Wrench Tectonics in The Carboniferous Coals of The Midland Valley of Scotland

Masters Thesis
Ben Craven BSc (Hons)
Wrench Tectonics in the Carboniferous Coals of The Midland Valley of Scotland

Implications for Coal Extraction in East Ayrshire
School of Earth and Environment, University of Leeds, Leeds, United Kingdom, LS2 9JT
August 14, 2013 Word Count: 10000
Abstract Resolution is often poor when studying the 3D nature of geological structures, as usually seismic is the only imaging method that can be used. This project aims to change this, by demonstrating high resolution (2-5 m) 3D coal seam data can be used to complement geological mapping and boreholing, producing a well-constrained structural model of key coal-bearing horizons and faults in East Ayrshire, Scotland. By creating a high-resolution model in Midland Valleys Move software, this method can determine locations of high coal extraction potential and gives a predictive tool to mining rms for fault trends and styles. The project also uses digital BGS data to determine an evolutionary model for the complex Midland Valley, where reversals in stresses have created multi-phase fault structures. These results complement recent studies in the area where The Midland Valley is determined to be a sinistrally reactivated graben in the Early Carboniferous, then dextrally oset in the Late Carboniferous. Folding in the coals also indicates far-eld N-S Variscan stresses may have penetrated as far as the Scottish Uplands. Applications for these results are numerous in other strike-slip settings, where often the 3D geometry is poorly conned. Additionally, as the coal-bearing rocks are sandstone/shale dominated, they can be used as an excellent analogue for conventional hydrocarbon reservoirs- particularly those with normal or strike-slip faults.
University of Leeds Department of Earth Sciences
Declaration of Academic Integrity
To be attached to any essay, Dissertation, or project work submitted as part of a University examination.
I have read the University regulations on Cheating and Plagiarism, and I state that piece of work is my own, and it does not contain any unacknowledged work from any other sources.
Name:
Signed:
Date:
Programme of Study:
Acknowledgements Acknowledgements and thanks must be extended to Kier Mining, ATH Resources, and British Coal who provided data for this project. Specic thanks go David Richardson and Giles Hemmings at Kier Mining who gave guidance and support, and had the unenviable job of the initial data validation during this project. Thanks also to Peter Rourke and Jenny Ellis at Midland Valley for their excellent training in the software, and their expertise during the project. Acknowledgements should also be extended to Taija Torvela, Douglas Paton, Toby Dalton and Geo Lloyd at the University of Leeds for their academic assistance during the project. Also thanks to Simon Libby for proof reading and help during the write-up.
Contents
List of Figures 1 Introduction 1.1 Introduction . . . . . . . . 1.2 UK Coal . . . . . . . . . . 1.3 Hydrocarbon Exploration 1.3.1 Conventionals . . . 1.3.2 Unconventionals . 1.4 Project Aims . . . . . . .
3 10 11 15 16 16 16 17 18 19 24 27 28 29 29 32 37 38 39 40 41 48 49 51 55 56 59 60 65 66 68 71
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
2 Geological Background 2.1 Research on The Midland Valley of Scotland . . . . 2.2 Ayrshire in the Carboniferous . . . . . . . . . . . . . 2.2.1 Stratigraphy and Tectonic History- Literature 2.2.2 Coal Seam Correlation in East Ayrshire . . . 2.3 Expected Structural Styles . . . . . . . . . . . . . . . 2.3.1 Normal Faulting . . . . . . . . . . . . . . . . 2.3.2 Strike-Slip Faulting . . . . . . . . . . . . . . . 2.3.3 Inter-Coal Faulting . . . . . . . . . . . . . . . 3 Methodology 3.1 Data Obtained for the Project . . . . . 3.2 Modelling Workow . . . . . . . . . . . 3.3 Modelling Workow- Large-Scale Data . 3.4 Modelling Workow- Small-Scale Data . 3.5 BGS Cross Section - New Cumnock 14E 3.6 Completing the 3D Model . . . . . . . . 3.7 Additional Modelling: Seam Depth . . . 3.8 Additional Modelling: Petrel . . . . . . 4 Results 4.1 Move Models . . . . . . . . . . . 4.2 Unfolding the BGS Cross Section 4.3 Dip Data . . . . . . . . . . . . . 4.4 Apparent Dip . . . . . . . . . . . 4.5 Fault Relationships . . . . . . . .
. . . . . . . . . . . . . . . . Compilation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
4.6 4.7 4.8 4.9 4.10 4.11 4.12 4.13
Fault Loss . . . . . . . . . . . . . . . . Fault Lengths: 5x5 km Move Model . Fault Lengths: 20x25 km Move Model Greenburn Coal Seams . . . . . . . . . Petrel Isochores . . . . . . . . . . . . . Seam Depth . . . . . . . . . . . . . . . Potential Coal Exploration Targets . . Correcting the BGS Map . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
73 76 76 78 81 83 86 88 89 90 91 93 94 96 100 103 104 106 111 111 113 115 117 118 119 119 121 121 123
5 Interpretation 5.1 Overview . . . . . . . . . . . 5.2 Folding . . . . . . . . . . . . 5.3 Faulting . . . . . . . . . . . . 5.3.1 Graben Interpretation 5.3.2 Flower Interpretation 5.4 Extraction . . . . . . . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
6 Discussion 6.1 Folding . . . . . . . . . . . . . . . . . . . 6.2 Faulting . . . . . . . . . . . . . . . . . . . 6.3 Underlying Fabric . . . . . . . . . . . . . 6.3.1 Stepover . . . . . . . . . . . . . . . 6.3.2 Boundary Faults . . . . . . . . . . 6.4 Inversion . . . . . . . . . . . . . . . . . . . 6.5 Implications for other Strike-Slip Settings 7 Conclusions & Synthesis 7.1 3D Data Modelling . . . . 7.2 Midland Valley Evolution 7.3 Faults and Coal Thickness 7.4 Coal Exploration Targets 7.5 Further Work . . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . . . . . Changes . . . . . . . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
References 124 7.6 Software Used . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129 A Petrel Edge Detection 130
B Full Word Count 134 B.1 Full Word Count . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135 C Digital Data (CD) 136
List of Figures
1.1 1.2 1.3 1.4 1.5 1.6
Map of The Midland Valley in relation to the British Isles, showing the two main bounding faults. Modied after Google Inc. (2013) and Smith (1995). . . . . . . . . . . . . . . . . . . . Maps showing location of study area. Area covers 20 km N-S and 25 km E-W. Maps compiled from Google Inc. (2013) and The British Geological Survey (1980, 2000). . . . . . . . . . . . Simplied version of the geological map. 25 km E-W, 20 km N-S. . . . . . . . . . . . . . . . . Map of UK showing the main Carboniferous coalelds. Modied after Cleal and Thomas (1996). A summary of potential shale gas sources across the whole UK. Modied after Selley (2012) . The red transparent colour highlights the erroneous sections of the BGS map identied by Kier. After (The British Geological Survey, 1980, 2000) . . . . . . . . . . . . . . . . . . . . . Model showing possible interactions between normal faults and transfer faults, and the structures that may form at the intersections. Modied after Gibbs (1990). . . . . . . . . . . . . . . . . Map of northern Britain showing the locations of LISBP lines ALPHA, BETA and GAMMA. Note the relatively shallow depth to basement within The Midland Valley. Modied after Bamford (1979). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Showing the general structure of the Iapetus suture, with specic reference to the faults that bound The Midland Valley- the Southern Upland fault and the Highland Boundary Fault. These may provide a similar underlying structure to that shown in Figure 2.19 above. Section drawn from WINCH and LISPB lines, plus other geologic and geophysical data (see original references). From (Needham and Knipe, 1986). . . . . . . . . . . . . . . . . . . . . . . . . . . A new model for stress regime that formed The Midland Valley, from interpreting seismic lines around the Firth of Forth. HBF: Highland Boundary Fault. SUF: Southern Uplands Fault. LM: Lammermuir Fault. Modied from From Ritchie et al. (2003). . . . . . . . . . . . . . . . A model for the formation of the Great Glen Fault. Modied from From Kennedy (1946). . . Location of seismic line is shown in Figure 2.5, note that it is not N-S across The Midland Valley but is orientated such that it crosses relevant stratigraphy and structures. Horizons are quoted thus: 1: sea bed; 2: rockhead; 3: ?intra-Upper Coal Measures; 4: intra-Middle Coal Measures; 5: intra-Lower Coal Measures; 6: top Lower Limestone Formation (top Dinantian); 7: intra-Strathclyde Group equivalent; 8: ?intra-Devonian. All horizons are shown in Figure 2.10 on page 27. From Ritchie et al. (2003), as the original data were not available for this project, the Figure has been left interpreted as presented in the original paper.
