SPE-175170-MS

Wellbore Stability Analysis for Trouble Free Drilling

S. A. Haidary, and H. A. Shehri, Saudi Aramco; A. Abdulraheem, King Fahd University of Petroleum & Minerals;
M. Ahmed, Schlumberger Saudi Arabia; M. H. Alqam, Saudi Aramco

Copyright 2015, Society of Petroleum Engineers

This paper was prepared for presentation at the SPE Kuwait Oil & Gas Show and Conference held in Mishref, Kuwait, 1114 October 2015.
This paper was selected for presentation by an SPE program committee following review of information contained in an abstract submitted by the author(s). Contents
of the paper have not been reviewed by the Society of Petroleum Engineers and are subject to correction by the author(s). The material does not necessarily reflect
any position of the Society of Petroleum Engineers, its officers, or members. Electronic reproduction, distribution, or storage of any part of this paper without the written
consent of the Society of Petroleum Engineers is prohibited. Permission to reproduce in print is restricted to an abstract of not more than 300 words; illustrations may
not be copied. The abstract must contain conspicuous acknowledgment of SPE copyright.

Abstract
This study systematically investigates the mechanical instability of a wellbore as a function of rock
mechanical properties and in-situ earth stresses, and validates the predicted failures with post-drilling
problematic wells. Results from this study will help to avoid wellbore instability related issues.
Extended Reach Wells (ERD) were drilled in field X through shale formations inter-bedded by
unconsolidated sandstone formations and carbonate formations which experienced severe wellbore
instabilities.
Rock mechanics lab tests were performed on preserved core samples and all the available well logs
were obtained and subsequently analyzed thoroughly to develop the rock mechanics properties profile.
Daily drilling reports were analyzed and this drilling history was used to evaluate the best failure
criterion for modeling purposes. Accordingly, Mogi-coulomb was found to be the most appropriate failure
criterion that matches the drilling history failures with predicted failures and can be used for more accurate
future wellbore problem predictions.
An industrial geomechanical software was used to build a Mechanical Earth Model (MEM). The
software output was calibrated and validated with the observed failed wellbore condition in selected
drilled wells.
Subsequently, the root causes of different wellbore instability problems were identified. After preliminary validation of the MEM, the safe mud weight window for drilling future wells was predicted with
respect to each formation by providing plots for different inclinations and azimuths.
The safe mud weight windows for drilling in any inclination, azimuth and measured depth for Aruma,
Rumaila, Ahmadi, Safaniya, Biyadh and Buwaib were found and reported. Furthermore, the safest mud
weight window that works for the entire section at any inclination, azimuth and measured depth was
discovered and reported for the fisrt time.
Results from this study indicate that the predicted mud weight window profile could be adopted in
future drilling operations for safe problem-free drilling in problematic formations.

Introduction
Wellbore stability analysis is essential for the success of any drilling operation. This is due to the
frequently encountered problems of wellbore instability and rock failures associated with drilling a hole

SPE-175170-MS

in a field. Examples of problems include shale/clay swelling, collapse, loss circulation and unintentional
formation fracturing.
Wellbore instability manifests itself in different ways like hole pack off, reaming, overpull, torque and
drag, which sometimes leads to stuck pipe that may require plugging and side tracking. This requires
additional time to drill a hole, driving up the cost of development. In offshore fields, loss of hole is more
critical due to the limited number of holes that can be drilled from a platform.
It has been reported in the literature [1] that the wellbore instability is caused mainly by mechanical
factors. Conventional approaches for the wellbore instability events, such as switching to oil-based mud
(OBM) or high-salinity water-based mud (WBM) to combat suspected reactive, water-sensitive shales has
not worked; neither has the increase of mud weight in some tight hole, or shale spalling situations.
Design of wells using rock mechanical principle is well reported [3, 4]. A case study of designing a
horizontal well in Vlieland sand in the Dutch sector of North Sea is reported by Fuh et al.[5]. Rock
mechanical parameters of the Vlieland sand and shale which overlies it were computed from the back
analysis of drilling problems from previous eight vertical wells drilled in the area. A detailed rock
mechanics study of Nigerian shales was done to characterize the state of in-situ stress, rock strength, and
formation pore pressure[6]. These parameters were used to perform a geomechanical simulation and
calibrate it to field evidence. The usage of predicted mud weights resulted in a marked improvement in
wellbore stability.
Mohiuddin et al. explained in his paper that the experience of the vertical drilling is not effective
enough to be redeveloped and used in drilling horizontal wells [10]. Consequently the rate of failure was
very high; 3 out 4 wells drilled failed. A new method of analysis was introduced using field drilling
parameters like initial mud weight, mud weight increment, and problems per well. Geomechanical
software was used with inputs of rock mechanical properties and in-situ stress information and the mud
weight window was estimated. When wells were drilled outside the safe mud weight window and had
sidetrack operation, 40% wells had cases of stuck pipe. On the other hand, only 10% of the wells drilled
within the safe mud weight window faced problems. This shows the importance of operation within the
safe mud weight window. A list of many other studies have reported the importance of calculating the safe
mud weight window for problem free drilling for wells in Abu Dhabi, Gulf of Suez, Mediterranean Sea,
offshore Egypt, Western Canada, Malaysia and Saudi Arabia [1, 7-15]
Consequently, researchers from the academia as well as industry R&D institutions were called in, and
it was soon found that rock mechanics had a much bigger role than anticipated in predicting mud weight
windows. In particular, the various types of in-situ stress regimes and rock failure mechanisms warranted
a much deeper understanding and coupling with rock strength in order to assess the various manifestations
of borehole instabilities. As research continued, a few commercial geomechanical simulation packages
emerged which have helped in mitigating borehole instabilities. One such simulator has been extensively
used in this study.
Geomechanical simulation of instability problems requires that the input data, especially the mechanical properties are close to those of in-situ values. These properties are obtained from sonic and density
logs. The properties thus obtained, called dynamic properties, have to be calibrated using those obtained
in the laboratory, i.e., static values. Through several studies [16] it is now known that the static properties
are more representative for modeling the rocks in-situ stresses. Therefore the industry standard method
is to calibrate the dynamic properties calculated using sonic logs with the static measurements on cores.
The industry has relied on the simple regression technique to calibrate sonic log derived mechanical
properties with static measurements [17, 18].
For many of the wells drilled in one of the fields in Saudi Arabia, significant rig downtime was
experienced due to wellbore instabilities. Most of the instability problems originated from the encountered
shale formations. While drilling, many problems were encountered such as stuck pipe tight hole, over pull

