OR/17/006 Bedrock geology: Difference between revisions

From MediaWiki
Jump to navigation Jump to search
m (1 revision imported)
 
Line 56: Line 56:
The deepest mineworkings in the area are shown on the plan of the Kiltongue Coal (Lower Coal Measures) as recorded at depths of -268.5 m. The depths marked on mine plans were surveyed in, and are believed to be accurate to within a metre or less. Accuracy will decrease with distance from a shaft. If underlying seams were mined subsequently to a particular plan, then the depths on the older plan may have been affected by subsidence from younger, underlying workings. This cumulative subsidence may need to be taken into account at a site specific level for borehole planning purposes.
The deepest mineworkings in the area are shown on the plan of the Kiltongue Coal (Lower Coal Measures) as recorded at depths of -268.5 m. The depths marked on mine plans were surveyed in, and are believed to be accurate to within a metre or less. Accuracy will decrease with distance from a shaft. If underlying seams were mined subsequently to a particular plan, then the depths on the older plan may have been affected by subsidence from younger, underlying workings. This cumulative subsidence may need to be taken into account at a site specific level for borehole planning purposes.


The abandoned, flooded mine workings form the resource for ‘shallow’ geothermal heat and heat storage (Gillespie et al., 2013<ref name="Gillespie 2013">GILLESPIE, M R, CRANE, E J, and BARRON, H F. 2013. Study into the Potential for Deep Geothermal Energy in Scotland: Volume&nbsp;2. ''British Geological Survey Commissioned Report'', CR/12/131 Scottish Government Project Number AEC/001/11. ISBN: 9781782569862 (web only publication)</ref>). Roadways emanating from shafts may be of particular importance as they are believed to be open at the present time, greatly increasing  permeability and potential yield (Gillespie et al., 2013<ref name="Gillespie 2013"></ref>).
The abandoned, flooded mine workings form the resource for ‘shallow’ geothermal heat and heat storage (Gillespie et al., 2013<ref name="Gillespie 2013"></ref>). Roadways emanating from shafts may be of particular importance as they are believed to be open at the present time, greatly increasing  permeability and potential yield (Gillespie et al., 2013<ref name="Gillespie 2013"></ref>).


[[Image:OR17006fig7a.jpg|thumb|center|700px|  '''Figure 7'''&nbsp;&nbsp;&nbsp;&nbsp;Summary of known (above) and probable (below) mining extents BGS © NERC 2016 for the Clyde Gateway area.    ]]
[[Image:OR17006fig7a.jpg|thumb|center|700px|  '''Figure 7'''&nbsp;&nbsp;&nbsp;&nbsp;Summary of known (above) and probable (below) mining extents BGS © NERC 2016 for the Clyde Gateway area.    ]]

Latest revision as of 15:23, 12 December 2019

Monaghan, A A, Dochartaigh, B O, Fordyce, F, Loveless, S, Entwisle, D, Quinn, M, Smith, K, Ellen, R, Arkley, S, Kearsey, T, Campbell, S D G, Fellgett, M, and Mosca, I. 2017. UKGEOS - Glasgow geothermal Energy Research Field Site (GGERFS): initial summary of the geological platform. British Geological Survey Open Report, OR/17/006.

Overview

The bedrock strata beneath the Clyde Gateway area comprise proved Carboniferous sedimentary rocks of the Scottish Upper, Middle and Lower Coal Measures, Passage and Upper Limestone formations (Figure 2). Based on the surrounding geology and deeper boreholes within a few kilometres of the study area, older Carboniferous strata of the Limestone Coal and Lower Limestone formations (Clackmannan Group) are expected at depths of around 500–1600 m (Figures 3, 4). These successions are dominantly fluvio-deltaic to shallow marine cyclic mudstones, siltstones, sandstones, coals and limestones.

Figure 2    Clyde Gateway area summary of bedrock geology from BGS © NERC 2016 DigMapGB 1:10 000 scale data.

At deeper levels, the geological sequence is poorly constrained with strata of the West Lothian Oil-Shale Formation or the laterally equivalent Lawmuir Formation predicted to lie above the Kirkwood and Clyde Plateau Volcanic formations. The early Carboniferous Clyde Sandstone, Ballagan and Kinnesswood formations underlie the volcanic rocks to the north and south of Glasgow and would be predicted beneath the Clyde Plateau Volcanic Formation (Figure 3).

Figure 3    Summary of proved and possible lithology and stratigraphy in the Clyde Gateway area. Possible intervals for consideration of geothermal potential indicated.

Potential geothermal resources are situated within the extensive abandoned flooded coal mine system of the Scottish Middle and Lower Coal Measures formations, with tunnels, shafts and workings across the entire eastern Glasgow conurbation as far as Wishaw (Campbell et al. 2010[1]; Gillespie et al., 2013[2], Kearsey et al., in press[3]).

Potential hot sedimentary aquifer geothermal resources have been considered within the sandstone-dominated Passage Formation in the wider Glasgow and central Scotland area (Browne et al., 1985[4]; Hall et al., 1998[5]), though temperatures linked to relatively shallow burial depths were a concern. Multiple channelised sandstones in the Upper Limestone, Limestone Coal and Lower Limestone formations are known in the wider Glasgow area. Deeper, lower Carboniferous-upper Devonian sandstones (potentially at depths of 2 km or more) beneath the Clyde Plateau Volcanic Formation were identified as geothermal targets by Browne et al (1985)[4] but lack of data precludes any knowledge of their presence or character at depth in the vicinity of Clyde Gateway.

Bedrock faulting and structure

Complex faulting, comprising mainly steep faults, pervades the area. 45 faults on a variety of trends have been modelled in the current bedrock model of the 10 km by 10 km area which contains the Clyde Gateway area (Monaghan et al., 2014[6]). The sedimentary rocks are cut by Late Carboniferous igneous intrusive sills and are also folded. The most common larger faults are roughly E-W-trending (Figure 2). The major Dechmont Fault trends north-west and downthrows Coal Measures to the north-east against Clackmannan Group strata. There is a major easterly-plunging, E-W-striking open fold of Coal Measures strata in the hanging wall of the Dechmont and Rutherglen faults; this structure covers the southern part of the Clyde Gateway area. Clackmannan Group strata are folded into c.north-north-east- to north-east-trending synclines and anticlines in areas surrounding the Clyde Gateway area. All areas are cut by north-west- to east-north-east-trending faults.

The Dechmont Fault (Figure 2, 4) has been interpreted as a deep and long-lived north-west-trending lineament (Forsyth et al., 1996[7]; Hall et al., 1998[5]). It divides two Midland Valley Upper Carboniferous structural styles — a north-east-trending half-graben/graben block and basin to the west (Ayrshire) and north-north-east-trending growth folds to the east (Central Coalfield and Fife; Hall et al., 1998[5]). Strike-slip to extensional tectonism was active during the Carboniferous (Forsyth et al., 1996[7]; Hall et al., 1998[5]; Read et al., 2002[8]; Underhill et al., 2008[9]), so stratal thickening and thinning across fault and fold structures is expected.

Figure 4    Visualisation of possible geology and geothermal targets beneath the Clyde Gateway area. Geology within the cube based on 3D bedrock model. ILS = Index Limestone, KDG = Knightswood Gas Coal, GU = Glasgow Upper Coal, GELL = Glasgow Ell Coal, KILC = Kiltongue Coal, PGP = Passage Formation, TOHO = Top Hosie Limestone, HUR = Hurlet Limestone.

Datasets

Boreholes

Figure 5    All boreholes held in the BGS SOBI database (above). Boreholes greater than 100 m drilled depth (below). Clyde Gateway area in blue. Key to geology map on Figure 2.

BGS holds over 3,400 borehole records within the Clyde Gateway area (Figure 5 above). However the majority of these penetrate to depths of less than 30 m beneath the ground surface, with only ten borehole records having a drilled length greater than 100 m (Figure 5 below).

The abundant borehole records adequately constrain the geology of the superficial deposits, and provide important information concerning the position of unmined and mined coal seams close to rockhead in the northern half of the Clyde Gateway area. Mined coal seams are commonly encountered in boreholes as ‘waste’ (collapsed workings) or occasionally as open ‘voids’.

The BGS.SOBI and BGS.Borehole_Geology databases were updated in autumn 2016 with borehole records received since the previous phases of work in 2009 and 2011. Updated borehole interpretations were fed into revised versions of the bedrock and superficial deposits models (see below).

Hydrocarbon wells and geophysical well logs

Limited geophysical well log data is available, all from outside the Clyde Gateway area (Figure 6). Three IGS (BGS) wells drilled at Maryhill (1983, 306 m, Limestone Coal and Lower Limestone formations), Alexandra Parade (1976, 222 m, Scottish Lower Coal Measures, Passage, and Upper Limestone formations) and Hallside (1976; 348 m; Scottish Upper and Middle Coal Measures formations) recorded a mixture of gamma, density, SP, resistivity, temperature, caliper logs. The Bargeddie 1 hydrocarbon exploration well (1989, 1046 m, Scottish Coal Measures to West Lothian Oil-Shale formations) has a large suite of geophysical logs (Teredo, 1989), plus some Rock-Eval source rock analysis (Monaghan, 2014[6] Appendix D). Gas shows within sandstones of the West Lothian Oil-Shale Formation showed initially promising pressures and estimated volumes but further fracturing and pressure testing showed a decline in reservoir pressure, suggesting a ‘small bounded reservoir model’ (1989 well report, Teredo and Oilfeld Production Consultants).

Figure 6    Location of the three IGS (BGS) boreholes with geophysical well logs in the wider Glasgow area, and the Bargeddie 1 conventional hydrocarbon exploration well. Includes mapping data licensed from Ordnance Survey. © Crown Copyright and/or database right 2017. Licence number 100021290 EUL.