11 13 14 15 16 17
2.1 2.2
19
20
2.3
20
2.4
2.5 2.6
21 21
22
List of Figures 2.7 Structural evolution of Scotland, with focus on The Midland Valley (shown in pale green). Black arrows show plate movements, light arrows show the proposed movement of the Northern European Block (NEB). a: Dinantian: NE trending sinistral strike-slip. b: Silesian: dextral movement on NE-trending faults and sinistral on NW-trending faults. The strain ellipse is similar to that Ritchie et al. (2003) in Figure 2.5, but rotated anti-clockwise by 40. c,d,e,f: Cross sections through a and b. Modied after Caldwell and Young (2013), originally modied after Coward (1993) and Leeder (1982). . . . . . . . . . . . . . . . . . . . . . . . . . Map of the distribution and source for Limestone Coal Formation. The structural highs of the Highlands and the Uplands stayed the same during deposition of the Coal Measures. From Read et al. (2002). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Map of alluvial fan deposits found in The Midland Valley. From Cameron and Stephenson (1985). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Compilation chart relating standard geological ages to the map units. These are then related to dierent interpretations of the palaeogeography and tectonic environment. Compiled from The British Geological Survey (1980), Browne et al. (1999), Ritchie et al. (2003), Rohde (2005), Heckel and Clayton (2006), Waters et al. (2012), Walker et al. (2012), and Caldwell and Young (2013). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . An updated version of the chart originally made by John Crone from Crouch Mining Ltd in 2000 and used by Kier Mining. The BC Code refers to the British Coal codes that are supposed to relate the various naming schemes across England and Wales. A column displaying the BGS coal seam map names has been added, and the overall graphical style improved. The Taits Marine Band is the second Lingula band beneath the Pathhead Coal (Cameron and Stephenson, 1985), and is not recognised as widely as the Aegiranum Marine Band which stretches even as far as South Derbyshire (Waters and Davies, 2006). . . . . . . . Model showing large-scale deformation of the Earths crust through extension and normal faulting. Upbending of the crust results from displacement in the lower layer. From Bott (1976). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Simple models for extensional faulting showing no 3D along-strike variation. From Fossen (2010). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Evolution of syn-rift extensional depocentres. From Cowie, Gupta and Dawers (2000). . . . . Diagram of a normal fault cutting a strong-weak-strong layering of rock beds. The weak layer is deformed and incorporated into the fault plane. This allows some exure within the strong bed. From Fossen (2010). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Large 3D block model Modied from (Cunningham and Mann, 2007). . . . . . . . . . . . . . 4D sandbox model evolution of a dextral spindle-shaped strike-slip transtensional ower structure. Basal layer is 1.5 cm thick SGM 36 polymer with similar scaled properties to the brittle-ductile transition zone. Overlaying this is 7.5 cm of dry quartz sand. Model scales at roughly 1:100,000 (1 cm = 1 km). From Wu et al. (2009) . . . . . . . . . . . . . . . . . . . Line drawings of experimental sand box modelling. a) Underlapping stepover. b) Neutral stepover. c) Overlapping stepover. Numbers relate to master fault segments. From (Dooley and Schreurs, 2012). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Line drawings of dextral strike-slip sandbox models overlying a 90 dipping normal fault. [1]: A/B: purely brittle deformation. C/D: 5 cm of sand on 1 cm of silicone (acting as a d ecollement horizon). E/F: 4 cm of sand on 2 cm of silicone. [2]: A similar dextral system, however here the underlying fabric is unactivated. Compiled from Richard and Krantz (1991).
23
2.8
24 25
2.9 2.10
27
2.11
28
2.12
29 29 30
2.13 2.14 2.15
2.16 2.17
31 32
33
2.18
34
2.19
35
List of Figures 2.20 The eects of simple-shear on a strain ellipse. Demonstrates that normal faults can form oblique to a dextral strike-slip zones, and if enough deformation occurs, can be made to invert. From (Waldron, 2005), originally after Harding (1974). . . . . . . . . . . . . . . . . . 2.21 Diagram showing the structures that are associated with a pull-apart basin, such as in Figure 2.16. From (Rodgers, 1980). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.22 Diagram showing some of the features that can be expected in a sheared coal seam. Illustration of features in the Wu seam of the Pingdingshan coaleld, northern China. Whilst this coaleld does penetrate mineable Carboniferous coal, the Wu seam is Permian in age. Crosses indicate primary axes of strain ellipses. Colour added to image from (Li, 2001). . . . . . . . . . . . . . 3.1 3.2 Workow summary. Covers both small-scale (upper path) and large-scale (lower path) data sources. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A Coloured DEM of area. B Draped extent of the scanned BGS solid maps. Note the arrow in the bottom-right indicating North with the red point. (The British Geological Survey, 1980, 2000) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Limits of map area within ArcMap 10, projected in the British National Grid co-ordinate system. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Depending on which way the fault is dipping, a specic colour is assigned. This produces synthetic pairs of NE-SW trending and NW-SE faults, similar to the research by Caldwell and Young (2013) and Hooper (2013). It is also very similar to a method used by Wilson et al. (2006) when they looked at strike-slip systems on the Norwegian Margin. . . . . . . . . BGS-mapped faults are coloured based on their dip direction. Draped on maps from The British Geological Survey (1980, 2000) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A 3D oblique view of coloured faults. NB: glow eect is added post-process in Photoshop. . . 3D Extrapolation of all faults. Dierent colours relate to dip direction. Red and Blue (major) faults dip at 80 to a depth of 800 m. Green and Yellow (minor) faults dip at an angle of 70 to a depth of 600 m. Also shown are where faults are clipping through each other, and where they have been manually cut to join. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Dips displayed on the map. Maps from The British Geological Survey (1980, 2000). . . . . . Location map of coal seam data from ATH Resources (Glenmuckloch, Netherton, Skares); Kier Mining (Braehead, Greenburn, Grieve Hill, Rough Hill) and British Coal (Lanehead). BGS map is shown for scale. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Full map and interpreted cross section scanned from the BGS map of New Cumnock, 14E. Note 1.5x vertical exaggeration on section. After The British Geological Survey (1980). . . . Traced section, with bases of UCMS, MCMS, LCMS and LSC shown. Vertical exaggeration has been removed. A: Thickness variation between the UCMS and the LCMS. B: No sense of reactivation here where C has clear reactivation. D: Large (200) normal oset on this fault. Maps from The British Geological Survey (1980, 2000) with the location of the 5x5 km study area shown bound by vertical surfaces. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Modied map showing only the key units from the stratigraphic compilation in Figure 2.10. Sections 01 and 02 are shown crossing through the units of interest (the Coal Measures). Modied from The British Geological Survey (1980, 2000). . . . . . . . . . . . . . . . . . . . 3D Model Bounds. 5000x5000x1100 m. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3D model of faults in the 5x5 km study area. . . . . . . . . . . . . . . . . . . . . . . . . . . . 3D model showing the modelling of the volcanic intrusions that penetrate the 5x5 km study area. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
35 36
37
40
41 42
3.3 3.4
42 43 44
3.5 3.6 3.7
3.8 3.9
45 47
48 49
3.10 3.11
50 51
3.12 3.13
3.14 3.15 3.16
52 53 53 54
List of Figures 3.17 3D model with topography shown in green, and a red surface 200 m below this to reect the maximum depth to which a surface mine can typically reach. . . . . . . . . . . . . . . . . . . 3.18 Oblique view from the South of the Greenburn coal seams 6500-6600-6800. Note that due to software bugs the seams are not shown in the right vertical order. Map cropped from The British Geological Survey (1980, 2000). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.19 The boundary polygon ts the most amount of data across all 3 seams. Data holes are due to outcrop. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.20 RDRs edge-detection algorithm applied to a 2 m spaced convergent interpolation surface on the 6800 seam. Hotter colours indicate edges- e.g. curvier surfaces with steeper dips. Same scale as previous Figure. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1 4.2 4.3 4.4 4.5 4.6 4.7 3D model showing the modelling of the Passage Group within the 5x5 km study area. . . . . 3D model showing the modelling of the Passage Group within the 5x5 km study area, coloured by depth. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3D model showing the modelling of the Lower Coal Measures (LCMS) within the 5x5 km study area. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3D model showing the modelling of the Lower Coal Measures (LCMS) within the 5x5 km study area, coloured by depth. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3D model showing the modelling of the Middle Coal Measures (MCMS) within the 5x5 km study area. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3D model showing the modelling of the Middle Coal Measures (MCMS) within the 5x5 km study area, coloured by depth. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3D model showing the modelling of the Upper Coal Measures (MCMS) within the 5x5 km study area. As all exposure UCMS is lost in the North-West of the area, it has not been modelled there. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3D model showing the modelling of the Upper Coal Measures (MCMS) within the 5x5 km study area, coloured by depth. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Unfold whole of LCMS to 0 m datum. Beds keep their parallel shape, however the geometry around the faults becomes unreasonable. This technique has actually, by restoring all line segments, produced a move-on-fault restoration. . . . . . . . . . . . . . . . . . . . . . . . . . Unfold only fold-hinge LCMS to 0 m datum. By only restoring the fold hinges, the fault blocks maintain their osets but are moved to a more horizontal position. The technique breaks in the North (left) where there is no further hinge to pull the beds down. Note that the faults do not warp considerably. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Stereographical analysis of all dip data from map. From Move 2013. . . . . . . . . . . . . . . Average dip for each unit on the 25x20 km map, plotted against maximum age. Error bars reach the maximum and minimum dips found on the map. . . . . . . . . . . . . . . . . . . . . Diagram showing the eect of dip on recorded seam thickness. A vertical borehole cuts through a coal seam (grey) and the thickness is recorded (a). The true thickness is actually the length at line b. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Graph showing what eect various dip angles will have on the recorded seam thickness down a vertical borehole. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3D model showing a dip attribute map for the prospective sub-surface coal seams, pink ellipses highlight steeply dipping areas. Note that the Middle Coal Measures surface has had the area that outcrop above topography removed. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
55
56 57
58 61 61 62 62 63 63
64 64
4.8 4.9
65
4.10
4.11 4.12 4.13
65 66 67
68 69
4.14 4.15
70
List of Figures 4.16 Plan view of fault model in Move. Note that NW-SE dipping (Blue & Red) faults appear to partition o the NE-SW dipping (Yellow & Green) faults. Also note that these fault sets do not abut perpendicular, but rather at a similar acute angle across the area. . . . . . . . . . . 4.17 By assigning a value to each node for the fault angle between the PDZ faults and the CBFs, a surface could be created to highlight areas of more acute or right-angled joins. There appears, however, to be no pattern. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.18 Schematic diagram demonstrating the eect of normal faulting on coal exploration. . . . . . . 4.19 Schematic diagram demonstrating the eect of normal faulting on coal exploration, with a more realistic model invoking fault drag. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.20 Fault loss map for Middle Coal Measures in 5x5 km area. . . . . . . . . . . . . . . . . . . . . 4.21 Fault loss map for Lower Coal Measures in 5x5 km area. . . . . . . . . . . . . . . . . . . . . . 4.22 Graph showing normal fault lengths plotted by their dip-direction, against the base-10 logarithm of their overall length. The average (mean) length is also shown. note how the measures lengths have poor data resolution due to fractal scaling of the faults- the largest and smallest simply cannot be imaged as well as the medium-sized ones. . . . . . . . . . . . . . . . . . . 4.23 Feature A in Figure 4.24. Photo taken 2013-05-28 at the Greenburn surface mine owned by Kier Mining Ltd. Fault drag is visible in the sandstone beds. Also note that the kinks in the fault are due to horizontal benches, rather than actual deviations in the plane. . . . . . . . . 4.24 Photo provided by David Richardson from Kier Mining Ltd. Seams from Figure 2.11 on page 28 have been added. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.25 Petrel model of Greenburn seams, created using the edge detection algorithm outputs such as Figure 3.20. Trace beds of 1 m thick coal are shown on the 6500 and 6600 seam, with the 6800 seam at the top. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.26 Cross-section through the Petrel model in Figure 4.25. Some fault reactivation is visible in the South. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.27 Isochores from Petrel, thickness between 6800 and 6600 seams. Hot = thick, cold = thin. . . 4.28 Isochores from Petrel, thickness between 6600 and 6500 seams. Note dierent scale from previous gure. Hot = thick, cold = thin. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.29 3D model with topograph shown in green, and a red surface 200 m below this to reect the maximum depth to which a surface mine can typically reach. . . . . . . . . . . . . . . . . . . 4.30 Plan view of 3D Model, showing the depths to the bases of the prospective coal measures that are below the surface. A lot of MCMS has been eroded, hence it is only really present in the South-East. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.31 Plan view of 3D Model with red maximum extraction depth slice present from Figure 3.17. Note that the volcanic intrusions are still shown for reference. . . . . . . . . . . . . . . . . . . 4.32 Depth-to-base for the Prospective Coal Measures. Colour map is only applied between 0 and 200 m, as that is the maximum extent of potential coal extraction. . . . . . . . . . . . . . . . 4.33 Isochore thickness mapping vertically between the topography and the base of the prospective coal measures. Note that the syncline has caused the underlying to become buried too deep to extract. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.34 New version of the BGS map created using the 3D Move model. Modied from The British Geological Survey (1980, 2000). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.1 Map from Figure 1.3 with the stereonet of all dips from Figure 4.11 overlain. 25 km E-W and 20 km N-S. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
71
72 73 73 74 75
77
78 79
80 80 81 82 83
84 85 86
87 88
90
List of Figures 5.2 5.3 5.4 5.5 5.6 5.7 5.8 5.9 5.10 5.11 5.12 5.13 5.14
Unit MCMS coloured by depth, with the colour map extending to cover only the extent of the unit, rather than the whole Z range like in the results. . . . . . . . . . . . . . . . . . . . . 91 Average dip of each unit (with maximum/minimum dip error bars) plotted by ages from Figure 2.10. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92 Normal faults are partitioned between strike-slip faults. . . . . . . . . . . . . . . . . . . . . . 93 A potential interpretation of a graben structure for the study area. Drawn to 6 km depth, as suggested by the LISPB survey in Figure 2.2. . . . . . . . . . . . . . . . . . . . . . . . . . . . 94 The red and blue faults from Figure 4.16 on page 71. Extension is NW-SE. Note orientation of North arrow in bottom right. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95 The green and yellow faults from Figure 4.16 on page 71. Extension is NE-SW. Note orientation of North arrow in bottom right. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95 A potential interpretation of a ower structure for the study area, using the same cross-section line as Figure 5.5. Again, base is determined by LISPB from Figure 2.2. . . . . . . . . . . . . 96 The location of the Greenburn seams from Figure 3.18 on page 56 within the 5x5 km study area. Note that the seam is not mined across the major blue/red faults. . . . . . . . . . . . . 97 Large fault damage zone in the normal faults seen in the Greenburn mine. Photo taken 2013-05-29 by Ben Craven at location Y on Figure 4.9. . . . . . . . . . . . . . . . . . . . . . 98 Schematic diagram demonstrating the eect of normal faulting on coal exploration, with a more realistic model invoking fault drag. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99 3D model with topograph shown in green, and a red surface 200 m below this to reect the maximum depth to which a surface mine can typically reach. . . . . . . . . . . . . . . . . . . 100 Areas of thick productive coal measures. Maps modied from The British Geological Survey (1980, 2000). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101 Green ellipses highlight areas that overlap from Figure 5.13. These will have the least risk during mining. Map modied from The British Geological Survey (1980, 2000). . . . . . . . . 102 Map of The Midland Valley showing major folds and faults. Study area(s) shown in red; trace of Leven-Midlothian (or Midlothian-Leven) Syncline highlighted in green. Modied after Rippon, Read and Park (1996). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104 Folding in an inverted strike-slip regime. From Allen, Alsop and Zhemchuzhnikov (2001). . . 105 Plan view of the 20x25 km area modelled in the project. North is the top-right. . . . . . . . . 106 Two styles of en- echelon faulting, compressive and extensive. Angle x is 40-45. Modied after Ramsay and Huber (1987). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107 Strain ellipse with labelled stress tensors. Modied after Sylvester (1988) with orientations and labels from Hooper (2013). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107 Modied after Wilcox, Harding and Seely (1981). Sub-gures are as follows: B: Dasht-e Ba yaz (Iran). C: Lake Basin fault zone, Montana (USA). D: Cottage Grove, Illinois (USA). . . . . 108 Figure 6.5 overlying the 3D fault model. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109 Model showing possible interactions between normal faults and transfer faults, and the structures that may form at the intersections. Modied after Gibbs (1990). . . . . . . . . . . . . . . . . 110 Line drawings of experimental sand box modelling. a) Underlapping stepover. b) Neutral stepover. c) Overlapping stepover. Modelled with a 10 cm thick layer ofbrittle overburden of silica sand. Numbers relate to master fault segments. From (Dooley and Schreurs, 2012).111