SPE-175170-MS

and lost circulation which were caused due to lack of understanding of the mud weight window to be used
while drilling.

Objective
The objective of this work is to build a 1-D geomechanical model for five selected vertical wells and
characterize and measure the rock mechanical properties, formation pressure, full in situ stress magnitudes
and stress orientations. The 1-D geomechanical model is then validated with the failure history. This will
help in understanding the wellbore stability related issues for a particular field in Saudi Arabia.
The aim of this study was to first measure the rock mechanical properties governing shale strength
using preserved core samples. Thereafter, the profile of rock properties was calibrated and constructed.
Next, the in-situ stresses, stresses around the wellbore and the pore pressure were estimated. Subsequently, the rock properties profile was integrated into a wellbore instability model along with the in-situ
stress profile. The critical mud weight window with respect to depth, azimuth and inclination were
predicted and different options in the modeling were explored to get best match between predictions and
observations. Next the model was validated using field failure cases and field data was analyzed to address
any discrepencies between the model predictions and field observations. Lastly, the safe mud weight
windows were found for all the formations drilled in the 12 hole section.

Results and Discussion

Rock Mechanical Properties
Rock mechanical properties are therefore basic and essential parameters for any geomechanical analysis
and prediction. The rock mechanical properties include elastic constants (Youngs modulus and Poissons
ratio), and rock strength data such as uniaxial compressive strength (UCS) and friction angle (FANG).
Log data, namely compressional and shear slowness and bulk density, are utilized to compute Youngs
Modulus, Poissons ratio and UCS along the depth in a given formation. The continuous profiles of these
data along the depth of a wellbore give important indication of natural variations of the formations
stiffness and strength in different layers.
Lab Data
Triaxial tests of core samples from different formations in WELL L were conducted and another rock
mechanical lab test was performed on samples obtained from an exploration well in 2004 of WELL AA.
The data from WELL AA was used to calibrate the estimated rock strength and elastic properties in
Safaniya, Khafji and Shuaiba formations. The lab test results from WELL AA is summarized and
presented in Table 4. The WELL L core samples were mainly shale and sand plugs from Wara, Rumaila,
Biyadh and Buwaib. The majority of the tests were conducted on the sample of WELL L are shale plugs
while few tests were performed on sand and carbonates samples as seen in Table 5.
Rock mechanical properties
Different methods were used to find a correlation between static elastic properties and log data and rock
strength and log data. Several correlations were developed to estimate the static elastic and rock strength
properties as seen in Figure 18 in the appendix and Table 1. Static elastic properties (Youngs Modulus
(E)) & Poissons Ratio (PR)) can be determined from the slope of stress-strain curves of laboratory
compressive tests. On the other hand, dynamic elastic properties such as Dynamic Shear Modulus,
Dynamic Bulk Modulus, Dynamic Youngs Modulus and the Dynamic Poissons Ratio can be calculated
from compressional and shear slowness as shown in equations (1)-(4)[21]. These dynamic properties will
be used to find a correlation with lab tested static elastic properties and develop a complete profile for
static eleastic properties. Rock strength properties are the Unconfined Compressive Strength (UCS) and
friction angle FANG. The UCS was messured in the lab with zero confining pressure and the sample was
allowed to fail. The (FANG) depends on several factors including mineralogy, lithological variations,

SPE-175170-MS

porosity and shaliness. In the present study constant values of FANG were used for different facies. For
shale, sand and carbonate samples from WELL L, FANG values are approximately 25, 35 and 42
respectively. These values are in agreement with the lab data for cores obtained from WELL L.
Table 1Correlations used for calibration and construction of rock mechanical properties.