Mine abandonment plan data

There are historical mine workings for coal in eight seams beneath a large part of the Clyde Gateway area, and related shafts and interconnecting underground roadways (Figure 7).

No records of mine workings for materials other than coal (ironstone etc) within the Clyde Gateway area have been located.

Workings recorded in coal mine abandonment plans dating from 1810–1934 have been captured by BGS for the following seams:

  • Glasgow Upper Coal (Scottish Middle Coal Measures Formation)
  • Glasgow Ell Coal (Scottish Middle Coal Measures Formation)
  • Glasgow Main Coal (Scottish Middle Coal Measures Formation)
  • Humph Coal (Scottish Middle Coal Measures Formation)
  • Glasgow Splint Coal (Scottish Middle Coal Measures Formation)
  • Glasgow Virgin Coal (Scottish Middle Coal Measures Formation)
  • Airdrie Virtuewell Coal (Scottish Lower Coal Measures Formation)
  • Kiltongue Coal (Scottish Lower Coal Measures Formation)

Extents of all mined areas, including sub-areas of stoop and room workings, have been digitised to GIS from the abandonment plans (Figure 7), as have spot heights, contours, shafts and roadways. Faults have been extracted from plans where marked and proved, and/or interpreted from offsets in spot height or contour data, combined with gaps in worked panels. Areas of probable workings have been interpreted where there is no mine abandonment plan record but where workings have been proved in boreholes as ‘waste’ or ‘voids’.

The deepest mineworkings in the area are shown on the plan of the Kiltongue Coal (Lower Coal Measures) as recorded at depths of -268.5 m. The depths marked on mine plans were surveyed in, and are believed to be accurate to within a metre or less. Accuracy will decrease with distance from a shaft. If underlying seams were mined subsequently to a particular plan, then the depths on the older plan may have been affected by subsidence from younger, underlying workings. This cumulative subsidence may need to be taken into account at a site specific level for borehole planning purposes.

The abandoned, flooded mine workings form the resource for ‘shallow’ geothermal heat and heat storage (Gillespie et al., 2013[2]). Roadways emanating from shafts may be of particular importance as they are believed to be open at the present time, greatly increasing permeability and potential yield (Gillespie et al., 2013[2]).

Figure 7    Summary of known (above) and probable (below) mining extents BGS © NERC 2016 for the Clyde Gateway area.

The coal authority datasets

The Coal Authority is the definitive source of coal mining information in the UK. Some datasets are openly available on its website, such as the position of shafts and mine entries, and areas of known and probable shallow mining (Figure 8, and see www.mapapps2.bgs.ac.uk/
coalauthority/home.html
). GIS versions of mine plan datasets can also be licensed (see www.gov.uk/guidance/coal-mining-records-data-deeds-and-documents).

Figure 8    Screenshot of The Coal Authority interactive viewer for eastern Glasgow highlighting the probable shallow coal mine workings in the north and east of the Clyde Gateway area.

Additional coal mine datasets (e.g. mine discharges, hydrogeochemistry, coal properties) may be held in the archives of The Coal Authority and BGS. Future work will investigate this further.

Legacy 2D seismic data interpretation

Introduction

A set of legacy 2D seismic lines from a 1985 survey located adjacent to the proposed Clyde Gateway area were interpreted with a view to: integrating the information into the Clyde Gateway geological platform; and through this gain a greater understanding of the subsurface in the area. The depth converted seismic horizons were imported into an existing 3D model to help constrain the surfaces. One deep commercial exploration well, Bargeddie 1, was used to tie the seismic reflectors; the well is 190 m off the nearest seismic profile. Velocity data from the well were used to obtain a velocity/depth relationship and to depth convert the seismic events.

Location and dataset

There are no seismic profiles within the Clyde Gateway area but a series of profiles with spacings of >1 km to >3 km and general north–south, east–west orientation are located east and north-east of the project area (Figure 9).

Figure 9    Location of Clyde Gateway area (blue outline), seismic lines (green) and Bargeddie 1 exploration well.

The Bargeddie 1 commercial hydrocarbon exploration well, drilled in 1989, is located 190 m south of seismic profile SAX-85-40 and provides information on the geological succession and velocity data that were used in the interpretation and depth conversion.

The DIGMapGB 1:50 000 and 1:10 000 bedrock and fault shapefiles were imported to Decision SpaceTM to guide and provide quality control on the subsurface interpretation.

A detailed Excel spreadsheet, containing time and depth information data from Bargeddie 1 exploration well was created.

Interpretation methodology

The seismic interpretation was carried out using Decision SpaceTM software. The seismic interpretation is based upon the Two-Way-Travel-Time (TWTT) to selected seismic events (Table 1) taken from the Bargeddie 1 commercial exploration well (Figure 10) and tied to the nearest (~190 m offset) seismic profile SAX-85-40 (Figure 11). Faults identified on the seismic data were verified on the 1:10 000-scale geological map and adjusted to fit where possible. The 1:10 000-scale map was also used to distinguish faults on the seismic data where initial interpretation had failed to identify them. The interpretation was guided by the 1:10 000-scale geological map showing formations mapped at outcrop as these are generally well-constrained by borehole data across the study area. The seismic events tied to the Bargeddie 1 well on seismic profile SAX-85-40 were then tied to key intersecting seismic profiles: SAX-85-38; SAX-85-05; SAX-85-37; IGS82-MV1 and SAX-85-01 that were closest to the Clyde Gateway area. An initial interpretation of seismic profiles adjacent to the Clyde Gateway area was imported to the existing 3D model which includes coal mine abandonment plan data for comparison.

Table 1    Seismic events interpreted.
Seismic TWTT event TWTT in seconds from Bargeddie 1 Seismic Depth event TVDSS in metres from Bargeddie 1
MFQ_ESIOS_TopPassageFm_v2 0.1478 MFQ_ESIOS_TPF_v2_DEPTH 211.20
MFQ_ESIOS_TopUpperLmstFm_v2 0.2122 MFQ_ESIOS_TULF_v2_DEPTH 303.20
MFQ_ESIOS_TopLmstCoalFm 0.4481 MFQ_ESIOS_TLCF_DEPTH 578.20
MFQ_ESIOS_TopLwrLmstFm 0.4662 MFQ_ESIOS_TLLF_DEPTH 606.40
MFQ_ESIOS_BaseLwrLmstFm 0.5194 MFQ_ESIOS_BLLF_DEPTH 698.20
MFQ_ESIOS_BaseResolvablePackage n/a MFQ_ESIOS_BRP_DEPTH n/a
  • Figure 10    Bargeddie 1 well, reproduced from Monaghan (2014)[6] © DECC.
  • Figure 11    Detail from seismic profile SAX-85-40 showing location of Bargeddie 1 exploration well and interpreted seismic events. Seismic data reproduced with permission of UKOGL.

The Top Passage and Top Upper Limestone formations lie above the first sonic velocity recordings in the Bargeddie 1 well. TWTT ties were calculated from their recorded depths in the well using the equation (y = 0.0007x) derived from T/D relationship deeper in section and extrapolated back to zero intercept at Ground Level (GL) (Figure 12; and see Depth Conversion section below). These horizons occur at the base of a seismic package exhibiting relatively high amplitude and continuous seismic reflections representing the Scottish Coal Measures Group (Figure 11; Figure 12; Figure 15).

The Limestone Coal Formation has been significantly faulted out in the Bargeddie 1 well and the Base Upper Limestone Formation is also obscured by the fault plane (Figure 10). A seismic package above the Top Lower Limestone Formation was inferred to represent the Top Limestone Coal Formation seismic horizon. The seismic package tends to be more transparent, containing weaker and less continuous seismic amplitude reflections (Figure 11; Figure 12).

The Top Lower Limestone Formation seismic event was taken at a velocity increase, closest to its interpreted boundary in the Bargeddie 1 well (Figure 11; Figure 12).

The Base Lower Limestone Formation seismic event was taken at a velocity decrease, closest to the interpreted boundary in the Bargeddie 1 well. This event marks the top of the West Lothian Oil Shale (Figure 11; Figure 12).

The Base Resolvable Package, the deepest seismic reflector picked, marks the perceived base of the seismic package that contains interpretable seismic events. This seismic package is interpreted to include the West Lothian Oil-Shale Formation. The seismic reflector picked does not represent a unique seismic event marked by a velocity/density contrast between two different rock formations; it simply marks the base of a seismic package that contains a variety of different geological successions (Figure 12).

Depth conversion

TWTT information from check shots and key formation boundaries were taken from the calibrated velocity and composite logs of the Bargeddie 1 well and plotted against True Vertical Depth SubSea (TVDSS) (Figure 12). There is a straight line relationship through all of the T/D points; however, the extrapolation of this line to Ground Level does not intercept at zero. The equation for this line (y=0.0006x+0.1128) was initially used to depth convert the deeper seismic events in the Bargeddie 1 well (Figure 12).

In order to identify the TWTT of the Top Passage and Top Upper Limestone Formations in the Bargeddie 1 well it was necessary to convert their known TVDSS to TWTT as no time measurements were taken at this level in the Bargeddie 1 well. However, their tops lie at 211 and 303 m (TVDSS) respectively, and it was decided that the equation used for the deeper horizons and derived from a line (if extrapolated) that does not pass through the zero intercept would put these formation tops too deep in terms of TWTT. A different line was generated, specifying a zero intercept, and the equation for that line (y = 0.0007x) was applied to give the TWTT for the top Passage and Upper Limestone formations (Figure 12).