6.1
6.2 6.3 6.4 6.5 6.6 6.7 6.8 6.9
List of Figures
6.10 Comparison between a sandbox strike-slip fault system and the faults modelled in Move seen in Figure 4.16 on page 71. 3 boxes are highlighted on the model of The Midland valley, these are then expanded with fault throw indicators added, then related to the Figure inset from (Dooley and Schreurs, 2012). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112 6.11 Figure 6.11 on page 113 repeated for convenience: line drawings of dextral strike-slip sandbox models overlying a 90 dipping normal fault. [1]: A/B: purely brittle deformation. C/D: 5 cm of sand on 1 cm of silicone (acting as a d ecollement horizon). E/F: 4 cm of sand on 2 cm of silicone. [2]: A similar dextral system, however here the underlying fabric is unactivated. Compiled from Richard and Krantz (1991). . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113 6.12 Schematic diagram showing the northerly subduction of the Variscan microplate, causing N-S back-arc subduction of The Midland Valley. The back-arc extension may have created the normal fault fabric that the strike-slip PDZ faults then follow during strike-slip motion. Modied after (Caldwell and Young, 2013). . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114 6.13 A re-drawn Figure 6.4, to reect the geometry and history of The Midland Valley. Modied after (Ramsay and Huber, 1987). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115 6.14 Schematic diagram showing the dextral strike-slip motion of the PDZ during the Late Carboniferous. This inverts the main through-going strike-slip faults. Modied after (Caldwell and Young, 2013). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116 6.15 3D schematic model of the Aryskum Graben in the South Turgay Basin. Area is roughly 30x20 km with depth to roughly 5 km (no original diagram scale). From (Yin et al., 2012). . 117 7.1 Full evolutionary model, determined from interpretations during the project and literature reviews. Evidence is present in the East Ayrshire area of Scotland for all the labels present. 01: In the Late Devonian the collision and subduction of the Variscan plate due to the slow ongoing collision between the Gondwana and Laurasia continental plates (Soper et al., 1992). NE-trending normal faults form as a result of back-arc extension. 02: During the Early Carboniferous, the wrenching motion of the closing Proto-Tethys Ocean and the continental block of Laurasia set up a sinistral strike-slip zone across The Midland Valley. This activated along the fabric of the normal faults. 03: Late Carboniferous motions of the Northern European Block (Caldwell and Young, 2013) caused reactivation of the sinistral PDZ into a dextral one. This also meant that the geometry of overlapping stepovers was created by the previous CBFs. 04: Continued motion of the Variscan plate causes contraction of The Midland Valley in a NW-SE motion, tightening folds that have been created during the Carboniferous. 120 New BGS map with locations of potential mining sites shown. Map modied from The British Geological Survey (1980, 2000). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121 New BGS map with locations of current and past coal minings. Taken from Kier Mining seam locations and The British Geological Survey (2013). Map modied from The British Geological Survey (1980, 2000). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122
7.2 7.3
A.1 RDRs edge-detection algorithm applied to a 2 m spaced convergent interpolation surface on the 6800 seam. Hotter colours indicate edges- e.g. curvier surfaces with steeper dips. . . . . . 131 A.2 RDRs edge-detection algorithm applied to a 2 m spaced convergent interpolation surface on the 6600 seam. Hotter colours indicate edges- e.g. curvier surfaces with steeper dips. . . . . . 132 A.3 RDRs edge-detection algorithm applied to a 2 m spaced convergent interpolation surface on the 6500 seam. Hotter colours indicate edges- e.g. curvier surfaces with steeper dips. . . . . . 133
Introduction
10
1. INTRODUCTION
11
1.1
Introduction
Last year, over forty percent of energy generation in the UK was provided by coal-burning power stations (Gosden, 2013). However this rising proportion of coal-based energy production is occurring against a background of increasing external coal imports, where 80% of steam-coal hails from abroad (CoalImp, 2013). This is despite there being, in March 2012, an estimated 728 million tonnes of coal remaining in known UK reserves (The Coal Authority, 2012). This contribution from outside is only compounding the profound decline of the UK coal industry, causing both industry commentators and insiders to refer to the industry as dying (Carrington, 2011). Between 2004 and 2012, a third of all the UKs saleable coal was extracted from the East Ayrshire district of Scotland (East Ayrshire Council, 2012). This area lies on the south west ank of The Midland Valley, a NE-SW trending structure that crosses the width of the country; delineated by the Highland Boundary Fault on the northern edge, and the Southern Uplands Fault on the other; shown in Figure 1.1:
Figure 1.1: Map of The Midland Valley in relation to the British Isles, showing the two main bounding faults. Modied after Google Inc. (2013) and Smith (1995).
Across the whole of the UK, coal mining rms are struggling against imports of cleaner, cheaper coal. Ayrshire is being particularly badly hit by job losses, however this was compounded in May 2013 when two major employers in the area went out of business: rst Scottish Coal went into liquidation (BBC, 2013), quickly followed by ATH Resources (ATH Resources PLC, 2013). This has meant that Kier Mining is the only coal extraction company left (Coal Action Scotland, 2012).
1. INTRODUCTION
12
Although in the past deep mining allowed seams as deep as 800 m to be worked, since the closure of the Barony mine in 1988 (Smith et al., 2008) surface mining is the only coal extraction technique used in Scotland (Fraser, 2013). This has greatly reduced the depth from which coal is extracted, and currently Kier mines down to no more than 200 m. Whilst there are huge surafce mines abroad, this is not unusual for the UK, where the deepest recorded surface coal mine was the Westeld Opencast Coal Mine in North-East Fife, now closed. It was just over 200 m deep (McIntosh, 2013). Many issues aect the depth to which a surface mine can operate. Firstly for receiving planning permissions; project size, overburden volume, and the local impacts need to be considered. Geotechnical issues such as slope stability and water management also need to be factored in. Protability of the mine operation is also hard-linked to ratio, literally the ratio between how much coal is being extracted versus how much non-coal (overburden and interburden) (Richardson, 2013). The vast majority of coal mined in the United Kingdom is of Carboniferous age (Waters and Davies, 2006), specically from the Westphalian Stage (315-306.5 Ma (Rohde, 2005)). These Carboniferous coals were formed in swamps on deltas that were lling the area during the uplift of the Caledonian Mountains (Cameron and Stephenson, 1985). They are common, and easily accessible. Indeed, records of coal extraction in The Midland Valley can be found from the twelfth century (Raymond, 1991). This project focuses specically on the area around mines already being operated by Kier Mining, west of New Cumnock (Figure 1.2:B). An area of interest has been highlighted by Kier for investigation on the British Geological Survey (BGS) maps of the area, south-west of a syncline, shown in Figure 1.2:C. A simplied version of this geological map is shown in Figure 1.3 where all the various minor units have been removed and only those specied in Figure 1.2 are coloured.
1. INTRODUCTION 13
Figure 1.2: Maps showing location of study area. Area covers 20 km N-S and 25 km E-W. Maps compiled from Google Inc. (2013) and The British Geological Survey (1980, 2000).
1. INTRODUCTION
Figure 1.3: Simplied version of the geological map. 25 km E-W, 20 km N-S.
14
1. INTRODUCTION
15
1.2
UK Coal
There are 5 major identied coal elds in Great Britain, described by Cleal and Thomas (1996) below and in shown Figure 1.4. Lithological correlation can be made between them, as shown by Waters and Davies (2006), however the formational structural style is dierent in The Midland Valley. Variscan deformation may be comparable across the elds (as shown by (Coreld and Gawthorpe, 1996)), however it will be much less pronounced in the North. 1. Culm Trough: Basin inversion and tectonic deformation earlier than the rest of Britain. Probably middle Westphalian. 2. Kent Coaleld: Only known through boreholes and deep mining. No exposure. 3. South-Wales: Progressive inll of shallow marine deltaics. Variscan front basins. 4. Central England: Northerly-derived deltaic deposits lling shallow-marine basin. Variscan deformation. 5. Midland Valley: Coal-bearing deposits in the lower Westphalian, with red beds in the upper formations. Transtensional strike-slip.
Figure 1.4: Map of UK showing the main Carboniferous coalelds. Modied after Cleal and Thomas (1996).