An estimated average tensile strength of 1/10 of UCS was used for this study. The core plugs from
WELL L were characterized using Gamma Ray (GR) and were compared with the measured GR from logs
to determine the correction due to depth shift that normally arises during well coring operations. The
measured depth shift between core and log for Rumaila, Wara, Biyadh, top Buwaib and bottom Buwaib
were approximately 1.42, 1.28, 1.91, 1.0, 0.66 feet respectively. The rock mechanical properties profile
was calibrated and constructed for the studied wells.
(1)
(2)

(3)
(4)

In-Situ Stress
In situ-stresses are experienced by rocks at different depths in multiple formations. These in-situ stresses
will be found and plugged into the mechanical earth model at a later stage to determine the stress
distribution around the wellbore.

SPE-175170-MS

Vertical (Overburden) Stress

The overburden stress at a given depth below ground surface is the weight of overlying earth material.
This stress can be calculated by integrating the density log profile of the formations is expressed by the
following equation:
(5)
Where z is overburden stress, is formation density and g is gravitational constant. Unavailability of
data for certain formations and quality of the density log can affect the calculated overburden stress. In
the interval where the density was missing, the density profile was extrapolated using a power law curve
shown below:
(6)
Where sur, A0 and are three fitting parameters, TVD is true vertical depth, WD (0 in this case) is
water depth and AG is air gap. Three reference points (b) from known density profile were plugged into
the power law equation to determine the fitting parameters (sur, A0 and ) for the wells with missing
density log profiles. This power law curve was extrapolated to determine the density profile for certain
formations in these wells.
Pore Pressure
Pore pressure is an important component in a Mechanical Earth Model (MEM), and critical to the
calculation of horizontal stresses, wellbore stability analysis and other geomechanics applications. Sonic
or resistivity logs can be used to identify pore pressure trends in shale which can then be used to estimate
the pore pressures. However, the estimated pore pressure needs to be calibrated by pore pressure
measurements.
Eatons method [23] was used in the current study to estimate the pore pressure for WELL L. The
Eatons method has been originally developed for the Gulf of Mexico and needs to be adjusted to match
the overpressure profile in the Middle East. The adjusted power used in the Eatons equation was n2.0
to fit our pore pressure profile. In non-shale formations, the interpolation method was implemented to
develop a profile that was subsequently calibrated against the formation pressure measurements (refer to
Table 2).
Table 2Formation pressure for WELL L from RDT test
Formation Pressure Data
Well

Formation

Depth (MD-ft)

(psi)

(pcf)

WELL L

Wara
Safaniya
Khafji

5372
5700
5861
6237
6832
7234

2547
2684
2759
2937
3265
3535

68.3
67.8
67.8
67.8
68.8
70.4

Biyadh
Buwaib

Horizontal Stress Magnitudes

The poro-elastic horizontal strain model [1] was used to model the magnitudes of the minimum (Equation
7) and maximum (Equation 8) horizontal stresses.
(7)

SPE-175170-MS

(8)
Here, the two horizontal strains x and y are compressional (for tectonic compression) or extensional
(for lateral spreading), and can be treated simply as calibration factors that are adjusted to best-match the
resulting stress estimates to micro-frac test data or/and rock failure seen in image logs.
The most comprehensive process available for estimating the magnitudes of horizontal stresses at a
single well location, and particularly to determine the magnitude of H., involves initial estimation of the
minimum and the maximum horizontal stresses. The single well approach (illustrated in Figure 1) was
used in the current study to quantify the horizontal stress magnitudes using micro-frac data, caliper and
image logs. The horizontal stress magnitudes were estimated using equations (7) and (8). The magnitude
of minimum horizontal stress (Sig-h) was calibrated using closure pressure values from the micro-frac
tests in WELL L, while the magnitude of maximum horizontal stress (Sig-H) was calibrated indirectly
using breakdown pressure from micro-frac data (refer to Table 3). Fine tuning of the models was achieved
by adjusting the strain terms in the equations (7) and (8) for horizontal stress to best match the occurrences
of breakout seen in caliper and image logs in the selected wells.