A quick comparison of depth converted values calculated from the Bargeddie 1 well data only with those derived using time/depth data from the a BGS well database, that covers the whole Midland Valley area, shows the latter to be around 30% deeper.

Further consideration of the time/depth relationship resulted in the application of a polynomial relationship, with an intercept of zero, to carry out a final depth conversion of the data; this curve provides an equation describing a time/depth relationship from zero intercept to 1200 m TVDSS. The equation for this line (y = 795.5x2 + 878.29x) was used in a second depth conversion of all the seismic events in the Bargeddie 1 well where y = TVDSS (m) and x = TWTT (seconds) (Figure 13).

Figure 12    TWTT in seconds plotted against TVDSS in metres. The graph shows a straight line relationship between the Time/ Depth pairs measured in the well and described by Equation: y = 0.0006x + 0.1128. The graph also shows the straight line relationship but with Intercept = Zero, described by equation y = 0.0007x. This latter equation was used to calculate TWTT from TVDSS of the Top Passage and Top Upper Limestone formations.
Figure 13    TWTT in seconds plotted against TVDSS in metres. The equation for this line was used in the final depth conversion.

Observations resulting from seismic interpretation

Structural/stratigraphic — The seismic data used in this study were acquired in 1985 and show coherent reflections down to about 1.5 seconds TWTT. Displacements of seismic reflections can generally be tied to faults mapped on the 1:10 000-scale geological map. The map shows a series of dominantly east–west– to west-north-west–east-south-east–ESE–trending relatively continuous faults; these faults locally bound sets of shorter faults that have been mapped trending north-west to north-east.

Hooper (2003)[10] describes a preferred tectonic model for the Carboniferous that begins with a predominantly transtensional dextral tectonic regime during the Namurian and Visean. Hooper (2003)[10] suggests faults formed during this time would have oblique-slip (rather than normal) movements. Regional extension during the early Carboniferous resulted in the development of normal faults and reactivation of pre-existing (Caledonian orientated) faults. Regional shortening in the late Carboniferous resulted in the development of reversed faults (Hooper 2003[10]). For this interpretation, the expected complicated nature of the subsurface, with potentially dip-slip, strike-slip, fault reversal and fault reactivation having occurred, has resulted in a variety of possible hanging-wall/foot-wall stratigraphic relationships that the wide spacing and quality of the seismic data could not resolve. However, seismic interpretation of the seismic data does show evidence of post-depositional compression in the form of anticlinal features e.g SAX-85-40 (Figure 11), SAX-85-37 (Figure 14).

Seismic line SAX-85-37 (Figure 14) shows possible reverse displacement and compression on a southerly dipping fault at shotpoints 350 to 440 and its associated antithetic fault, resulting in the Upper Limestone Formation succession at outcrop juxtaposed with Coal Measures. This is interpreted as being followed by reactivation of the fault in a normal or dip-slip displacement (Figure 14).

Figure 14    Seismic profile SAX-85-37 showing evidence for possible reverse movement on main faults. Seismic data reproduced with permission of UKOGL.

Seismic reflectors deepen gradually southwards. For instance, on seismic Line SAX-85-01, the Base Lower Limestone Formation is interpreted at around 400 msecs TWTT in the north deepening to nearly 800 msecs TWTT in the south. On seismic line SAX-85-37 the Top Lower Limestone Formation is interpreted at less than 400 msecs TWTT in the north deepening to 560 msecs TWTT southwards. In addition, possible onlap is observed within the West Lothian Oil-Shale Formation at shotpoints 310 to 250 (Figure 14).

Igneous features — Some seismic profiles show very high amplitude reflectors whose relationship with surrounding seismic events within the seismic package indicate they may be igneous sills; for instance, profile SAX-85-38 (Figure 15a) and SAX-85-01 (Figure 15b). Igneous sills have been mapped at outcrop in the study area and this interpretation of the seismic data indicates that sills could be expected to be present at different levels within the subsurface.

The Base Resolvable Package seismic reflector may, in places lie close to the top of the Clyde Plateau Volcanic Formation lavas, where these are expected to be present. On some seismic lines, continuous seismic reflectors occur beneath this seismic reflector, most noticeable on SAX-85-04 shotpoints 360 to 100 below 1 second TWTT) indicating a possible eastwards change in the succession at depth. However, sparsity of data meant this could not be followed on other lines.

Forsyth et al. (1996)[7] describe a Penn et al. (1984)[11] seismic interpretation of a c.600 m thick Clyde Plateau Volcanic Formation on two IGS lines situated over 10 km to the north of Clyde Gateway, as well as a top Lower Devonian interpretation at c.1 second.

Figure 15    Above: Detail from seismic profile SAX-85-38, high amplitude events within the Limestone Coal and possibly lower part of Upper Limestone formations. These could be igneous sills (around 600 milliseconds, shotpoints 800 to 650). Below: Detail from seismic profile SAX-85-01, high amplitude events within and just below the Passage Formation could be igneous sills (around 250 milliseconds, shotpoints 107 to 196]. Seismic data reproduced with permission of UKOGL.

Conclusions

Interpretation of selected seismic profiles immediately adjacent to the Clyde Gateway area has confirmed the complex, faulted Carboniferous geology present. The seismic reflectors picked were identified and tied from the Bargeddie 1 exploration well that lies approximately 190 m from the nearest seismic profile. Confidence in the interpretation of the reflectors away from the well decreases as the stratigraphic relationships across faults are not always clear and are likely to be misleading due to the number of possible styles of fault movement that have occurred through time.

Another source of error is introduced in the depth conversion method utilised to allow inclusion into the 3D geological model. While the depth conversion may cause observed error when compared to surveyed-in mine plan data, and absolute depths may be wrong, structural and stratigraphic relationships should remain, provided there are no errors in the TWTT interpretation of seismic reflectors.

A better seismic interpretation of the subsurface would be facilitated by a denser grid of newer seismic lines. However, another cheaper option, if original field tapes are still available, would be to carry out re-processing of the original data. This proved to be successful for a similar vintage of legacy seismic data that imaged the Carboniferous strata in the Firth of Forth (Smith et al., 2011[12]).

There is scope for the seismic interpretation to be improved using the existing dataset coupled with an iterative approach with the 3D model. The initial depth converted horizons, imported into the 3D model, were compared where possible with surfaces generated from mine working data. Potential next steps in the seismic interpretation would be to record the differences observed in the 3D model and then investigate possible reasons for these discrepancies. Closer examination of the events, particularly where they cross faulted boundaries, may reveal errors in the interpretation. In addition, the depth conversion method may also be a source of error that could be improved with selective inclusion of additional Time/Depth information.

Map data

The Clyde Gateway area is located on the BGS 1:50 000-scale map sheets 30E Glasgow (1993 bedrock, 1994 superficial deposits) and 31W Airdrie (1992 bedrock, 1992 superficial deposits). BGS Memoirs form a definitive reference source for the geology of this area (Hall et al., 1998[5]; Forsyth et al., 1996[7]).

The central and eastern side of the study area is located on 1:10 000-scale map NS66SW that was revised in 2008. The western side is located on 1:10 000-scale map NS56SE. BGS hold some updated bedrock linework from around 2007 but this was not incorporated into a revised published map; the published version is from 1995.

Map data is available as GIS shapefiles (DigMapGB) or as scans (Mapviewer).

Gravity and magnetic data

BGS regional gravity and magnetic gridded datasets are based on data points between 1–2 km apart, there are only a handful of data points within the Clyde Gateway area and any interpretation at this scale is not appropriate. Over a larger area, Forsyth et al. (1996)[7] suggest that sedimentary rocks may be underlain by the Clyde Plateau Volcanic Formation in the Airdrie (31W) map district. In the Glasgow map district (30E), Hall et al. (1998)[5] describe gravity lows, and low frequency, low amplitude magnetic anomalies over Upper Palaeozoic sedimentary rocks (such as over the Clyde Gateway area). Areas with Clyde Plateau Volcanic Formation at surface are characterised by gravity and magnetic highs that are interpreted to be caused by hundreds of metres of lavas and underlying basin intrusions (Hall et al., 1998[5]).

No site specific or local surveys were located in the Clyde Gateway area using the BGS Geoscience Data Index (GDI).

Various bedrock property data

Data are lacking on bedrock properties at depths greater than a few tens of metres (see sections 4, 5, 7) but data exist from Carboniferous strata from deep boreholes and hydrocarbon exploration wells elsewhere across central Scotland.

Further data on rock properties at depths greater than a few hundred metres could be extracted from hydrocarbon exploration well records. Monaghan (2014[6], Appendix E) gives some XRD compositional data from deep core samples. Porosity, permeability rock strength and mineralogy data on from some deep core samples was measured by Heriot-Watt and BGS as part of the CASSEM CCS project in 2008/9 and is available from BGS NGDC. Whilst porosity of up to 17% was measured in Upper Devonian-Lower Carboniferous sandstone core samples from depths up to 350 m, the highest permeability measured was 16 mD (where this unit is utilised as an aquifer near surface, much flow is via fractures). A mineralogical and petrological study of the same sandstone core samples highlighted porosity reduction by compaction processes, by authigenic chlorite and by a diagenetic ferroan dolomite and ankerite pore filling cement (Milodowski and Rushton, 2008[13]); however dissolution of feldspar created secondary porosity.

3D deterministic model

Bedrock models were produced in GOCAD by interpolation of all available borehole, mine plan and map outcrop data in 2008 and 2011 (documented in Merritt et al., 2009[14]; Monaghan et al., 2012[15]). The bedrock model of the Clyde Gateway area was updated in 2016–17 to include any new borehole data and migrated to an improved workflow in GOCAD-SKUA 15.5 (Kearsey, 2017[16]; Figures 16, 17, 18).