1. INTRODUCTION
16
1.3
1.3.1
Hydrocarbon Exploration
Conventionals
Although the eastern Midland Valley has a proven hydrocarbon system (Maddox, Blow and OBrien, 1997; Smith et al., 2008; Selley, 2012), Ayrshire in the West contains no hydrocarbon exploration wells: the volume of volcanic intrusion present in the area has replaced the main prospective horizon that is prospective further East- the oil-shales within the Strathclyde facies (Smith et al., 2008). Additionally, any hydrocarbons that may have been produced from the Carboniferous sediments have likely been mostly lost during the erosion of the Permian sediments.
1.3.2
Unconventionals
A report from 2008 stated no [unconventional hydrocarbons] exploration is currently planned for the Ayrshire region (Smith et al., 2008). Although Figure 1.5 demonstrates that the lithologies present in The Midland Valley would ordinarily be an excellent source rock, it is the same activity associated with the Permian Volcanics (Shelton, 1995) that has burnt o volatiles within the Coal Measures, making prospectivity low. Also, as much of the Upper Coal Measures have been converted to limestone, they too are almost wholly non-prospective (Smith et al., 2008).
Figure 1.5: A summary of potential shale gas sources across the whole UK. Modied after Selley (2012)
1. INTRODUCTION
17
1.4
Project Aims
This project will focus on the specic problems listed below:

Display Kiers coal seam data in 3D Produce a structural evolutionary model for the faulting around the mines and relate this to the evolution of The Midland Valley Correct the BGS map in the area around Kiers current mining operations (Figure 1.6, below) Relate the faults to coal seam thickness changes Identify any new surface mining targets based on these new interpretations
Figure 1.6: The red transparent colour highlights the erroneous sections of the BGS map identied by Kier. After (The British Geological Survey, 1980, 2000)
Geological Background
18
2. GEOLOGICAL BACKGROUND
19
2.1
Research on The Midland Valley of Scotland
The Midland valley was recognised as a specic post-Caledonian structural terrane by Kennedy (1958) and George (1960), both invoking what would become known plate tectonics (Bluck, 2002). Georges model for the structural evolution of The Midland Valley of Scotland was described in a recent paper by Underhill, Monaghan and Browne (2008) as a simple rift valley formed by NW-SE extension. The basin boundaries are the Highland Boundary Fault in the North, and the Southern Uplands Fault in the South. This extensional model was expanded upon by Leeder (1982) when he proposed that it was Iapetus-closure N-S back-arc extension that had created the rift. This was corresponding to the styles of deformation he had seen in the other Carboniferous basins across the UK (e.g. Bowland, Northumberland and Solway) (Underhill, Monaghan and Browne, 2008). However, Haszeldine (1988) and Stedman (1998) didnt agree with Leeders interpretation of N-S trending structures found across the area, instead both proposing models of E-W extension. This was backed with evidence from Gibbs (1984, 1990) who modelled the interaction between normal faults in basin-forming settings and found a variety of structures that were also present in The Midland Valley (Figure 2.1).
Figure 2.1: Model showing possible interactions between normal faults and transfer faults, and the structures that may form at the intersections. Modied after Gibbs (1990).
These newly-named transfer faults were also seen by Read (1988), and later by Rippon, Read and Park (1996) in their studies of The Midland Valley. The resultant models for basin evolution were complex, invoking both dextral and sinistral slip vectors produced in the overall rifting/extension to explain the geometries seen. This meant that by the late 90s the upper-crustal structure of The Midland Valley terrane was becoming clearer. Linking this to the deep crust had already been attempted, but progress was slow.
The overall structure of The Midland Valley was unknown for a long time due to the lack of seismic information available in the area. The best available, and still one of the few deep-crust surveys, was the 1974 Lithospheric Seismic Prole in Britain (LISPB), rst analysed by Bamford (1979), and shown in Figure 2.2 on the next page. The survey determined that the crust around the Caledonian Orogeny exhibits clear... lateral variations in the deep structure, and that there was a certain asymmetry across the orogenic belt (Bamford, 1979).
20
Figure 2.2: Map of northern Britain showing the locations of LISBP lines ALPHA, BETA and GAMMA. Note the relatively shallow depth to basement within The Midland Valley. Modied after Bamford (1979).
By combining this LISBP survey with another line (WINCH, interpreted from (Brewer et al., 1983)), Needham and Knipe (1986) developed a crustal overview for The Midland Valley, shown in Figure 2.3:
Figure 2.3: Showing the general structure of the Iapetus suture, with specic reference to the faults that bound The Midland Valley- the Southern Upland fault and the Highland Boundary Fault. These may provide a similar underlying structure to that shown in Figure 2.19 above. Section drawn from WINCH and LISPB lines, plus other geologic and geophysical data (see original references). From (Needham and Knipe, 1986).
There was still reason to further investigate the deep crustal features of the area, so the Midland Valley Investigation by Seismology (MAVIS) survey was produced in the mid 80s and analysed by Dentith and Hall in 1989. The authors proposed that instead of the 3-layer LISPB crustal model shown in Figure 2.2, a 4-layer model was more appropriate: 1 Carboniferous and Upper Old Red Sandstone, 2 Lower Old Red Sandstone and Lower Palaeozoic, 3 crystalline basement and 4 higher velocity crystalline upper crust Dentith and Hall (1989). The paper concluded that because the crystalline lower basement is virtually planar the surface structures result from thin-skinned tectonic packages that detach either internally, or upon the Lower Palaeozoic sequence.
21
The purely extensional interpretation began to lose support in 2003 with the publication of a new set of commercially available seismic lines across the Firth of Forth interpreted by Ritchie et al.. An example interpreted line is shown in Figure 2.6. The group found evidence of oblique-slip and strike-slip faulting, leading them to propose an oblique compressive stress eld; shown in Figure 2.5. The displacement on the Southern Uplands fault was found to be no more than a few hundred kilometres, possibly much less by Owen and Clarkson (1992). These strike-slip faults may be reactivating an underlying suture (Coward, 1993). Also seen were large open synclines, such as the Leven Syncline that were interpreted as growth structures (Ritchie et al., 2003).
Figure 2.4: A new model for stress regime that formed The Midland Valley, from interpreting seismic lines around the Firth of Forth. HBF: Highland Boundary Fault. SUF: Southern Uplands Fault. LM: Lammermuir Fault. Modied from From Ritchie et al. (2003).
This wrench basin model was not foreign to the Scottish Highlands, however, as one of the rst papers to recognise strike-slip motion from outcrop was published by William Kennedy in 1946 regarding the Variscan formation of the Great Glen Fault, less than 100 miles to the North:
Figure 2.5: A model for the formation of the Great Glen Fault. Modied from From Kennedy (1946).
2. GEOLOGICAL BACKGROUND 22
Figure 2.6: Location of seismic line is shown in Figure 2.5, note that it is not N-S across The Midland Valley but is orientated such that it crosses relevant stratigraphy and structures. Horizons are quoted thus: 1: sea bed; 2: rockhead; 3: ?intra-Upper Coal Measures; 4: intra-Middle Coal Measures; 5: intra-Lower Coal Measures; 6: top Lower Limestone Formation (top Dinantian); 7: intra-Strathclyde Group equivalent; 8: ?intra-Devonian. All horizons are shown in Figure 2.10 on page 27. From Ritchie et al. (2003), as the original data were not available for this project, the Figure has been left interpreted as presented in the original paper.
23
The most recent paper to look at The Midland Valley is by Caldwell and Young (2013). After detailed mapping in the Firth of Clyde, and compiling previous studies, they propose that the main sets of faults in the area that can be linked by invoking two major tectonic events in the Carboniferous: back-arc subsidence during proto-Tethys closure, then compression during the closure of the Ural Ocean. Their model is shown in Figure 2.7 below:
Figure 2.7: Structural evolution of Scotland, with focus on The Midland Valley (shown in pale green). Black arrows show plate movements, light arrows show the proposed movement of the Northern European Block (NEB). a: Dinantian: NE trending sinistral strike-slip. b: Silesian: dextral movement on NE-trending faults and sinistral on NW-trending faults. The strain ellipse is similar to that Ritchie et al. (2003) in Figure 2.5, but rotated anti-clockwise by 40. c,d,e,f: Cross sections through a and b. Modied after Caldwell and Young (2013), originally modied after Coward (1993) and Leeder (1982).
The study thus explains the complexity of The Midland Valley with only two tectonic events. It may be that the area appears more complicated due to the various overprinting stress regimes that have aected the British Isles, resulting in a complex poly-phase tectonic history of extension, transtension, transpression and inversion (Macdonald and Fettes, 2007; Caldwell and Young, 2013).
