URTeC: 15

Anisotropic Borehole Stability Analysis for the UK’s First Horizontal

Shale Gas Well in the Bowland Basin
Huw Clarke1, Hamed Soroush*2; 1. Cuadrilla Resources Ltd, 2. Petrolern LLC.
Copyright 2019, Unconventional Resources Technology Conference (URTeC) DOI 10.15530/urtec-2019-15

This paper was prepared for presentation at the Unconventional Resources Technology Conference held in Denver, Colorado, USA,
22-24 July 2019.

The URTeC Technical Program Committee accepted this presentation on the basis of information contained in an abstract
submitted by the author(s). The contents of this paper have not been reviewed by URTeC and URTeC does not warrant the
accuracy, reliability, or timeliness of any information herein. All information is the responsibility of, and, is subject to corrections by
the author(s). Any person or entity that relies on any information obtained from this paper does so at their own risk. The information
herein does not necessarily reflect any position of URTeC. Any reproduction, distribution, or storage of any part of this paper by
anyone other than the author without the written consent of URTeC is prohibited.

Abstract
The Bowland Shale is a Carboniferous formation of Asbian to Yeadonian age located in Northern
England. It is estimated to have a shale gas section with more than 5,000 ft thickness holding over 1300
TCF of total original gas in place. Drilling in the Bowland Basin started in August 2010 and by the end of
2011, three vertical wells (PH-1, GH-1 and BS-1) were drilled to a depth of 8,860 to 10,500 ft with
objective of logging and coring the potential shale gas formations including Upper Bowland, Lower
Bowland, Hodder Mudstone and Sabden Shale. All the drilled wells encountered several borehole
stability problems, such as tight-hole, pack-off, overpull and excessive cutting, causing significant non-
productive time (NPT) during drilling. Specifically, in GH-1 and BS-1, side-tracking was required to
reach the target depth which imposed significant cost to the project. Careful investigation of the recorded
drilling problems showed that majority of them were associated with formation collapse due to
insufficient drilling fluid pressure. Fluid losses also occurred in some of the formations due to either too
high of downhole pressure or presence of critically stressed natural fractures. These incidents implied that
the applied casing design and mud weight program were not appropriate for the current-day state of stress
and rock properties.
A comprehensive experimental and analytical geomechanical study was carried out to develop a reliable
borehole stability model that can firstly explain the observed drilling incidents and secondly provide
guidance for design and drilling of the planned wells. The plan was to drill a S-shape appraisal well
(PNR-1) in the Preston New Road exploration site to log and core the Bowland Shale sequence and select
the optimum landing depths for subsequent horizontal sections (PNR-1z and PNR2) to be completed for
multi-stage hydraulic fracturing. The study recognized intrinsic shale anisotropy as a primary causative
factor for borehole stability issues and formation collapses in the offset wells. Other important factors
were identified to be the abnormal pore pressure regime and the presence of a tectonic strike-slip stress
regime with large horizontal stress anisotropy. The anisotropy of the Bowland Shale was characterized in
both laboratory and field scales, and anisotropic wellbore stability models were developed for offset and
planned wells. As a result of this study, the PNR-1, PNR-1z and PNR2 wells were successfully drilled
and completed with no notable borehole stability problems despite the presence of narrow operating mud
weight window in several stratigraphic intervals. This paper presents a summary of the conducted
borehole stability analysis aiming at risk-free and successful drilling of horizontal wells in the Preston
New Road exploration site with emphasis on the effect of shale anisotropy.
URTeC 15 2

Introduction
The Bowland Basin in Northern England (Figure 1, left) is estimated to hold over 1,300 TCF of total
original gas in place of shale gas resource (Andrews, 2013). The gas-bearing shale section is evaluated to
be extremely thick (>5,000 ft) and intensely fractured (Clarke et al., 2018). Analyzing the extensive
amount of core and log data acquired in the PH-1 discovery well and the subsequent GH-1 and BS-1
wells proved approximately 1 tcf per square mile (TOGIP) of methane-rich gas resource, with a gas
density comparable with those produced from North America shale gas. Figure 1 (right) shows the
locations of the three vertical wells drilled during 2010-2011in the Bowland Basin to a depth of 8,860-
10,500 ft. Figure 2 illustrates the lithology correlation among the three wells.