Figure 16    Overview image of the 7 by 4 km bedrock model to depths of -500 m. The Scottish Coal Measures Group is represented in the black to orange colours. The Passage, Upper Limestone and Limestone Coal formations are in blue.
Figure 17    Depth grid to the Glasgow Main Coal from the bedrock model described in Kearsey (2017)[16]. Includes mapping data licensed from Ordnance Survey. © Crown Copyright and/or database right 2017. Licence number 100021290 EUL.
Figure 18    Depth grid to the Kiltongue Coal from the bedrock model described in Kearsey (2017)[16]. Includes mapping data licensed from Ordnance Survey. © Crown Copyright and/or database right 2017. Licence number 100021290 EUL.

Model revision

The bedrock model was revised to include a limited amount of new borehole information and to utilise up-to-date 3D modelling software (Kearsey, 2017[16]). The result is a consistent faulted geological model including in areas poorly constrained by data. A limitation of the model is that it is a best fit of highly variable borehole and mining data, rather than an exact fit of data points.

Seismic data are not currently used in the 3D bedrock model because whilst largely in agreement with borehole and mine data, there are inconsistencies in either seismic interpretation or depth conversion. These result in a mismatch between the borehole/mine and depth-converted seismic interpretation picks which could be resolved in an iterative interpretation and modelling process. However, since only a limited section of a seismic line is within the smaller area modelled in detail, and it is to the north-east of the main area of interest, this iterative interpretation and modelling step was not deemed to be a top priority at the current time.

Bedrock descriptions, conceptual models and uncertainties

Figure 19    Illustration of the depth limit of borehole data within the Clyde Gateway area in the 2011 bedrock geology model, and the single hydrocarbon well available to the east. Interpretation of Bargeddie 1 from Monaghan, 2014[6].

The deepest borehole drilled from surface within the Clyde Gateway area is the Dalmarnock borehole/shaft [NS66SW BJ236] to a drilled depth of 294 m. This is interpreted to penetrate beyond the base of the Scottish Lower Coal Measures Formation and into the top of the Passage Formation (Figure 19). A 173 m long underground borehole drilled from the pavement of the mined Kiltongue Coal in the Govan no.5 pit [NS66SW BJ197] is interpreted to prove the base of the Scottish Lower Coal Measures, the Passage Formation and the top part of the Upper Limestone Formation and documents the bedrock geology for the area to around 430 m.

Whilst the information is very sparse, the records that exist indicate the ‘expected’ Carboniferous sequence.

The Bargeddie 1 hydrocarbon exploration well, situated 5.5 km to the east of the study area, records over 1000 m of the Carboniferous succession, terminating in the West Lothian Oil-Shale Formation (Figure 19). There are also a small number of boreholes to the east and north of the Clyde Gateway study area that penetrate as deep as the Index Limestone (base Upper Limestone Formation).

The bedrock geology at depths greater than 430 m in the Clyde Gateway area is therefore interpreted from adjacent areas where strata beneath the Coal Measures are present at shallower levels, and are extensively penetrated by boreholes, plus occasional rock outcrops. The Carboniferous succession is proved to be of variable thickness (Table 2) and borehole prognosis requires detailed work on local thickness variations and trends.

Table 2    Thickness estimates (metres) and variability to inform deep borehole prognosis end members. The study area is located in the southern corner of the Airdrie and Glasgow 1:50 000-scale geological maps. The yellow highlighted cells represent the estimated maxima and minima for a 2 km deep borehole which could reach volcaniclastic deposits above the Clyde Plateau Volcanic Fm., or the Lower Devonian in a more condensed succession.
Unit Bargeddie 1 well Boreholes in area NE56SE 10k sheet GVS NE66SW 10k sheet GVS Glasgow 50k GVS Glasgow Memoir Airdrie 50k GVS Airdrie Memoir Cumulative minimum thickness (to base) Cumulative maximum thickness (to base)
Upper Coal Measures 275 (Coal Measures Group) 230+ Coal Measures Group 88+ 86+ 100 100 264 270 100 270
Middle Coal Measures 150 157 162 100 192 195 200 465
Upper Coal Measures 110 96 105 100 129 135 300 600
Passage Fm 90 84 m NS66SW197 44 81 78 80 189 215 378 815
Upper Limestone Fm 275 55+ NS56SE227 234 33+ 270 270 294 270 648 1085
Limestone Coal Fm 21.5 (faulted) 180+ NS56SE255 409 312 300 345 350 948 1494
Lower Limestone Fm 95 172 NS56SE255 183 108 80 180 180 1028 1677
Lawmuir, West Lothian Oil-Shale Fm 265.5+ (WLOS) 189 0–330 192 195 1028 1943
Kirkwood Fm 141 0–70 66 65 1028 2084
Clyde Plateau Volc. Fm 450 450 804 c. 700 1478 2888
Clyde Sandstone Fm 72 0–80 90 0-110 1478 2998
Ballagan Fm 102 0–180 204 0-380 1478 3378
Kinnesswood Fm 171 180 135 140 1618 3558
Stratheden Group/Stockiemuir Sandstone Fm 402 400 35+ 1653 3960
Strathmore Group/Teith Sandstone Fm 201 200+ 1853 4161

Unit descriptions

This section provides a summary of each stratigraphic unit to inform borehole planning.

Scottish coal measures group

The Scottish Coal Measures Group is characterised by coal-bearing fluvio-deltaic sedimentary rocks in cyclical sequences of mudstone, siltstone, seatearth (rootlet-bearing palaeosol), sandstone and coal. The depositional environment was interpreted to be a broad flat, coastal deltaic plain in which coal swamp conditions occurred frequently. The Group is divided into three formations, based on lithological variations, marine band biostratigraphy and non-marine macrofossil and palynological assemblages.

  1. The Prospecthill borehole [NS56SE BJ393] is representative of the Scottish Upper Coal Measures Formation (UCMS, Bolsovian, Westphalian C, up to c.80 m thick) succession, containing a 3 m mudstone that includes the Aegiranum Marine Band, the 60 m thick Westmuir Sandstone, and 23 m of argillaceous rocks including a thin coal and the Bothwell Bridge Marine Band (Hall et al., 1998[5]). Red-purple staining is characteristic of the formation, though some beds retain their original pale grey colour. The formation is not known to contain any mined coals within the Clyde Gateway area.
  2. The Scottish Middle Coal Measures Formation (MCMS, Duckmantian, Westphalian B) is described by Hall et al. (1998)[5] as being 160 m thick (thinner than in most parts of the Scottish coalfields), and to contain an unusually high proportion of mudstone and siltstone with rare fluvial sandstones. This unit contains the thickest mined coals such as the Glasgow Upper, Glasgow Ell, Glasgow Main, Splint and Virgin (Figure 20). Further details of the succession are given in Hall et al. (1998)[5], Hinxman et al. (1920)[17], and Clough et al. (1926)[18].
  3. The Scottish Lower Coal Measures Formation (LCMS, Langsettian, Westphalian A, c.100 m thick) consists of a basal sandstone-seatearth sequence, a middle part with coals up to 0.75 m thick and an upper sandstone-seatearth dominated sequence with several coals (Hall et al., 1998[5]). The Kiltongue and Airdrie Virtuewell coals were mined from Govan no.5 pit (in the south-west of the Clyde Gateway area).
Figure 20    Vertical section of the Scottish Middle Coal Measures Formation from Hall et al. (1998)[5].

Passage formation

The Passage Formation (PGP, Namurian-Westphalian A) is thinner in the Clyde Gateway area than in other parts of central Scotland, reaching around 85 m in an underground borehole [NS66SW BJ197] in Govan no.5 pit. It is dominated by sandstone, some coarse-grained, with red-purple-green-yellow mudstone, seatclay and fireclay. Records of bedded mudstone, coal and marine bands are rare; however the no.3 Marine Band is interpreted in the Alexandra Parade borehole [NS66NW BJ408] (Forsyth et al., 1996[7]) to the north of the Clyde Gateway area. The Passage Formation is interpreted as a dominantly fluvial system with many minor unconformities (Hall et al., 1998[5]).

The Passage Formation represents a target geothermal aquifer for this study.

Upper limestone formation

The Namurian (Pendleian-Arnsbergian) Upper Limestone Formation (ULGS) is proved in boreholes outside the Clyde Gateway area and interpreted in the base of the Govan no.5 pit underground borehole [NS66SW BJ197]. It reaches around 270 m in thickness. The unit is characterised by cyclical sequences of mudstone, siltstone, seatrock, coal, sandstone and limestone. In places, cyclic sequences are replaced by thick, medium- to coarse-grained sandstones with markedly erosive bases; for example, the Upper Drumbreck and Cadgers Loan sandstones north of the River Clyde, and the Barrhead Grit up to 55 m thick in south-west Glasgow (Hall et al., 1998[5]). Depositional environments are interpreted as ranginf from shallow seas, to deltaic and alluvial plains and river channels (Hall et al., 1998[5]).

Limestone coal formation

Proved in boreholes outside the Clyde Gateway area, the Limestone Coal Formation (LSC, Namurian, Pendleian) comprises cycles of coal, mudstone, siltstone, sandstone, seatearth and coal with some thicker mudstone intervals (e.g. Black Metals Marine Band), ironstone and limestone. Thickness across the wider Glasgow area varies from 270 to 350 m (Hall et al., 1998[5]). The numerous coals range from 0–1.8 m in thickness and have been mined extensively in areas from the west to north-east of the Clyde Gateway area. The majority of sandstones within the succession are fine grained. Rarely, medium- and coarse-grained channel fill sandstones with erosive bases which cut out coals, are up to around 20 m thick (e.g. Nitshill Sandstone, Cowlairs Sandstone; Hall et al., 1998[5]).