Figure 1Process for determining in-situ stress magnitudes at a single well location.
Table 3Closure and breakdown pressure data for WELL L from micro-frac (RDT) test
Closure Data
Well

Formation

Depth
(MD-ft)

WELL L

Rumaila
Wara
Safaniya
Biyadh

5097
5390
5595
6886

Breakdown Data

(psi)

(pcf)

(psi)

(pcf)

4200
4480
4115
5725

119
120
106
120

4740
6125
6500
7385

134
164
167
154

Wellbore Stability
Image interpretation analysis was performed on the image log acquired from WELL L. Figure 2 shows
that the manual dip picking of borehole breakouts is toward NE-110. The breakout has occurred in the
direction of the minimum horizontal stress. Therefore, it can be concluded that, the orientation of the
minimum horizontal stress is NE-110. Consequently, the orientation of the maximum horizontal stress

SPE-175170-MS

was calculated to be NE-200 as the maximum horizontal stress is always perpendicular to the minimum
horizontal stress.

Figure 2OMRI image with manual dip picking of the borehole breakouts, showing strike orientation towards NE-110, which represent
the direction of minimum horizontal stress (Sig-h).

Analysis of Daily Drilling Reports

The occurrence of failures in the drilling history in the existing wells provides good reference points to
calibrate results from the wellbore stability analysis model. This analysis can lead to a better understanding of the nature of failures.
The assessment of drilling history of the selected 27 wells from Field X (offshore and onshore) lead
to identifying incidents of wellbore instability. 40 wellbore instability events (mud losses, stuck pipe, tight
holes, pack-off and over-pull) that were experienced while drilling 27 wells have been extracted and used
for validation of the MEM.
The wellbore instability events such as the mud losses for selected lateral wells were either partial or
complete while drilling Aruma, Maudud, Wara, Safaniya, Shuaiba and Biyadh formations. Partial losses
were mostly cured by pumping lost circulation materials (LCM) (e.g. Marble, Barofiber, Mica etc.), while
total/complete losses were cured through a combination of LCM and cements plugs. Drilling mud weights
ranging from 70-76 pcf (Oil Base Mud) were used to drill 12- hole sections of all the selected wells.
In the selected lateral wells, no real impact of wellbore azimuth and deviation on mud losses and stuck
pipe/tight hole incidents was observed as shown by the rose plot in Figure 3.

SPE-175170-MS

Figure 3Rose plots showing wells that experienced mud loss/no mud loss and wellbore stability related events (stuck pipe, tight
holes etc.) for selected wells

No severe mud losses were experienced in selected vertical wells except in WELL L where total losses
occurred while drilling through Maudud formation. Several cement plugs were pumped to cure the losses
in WELL L. However, the image log showed natural fractures in the formation and the losses were not
caused by high mud weight. Therefore, presence of fracture in the Maudud formation was the main reason
for the losses.
Data assessment shows that most of stuck pipe, tight holes, pack-off and over-pull incidents were
observed during pull out of the hole (POOH) operation. These incidents could possibly be due to low mud
weights used to drill relatively weak shaly layers across Aruma, Ahmadi, Safaniya, Khafji, Biyadh and
Buwaib formations.
The problems encountered during drilling are later plotted on the final mud weight window profile
(contour plot) to validate the MEM.
Rock Failure Criteria
The two types of stress induced failure of rock around the wellbore are encountered during drilling are
shear failure or tensile failure. Shear failure is usually caused by low mud weight while tensile failure is
caused by high mud weight. Several methods exist for predicting rock failure and wellbore instability. The
most commonly used failure criteria are Mohr Coulomb or Mogi Coulomb to determine shear failure, and
Maximum Tensile Stress criteria to determine tensile failure.
To determine the most accurate failure criterion for drilling events, historical data was plugged into,
both, Mogi Coulomb and Mohr Coulomb criteria. Extensive analysis was done to determine the best
match of historical drilling data with the model prediction. It was discovered that Mogi Coulomb failure
criterion provided the best match and hence it has been used throughout this thesis to determine the shear
failure. The simulated failure at different depths in a wellbore using Mogi coulomb criterion and the actual
caliper based failure readings show very good match which could not be obtained by modeling using Mohr
Coulomb approach. This can clearly be seen in Figure 4.

SPE-175170-MS

Figure 4 Predicted failures vs. actual failures for the Mohr Coulomb and Mogi Coulomb criteria.

Failure Analysis and History Matching

The approach for wellbore stability analysis and mud weight predictions is based on comprehensive
understanding of current drilling history. Drilling and completion reports are used to extract wellbore
instability-related events while drilling which are then used to validate predicted failures in the MEM.
A mud weight window found using the MEM is calibrated using the extracted event data. The model
predicts presence or absence of breakouts and/or drilling induced tensile fractures along the wellbore. The
prediction is compared to actual rock failure shown on image and/or caliper logs and the drilling events
in order to ensure that all model parameters are well-constrained and are of reasonable accuracy. The
calibrated model can be used to predict safe mud weight window to drill future vertical, horizontal and
deviated wells.
The mud weight windows developed for the five vertical wells that are analyzed were based on MEMs
and were further calibrated using data from caliper, image and drilling experience.
Mud Weight Window Analysis
Analysis was conducted to evaluate the impact of inclination and azimuth on the safe mud weight window,
determine breakout and breakdown mud weights, and develop recommendations for mud weights to safely
drill wells of different azimuth and inclination.
A change in breakout (shear failure) and breakdown (tensile failure) mud weights with azimuth and
inclination is shown in plots generated for several depths in the problematic formations in Figures 5 to 18.