Figure 1. Location of the Bowland basin in the Northern England (Left - Andrews, 2013); and surface location of PH-1, GH-1 and BS-1 appraisal
wells (right).

Figure 2. Lithology correlation in PH-1, GH-1 and BS-1 wells.

URTeC 15 3

Delays were encountered in the drilling of all the three wells due to multiple borehole stability problems.
Specifically, in GH-1 and BS-1, the wells required side-tracking to reach the target depths. The
geomechanical related incidents in the offset wells were extracted and summarized for better
understanding of the formations’ response to the drilling operation. As an example, summary of the
drilling problems in the GH-1 well is presented in Figure 3 showing several tight-hole, pack-off, overpull
and excessive cutting issues which are associated with formation collapse due to insufficient drilling fluid
pressure. In some of the formations, fluid losses have been reported indicating that either fluid pressure is
too high or critically stressed natural fractures exist. These incidents imply that the casing designs and the
applied mud weight programs were not appropriate for the existing state of stress and rock properties.

Figure 3. Summary of drilling incidents in GH-1 vertical well.

The surface location of the planned S-shaped pilot well (PNR-1) is located within the Bowland Basin,
Lancashire, Northern England, approximately 12,139 ft south of PH-1 well (Figure 1, right). It is within
one of the two sites which are planned for drilling, hydraulic fracturing and flow testing the carboniferous
shale gas layers. The trajectory of the well is depicted in Figure 4 (left). The plan for the pilot well was to
facilitate data acquisition including wireline logging and coring throughout the prospective Bowland shale
in addition to performing an extended leak-off test (XLOT) close to the target formation. PNR-1 was then
to be plugged back to a selected optimum depth in the Lower Bowland Shale and side-tracked to the first
of two horizontal lateral boreholes (PNR-1z) of up to 3,280 ft lateral length followed by the second
horizontal well (PNR-2) targeting the Upper Bowland Shale, both wells to be completed for multi-stage
hydraulic fracturing and flow testing. The planned trajectories are shown in Figure 4 (right).
The primary targets were the gas-charged Carboniferous sections in the Upper and Lower Bowland shales
with Hodder Mudstone and Sabden Shale being the secondary and tertiary targets.
The objectives of the drilling program were to:
• obtain sufficient core and log data from the pilot PNR-1 in order to identify the most
prospective zones for the lateral wells,
• drill the horizontal PNR-1z well, stimulate and flow test the most prospective zone identified
in the pilot hole,
URTeC 15 4

• evaluate the commerciality of the well with flow testing,

• tie well data to seismic data,

The main goal of the program was to meet all the legal and HSE requirements and execute the plan with
zero safety, environmental and community related incidents. This goal was recognized and appreciated by
Petrolern while conducting the geomechanical and wellbore stability studies.

Figure 4. Planned trajectories and lithology columns for PNR-1 (left); and the locations of the horizontal laterals in compare with PNR-1 (right).

Developing Geomechanical Model for Preston New Road

Based on the available data from offset wells after an extensive data review and quality control scheme, a
pre-drill geomechanical model for the Preston New Road site was constructed, containing depth-
continuous profiles for pore pressure, principal stresses (i.e., vertical stress, minimum and maximum
horizontal stresses), and formation mechanical properties (i.e., Young’s modulus, Poisson’s ratio, UCS,
cohesion, friction angle and brittleness). Please refer to (Clark et al., 2019) for detailed information about
the methodologies and workflow used to develop the geomechanical model.