Lower limestone formation

In the wider Glasgow area, the Lower Limestone Formation (LLGS, Brigantian) thickens eastwards from 110 m to around 180 m. The unit is dominated by bedded mudstone with siltstone, sandstone, limestone and thin coal and ironstone. The depositional environment is interpreted as a quiet marine and non-marine backwater with sandy lobes extending from the east (Hall et al., 1998[5]).

Strathclyde group

In areas west and north-west of the Clyde Gateway area, the Visean age Strathclyde Group comprises sandstone, mudstone, thin coal, seatearth and conglomerate of the Lawmuir Formation (LWM) and the underlying volcanic detritus of the Kirkwood Formation (KRW; Hall et al., 1998[5]). The unit ranges from thicknesses of 10’s to 200 m and is interpreted as having been deposited in fluvial and other environments with a series of marine incursions. East of the Clyde Gateway area, similar Brigantian age strata proved below 780 m in the Bargeddie 1 hydrocarbon exploration well are assigned to the West Lothian Oil-Shale Formation (WOLS), as a basinal equivalent of the Lawmuir Formation (Monaghan, 2014[6]). Lack of data creates uncertainty in the character of Visean sedimentation underlying the Clyde Gateway area (see section 2.4.3 below).

Clyde plateau volcanic formation

A succession of up to 1000 m of basaltic to trachytic lavas and volcaniclastic deposits constitutes the Clyde Plateau Volcanic Formation. This forms the Campsie, Kilpatrick, Renfrewshire and Beith-Barrhead hills surrounding Glasgow. Based on seismic reflection data from c.6 km to the north of the Clyde Gateway area, the lavas are inferred to extend southwards from the Campsie Fells, beneath the upper Carboniferous sedimentary basin. The formation is present at outcrop some 2 km to the south of the Clyde Gateway area, in the footwall block of the major Dechmont Fault structure. However, the extent of these volcanic rocks beneath the upper Carboniferous sedimentary basin in the vicinity of the Clyde Gateway area is unknown. This is an important potential constraint, and significant uncertainty, for any deep drilling in the Clyde Gateway area.

Inverclyde group

The character of the Inverclyde Group is inferred from outcrops beneath the Clyde Plateau Volcanic Formation, some distance north of the Clyde Gateway area. The fluvial Clyde Sandstone Formation (CYD, up to 80 m, Tournaisian-Visean) is underlain by interbedded mudstone and dolomitic limestone of the marginal marine-lagoonal Ballagan Formation (BGN, up to 180 m, Tournaisian) and sandstone and concretionary carbonates of the fluvial Kinnesswood Formation (KNW, up to 25 m, Tournaisian).

Stratheden group

Upper Devonian sandstones are exposed beneath the Inverclyde Group to the north of the Campsie and Kilpatrick Hills and west of the Renfrewshire Hills, at significant distances (25–40 km) from the Clyde Gateway area. The >400 m Stockiemuir Sandstone Formation comprises largely fluvial and aeolian sandstone, is a partial lateral equivalent of the Knox Pulpit Formation of Fife and, along with the Kinnesswood Formation is considered as a potential hot sedimentary aquifer geothermal resource (Hall et al., 1998[5], Browne et al., 1985[4]). The presence, depth, thickness and character of such a unit beneath the Clyde Gateway area is very uncertain.

Igneous intrusions

Olivine dolerite sills are proved at various levels within Carboniferous sedimentary strata in the vicinity of the study area:

  • Within the basal Passage Formation to upper Upper Limestone Formation and within upper parts of the Limestone Coal Formation on NS56SE (BGS 1995[19]), proved outwith the Clyde Gateway area.
  • Within lower parts of the Middle Coal Measures and upper parts of the Lower Coal Measures on the generalised vertical section of NS66SW (BGS 2008[20]) but not present at surface on the map sheet (mapped at surface in the sheet to the north). A relatively small number of boreholes record igneous intrusions to the north of the Clyde Gateway area e.g. four intervals of dolerite from 4–22 m in thickness in the Crown Brewery borehole [NS66SW BJ 34], at Camlachie.

No dykes are recorded on the BGS 1:10 000-scale map of the Clyde Gateway area, or in mine plan data, though these are common in some other parts of central Scotland. A systematic examination for records of igneous rocks within the existing borehole record dataset should be undertaken once potential drilling sites are identified.

Sand body geometry

The likely volume of sand bodies within the heterolithic Carboniferous succession of central Scotland is one factor in the sustainability of a hot sedimentary aquifer geothermal resource. The sandstones are interpreted to have been deposited in a variety of environments from marine, deltaic to fluvial and aeolian (e.g. Hall et al., 1998[5]; Browne et al., 1999[21]; Read 1988a[22], 1989[23]) Sandstone beds/units range in thickness from millimetres to tens of metres within a mixed lithology succession and, where the thickest sandstone units are exposed or correlated in boreholes, commonly have localised spatial extents interpreted as fluvial or fluvio-deltaic channel systems (e.g. Hall et al., 1998[5]). The exception in the upper Carboniferous sequence is the sandstone-dominated Passage Formation that is interpreted as a dominantly fluvial succession containing many minor unconformities (Hall et al., 1998[5]; Read 1988b[23]).

Examples of sandstone bodies in the vicinity of the Clyde Gateway area are summarised in Table 3 below and in photos P708158 and P219897.

Table 3    Summary of main sandstone units documented beneath the Coal Measures Group in the vicinity of the Clyde Gateway area.
Stratigraphic Unit Summary of known thickness/extent of sand body in Glasgow area
Passage Formation Govan no 5 pit shaft proves mainly sandstone (84 m thick), interpreted as fluvial with minor unconformities (Hall et al., 1998[5]).
Upper Limestone Formation Barrhead Grit (55 m, in west of Glasgow), Giffnock sandstone, Cadgers Loan Sandstone.
Limestone Coal Formation Nitshill Sandstone, Cowlairs Sandstone (beneath the Index Limestone). Maximum 20–25 m.
Lower Limestone Formation Various levels but laterally discontinuous.
Lawmuir Formation Hall et al. (1998)[5] document some thick sandstones to the west of the study area, laterally variable.
West Lothian Oil-Shale Formation and equivalents Sandstones were conventional hydrocarbon targets across Central Scotland e.g. Bargeddie 1, Salsburgh etc.
P708158    Passage Formation sandstone, Levenseat silica sand quarry, Fauldhouse, West Lothian. BGS © NERC.
P219897    Example of variable sand body geometry (centre) within Limestone Coal Formation strata, Blindwells Opencast Coal Site, East Lothian. BGS © NERC.

Further research on the geometry of sand bodies within the Passage and Upper Limestone Formation, linked to their physical properties and measured flow yields would be beneficial for improved understanding of potential hot sedimentary aquifer geothermal targets likely to be encountered in the Clyde Gateway area. This could involve combining the legacy borehole and well dataset, previous work, outcrop exposures with the significant worldwide knowledge of fluvial hydrocarbon reservoirs and reservoir properties.

Conceptual understanding of geology at depth

As has been highlighted above, uncertainty on the likely lithological variations at depth within the subsurface beneath the Clyde Gateway area can be reduced by using regional data and knowledge. Palaeogeographic reconstructions of the depositional environment through time form an effective tool in synthesis of the complex tectono-stratigraphic setting that is interpreted for the Carboniferous of central Scotland.

As well as the overarching regional picture described in overview publications and BGS memoirs (Forsyth et al., 1996[7]; Hall et al., 1998[5]; Browne et al., 1999[21]; Read et al., 2002[8]), key papers documenting palaeogeographic reconstructions include:

Scottish Coal Measures Group: Read (1989)[24] shows a uniform Westphalian A succession to the south-east of Glasgow.

Passage Formation: Read (1988b)[23] shows coarse clastic input towards Glasgow and a basal disconformity towards the Dechmont Fault.

Upper Limestone, Limestone Coal, Lower Limestone formations: Goodlet (1957)[25] showed a shale-dominated lithofacies map of the Lower Limestone Formation in the vicinity of Glasgow. Wilson (1989)[26] used macrofossil assemblages to deduce the depositional environments of limestone beds and highlighted an eastern origin for marine transgressions. In the Limestone Coal and Upper Limestone formations, Read (1988) highlights coarse clastic input from the east-north-east towards the south of Glasgow. The response of the channel systems to sea level variations is illustrated in Read (1994)[27] in which there is a continued channel system towards the Clyde Gateway area, but which at relative high sea level is shown as submerged beneath brackish water. Using deep hydrocarbon wells and boreholes, Monaghan (2014)[6] highlights a high percentage of shale (mudstone/siltstone) within the succession in the Glasgow area. Hooper (2003)[10] provides details of sedimentology and notes the lack of basin marginal alluvial fans.

West Lothian Oil-Shale Formation and lateral equivalents of the Strathclyde Group: Loftus and Greensmith (1988) highlighted the facies variability of an oil-shale lake, volcanism, fluvio-deltaic and marine sedimentation. Monaghan (2014[6]; Figure 23) produced additional time slices and highlighted the westward extension of the West Lothian Oil-Shale Formation towards the Glasgow area, as proved in the Bargeddie 1 well drilled in 1989.

Figure 23    Regional paleogeographical reconstruction, Latest West Lothian Oil-Shale Formation times (near the top of the unit, c.331 Ma, NM palynomorph zone). Evidence is patchy and the reconstruction is tentative. Dashed lines are faults and folds with evidence for active growth from Monaghan 2014[6] © DECC.