10

SPE-175170-MS

Figure 5(a) Breakout and breakdown mud weight vs orientation and deviation diagrams in Lower Aruma formation for WELL A (@
4601 MD) and (b) Problems occurrence plot for Aruma from drilling history.

SPE-175170-MS

11

Figure 6 Breakout and breakdown (a&b) mud weight vs. the inclination for minimum and maximum horizontal stress and (c&d) mud
weight vs. azimuth for fixed inclinations of 45 and 90 for Aruma formation.

12

SPE-175170-MS

Figure 7(a) Breakout and breakdown mud weight vs orientation and deviation diagrams in Rumaila formation for WELL L (@ 4982
MD) and (b) Problems occurrence plot for Rumaila from drilling history.

SPE-175170-MS

13

Figure 8 Breakout and breakdown (a&b) mud weight vs. the inclination for minimum and maximum horizontal stress and (c&d) mud
weight vs. azimuth for fixed inclinations of 45 and 90 for Rumaila formation.

14

SPE-175170-MS

Figure 9 (a) Breakout and breakdown mud weight vs orientation and deviation diagrams in Ahmadi formation for WELL L (@ 5179
MD) and (b) Problems occurrence plot for Ahmadi from drilling history

SPE-175170-MS

15

Figure 10 Breakout and breakdown (a&b) mud weight vs. the inclination for minimum and maximum horizontal stress and (c&d) mud
weight vs. azimuth for fixed inclinations of 45 and 90 for Ahmadi formation.

16

SPE-175170-MS

Figure 11(a) Breakout and breakdown mud weight vs orientation and deviation diagrams in Safaniya formation for WELL L (@ 5513
MD) and (b) Problems occurrence plot for Wara, Safaniya and Khafji formations from drilling history.

SPE-175170-MS

17

Figure 12Breakout and breakdown (a&b) mud weight vs. the inclination for minimum and maximum horizontal stress and (c&d) mud
weight vs. azimuth for fixed inclinations of 45 and 90 for Safaniya (Sand) formation.

18

SPE-175170-MS

Figure 13Breakout and breakdown (a&b) mud weight vs. the inclination for minimum and maximum horizontal stress and (c&d) mud
weight vs. azimuth for fixed inclinations of 45 and 90 for Safaniya (Shale) formation.

SPE-175170-MS

19

Figure 14 Breakout and breakdown mud weight vs. orientation and deviation diagrams in Biyadh formation for WELL B (@ 7605 MD)
and (b) Problems occurrence plot for Biyadh from drilling history.

20

SPE-175170-MS

Figure 15Breakout and breakdown (a&b) mud weight vs. the inclination for minimum and maximum horizontal stress and (c&d) mud
weight vs. azimuth for fixed inclinations of 45 and 90 for Biyadh formation.

SPE-175170-MS

21

Figure 16 Breakout and breakdown mud weight vs. orientation and deviation diagrams in Buwaib formation for WELL A (@ 7577 MD)
and (b) Problems occurrence plot for Buwaib from drilling history.

22

SPE-175170-MS

Figure 17Breakout and breakdown (a&b) mud weight vs. the inclination for minimum and maximum horizontal stress and (c&d) mud
weight vs. azimuth for fixed inclinations of 45 and 90 for Buwaib formation.

SPE-175170-MS

23

Figure 18 Breakout and breakdown safe mud weights for all inclinations and azimuths for different formations.