Anisotropy Characterization of the Bowland Shale

Shale anisotropy has a significant effect on borehole stability. The severity of this effect depends on the
relative orientation of the well to the anisotropy surfaces (e.g., bedding planes, laminations, foliations,
etc.). Overall, the Bowland Shale is characterized as a heterogeneous interbedded sequence of turbidite
flows, with thin beds of calcareous detrital facies, extremely fine-grained silica classics, and hemipelagic
organic rich siltstones (Figure 5, top). Magraw and Ramsbottom (1957) reported that interbedded thin
beds of limestone with a film of pyrites on the bedding planes are characteristic of the Bowland Shale
sequences. The shales immediately adjacent to the limestone are frequently slicken-sided, brecciated and
veined with calcite (Figure 5, bottom). They interpreted as the genesis of the slicken-siding and
brecciation as folding of the “relatively incompetent shales with thin limestones in close contact with beds
of competent, massive limestone”. The interbedding of calcareous‐rich and hemipelagic sediments, often
at a fine scale, implies strong mechanical anisotropy. The distribution of the damage to the casing in the
URTeC 15 5

PH-1 well demonstrates that bedding slips commonly occur at or very close to interfaces between
limestone‐rich and mudstone‐rich layers.
Study of the image logs from offset wells (Figure 6, left) revealed that the dip and dip direction of the
bedding planes rotated in some intervals. Examples of these rotations within the Bowland Shale in PH-1
and GH-1 wells are shown in Figure 6 (right). This study concluded the average dip and dip direction of
the bedding planes as below:
• Average over the whole imaged interval: Dip = 28.7°; Dip Azi = 265°
• Average over the Upper Bowland Shale interval: Dip = 12.2°; Dip Azi = 180°
• Average Over the Lower Bowland Shale interval: Dip = 40°; Dip Azi = 288°

Figure 5. Outcrop of the Hodder Formation in the Bowland Basin showing anisotropic structure of the shale (top left - Andrews, 2013); thin
section from a typical Bowland Shale lithofacies (top right); and the typical slickenside anisotropy surfaces in the Bowland Shale (bottom).

The cross-dipole sonic logs in the Bowland intervals of PH-1 and GH-1 show varying anisotropy degrees
from nearly isotropic to strongly anisotropic shale (Figure 7, left). Surprisingly, the correlation made
between the fast and slow shear travel times showed a correlation coefficient of close to 1 (Figure 7,
right) meaning that the intervals with very low anisotropy in sonic wave velocities are dominant. This
also implied that there were insignificant differences between shale stiffness (i.e., Young’s Modulus) in
the orientations parallel and perpendicular to the bedding planes. Figure 8 (top) shows cross plots of fast
and slow Young’s Moduli versus fast and slow Poisson’s Ratios calculated from the cross-dipole sonic
logs from PH-1. This conclusion was also supported by the rock mechanics testing results as no
meaningful anisotropy was observed in the Young’s Moduli and Poisson’s Ratios measured parallel and
perpendicular to the bedding planes. Figure 8 (bottom) presents an example of ultra-sonic wave velocities
measured on a core plug from the Lower Bowland Shale showing insignificant differences between fast
and slow shear waves (S1 and S2).
URTeC 15 6

Figure 6. Shale anisotropy observed on image log from PH-1; and rotation of bedding planes in the Bowland Shale in PH-1 and GH-1.

Figure 7. Anisotropy analysis using cross-dipole sonic log in PH-1 (left); and correlation of DTSFast with DTSSlaow showing small difference in the
fast and slow wave velocities (right).

Unlike the rock elastic parameters (stiffness), the strength parameters, i.e., UCS, friction angle and
cohesion, were significantly affected by the anisotropy surfaces. This was concluded from a
comprehensive rock mechanics testing program including Pseudo-UCS tests (PUCS) under 0.5 MPa
confining pressure on plugs taken parallel, perpendicular and at 45° to the bedding planes, single-stage
triaxial tests on plugs taken parallel and perpendicular to the bedding planes, and direct shear tests on the
surface of bedding planes. The PUCS test results showed that core plugs loaded at an angle of 45° to the
anisotropy surfaces were significantly weaker than those loaded parallel or perpendicular to the
URTeC 15 7

anisotropy surfaces. While Figure 9 (top) shows the anisotropy observed in the core plugs photos and CT
scans, Figure 9 (bottom) depicts the results of PUCS tests conducted on the Upper Bowland Shale core
plugs loaded perpendicular and at 45° to the bedding planes. It shows 113 MPa difference between the
two cases.
To indicate the differences between the cohesions and friction angles measured using triaxial tests (on
rock matrix) and direct shear tests (on bedding planes), a summary of the test results in three different
intervals are presented in Table 1. As seen, the strength parameters measured by direct shear test on the
surface of bedding planes are significantly lower than those measured on rock matrix by triaxial test. The
table also provides a comparison between the tensile strengths measured by Brazilian test perpendicular
and parallel to the bedding planes.