Further work on semi-regional palaeogeographic reconstructions using data from in and around the Glasgow area would be highly beneficial to better understand predicted lithological variability at depth in the Clyde Gateway area, as a tool to inform pre-drill borehole prognosis.

Common to the majority of published work are the long-lived high areas underlain by the Clyde Plateau Volcanic Formation, that form an enclosed western end of Carboniferous sedimentary basins. An initial conceptual understanding of the generalised setting (Figure 24) highlights possible end members of deposition adjacent to the Dechmont Fault structure and within the upper Carboniferous basin, bounded by the volcanic uplands of the Beith-Barrhead hills to the south and the Campsie Fells at some distance to the north.

Figure 24    Initial 3D conceptual ideas on palaeogeography in the Clackmannan Group (e.g. Lower Limestone Formation) with a bearing on location and thickness of sandstones. a) an interpretation of palaeogeography with fault planes highly exaggerated b) potential fluvial end member of deposition, though note this kind of fault-controlled sedimentation is rarely evidenced in the Carboniferous of central Scotland c) potential lacustrine/brackish siltstone/mudstone-dominated end member with little fault relief.

Further development of the geological platform should focus on documenting and reducing uncertainty in the deep bedrock geology of the Clyde Gateway area.

Fault damage zones

Faults can behave as pathways or baffles to fluid (or both simultaneously) and can have a significant impact on subsurface temperatures due to their influence on rock permeability and resulting impacts on fluid flow pathways and heat transport. Fault damage zones in brittle rocks tend to fracture, creating breccia or joints (Bense et al., 2013[28]). Often, the damage zone around the fault core can be much more extensive and hydraulically more important, contributing more towards bulk permeability than the fault core (10’s to 1000’s of meters in thickness perpendicular to the fault strike compared with 0.1 to 10 m thickness).

Hydrogeological and geothermal considerations relating to faults and fracture zones are given in Faults and fluid flows, this section presents preliminary analysis of possible fault damage zone widths based on faults included in the approved version of the bedrock model (Monaghan et al., 2013[29]) that covers the Clyde Gateway area.

Background

There is generally a broad positive relationship between fault zone width and fault throw (e.g. Beach, 1999[30]; Fossen and Hesthammer, 2000[31]; Savage and Brodsky, 2011[32]; Shipton and Cowie 2001[33]), with lower displacement faults typically having a smaller damage zone. However, field evidence has proved that fault thickness can vary by three orders of magnitude along the same fault, regardless of displacement value or host rock lithology (e.g. Shipton et al., 2006[34]). Beach (1999)[30] observed maximum damage zone values of 80 m in sandstone, even along faults with several hundred metres of displacement. Shipton and Cowie (2001)[33] found that faults in sandstone had a damage zone width approximately 2.5 times that of the fault throw, though with the caveat that predicted damage zones could vary by about as much as 10–20% for any throw value. Knott et al. (1996)[35] found damage zone widths in mixed lithologies in the hangingwall and footwall had different distributions, whilst still scaling with offset. Beach et al., (1999)[30] and Antonellini and Aydin (1995)[36] found that damage zones are narrower in finer-grained argillaceous rocks, but wider in coarser sandstones. Field observations from surface coal mines in Scotland show that fault damage zones hosted in Carboniferous carbonates such as limestones are typically fracture dominated, and extend over a larger area than those in sandstone, mudstone or coal (Figure 25). Damage zones are hypothesised to decrease in size with increasing depth, due to a presumed increase in the strength of rock surrounding the fault zone (e.g. Scholz, 2002[37]). Other parameters such as associated diagenesis, depth of faulting and tectonic environment have been suggested as other factors responsible for scattering of data within the overall broad positive trend of fault displacement vs fault thickness (Choi et al., 2016[38]).

Figure 25    Faulted McDonald limestone pavement exposed in the Spireslack surface coal mine, East Ayrshire. The faults and fractures here form part of a broader damage zone of an oblique-slip fault with 25 m displacement. The damage zone is composed of subsidiary synthetic faults, brittle fractures, and mineral veins.

Modelling damage zones beneath the Clyde Gateway area

The nature and width of fault damage zones beneath the Clyde Gateway area is not known with any certainty, as we have no field and only limited subsurface data. Mine plans from beneath the site accurately record the position, dip and often the throw of the faults, and their dip separation as they displace a coal seam, but there is no information on fractures relating to the faults on these plans. During mining ,coal would have extracted up to fault planes that significantly displaced the coals, and regardless of fracture content, and so mine plans cannot be used to determine damage zone width.

To model potential damage zone widths across faults in the Clyde Gateway area, data from fault widths in siliciclastic rocks from other studies are used (Beach et al., 1999[30]; Shipton and Cowie, 2001[33]; Fossen and Hesthammer, 2000[31]; Shipton et al., 2006[34]; Savage and Brodsky, 2011[32]). The BGS Central Glasgow bedrock 3D model (Monaghan et al., 2013[29]) is used to model likely damage zone widths around faults cutting the Carboniferous strata. The damage zone widths are applied based on the smallest likely damage zone width and the largest likely damage zone width, related to the average throw along the fault. For example, a fault with a recorded displacement of 10 m may have a damage zone as small as 1 m or as large as 100 m based on data from Fossen and Hesthammer (2000)[31], Beach et al., (1999)[30] and Shipton and Cowie (2001)[33]. For this study, the throws recorded along each fault in the Clyde bedrock model were arranged by order of magnitude (i.e. 1–10 m throw, 10–100 m throw, etc). Most fault throws fall between the range of 10 and 100 m, but the Shettleston and Dechmont faults have throws of over 100 m. Therefore, the likely ranges of damage zones for faults with throws of between 10 and 100 m, and 100 and 1000 m, are taken from existing data (see Table 4). Faults with throws of between 1 and 10 m were not included in the original 3D geological model but their ranges are included here also.

Table 4    Fault thickness ranges organised by fault throw.
Source: Shipton et al., 2006[34] (fault core and damage zones) Beach et al., 1999[30] (damage zones) Shipton and Cowie 2001[33] (damage zones) Fossen and Hesthammer 2000[31] (damage zones) Savage and Brodsky, 2011[32] (fault zone thickness)

Fault thickness ranges (m)

Throw (m) Min Max Min Max Min Max Min Max Min Max
1–10 <10 50 0.3 30 5 35 10 50 n/a n/a
10–100 6 100 1 100 25 >80 6 100 n/a n/a
100–1000 0.7 100 5 n/a n/a n/a 50 n/a 5 100

From these, the most likely ranges were taken and used for modelling damage zone widths within the Clyde Gateway area. For example, the data from Shipton et al. (2006)[34] included undifferentiated fault core and damage zone width thicknesses: therefore the value of 0.7 m as a minimum thickness for damage zone width is not used as that is likely to be a data point from a fault core reading. The modelled widths are described in Table 5.

Table 5    Damage zone ranges used in damage zone thickness modelling of faults in the Clyde Gateway. See Table 4 for source data and section 2.5.2.1 for caveats.
Throw (m) Modelled Widths (m)
1–10 (not modelled in this study) 0.3 min, 50 m max
10–100 1 m min, 100 m max
100–1000 5 m min, 100 m max

For modelling purposes, the damage zone is represented by a copy of the originally modelled fault plane, placed at a distance of 100 m on either side of the original fault (see Figure 26 and Figure 27). All faults have been modelled with a maximum damage zone width of 100 m. Due to the resolution of the model, the minimum of 1 m wide for damage zones associated with faults of 10–100 m displacement faults has not been modelled, but the 5 m wide minimum for damage zones associated with faults with displacements of 100–1000 m has been modelled.

Figure 26    Damage zone limits modelled on either side of the existing fault network in the BGS 3D Clyde geological model. The originally modelled faults are coloured in red, and their maximum likely damage zone widths are in grey in both the hangingwall and the footwall of the fault. The rectangular outline of the Clyde Gateway area is included.
Figure 27    Cross-section across the Clyde Gateway area, intersecting the Dechmont Fault and the Shettleston Fault. Likely maximum damage zone widths are outlined in grey, on either side of the red fault plane.

Model caveats

The likely damage zone widths of a particular fault within the Clyde Gateway area are based on data collected from exhumed fault zones across the world. However, this maximum damage zone width value is likely to exceed the true damage widths of the faults in the Clyde Gateway area due to a number of factors:

- Damage zones are often asymmetric, with damage zones having different widths on either side of the fault. This asymmetry has been reported in many previous publications, and is partly attributed to the differing rock properties on either side of the fault, and to differing stress conditions in the fault walls (see review paper of Choi et al., 2016[38]). For example, the hangingwall damage zone is more than three times wider than the footwall damage zone in the Moab Fault in Utah (Berg and Skar, 2005[39]). The extent of damage in the footwall and hangingwall of faults in the Clyde Gateway area is not known; therefore further data are required to model the damage zones more accurately within the mixed lithologies in the Clyde Gateway area.
- Displacements along the faults in the Clyde Gateway area vary laterally, and with depth; it is likely therefore that the damage zone widths will vary along and down faults. Future work could focus on modelling damage zone widths with fault displacement along faults, and with depth.
- The influence of host rock lithology on damage zone width and type is not known within the Clyde Gateway area. Published research has shown that host rock lithology has an effect on both the width and nature of the damage zone: Beach et al., (1999)[30] and Antonellini and Aydin (1995)[36] found that damage zones are narrower in finer-grained argillaceous rocks, but wider in coarser sandstones. The damage zone maximum widths modelled in this study are based on research carried out predominantly in sandstone-hosted faults. However, the strata beneath the Clyde Gateway area are mudstone-rich, and in a mudstone hosted fault, it is possible that fault zone width will be narrower. Further data are required to model mudstone-hosted faults.
- Scholz (2002)[37] hypothesised that damage zone width should narrow with depth due to a presumed increase in the strength of rock surrounding the fault zone. Whilst this has not been modelled here due to a lack of data, future work could investigate this aspect further.

