LOT tests are conducted at these depths, an estimate of the stress state can be established at these depths only. Figure 4 shows the resulting horizontal stress and the fracture gradient for a vertical well, in the depth interval 1000-2000 m. Please observe that these are continuous throughout the entire depth interval. There is no longer a need to extrapolate between two casing points. Also observe that the fracture prognosis shows a continuous increase with depth. The results of Fig. 4 gives a significant improvement to the stress analysis performed over the field for the past decade.

Wells are drilled at different inclinations at this depth level. One horizontal section is drilled confirming that it is possible to drill wells in any direction. Evaluation of well data resulted in that the minimum difference between the fracturing and collapse pressure is estimated as = 0.173 s.g. Inserting the data into Table 4, the following bounds are established, assuming a normal fault stress state:

Conclusions

1 h 0.80, 1 H 0.80, v v

H h v v

(12)

These conditions guarantees that the minimum difference between the critical fracturing pressure and the critical collapse pressure never exceeds . To determine the magnitude of the two horizontal in-situ stresses, one must use other methods. For this case, a number of LOT data were analyzed using the inversion technique6,7,8. These results are in agreement with Eqn.12.

Based on the observation that non-physical results sometimes appear in wellbore stability analysis, this publication identifies the cause, which is the relative magnitude of the principal insitu stress tensor. By analyzing the critical fracturing and collapse pressure for a borehole that is transformed into all three principal in-situ stress directions, a stability margin is determined. Furthermore this leads to bounds on the stress state. Bounds for normal, strike/slip and reverse fault stress states are presented. Field cases demonstrates the application of the in-situ stress bounds. One case demonstrates the minimum stresses required to predict a stable horizontal well. The other field case uses the bounds to create a continuous stress profile also between the casing shoes where stresses are known.

Field case B Figure 4 presents the overburden gradient and the pore pressure gradient for a production field in the North Sea. The intermediate casing is set in the depth interval 1100 m - 1900 m. This interval is characterized by a pore pressure increase from 1.03 s.g. at top to 1.4 s.g near bottom. Most of the interval consists of young clays. Well inclinations are low in most wells, resulting in a limited amount of data to develop a general model. It is difficult to establish stress magnitudes that provide a realistic frac prognosis over the entire interval. Two problems were encountered. By changing the inclination towards horizontal, the aforementioned crossing of fracturing and collapse curves resulted, questioning the quality of earlier used in-situ stress data. Secondly, LOT data exists only at the 1200 m depth and 1800-1900 m depths because these are the setting dephts for the 18-5/8 in. and 13-3/8 in. casing strings. A well optimization study suggested to change the setting depth to somewhere in between these two depths. However, no data exist here. The stress states at the top and bottom are different. Using eithter of these gives good results in one end but poor results in the other end. We will use the method of this paper to

www.petroman.ir

IADC/SPE 87223

Nomenclature = normal stress = shear stress v = overburden stress H = maximum horizontal stress h = minimum horizontal stress = borehole inclination from vertical = borehole azimuth = angular position on borehole wall from x-axis Po = pore pressure gradient Pw = borehole pressure gradient t = tensile rock strength = stability margin: minimum distance between critical fracturing and collapse pressures LOT = Leak-Off-Test

References
1. Bradley,W.B.: "Failure of Inclined Boreholes". Trans. Of the ASME, Vol.101, Dec. 1979, pp. 232-239. 2. Aadny, B.S.: Modern Well Design. A.A.Balkema/Rotterdam/Brookfield/1996, ISBN: 90 5410 633 6. 3. Raaen, A.M. and Brudy, M.: "Pump-in/Flowback Tests Reduce the Estimate of Horizontal In-Situ Stress Significantly", paper SPE 71367 presented at the 2001 SPE Ann. Tech. Conf. And Exhibition, New Orleans, Louisiana, 30 Sept. 3 Oct. 4. Addis, M.A., Hanssen, T.H., Yassir, N., Willoughby, D.R. and Enever, J.: "A Comparison of Leak-Off-Test and Extended LeakOff-Test for Stress Estimation", paper SPE/ISRM 47235 presented at Eurock 98, Trondheim, Norway, 8-10 July. 5. Aadny, B.S. and Chenevert, M.E.: "Stability of Highly Inclined Boreholes", SPE Drilling Engineering, Vol. 2, No. 4, 364-374. 6. Aadny, B.S.: "Inversion Technique to Determine the In-Situ Stress Field from Fracturing Data", Journal of Petroleum Science and Engineering, 4(1990), pp. 127-141. 7. Aadny, B.S., Bratli, R.K. and Lindholm,C.: "In-Situ Stress Modeling of the Snorre Field", paper presented at Eurock 94, Delft, the Netherlands. Proceedings: Rock Mechanics in Petroleum Engineering. A.A.Balkema, 1994, pp. 871-878. 8. Djurhuus, J. and Aadnoy, B.S.:"In Situ Stress State from Inversion of Fracturing Data from Oil Wells and Borehole Image Logs. Journal of Petroleum Science and Engineering 38 (2003) 121-130. 9. Tan, C.P., Willoughby, D.R., Zhou, S. and Hillis, R.R.: "An Analytical Method for Determining Horizontal Stress Bounds from Wellbore Data". Int. J. Rock Mech. Min. Sci. 30:7, 1993, 1103-1111.

www.petroman.ir

IADC/SPE 87223

Appendix A: Equations for stresses The following equations are required for the analysis: The following matrix transforms the in-situ stresses to the borehole coordinate system:
2 2 cos cos sin 2 x y 2 2 z cos sin = xy 1 2 sin(2 ) cos xz 1 yz cos 2 sin 2 2 1 sin 2 sin 2 sin 2 cos 2 cos 2 sin 2 sin 2 1 sin(2 ) cos 2 1 2 sin sin 2 2 1 sin 2 sin 2 sin 2 0 cos 2 0 1 sin 2 2 0 H h v

k = ( + z )

1 2

1 2

( z )

+ 4 z

After each computation the indices are rearranged according to convention, i, j, k assumes values: 1 is largest, 3 is the least principal stress.