Figure 8. Cross plots of fast Young’s Moduli versus fast Poisson’s Ratios (top left) and slow Young’s Moduli versus slow Poisson’s Ratios (top
right) calculated from the cross-dipole sonic logs from PH-1, showing similar trend and ranges for shale stiffness; and an example of ultra-sonic
wave velocities measured on a horizontal core plug taken from the Lower Bowland Shale (bottom).
URTeC 15 8

Figure 9. Photos and CT scans of core plugs taken perpendicular, parallel and at 45° to the bedding planes (top); comparison of the shale pick
strength under PUCS tests on plugs perpendicular to bedding planes (bottom left, UCS = 190 MPa); and at 45° to the bedding planes (bottom
right, UCS = 77 MPa) showing a strong effect of anisotropy on shale strength.

Table 1. Effect of anisotropy (bedding planes) on the strength parameters of the Bowland Shale.
 URTeC 15 9

Borehole Stability Analysis

The cross-dipole sonic logs, density logs, and the rock mechanics testing results enabled developing
anisotropic rock property models for the offset wells. The models include dynamic and static elastic
property profiles, cohesion in both parallel and perpendicular to bedding orientations, and friction angle
for both matrix and anisotropy surfaces. These models were used to generate lithology-dependent stress
profiles which formed the geomechanical model for each well along with the developed pore pressure
profiles. Figure 10 illustrate an example of the rock property (left) and stress models developed for PH-1
(right).

Figure 10. Developed anisotropic rock property model for the Upper and Lower Bowland shale formation in PH-1 well (left); and pore pressure
Developed rock propertyand
model formodel
stress the Upper and Lower
for PH-1 Bowland
in both gradientshale
and formation in PH-1units
absolute pressure well. (right).

Using the developed geomechanical model and based on the results of the anisotropy study on the image
logs, following three scenarios were modelled to show the effect of anisotropy on the borehole stability
condition in the PNR-1 well:
1. Bedding planes orientation follows the average orientation over the whole imaged interval:
Dip: 28.7° and Dip Azi: 265°
2. Bedding planes orientation follows the average orientation over the Upper Bowland Shale:
Dip: 12.2° and Dip Azi: 280°
3. Bedding planes orientation follows the average orientation over the Lower Bowland Shale:
Dip: 40° and Dip Azi: 288°
Figure 11 shows the results of this analyses and makes comparison between the severity of the breakout
zones developed in the three scenarios. It is noted that in all the scenarios, identical mud weight program
and casing design have been used and only the dip and dip direction of bedding planes have been
changed. The right plot for each scenario shows the width of breakout developed along the well (red
curves) due to using a given mud weight. The blue curve is the width of the developed breakouts when
anisotropy is ignored, for comparison. The top right plot for each scenario is the cross section of the well
at 6,880 ft MD showing the wellbore geometry and the developed 4-lobbed breakouts (the typical type of
breakouts in anisotropic formation). The bottom right plot for each scenario is the polar plot colour coded
to breakout width, showing that the high deviated to horizontal wells (60-90° deviation) with azimuths of
60-120° and 210-300° are the least stable well trajectories (breakouts larger than 90° are expected).
According to this comparison, scenario 2 creates the most stable borehole, where all the developed
URTeC 15 10

breakouts are below the manageable range (black curve). In contrary, scenario 3 is the worst case where
several intervals exceed the allowable breakout width and cause borehole stability problems.
Figure 12 shows the existing mud weight window and predicted breakout extension for the best and worst
scenarios (2 and 3, respectively).

Figure 11. Predicted breakout zone extension for scenario 1 (top left), scenario 2 (top right) and scenario 3 (bottom) using identical mud weight
and casing design.