Further caveats and questions related to the modelling of fault damage zones include:

- How do overlapping damage zones affect the widths and nature of the damage zones (e.g. Figure 28). Is this area more intensely damaged than damage zones which are not linked?
- What effect will the damage zones have on hydraulic behaviour in the subsurface?

Future work should seek to address these problems specific to the Clyde Gateway area.

Figure 28    Example of overlapping damage zones within the Clyde Gateway area.

In situ stress

Understanding the geomechanics and stress state of the bedrock is important for any subsurface activity where the injection or extraction of fluid from the rock volume has the potential to change pore pressure. Changes in pore pressure have the potential to cause re-activation of fault structures or compaction.

Baptie (2010)[40] and Baptie et al. (2016) describe current understanding of the stress field across central Scotland and the fact that there are no good stress data. In a regional context, faults striking east–west or east-north-east–west-south-west are considered to have a low reactivation potential, while faults striking north-east–south-west or north-west–south-east have the relatively highest reactivation potential based on the very limited existing dataset (Baptie et al., 2016[41]). Site specific study and local data are required to assess further the potential for fault reactivation.

Stress is a six component tensor; however it can be simplified to four components for a stress field characterisation: vertical stress (Sv), minimum horizontal stress (Shmin) and maximum horizontal stress (SHMax) and stress field orientation. The relative magnitudes of these stress components vary depending on the fault environment (Table 6).

Table 6    Summary of stress states associated with fault environments; after Zobach et al. (2003)[42].
Faulting Environment σ1 σ2 σ3
In normal faulting Sv

SHMax

Shmin
In strike slip SHMax

Sv

Shmin
In reverse faulting SHMax

Shmin

Sv

Stress field orientation

The stress field orientation typically defines the orientation of SHMax, which is perpendicular to the orientation of Shmin. Borehole breakouts which define the orientation of Shmin can be identified using imaging tools and 4-Arm calipers (Reinecker et al., 2003[43]; Tingay et al., 2008[44]; Kingdon et al. 2016[45]). The Bargeddie 1 well has 4-Arm Caliper data, but no borehole imaging. Kingdon et al. (2016)[45] showed that the use of imaging tools over 4-Arm caliper data can reduce uncertainties in the stress field orientation. Given the lack of borehole imaging data, however, no study of the stress orientation has been attempted.

Vertical stress

The vertical stress is often used to predict fracture gradients and pore pressure in the absence of in situ data (Tingay et al., 2003[46]). It has been documented in multiple studies that the state of stress can be highly variable and where possible the vertical stress should be determined using in situ data (Tingay et al., 2003[46]; Williams et al., 2015[47], 2016[48]; Verweij et al., 2016[49]).

Vertical stress can be calculated using density logs, after the method of Zoback et al. (2003). This method integrates measurements from density logs from the surface to the area of interest — the total depth of the well in this case. Using Equation 1:

Sv = ∫z ρ(z)g dz ≈ ρ̅gz            (1)

where ρ̅ is the mean overburden density, ρ(z) is the density as a function of depth and g is the acceleration due to gravity. Figure 29 shows the two wells in the greater Glasgow area, Bargeddie 1 and Maryhill, for which data are available.

Figure 29    Location of wells in the Glasgow area with data that could be used to calculate vertical stress from density logs. Includes mapping data licensed from Ordnance Survey. © Crown Copyright and/or database right 2017. Licence number 100021290 EUL.

Maryhill has a total depth (TD) of 306 m with a density log from 55 m. Bargeddie 1 has a density log from 440 m to 1040 m. To calculate the vertical stress for Bargeddie 1 the density had to be estimated from 0–440 m. The density from ground level to 55 m was assigned a value of 2.1 G/CC. From 55–440 m an average density was calculated using the density from Maryhill (2.57 G/CC). The vertical stress gradient for Bargeddie 1 calculated using this method is 25 MPa km-1.

Minimum horizontal stress

The minimum horizontal stress (Shmin) is also the minimum principle stress (σ3) in both normal and strike slip environments. Baptie (2010)[40] demonstrated that the UK is predominately a strike slip/reverse environment with north-west–south-east compression driven by the Mid-Atlantic Ridge. The minimum horizontal stress can be estimated from formation integrity tests or leak off tests. Extended leak off tests are required to determine the value of Shmin(XLOT’s). There are no XLOT’s available for this area and only one leak off test from Bargeddie 1 which is not sufficient to estimate the magnitude of the minimum horizontal stress.

Maximum horizontal stress

The data required to quantify a maximum horizontal stress (e.g. tensile strength, compressive strength, reservoir pressure, minimum horizontal stress) was not located in the available datasets and no legacy hydraulic fracturing or overcoring data have been found in the area.