24
2.2
Ayrshire in the Carboniferous
Carboniferous rocks occupy much of The Midland Valley of Scotland, but are commonly obscured at the surface by Quaternary deposits (Waters et al., 2012). During the Carboniferous, The Midland Valley lay near the equator in the closing Iapetus Ocean between the Gondwanan and Laurasian continental plates (Soper et al., 1992); a wet equatorial climate, with swamps and deltas dominating the geography- ideal for the formation of coal (Browne et al., 1999). The thickest and most complete coal deposits in Scotland are exposed in The Midland Valley, known as The Coal Measures. These are units LCMS to UCMS in Figure 2.10. Cyclic uviodeltaics make up the Lower Coal Measures, deposited on top of the braided river systems of the latest Passage Formation (PGP) members (Read et al., 2002).
Figure 2.8: Map of the distribution and source for Limestone Coal Formation. The structural highs of the Highlands and the Uplands stayed the same during deposition of the Coal Measures. From Read et al. (2002).
25
Many individual coals can be traced from Glasgow eastwards toward Falkirk and westwards to Ayrshire (Forsyth and Brand, 1986), possibly suggesting an allocyclic mechanism-perhaps glacioeustatic (Read et al., 2002). There is also evidence of alluvial fan deposits at the basin margins (e.g. the Southern Uplands Fault), indicating the whole Midland Valley trough may have been non-terrestrial at the same time (Figure 2.9). It is inferred that accommodation roughly kept pace with sedimentation (Hooper, 2013), the sediment itself being sourced from the surrounding highs of the Southern Uplands High and the and the Highland High:
Figure 2.9: Map of alluvial fan deposits found in The Midland Valley. From Cameron and Stephenson (1985).
This made it easy for marine incursions to spread across a wide area, such as the Aegiranum Marine Band (unit AGMB) (Figure 2.10). One interpretation for the drive on sediment distribution in the Ayrshire coalelds is that it was mainly controlled by periodically-moving NE-trending faults due to regional subsidence (Hooper, 2013). Other authors suggest strike-slip or oblique-slip on faults controlled by larger scale en- echelon stresses due to microplate movements, as seen in Figure 2.7 (Caldwell and Young, 2013). The major deposition is certainly within a graben-shaped area, seen in Figures 2.3 and 2.8. Whether through subsidence, or due to faulting, the sedimentation is punctuated, explaining the presence of Yoredale cyclothems within the Mid-Late Carboniferous (Browne et al., 1999).
26
The folding that aects the area is determined to have been caused by the Variscan orogeny, after the deposition of the Coal Measures (Coreld and Gawthorpe, 1996; Hooper, 2013). The SW-NE trending synclines in the Ayrshire coal elds are a later feature as they show no hinge thickening (Hooper, 2013). This is dierent to the eastern Midland Valley where compressional NNE-SSW folds such as the Midlothian Leven Syncline have been shown to be active during the deposition of the passage Group (Underhill, Monaghan and Browne, 2008). As the majority of the folding follows the overall trend of The Midland Valley, it may be due to the Variscan orogeny re-activating the underlying suture that allowed the Valley to form in the rst place. This movement also likely reactivated some of the faults, but this will be dicult to distinguish from any inversion that occurred during formational transtension of the area (the idea that compression can occur in transtension was rst explored by De Paola et al. in 2005 and was applied to the Northumberland basin, less than 150 km to the East).
2.2.1
Stratigraphy and Tectonic History- Literature Compilation
Figure 2.10 below shows a chart compiled from various sources to compare dierent interpretations and histories of The Midland Valley:
Figure 2.10: Compilation chart relating standard geological ages to the map units. These are then related to dierent interpretations of the palaeogeography and tectonic environment. Compiled from The British Geological Survey (1980), Browne et al. (1999), Ritchie et al. (2003), Rohde (2005), Heckel and Clayton (2006), Waters et al. (2012), Walker et al. (2012), and Caldwell and Young (2013).
27
2.2.2
Coal Seam Correlation in East Ayrshire
Correlation of coal seams between the various coalelds across Easy Ayrshire, the Midland valley and even Scotland and England has been attempted in the past. A compilation of the East Ayrshire coal seams is shown below in Figure 2.11, produced for Kier Mining (Crone, 2000).
Figure 2.11: An updated version of the chart originally made by John Crone from Crouch Mining Ltd in 2000 and used by Kier Mining. The BC Code refers to the British Coal codes that are supposed to relate the various naming schemes across England and Wales. A column displaying the BGS coal seam map names has been added, and the overall graphical style improved. The Taits Marine Band is the second Lingula band beneath the Pathhead Coal (Cameron and Stephenson, 1985), and is not recognised as widely as the Aegiranum Marine Band which stretches even as far as South Derbyshire (Waters and Davies, 2006).
29
2.3
2.3.1
Expected Structural Styles

Normal Faulting
As Section 2.1 explained, it was thought that The Midland Valley of Scotland could be modelled as a simple graben, as in Bott (1976). This would give The Midland Valley an overall shape such as that in Figure 2.12:
Figure 2.12: Model showing large-scale deformation of the Earths crust through extension and normal faulting. Upbending of the crust results from displacement in the lower layer. From Bott (1976).
Whilst this theory was later surpassed by a strike-slip interpretation (Ritchie et al., 2003; Caldwell and Young, 2013), there is still normal faulting within The Midland Valley. These will grossly follow established normal fault geometries, such as in Figure 2.13:
Figure 2.13: Simple models for extensional faulting showing no 3D along-strike variation. From Fossen (2010).
30
If a normal fault array is viewed in 3D, the linkage between the faults will produce recognisable patterns such as relay ramps and linked faults. Figure 2.14 shows the evolution of normal faults in a purely extensional setting. This model may well apply to The Midland Valley where normal faulting is certainly prevalent enough to initially be confused as the main proponent of basin formation (Underhill, Monaghan and Browne, 2008).
Figure 2.14: Evolution of syn-rift extensional depocentres. From Cowie, Gupta and Dawers (2000).
31
This normal faulting will also aect the rocks at a small-scale with, with the interaction between dierent lithologies producing interesting results. The coal of the Coal Measures is deposited within cyclothems (Browne et al., 1999), dominated by mixtures of shale and sand. This dierence in rock competency, when faulted, may create structures such as fault smear and fault drag, as seen in Figure 2.15:
Figure 2.15: Diagram of a normal fault cutting a strong-weak-strong layering of rock beds. The weak layer is deformed and incorporated into the fault plane. This allows some exure within the strong bed. From Fossen (2010).
2.3.2
Strike-Slip Faulting
As the structural displacement within The Midland Valley is believed to have a considerable strike-slip component, there may be any number of structures associated with such settings in the area. Figure 2.16 demonstrates a wide variety of structures that can form in a strike-slip regime, including transtensional and transpressional features:
Figure 2.16: Large 3D block model Modied from (Cunningham and Mann, 2007).
32
Figure 2.17 from Wu et al. (2009) demonstrates what cross-sectional features may be expected in a negative strike-slip ower structure, such as the one in the bottom-left of Figure 2.16 on the preceding page.
Figure 2.17: 4D sandbox model evolution of a dextral spindle-shaped strike-slip transtensional ower structure. Basal layer is 1.5 cm thick SGM 36 polymer with similar scaled properties to the brittle-ductile transition zone. Overlaying this is 7.5 cm of dry quartz sand. Model scales at roughly 1:100,000 (1 cm = 1 km). From Wu et al. (2009)
34
In 2012, Dooley and Schreurs modelled strike-slip faults in sandboxes with quartz sand to attempt to create analogue models of such zones during brittle deformation. Figure 2.18 shows the result of one of the experiments. These models have a variable component of underlying basement fabric, with the three columns indicting dierent amount of stepover.
Figure 2.18: Line drawings of experimental sand box modelling. a) Underlapping stepover. b) Neutral stepover. c) Overlapping stepover. Numbers relate to master fault segments. From (Dooley and Schreurs, 2012).
Modelling of strike-slip faults with an underlying fabric has also been done experimentally by Richard and Krantz (1991). However, in their paper the researchers looked at the inuence of reactivation of normal faults before strike-slip motion. Once again, sandbox modelling was used. Figure 2.19 shows two sandbox experiments: [1] with a dip-slip normal fault, and [2] with an un-activated fabric.