Part a of each of Figures 5, 7, 9, 11, 14 and 16 show the contour graphs of five different problematic
shale formations namely Aruma, Rumaila, Ahmadi, Wara, Khafji, Safaniya Biyadh and Buwaib. These
contour plots show the breakout and breakdown mud weight windows for different azimuth and
inclination of drilling in these formations.
For all formations except Safaniya, Wara and Khafji, the mud weight windows for different wells and
depths were developed. It was found that these mud weight windows were similar for different wells and
depth intervals of the same formation. The result indicates that this mud weight profile could be used in
the future for safe drilling in these formations. Safaniya, Wara and Khafji have two different types of
geological layers which are shale and sand. Therefore, different mud weight windows were developed for
the shale and sand layers.
Part b of each of Figures 5, 7, 9, 11, 14 and 16 show the problem occurrence location of different wells
from actual drilling data. It was found from the contour graphs that the failed wells had been drilled with
a mud weight lower than the minimum recommended breakout mud weight for the respective azimuth and
inclination in these formations. The accuracy of the graphs is thus corroborated with actual wells that had
failed due to low mud weight being applied.
Part a and b of each of Figures 6, 8, 10, 12, 13, 15 and 17 show the breakout and breakdown mud
weight window for different inclination angles. Part (a) shows the mud weight window for the direction
of the minimum horizontal stress (Sig-h) which corresponds to NE 110. Part (b) shows the mud weight
window for the direction of the maximum horizontal stress (Sig-H) which corresponds to NE 200.
Part c and d of each of Figures 6, 8, 10, 12, 13, 15 and 17 show the breakout and breakdown mud
weight window for different azimuth for drilling inclinations of 90 (horizontal) and 45 (deviated) wells.
Figures 6a and 6b show that, for formation Aruma, a wider mud weight window needs to be employed
when drilling in the direction of minimum horizontal stress than the maximum horizontal stress. This
implies that drilling towards the direction of minimum horizontal stress will be relatively easier.
Furthermore, Figures 6c and 6d show that mud weight window for deviated wells is much wider than

24

SPE-175170-MS

horizontal wells, making deviated wells much easier to drill. Furthermore, it was found that the breakout
mud weight is not a function of the azimuth for drilling and is affected only by inclination for Aruma.
Figure 8 shows a similar mud weight profile of Rumaila to that of Aruma which signifies easier drilling
in the direction of minimum horizontal stress and for deviated wells. However, in general, Aruma gives
a slightly wider mud weight window which indicates that drilling in Rumaila would be slightly more
critical than Aruma.
The mud weight contour plot of Safaniya, Khafji and Wara is shown in Figure 11a. It is seen that many
wells failed in these formations due to low mud weight use. Figures 12 and 13 show the mud weight
window for drilling in the sand and shale layers of Safaniya, Khafji and Wara formations respectively. In
general, these formations should be the easiest to drill since they have the widest mud weight ranges when
compared to all other formations. However, due to lack of knowledge on safe mud weight windows,
drilling in these formations lead to the most failures. Furthermore, the shale layer (Figures 13a and 13b)
gives a wider mud weight window when compared to sand (Figures 12a and 12b) when drilling in the
directions of maximum and minimum horizontal stress.
On the other hand, Figures 12 and 13 show that the mud weight window for drilling show that the shale
layer has a larger mud weight window compared to the sand layer when drilling for drilling in the direction
of the maximum and minimum horizontal stress as well as drilling at an inclination of 45. However, the
mud weight window for shale and sand when drilling in the horizontal direction (90) is relatively the
same.
Figures 10 and 15 show a similar mud weight window profile for Ahmadi and Biyadh are similar to
those of Aruma and Rumaila. However, it is worth noting that the mud weight window for horizontal
drilling (Figure 10d and Figure 15d) is considerably narrow which tells us that drilling horizontal will be
very challenging in Ahmadi and Biyadh.
Figure 17 shows the mud weight window for Buwaib. One major point worth noting is that Buwaib
shows the largest mud weight window for drilling in the direction of the minimum horizontal stress.
Finally, the analysis done for all the formations was combined to find the safe mud weight windows
for different formations, regardless of which inclination or azimuth the well is drilled in. To find the lower
end of the safe mud weight windows, the maximum breakout mud weight for wells drilled in the direction
of the minimum horizontal stress was found for each formation. Similarly, to find the upper end of the safe
mud weight windows, the minimum breakdown mud weight for wells drilled in the direction of the
maximum horizontal stress was found. Figure 18 shows the safe mud weight window that can be used to
drill in any inclination, azimuth and measured depth for each of the formations studied. The breakout and
breakdown mud weight range for different formations is shown in Table 3. As a result, a mud weight
window of 99 - 103 pcf was found to be a safe window for drilling through the entire 12 hole section
at any inclination, azimuth and measured depth.

Table 3Mud weight range for breakout and breakdown in each formation.

Formation

Minimum
breakout mud
weight

Maximum
breakout mud
weight

Minimum
breakdown mud
weight

Maximum
breakdown mud
weight

Aruma
Rumaila
Ahmadi
Safaniya (Shale)
Safaniya (Sand)
Biyadh
Buwaib

67.31
70.28
71.93
71.5
72.5
73.63
73.4

91.37
93.92
97
90
90
99
95

106.23
107.66
103.81
120
120
108.3
108.6

135
136.5
136
140
145
140
141.3

SPE-175170-MS

25

Table 4 Results of Lab Test Conducted on Core Samples from Safaniya, Khafji and Shuaiba formation for WELL AA.