References

  1. CAMPBELL, S D G, MERRITT, J E, Ó DOCHARTAIGH, B E, MANSOUR, M, HUGHES, A G, FORDYCE, F M , ENTWISLE, D C, MONAGHAN, A A, and LOUGHLIN, S C. 2010. 3D modelling and related datasets for Urban Development - A case study in Glasgow- Clyde, UK 3D geological models and their hydrogeological applications: supporting urban development: a case study in Glasgow-Clyde, UK in Zeitschrift der Deutschen Gesellschaft fur Geowissenschaften, Vol. 161 pt/no.2, 251–262.
  2. 2.0 2.1 2.2 GILLESPIE, M R, CRANE, E J, and BARRON, H F. 2013. Study into the Potential for Deep Geothermal Energy in Scotland: Volume 2. British Geological Survey Commissioned Report, CR/12/131 Scottish Government Project Number AEC/001/11. ISBN: 9781782569862 (web only publication).
  3. KEARSEY, T I, WHITBREAD, K, ARKLEY, S L B, FINLAYSON, A, MONAGHAN, A A, MCLEAN, W S, TERRINGTON, R L, CALLAGHAN, E, MILLWARD, D and CAMPBELL S D G. In press. Creation and delivery of a complex 3D geological survey for the Glasgow area and its application to urban geology. Earth and Environmental Science: Transactions of the Royal Society of Edinburgh.
  4. 4.0 4.1 4.2 BROWNE, M A E, HARGREAVES, L, and SMITH, I F. 1985. The Upper Palaeozoic basins of the Midland Valley of Scotland. Investigation of the Geothermal Potential of the UK. Keyworth, Nottingham. British Geological Survey.
  5. 5.00 5.01 5.02 5.03 5.04 5.05 5.06 5.07 5.08 5.09 5.10 5.11 5.12 5.13 5.14 5.15 5.16 5.17 5.18 5.19 5.20 5.21 5.22 5.23 5.24 5.25 HALL, I H S, BROWNE, M A E, and FORSYTH, I H. 1998. Geology of the Glasgow district. Memoir of the British Geological Survey, Sheet 30E (Scotland). ISBN 0-11-884534-9.
  6. 6.0 6.1 6.2 6.3 6.4 6.5 6.6 6.7 6.8 MONAGHAN, A A, BROWNE, M A E, and BARFOD, D N. 2014 An improved chronology for the Arthur's Seat Volcano and Carboniferous magmatism of the Midland Valley of Scotland. Scottish Journal of Geology, Vol. 50, 165–172.
  7. 7.0 7.1 7.2 7.3 7.4 7.5 7.6 FORSYTH, I H, HALL, I H S, MCMILLAN, A A, ARTHUR, M J, BRAND, P J, GRAHAM, D K, and ROBINS, N S. 1996. Geology of the Airdrie district: Memoir of the British Geological Survey, Sheet 31W (Scotland). ISBN 011884508X
  8. 8.0 8.1 READ, W A, BROWNE, M A E, STEPHENSON, D, and UPTON, B J G. 2002. Carboniferous. In: Trewin N.H. (ed) The Geology of Scotland. Fourth Edition. The Geological Society, London, 251–300.
  9. UNDERHILL, J R, MONAGHAN, A A, and BROWNE, M A E. 2008. Controls on structural styles, basin development and petroleum prospectivity in the Midland Valley of Scotland. Journal of Marine and Petroleum Geology, 25, 1000–1022.
  10. 10.0 10.1 10.2 10.3 HOOPER, M. 2003. The Carboniferous evolution of the Central Coalfield Basin, Midland Valley of Scotland: implications for basin formation and the regional tectonic setting. Unpublished PhD thesis, University of Leicester.
  11. PENN, I E, SMITH, I F, and HOLLOWAY, S. 1984. Interpretation of a deep seismic reflection profile in the Glasgow area.
  12. SMITH, M, LAWRENCE, D, MCINROY, D, MONAGHAN, A, EDWARDS, M, NAYLOR, M, FORD, J, and BRICKER S. 2011. Chapter 3 Defining the Store; Geological Interpretation and storage modelling. in Smith, M. Campbell, D, Mackay, E, and Polson, D, (Eds). CO2 Aquifer Storage Site Evaluation and Monitoring (CASSEM). Heriot-Watt University, Edinburgh. ISBN 978-0-9571031-0-8
  13. MILODOWSKI, A E, and RUSHTON, J E. 2008. Mineralogical and porosity characterisation of potential aquifer and seal units for carbon capture and storage methodologies for the CASSEM Project. British Geological Survey Commissioned Report, CR/08/153. 86pp.
  14. MERRITT, J E, MONAGHAN, A A, LOUGHLIN, S C, MANSOUR, M, Ó DOCHARTAIGH, B É, and HUGHES A G. 2009. Clyde Gateway Pilot 3D Geological and Groundwater Model. British Geological Survey Commissioned Report, CR/09/005.
  15. MONAGHAN, A A, TERRINGTON, R L, and MERRITT, J E. 2012. Clyde Gateway Pilot 3D Geological Model, Version 2. British Geological Survey Commissioned Report, CR/12/010. 10pp.
  16. 16.0 16.1 16.2 16.3 KEARSEY, T. 2017. Model metadata report for Glasgow Geothermal Energy Research Field Site Bedrock Model. British Geological Survey Open Report, OR/17/044, in review.
  17. HINXMAN, L W, ANDERSON, E M, and CARRUTHERS, R G. 1920. The economic geology of the Central Coalfield of Scotland, Area IV, Paisley, Barrhead and Renfrew. Memoir of the Geological Survey, Scotland.
  18. CLOUGH, C T, HINXMAN, L W, WRIGHT, W B, ANDERSON, E M, and CARRUTHERS, R G. 1926. The economic geology of the Central Coalfield of Scotland, description of area V, Glasgow East, Coatbridge and Airdrie. Memoir of the Geological Survey of Scotland.
  19. BRITISH GEOLOGICAL SURVEY. 1995. Glasgow SE. Scotland Sheet NS56SE. Bedrock Geology. 1:10 000. (Keyworth, Nottingham: British Geological Survey.)
  20. BRITISH GEOLOGICAL SURVEY. 2008. Rutherglen. Scotland Sheet NS66SW. Bedrock Geology. 1:10 000. (Keyworth, Nottingham: British Geological Survey.)
  21. 21.0 21.1 BROWNE, M A E, DEAN, M T, HALL, I H S, MCADAM, A D, MONRO, S K, and CHISHOLM, J I. 1999. A lithostratigraphical framework for the Carboniferous rocks of the Midland Valley of Scotland. British Geological Survey Research Report RR/99/07.
  22. READ, W A. 1988a. Controls on Silesian sedimentation in the Midland Valley of Scotland. In: Besly, B M, and Kelling, G. (eds) Sedimentation in a synorogenic basin complex: the Upper Carboniferous of northwest Europe. Blackie and Son, Glasgow. 222–241.
  23. 23.0 23.1 23.2 READ, W A. 1988b. The interplay of sedimentation, volcanicity and tectonics in the Passage Group (Arnsbergian, E2 to Westphalian A) in the Midland Valley of Scotland. In: ARTHURTON, R S, GUTTERIDGE, P, and NOLAN, S C. (eds) The Role of Tectonics in Devonian and Carboniferous Sedimentation in the British Isles. Occasional Publications of the Yorkshire Geological Society, 6, 143–152.
  24. READ, W A. 1989. Sedimentological evidence for a major subsurface fracture system linking the eastern Campsie and the eastern Ochil faults. Scottish Journal of Geology, 25, 187–200.
  25. GOODLET, G A. 1957. Lithological variation in the Lower Limestone Group in the Midland Valley of Scotland. Bulletin of the Geological Survey of Great Britain, Vol. 12, 52–65.
  26. WILSON, R B. 1989. A study of the Dinantian marine macrofossils of central Scotland. Transactions of the Royal Society of Edinburgh, Earth Sciences, 80, 91–126.
  27. READ, W A. 1994. High frequency glacial-eustatic sequences in early Namurian coal-bearing fluviodeltaic deposits, central Scotland. Spec Publ. Int Ass Sediment., 19, 413–428.
  28. BENSE, V F, GLEESON, T, LOVELESS, S E, BOUR, O, and SCIBEK, J. 2013. Fault zone hydrogeology. Earth-Science Reviews, 127, 171–192.
  29. 29.0 29.1 MONAGHAN, A A, ARKLEY, S L B, WHITBREAD, K, and MCCORMAC, M. 2013. Clyde superficial deposits and bedrock models released to the ASK Network 2013: a guide for users Version 2. British Geological Survey Open Report, OR/13/025.35pp.
  30. 30.0 30.1 30.2 30.3 30.4 30.5 30.6 BEACH, A, WELBON, A L, BROCKBANK, P, and MCCALLUM, J E. 1999. Reservoir damage around faults: outcrop examples from the Suez rift. Petroleum Geoscience, 5, 109–116.
  31. 31.0 31.1 31.2 31.3 FOSSEN, H, and HESTHAMMER, J. 2000. Possible absence of small faults in the Gullfaks Field, northern North Sea: Implications for downscaling of faults in some porous sandstones, Journal of Structural Geology, 22(7), 851–863.
  32. 32.0 32.1 32.2 SAVAGE, H M, and BRODSKY, E E, 2011. Collateral damage: Evolution with displacement of fracture distribution and secondary fault strands in fault damage zones. Journal of Geophysical Research, 116, BO3405.
  33. 33.0 33.1 33.2 33.3 33.4 SHIPTON, Z K, and COWIE, P A. 2001. Damage zone and slip surface evolution over μm to km scales in high-porosity Navajo sandstone, Utah. Journal of Structural Geology, 23, 1825–1844.
  34. 34.0 34.1 34.2 34.3 SHIPTON, Z K, SODEN, A M, KIRKPATRICK, J D, BRIGHT, A M, and LUNN, R J. 2006. How thick is a fault? Fault‐displacement‐thickness scaling revisited, in Earthquakes: Radiated Energy and the Physics of Faulting, Geophysical Monograph Series, 170, edited by R E Abercrombie et al., 193–198 AGU, Washington, DC.
  35. KNOTT, S D, BEACH, A, BROCKBANK, P J, BROWN, J L, MCCALLUM, J E, and WELBON, A I. 1996 Spatial and mechanical controls on normal fault populations, Journal of Structural Geology, 18, 359–372, ISSN 0191-8141, https://dx.doi.org/10.1016/S0191- 8141(96)80056-3
  36. 36.0 36.1 ANTONELLINI, M, and AYDIN, A. 1995. Effect of faulting on fluid flow in porous sandstones: Geometry and spatial distribution. AAPG Bulletin 1995, 79:5, 642–671.
  37. 37.0 37.1 SCHOLZ, C H. 2002. The Mechanics of Earthquakes and Faulting, Cambridge University Press, Cambridge, UK.
  38. 38.0 38.1 CHOI, J H, EDWARDS, P K, YOUNGTAE, K, and YOUNG-SEOG, K. 2016. Definition and classification of fault damage zones: A review and a new methodological approach. Earth-Science Reviews, 152, 70–87.
  39. BERG, S S, and SKAR, T. 2005. Controls on damage zone asymmetry of a normal fault zone: Outcrop analyses of a segment of the Moab fault, SE Utah, Journal of Structural Geology, 27, 1803–1822.
  40. 40.0 40.1 BAPTIE, B. 2010. Seismogenesis and state of stress in the UK. Tectonophysics, 482, 1–4, 150-159.10.1016/j.tecto.2009.10.006.
  41. BAPTIE, B, SEGOU, M, ELLEN, R, and MONAGHAN, A. 2016. Unconventional Oil and Gas Development: Understanding and Monitoring Induced Seismic Activity. British Geological Survey Open Report, OR/16/042. 92pp.
  42. ZOBACK, M D, BARTON, C A, et al. 2003. Determination of stress orientation and magnitude in deep wells. International Journal of Rock Mechanics and Mining Sciences, 40, 1049–1076.
  43. REINECKER, J, TINGAY, M, and MÜLLER, B, 2003. Borehole breakout analysis from four-arm caliper logs, World Stress Map Project. WSM website: https://dcapp314.gfzpotsdam.de/pub/guidelines/WSM_analysis_guideline_breakout_caliper.pdf
  44. TINGAY, M, REINECKER, J, and MÜLLER, B. 2008. Borehole breakout and drilling-induced fracture analysis from image logs [online], in World Stress Map Project–Guidelines: Image Logs, Helmholtz Cent. Potsdam, GFZ German Research Centre for Geosciences, Potsdam, Germany. (Available at: https://dc-app3-14.gfz- potsdam.de/pub/guidelines/WSM_analysis_guideline_breakout_image.pdf).
  45. 45.0 45.1 KINGDON, A, FELLGETT, M W, and WILLIAMS, J D O. 2016. Use of borehole imaging to improve understanding of the in-situ stress orientation of Central and Northern England and its implications for unconventional hydrocarbon resources. Marine and Petroleum Geology, Vol. 73 May 2016, Pages 1–20. DOI: 10.1016/j.marpetgeo.2016.02.012
  46. 46.0 46.1 TINGAY, M, HILLIS, R R, MORLEY, C K, SWARBRICK, E C, and OKPERE, E C. 2003. Variation in vertical stress in the Baram Basin Brunei: tectonic and geomechanical implications. Marine and Petroleum Geology, 20, 1201–1212.
  47. WILLIAMS, J D O, FELLGETT, M W, KINGDON, A, and WILLIAMS, P J. 2015. In-situ stress orientations in the UK Southern North Sea: Regional trends, deviations and detachment of the post-Zechstein stress field. Marine and Petroleum Geology. DOI:10.1016/j.marpetgeo.2015.06.008
  48. WILLIAMS, J D O, FELLGETT, M W, and QUINN, M F. 2016. Carbon dioxide storage in the Captain Sandstone aquifer: determination of in situ stresses and fault-stability analysis. Petroleum Geoscience, Vol. 22, 211–222. DOI: 10.1144/petgeo2016-036
  49. VERWEIJ, J M, BOXEM, T A P, and NELSKAMP, S. 2016. 3D spatial variation in vertical stress in on- and offshore Netherlands; integration of density log measurements and basin modeling results. Marine and Petroleum Geology. DOI: 10.1016/j.marpetgeo.2016.06.016