35
Figure 2.19: Line drawings of dextral strike-slip sandbox models overlying a 90 dipping normal fault. [1]: A/B: purely brittle deformation. C/D: 5 cm of sand on 1 cm of silicone (acting as a d ecollement horizon). E/F: 4 cm of sand on 2 cm of silicone. [2]: A similar dextral system, however here the underlying fabric is unactivated. Compiled from Richard and Krantz (1991).
Models of faulting styles shows that a variety of fault types can result from a simple shear strike-slip, even including normal extensional faulting. These are present in Figure 2.16 within the releasing bend, transtensional rift, and extensional horsetail splay. Extensional faulting in strike-slip environments was recently analysed by Waldron (2005), with the results shown in Figure 2.20 below. The diagram shows that as displacement increases on the main transfer fault, normal faults can be made to invert.
Figure 2.20: The eects of simple-shear on a strain ellipse. Demonstrates that normal faults can form oblique to a dextral strike-slip zones, and if enough deformation occurs, can be made to invert. From (Waldron, 2005), originally after Harding (1974).
2. GEOLOGICAL BACKGROUND Many of the features associated with a strike-slip basin are summarised in Figure 2.21 below.
36
Figure 2.21: Diagram showing the structures that are associated with a pull-apart basin, such as in Figure 2.16. From (Rodgers, 1980).
37
2.3.3
Inter-Coal Faulting
In addition to the gross-scale faulting across The Midland Valley, East Ayrshire, and the area studied in the project, there may be small-scale faulting within the units. This is not to say those larger features will not be visible- the fractal nature of geology ensures that most will be likely evident within even the smallest scale data available. Examples of these structures are seen in Figure 2.22. Nevertheless, within the coal beds, predicted structures will be associated with soft-rock deformation in the mudstone sections of the lithological layering, but if the coal is suciently metamorphosed it may develop more brittle fracturing.
Figure 2.22: Diagram showing some of the features that can be expected in a sheared coal seam. Illustration of features in the Wu seam of the Pingdingshan coaleld, northern China. Whilst this coaleld does penetrate mineable Carboniferous coal, the Wu seam is Permian in age. Crosses indicate primary axes of strain ellipses. Colour added to image from (Li, 2001).
Methodology
38
3. METHODOLOGY
39
3.1
Data Obtained for the Project
A selection of various data sources were brought together from the three coal mining companies within East Ayrshire for this project. This was complemented with data and software from Midland Valley PLC. To tie-in all the various datasets, maps- both digital and paper- were used. A list of the various data sources and providers is provided below:
Kier Mining:
3D Models of worked coal seams Borehole depths Excel spreadsheets
ATH Resources PLC:

3D Models of worked coal seams
British Coal:
Lanehead area boreholes Excel spreadsheet
Midland Valley Ltd.:

Old (2012) geological models from previous work with Kier
EDiNA Digimap:
BGS shapeles, both bedrock geology and fault lines Digital Elevation Model (DEM) area
The British Geological Society:

Solid map: Cumnock 14E (lithological key, dips, cross-section) Solid map: New Cumnock 15W (lithological key, dips)
3.2
Modelling Workow
3. METHODOLOGY
In order to correctly build a 3D model in Midland Valley PLCs Move software a workow was devised. It was modied during progression (for instance the re-formatting of boreholes was not expected initially) but does demonstrate all the processes needed to build the 3D model.
Figure 3.1: Workow summary. Covers both small-scale (upper path) and large-scale (lower path) data sources.
40
3. METHODOLOGY
41
3.3
Modelling Workow- Large-Scale Data
The rst task was to audit all the available data, particularly the old models from Midland Valley. Once complete, it was realised that additional scans of the BGS maps (particularly the cross-section shown in Figure 3.10) were needed at a higher resolution. These scanned maps (from 1980 in the West and 2000 in the East) were then georeferenced against the downloaded data from EDiNA Digimap in ArcMap 10 (geology polygons for lithological units and fault lines). The combined image was then draped onto a DEM of the area in Midland Valleys Move software, allowing a 3D model to begin to take form. An image of this model is shown in Figure 3.2:
Figure 3.2: A Coloured DEM of area. B Draped extent of the scanned BGS solid maps. Note the arrow in the bottom-right indicating North with the red point. (The British Geological Survey, 1980, 2000)
3. METHODOLOGY The map extent chosen was based on the limits of data available, as shown in Figure 3.3.
42
Figure 3.3: Limits of map area within ArcMap 10, projected in the British National Grid co-ordinate system.
Onto this model could be projected the BGS fault lines from EDiNA Digimap, with their down-throw sides manually assigned. Also projected were dips traced from the scanned maps, and then dips in uncertain areas from structure contours. For analysis purposes, the faults were coloured based on their dip direction, as shown in Figure 3.4. This was calculated by looking at the ticks shown on the map and by looking at unit osets. The results on the full area are shown in Figure 3.5. These can also be displayed on separate layers, as in Figure 3.6.
Figure 3.4: Depending on which way the fault is dipping, a specic colour is assigned. This produces synthetic pairs of NE-SW trending and NW-SE faults, similar to the research by Caldwell and Young (2013) and Hooper (2013). It is also very similar to a method used by Wilson et al. (2006) when they looked at strike-slip systems on the Norwegian Margin.
3. METHODOLOGY
Figure 3.5: BGS-mapped faults are coloured based on their dip direction. Draped on maps from The British Geological Survey (1980, 2000)
43
3. METHODOLOGY
Figure 3.6: A 3D oblique view of coloured faults. NB: glow eect is added post-process in Photoshop.
44
3. METHODOLOGY
45
The model was then extrapolated into the third dimension in Midland Valleys Move software. This required knowledge of the geological history of the area, knowledge of the modelling software, and a good interpretation of the map. The approach also assumed that the British Geological Survey map was accurate, which is likely for the regional scale features on the map such as the faults and lithologies. There were some areas known to be inaccurate, highlighted by Kier during mining. These are shown in Figure 1.6 on page 17. Fortunately the local scale coal seam data was highly accurate and superseded the BGS survey when working at a higher resolution. Section 2.2 summarised that folding is determined to be later than the deposition of the Coal Measures, however due to software limitations and time constraints, planar faults were used in the model. Figure 4.10 on page 65 shows that should not have aected the unfolded fault geometries too greatly. Below, Figure 3.7 shows the result of the 3D extrapolation of the faults, with the caption describing the specics.
Figure 3.7: 3D Extrapolation of all faults. Dierent colours relate to dip direction. Red and Blue (major) faults dip at 80 to a depth of 800 m. Green and Yellow (minor) faults dip at an angle of 70 to a depth of 600 m. Also shown are where faults are clipping through each other, and where they have been manually cut to join.
Note: A method for nding the fault dips from the map data was attempted in Midland Valleys 3DMove using the Ribbon Tool : a best-t plane is t through the interaction between the fault lineament and the topography. This, however, proved fruitless as there was not enough dierence in slope between the topography and the intersecting fault lines, forcing the ribbon sticks to vary wildly in dip and dip-direction. The method is only useful in places with more extreme topography variation, as used in
3. METHODOLOGY
46
the study mentioned in Figure 3.4 where Wilson et al. (2006) performed the method at the Lofoten Ridge on the Norwegian Margin.
The lithological units were then added to the model. These were obtained in a digital format from EDiNA Digimap, and were put into Moves Stratigraphy tab after manipulation in Excel and ArcMap as 2D base-horizons. Dip data was then manually added from the map, using the plotted values and structure contours in sparser areas, shown in Figure 3.8 on the next page.
3. METHODOLOGY
Figure 3.8: Dips displayed on the map. Maps from The British Geological Survey (1980, 2000).
47
3. METHODOLOGY
48
3.4
Modelling Workow- Small-Scale Data
The coal seam data from the companies shown on page 39 had to be converted into surfaces in Move. Equally, borehole data had to be converted using Microsoft Excel to a format that could be imported. The locations of the various small-scale datasets are shown in Figure 3.9 below:
Figure 3.9: Location map of coal seam data from ATH Resources (Glenmuckloch, Netherton, Skares); Kier Mining (Braehead, Greenburn, Grieve Hill, Rough Hill) and British Coal (Lanehead). BGS map is shown for scale.
3. METHODOLOGY
49
3.5
BGS Cross Section - New Cumnock 14E
The only regional depth data available that was not related to the local scale geology, was the geological cross-section section published by the The British Geological Survey in 1980. The section and map location are shown below in Figure 3.10:
Figure 3.10: Full map and interpreted cross section scanned from the BGS map of New Cumnock, 14E. Note 1.5x vertical exaggeration on section. After The British Geological Survey (1980).