Figure 12. Mud weigh window and the predicted breakouts in the horizontal section (PNR-1z) using a given mud weight for scenario 2 (left) and
scenario 3 (right).
URTeC 15 11

Drilling Experience in PNR-1, PNR-1z and PNR-2

PNR-1 was spudded in September 2017 and successfully reached the target at 8,858 ft MD (8,538.7 ft
TVD) on the 2nd January 2018 with no notable drilling problems except some losses in the Lower
Bowland Shale which were proven to be through some critically stressed natural fractures according to
the image logs (Figure 13). A comprehensive data acquisition program was carried out including 114m of
coring from the mid Upper Bowland Shale through to the mid Lower Bowland Shale and logging the 16”,
12 ¼” and 8 ½” intervals. Also logging while drilling data and gas shows were recorded and a successful
Extended Leak-off Test (XLOT) was conducted at 7,673.5 ft TVD after the 7” casing shoe. These data
were used to update the pre-drill geomechanical model and resulted in higher accuracy borehole stability
models for PNR-1z and PNR-2 horizontal wells. Figure 14 shows the available mud weight window for
PNR-1z. According to the model, a mud weight of 12.1 ppg was recommended considering a manageable
breakout of 30° in the horizontal section, meaning that some formation failure might occur, but will be
manageable with a proper borehole cleaning practice.
The horizontal PNR-1z well was kicked off at 6,749 ft MD on the 11 January 2018 and successfully
reached the target at 11,233 ft MD on 21 March 2019 with no significant drilling problems. It penetrated
2,560 ft in the Lower Bowland Shale. The well was then completed with a sliding sleeve configuration
which included 41 sleeves set at 22m spacing ready for stimulation. Following this, a second horizontal
section (PNR-2) was also drilled successfully for 2,460.6 ft in the Upper Bowland Shale using a
customized borehole stability model for its own trajectory.

Figure 13. PNR-1 drilling summary plot.

URTeC 15 12

Figure 14. PNR-1z wellbore stability model with the available safe operating mud weight window in green.

Conclusions
The severity of drilling problems encountered during drilling of three exploration and appraisal wells
targeting the Bowland Shale confirmed requirement for a comprehensive geomechanical study in order to
develop reliable borehole stability models for the planned drilling program in the Preston New Road
exploration site. The geomechanical study identified three main reasons for the problematic drilling of the
offset wells: (1) abnormal pore pressure, (2) strike-slip stress regime with large horizontal stress
anisotropy, and (3) anisotropic characteristics of the Bowland Shales.
Developing a borehole stability model based on anisotropic characteristics of the Bowland Shale resulted
in successful and problem-free drilling of an appraisal S-shape well (PNR-1) and two horizontal laterals
(PNR-1z and PNR-2) through the Upper and Lower Bowland Shale in Preston New Road, becoming the
first horizontal shale gas wells ever being drilled in the UK.

Acknowledgement
The authors would like to acknowledge the support from Cuadrilla Resources and the permission to
publish the work. We would also like to extend our appreciation to all Petrolern’s and Cuadrilla’s
personnel who helped us to make this project a success.

References
Andrews, I. J. 2013. The Carboniferous Bowland Shale gas study: geology and resource estimation.
British Geological Survey for Department of Energy and Climate Change, London, UK.
Magraw, D. and Ramsbottom, W. H. C. 1957. A deep borehole for oil at Croxteth Park, near Liverpool.
Geological Journal 1(6): 512-535.
Clarke, H., Turner, P. and Bustin, R. M. 2014. Unlocking the Resource Potential of the Bowland Basin,
NW England. SPE/EAGE European Unconventional Conference and Exhibition, Vienna, 25–27 Feb.
Clarke, H., Soroush, H. and Wood, T. 2019. Preston New Road: The Role of Geomechanics in Successful
Drilling of the UK’s First Horizontal Shale Gas Well. The SPE Europec featured at 81st EAGE Annual
Conference & Exhibition. SPE-195563-MS.
Huw Clarke, Peter Turner, Marc Bustin, Nick Riley & Bernard Besly. (2018). Shale gas resources of the
Bowland Basin, NW England: a holistic study. Petroleum Geoscience. 24(3):287
Soroush, H. 2017. Preston New Road: Geomechanical Modelling and Wellbore Stability Study. Petrolern
LLC Report for Cuadrilla Resource. April 2017.