Formation
Safaniya
Khafji

Shuaiba

Sample
#

Core Depth (MD-ft)

Core
RHOB
(g/cc)

RM-3
RM-12
RM-16
RM-17
RM-19
RM-20
RM-25
RM-27
RM-29
RM-30
RM-38
RM-39

6,580.20
6,600.60
7,059.30
7,060.60
7,065.40
7,066.80
7,074.70
7,080.80
7,619.70
7,619.90
7,638.50
7,640.00

1.83
1.77
1.92
2.35
2.12
2.15
1.99
1.94
2.47
2.47
2.37
2.28

Average
Static
Youngs
Modulus
(E_STA)
Mpsi

Average
Static
Poissons
Ratio
(PR_STA)

Unconfined
Compressive
Strength
(UCS) psi

Friciton
Angle
(FANG)
degs

1.30
0.71
1.41
3.33
2.81
3.70
1.93
1.90
3.77
4.63
3.51
3.08

0.25
0.72
0.28
0.38
0.32
0.33
0.33
0.28
0.34
0.34
0.37
0.34

2068
2108
4617
8597
4804
6316
7138
5352
11184
10716
13411
8089

31
30
30
46
40
43
34
38
42
34
37
35

Table 5Results of Lab Test Conducted on Core Samples from Rumaila, Wara, Biyadh and Buwaib formation for WELL L.

Conclusion
There is good agreement between predicted and actual borehole failure in all the offset wells which is
further corroborated with matching drilling history in the problematic wells. An additional proof to this
aspect is shown by drilling history in the problematic wells. It was observed in the drilling history that the
increase in the mud weight during drilling lead to problem-free operation because the new mud weight
used was in the safe mud weight window of the model. The predicted model is further verified by the fact
that the drilled wells that had failed had used a mud weight lower than the minimum recommended
breakout mud weight for the respective azimuth and inclination in that formation. This is clearly seen in

26

SPE-175170-MS

the contour plots. Using the Mogi-coulomb failure criterion as a basis of the model has enhanced the
accuracy of the future prediction.
The in-situ stress profiles were constructed and the orientation of minimum horizontal stress and
maximum horizontal stress are at NE-110 and NE-200 respectively. A wider mud weight window in all
the formations has been noticed in the direction of the minimum horizontal stress.
The mud weight analysis for all the formation showed that Safaniya, Khafji and Wara formations
should be the easiest to drill since they have the widest safe mud weight ranges when compared to all other
formations. However, due to lack of knowledge on safe mud weight windows, drilling in these formations
have experienced the most failures. The large difference between the minimum breakout and breakdown
mud weights as well as the maximum breakout and breakdown mud weights observed in Ahmadi and
Biyadh formations shows a high level of anisotropy in these formations. When drilling through all the
studied formations, the minimum safe breakout mud weight was found in Aruma formation (67.31 pcf)
and the maximum safe breakdown mud weight was found in Safaniya formation (145 pcf). Ahmadi and
Biyadhn foramtions are predicted to have the most challenging horizontal (90) drilling due to very
narrow mud weight windows.
More importantly, the safe mud weight windows for all the studied formations were found, which could
be used to drill at any inclination, azimuth and measured depths in those respective formations only. The
mud weight windows for Aruma, Rumaila, Ahmadi, Safaniya, Biyadh and Buwaib were 91.37-106.23 pcf,
93.92 - 107.66 pcf, 97 - 103.81 pcf, 90 - 120 pcf, 99 - 108.3 pcf and 95 - 108.6 pcf respectively.
Furthermore, a mud weight window of 99 - 103 pcf was found to be a safe window for drilling in the entire
12 hole section at any inclination, azimuth and measured depth.
The root cause analysis of different wellbore instability problems revealed that the mud losses are due
to presence of open fractures. Furthremore, no real impact of wellbore azimuth and deviation on mud
losses was observed. The mud weights used to drill these carbonate formations were not expected to either
create drilling-induced fractures or open closed natural fractures. Also, stuck pipe, tight hole, and
over-pull incidents were observed during POOH across Aruma, Ahmadi, Safaniya, Khafji, Biyadh and
Buwaib. The occurrence of these problems could be explained by the use of low mud weightd. Moreover,
it is found that the breakout mud weight is not a function of the azimuth of drilling and is affected only
by the inclination in all formations.
Results from this study indicate that the predicted mud weight window profile could be adopted in
future drilling operations for safe problem-free drilling in problematic formations.

SPE-175170-MS

27

Figure 1Various correlations generated based on core and log data (shale and sand samples) were used to estimate static elastic and
rock strength properties.