Appendix B: Wellbore failure criteria

Borehole fracturing The fracturing condition for the borehole is defined as follows:

3' = 3 P0 = t

(B1)

(A1) The stresses at the borehole wall are given by the following equations. The tangential stress component is the so-called Kirch equation. The axial stress is for a plane stress conditions.

The minimum principal stress equal to the tensile rock strength defines this condition. Often the rock tensile strength is neglected assuming natural cracks and fissures.

Borehole collapse. For borehole collapse we assume a Mohr-Coulomb failure model. This is overned by the maximum and the minimum principal stresses. The failure model is:

r = Pw = x + y
2 ( x y ) cos(2 ) 4 xy sin(2 ) Pw

z = zz

(A2)

= 0 + ' tan
where :

r = 0 tz = 0
z = 2 xz sin + 2 yz cos
As failure is governed by the principal stresses, the following matrix defines planes of principal stress.
0 i r 0 0 0 z 0 z z 0 0 0 0 k

1 ' ( 1 3' ) cos 2 1 1 ' = ( 1' + 3' ) ( 1' 3' ) sin 2 2 ' where : = P0

(B2)

Combining the equations above, the collapse model becomes:

j
0

' 1

' 3 ) ( 1' + 3' ) sin = 2 0 cos

(B3)

Taking the determinant of the above matrix, the principal stresses are given by the following eigenvalue equation:
( r ) {( )( z ) 2z } = 0

For borehole collapse at low borehole pressures, j will be the maximum principal stress 1 whereas i will be the least principal stress 3 (Appendix A).

The solutions are:

i = Pw j = ( + z ) +
1 2 1 2

( z )

+ 4 z

(A3)

www.petroman.ir

IADC/SPE 87223

Method: Individual LOT Empirical, LOT Extended LOT Inversion of LOT Breakout analysis

h x x x x

Direction of h

x x

Table 1: Common methods to estimate the principal in-situ stresses.

Transformed in-situ stresses (A1): Boreholes stresses(A2): r = Pw x = H cos 2 + h sin 2 = H + h Pw = sin 2 + cos 2
y H h

Principal stresses(A3): i = r

z = v xy = sin 2 ( h H ) xz = yz = 0
1 2

2 ( H h ) cos 2 ( + )

j = k = z

z = v z = 0

Table 2: Stress state for a vertical hole, = 0.

Stress state: Normal Fault

Bound 1: Bound 2: Bound 3: h A H B + C H A vB + C h A vB + C

Strike/Slip Fault h A H B + C Reverse Fault

v A H B + C v A H B + C

h A vB + C v A hB + C

hA H B +C

Where: A = 7 - sin, B = 5 3sin, C = 2Po(1+sin) + 2(-ocos) Table 3: Bounds for in-situ stresses for boreholes located in the 3 principal stress directions.

www.petroman.ir

IADC/SPE 87223

Stress state: Normal Fault

Upper Bound

h H , 1 v v

Lower Bound h H B +C , A v v

Strike/Slip Fault H

AC B

H 1 v h B +C A v h H , 1 v v

h 1 v
Reverse Fault

H h AC , v v B

Where:A = 7 - sin, B = 5 3sin, C = {2Po(1+sin) + 2(-ocos)}/v Table 4: General bounds for in-situ stresses for various stress states. In all entries: H > h.

Depth (m): 1000 1200 1400 1600 1800 2000

Po(s.g.): 0.97 1.02 1.15 1.33 1.36 1.26

v(s.g.): 1.73 1.78 1.82 1.85 1.87 1.91

h/v, H/v,> h(s.g.): 0.73 1.26 0.74 0.77 0.81 0.81 0.78 1.31 1.40 1.50 1.54 1.50

Pwf(s.g.): 1.55 1.60 1.64 1.67 1.73 1.73

Table 5: In-situ stress bounds and fracture prognosis for Field Case B. Parameters: = 0.1 s.g., = 35, o = 0.5 s.g.

www.petroman.ir

IADC/SPE 87223

90

90

a) Example of incorrect stress state

b) Physically correct result

Figure 1: Fracturing and collapse pressures versus wellbore inclination

Figure 2: Wellbore orientations investigated.

www.petroman.ir

10

IADC/SPE 87223

2.0 a) Normal fault stress state

1.0

0 0 2.0 1.0 2.0

b) Strike/slip fault stress state

1.0

0 0 2.0 1.0

c) Reverse fault stress state

1.0

0 0 1.0 2.0

Figure 3: Bounds on the in-situ stresses Assumptions: 30, 0.866

www.petroman.ir

IADC/SPE 87223

11

Gradient (s.g.) 0,5 0 1 1,5 2

500

h
Depth (mTVD) 1000

1500

Po
2000

Pwf
2500

Figure 4: Gradient plot for Field case B.

www.petroman.ir