References
1. Hassan, S., et al. Optimizing drilling performance by wellbore stability evaluation and directional
drilling practices. in SPE/IADC Middle East drilling technology. Symposium. 1999.
2. Plumb, R., C. Ramshorn, and B. Zehner, Method and apparatus for generation of 3D graphical
borehole analysis. 2000, Google Patents.
3. Wong, S. and W. Heidug, Borehole stability in shales: A constitutive model for the mechanical
and chemical effects of drilling fluid invasion. Rock Mechanics in Petroleum Engineering, 1994.
4. Morita, N. and L. Whitebay. Rock mechanics aspects of drilling and completing highly inclined
wells in weak formations. in University of Tulsa Centennial Petroleum Engineering Symposium.
1994. Society of Petroleum Engineers.
5. Fuh, G., D. Deom, and R. Turner. Wellbore Stability and Drilling Results from the First
Horizontal Well in the Kotter Field Offshore The Netherlands. in SPE Annual Technical
Conference and Exhibition. 1991. Society of Petroleum Engineers.

28

SPE-175170-MS

6. Lowrey, J.P. and S. Ottesen, An assessment of the mechanical stability of wells offshore Nigeria.
SPE Drilling and Completion, 1995. 10(1).
7. Saidin, S. and S. Smith, Wellbore stability and formation damage considerations for Bekok field
K formation. IADC/SPE Asia Pacific Drilling Technology, 2000.
8. Cheatham, Jr, J., Wellbore stability. Journal of petroleum technology, 1984. 36(06): p. 889 896.
9. Abbas, H., et al. Wellbore instability of shale formation; Zuluf Field, Saudi Arabia. in Tech. l
Symp. of Saudi Arabia. 2006.
10. Mohiuddin, M., et al., Analysis of wellbore instability in vertical, directional, and horizontal wells
using field data. Journal of petroleum Science and Engineering, 2007. 55(1): p. 8392.
11. Santarelli, F., et al. Wellbore stability analysis made easy and practical. in IADC/SPE drilling
conference. 1996.
12. Santarelli, F., et al. Mechanisms of borehole instability in heavily fractured rock media. in
International journal of rock mechanics and mining sciences & geomechanics abstracts. 1992.
Elsevier.
13. Onaisi, A., J. Locane, and A. Razimbaud. Stress related wellbore instability problems in deep
wells in ABK field. in Abu Dhabi International Petroleum Exhibition and Conference. 2000.
Society of Petroleum Engineers.
14. Woodland, D.C., Borehole instability in the western Canadian overthrust belt. SPE drilling
engineering, 1990. 5(1): p. 2733.
15. Al-Buraik, K. and J. Pasnak, Horizontal drilling in Saudi Arabian oil fields: case histories. Middle
East Oil Show, 1993.
16. Warpinski, N., et al. In situ stress and moduli: Comparison of values derived from multiple
techniques. in SPE annual technical conference. 1998.
17. Ahmed, U., M. Markley, and S. Crary, Enhanced In-Situ Stress Profiling With Microfracture Core
and Sonic-Logging Data. SPE Formation Evaluation, 1991. 6(02): p. 243251.
18. Gatens, III, J., et al., In-Situ Stress Tests and Acoustic Logs Determine Mechanical Propertries
and Stress Profiles in the Devonian Shales. SPE Formation Evaluation, 1990. 5(03): p. 248 254.
19. Larsen, I., E. Fjr, and L. Renlie. Static and Dynamic Poissons Ratio of Weak Sandstones. in 4th
North American Rock Mechanics Symposium. 2000. American Rock Mechanics Association.
20. Rahim, Z. and M. Al-Qahtani, Sensitivity Study on Geomechanical Properties to Determine Their
Impact on Fracture Dimensions and Gas Production in the Khuff and Pre-Khuff Formations Using
a Layered Reservoir System Approach, Ghawar Reservoir, Saudi Arabia. paper SPE, 2001.
72142.
21. Fjar, E., et al., Petroleum related rock mechanics. Vol. 53. 1992: Elsevier.
22. Chang, C., M.D. Zoback, and A. Khaksar, Empirical relations between rock strength and physical
properties in sedimentary rocks. Journal of Petroleum Science and Engineering, 2006. 51(3): p.
223237.
23. Mouchet, J.P. and Mitchell, A. 1989. Abnormal Pressures While Drilling: Origins, Prediction,
Detection, Evaluation. Paris: Elf EP-Editions, Editions Technip, 255.
24. K. Khan. Understanding Overpressure Trends Helps Optimize Well Planning and Field Development In Tectonically Active Area In Kuwait, Paper SPE 122631 prepared to be presented at
2009 Asia Oil and Gas Conference and Exhibition held in Jakarta, Indonesia, August 4-6.
25. Bol, G., et al., Borehole Stability in Shales. SPE Drilling & Completion, 1994. 9(02): p. 8794
26. Hussain Rabia., Well Engineering and Construction
27. Khaksar, A., et al. Rock Strength from Core and Logs Where We Stand and Ways to Go. in
EUROPEC/EAGE Conference and Exhibition. 2009. Society of Petroleum Engineers.
28. Russell, K.A., et al., Predicting and Preventing Wellbore Instability: Tullich Field Development
North Sea. SPE Drilling & Completion, 2006.