3. METHODOLOGY
50
This cross-section was traced in Move so it could be analysed. The result is shown in Figure 3.11, with some signicant features labelled:
Figure 3.11: Traced section, with bases of UCMS, MCMS, LCMS and LSC shown. Vertical exaggeration has been removed. A: Thickness variation between the UCMS and the LCMS. B: No sense of reactivation here where C has clear reactivation. D: Large (200) normal oset on this fault.
3. METHODOLOGY
51
3.6
Completing the 3D Model
Having compiled all the data, and seen the rough structural shape of the area in Figure 3.11, a 3D model could be constructed. Following the initial aims, the study area was trimmed down to a 5x5 km squareshown in Figure 3.12 below. The area covers the parts of the BGS map known to be incorrect. Unit thicknesses were, however, rst taken from the few places on the map where this could be carried out for multiple units- shown in Figure 3.13. These thicknesses could then be used to construct multiple cross-sections across the focus area, using predictable fault osets and angles to extrapolate across the features. The small-scale coal seams were also used, particularly Greenburn, Lanehead and Braehead. These lay directly on the BGS section, immediately providing useful dip data and horizons to phantom against, as seen in Figure 3.9 on page 48.
Figure 3.12: Maps from The British Geological Survey (1980, 2000) with the location of the 5x5 km study area shown bound by vertical surfaces.
3. METHODOLOGY 52
Figure 3.13: Modied map showing only the key units from the stratigraphic compilation in Figure 2.10. Sections 01 and 02 are shown crossing through the units of interest (the Coal Measures). Modied from The British Geological Survey (1980, 2000).
3. METHODOLOGY
53
By covering the small study area with over 50 intersecting cross-sections, a 3D model could be built. The results follow: Firstly, Figure 3.14 shows a 3D box that contains the 5x5 km study area:
Figure 3.14: 3D Model Bounds. 5000x5000x1100 m.
This box is shown cut away below, with the faults now visible.
Figure 3.15: 3D model of faults in the 5x5 km study area.
The rst unit to be modelled was the orange volcanic unit ABFPV- Ayrshire basanitic and foiditic plugs and vents (The British Geological Survey, 2012). These are modelled as vertical, due to the extrusion method that Move uses to create surfaces from lines.
3. METHODOLOGY
54
Figure 3.16: 3D model showing the modelling of the volcanic intrusions that penetrate the 5x5 km study area.
The Passage Group Formation was used as the lowermost marker for the units. The base of the Lower Coal Measures was created in light grey above this, overlain by the Middle Coal Measures in dark grey. Lastly, the Upper Coal Measures were mapped for the southern area of the 5x5 km area. They are lost to erosion North of this, this there is no coal prospectivity. From the BGS map and cross-section, there is no evidence for packages thickening across the whole area. The coal seam data from the mining rms also showed little-to-no thickening on the faults, certainly not on the 10-100s of meter scale being modelled here. Surfaces were created in Move using Delaunay triangulation, as other methods could not cope with the faults or volume of cross-sections to join. It was also the format that the small-scale seam data was available. The method ensures that all data points are obeyed, as they were during surveying.
3. METHODOLOGY
55
3.7
Additional Modelling: Seam Depth
Figure 2.11 on page 28 demonstrates that the Lower and Middle Coal Measures are vastly more prospective than the upper, hence why they used to be known as the Barren Coal Measures (Underhill, Monaghan and Browne, 2008; Smith, 1999). These units are thus of the greatest importance in determining where to explore for coal in East Ayrshire. In addition, as surface mining is the only method of coal extraction available in East Ayrshire (Fraser, 2013), the units can only be mined to relatively shallow depths; for a mine to be protable there should be no more than 200 m of rock head above it (The British Geological Survey, 2010). In Kiers East Ayrshire mining operations, there are few mines deeper than 100 m and none more than 200 m. For the purposes of gathering results, a maximum limit of coal extraction has been placed at 200 m. This can be easily shown in the 3D model: the topographic surfaces can be copied and transposed 200 m vertically downwards:
Figure 3.17: 3D model with topography shown in green, and a red surface 200 m below this to reect the maximum depth to which a surface mine can typically reach.
Consequently once the units from the 3D Move model are projected through the area an indication of how much coal could be reached can be determined.
3. METHODOLOGY
56
3.8
Additional Modelling: Petrel
Schlumbergers Petrel software was chosen to create isochore maps and 3D models of the small-scale coal seam data at the Greenburn site (see Figure 3.9 on page 48). Familiarity and superior 3D fault interpretation meant a shift away from Midland Valleys Move. Figure 3.18 shows the Greenburn coal seams in relation to the BGS map. The isochores were created between the neighbouring seams; 6800 to 6600 and 6600 to 6500.
Figure 3.18: Oblique view from the South of the Greenburn coal seams 6500-6600-6800. Note that due to software bugs the seams are not shown in the right vertical order. Map cropped from The British Geological Survey (1980, 2000).
Firstly, as Petrel cannot accept .dxf les, the surfaces in Move were converted into points at each vertex corner. By passing this through Notepad++ to correctly format the headers, Petrel would accept the points as General ASCII data. Figure 3.19 shows this initial import of the 6800 seam, which complements the other data for seams 6600 and 6500 (see Figure 2.11 on page 28 for the relationship between these). A surface was created with 2m spacing and convergent interpolation. An edge detection algorithm from the RDR structural tools package was then used on the surface, with the result shown in Figure 3.20. The results for the other two surfaces are shown in Appendix A.
3. METHODOLOGY
Figure 3.19: The boundary polygon ts the most amount of data across all 3 seams. Data holes are due to outcrop.
57
3. METHODOLOGY 58
Figure 3.20: RDRs edge-detection algorithm applied to a 2 m spaced convergent interpolation surface on the 6800 seam. Hotter colours indicate edges- e.g. curvier surfaces with steeper dips. Same scale as previous Figure.
Results
59
4. RESULTS
60
4.1
Move Models
The results of the construction of the 3D Move model are shown over the following pages. On each page two images are presented, rst the horizon itself and then a depth-coloured horizon (coloured between 500 m and -600 m in every case). The horizons are as follows:
Passage Group (PGP) - page 61 Lower Coal Measures (LCMS) - page 62 Middle Coal Measures (MCMS) - page 63 Upper Coal Measures (UCMS) - page 64
As explained on page 54, Delauney triangulation was used to construct the surfaces. This does appear to have created some artefacts and ragged edges in the model, but they are minor and should not have too large an eect on the nal interpretations or conclusions.
4. RESULTS 3D Modelling - Base Passage Group Formation
61
Figure 4.1: 3D model showing the modelling of the Passage Group within the 5x5 km study area.
Figure 4.2: 3D model showing the modelling of the Passage Group within the 5x5 km study area, coloured by depth.
4. RESULTS 3D Modelling - Base Lower Coal Measures
62
Figure 4.3: 3D model showing the modelling of the Lower Coal Measures (LCMS) within the 5x5 km study area.
Figure 4.4: 3D model showing the modelling of the Lower Coal Measures (LCMS) within the 5x5 km study area, coloured by depth.
4. RESULTS 3D Modelling - Base Middle Coal Measures
63
Figure 4.5: 3D model showing the modelling of the Middle Coal Measures (MCMS) within the 5x5 km study area.
Figure 4.6: 3D model showing the modelling of the Middle Coal Measures (MCMS) within the 5x5 km study area, coloured by depth.
4. RESULTS 3D Modelling - Base Upper Coal Measures
64
Figure 4.7: 3D model showing the modelling of the Upper Coal Measures (MCMS) within the 5x5 km study area. As all exposure UCMS is lost in the North-West of the area, it has not been modelled there.
Figure 4.8: 3D model showing the modelling of the Upper Coal Measures (MCMS) within the 5x5 km study area, coloured by depth.
4. RESULTS
65
4.2
Unfolding the BGS Cross Section
The original BGS cross-section is shown in 3.10 on page 49, with the Move model in Figure 3.11. Using the fault modelling tools available in Move, a 2D restoration can be attempted. This can determine the original geometry of the faults for modelling purposes. Figure 4.9 shows an initial unfolding technique:
Figure 4.9: Unfold whole of LCMS to 0 m datum. Beds keep their parallel shape, however the geometry around the faults becomes unreasonable. This technique has actually, by restoring all line segments, produced a move-on-fault restoration.
This technique is modied to produce Figure 4.10:
Figure 4.10: Unfold only fold-hinge LCMS to 0 m datum. By only restoring the fold hinges, the fault blocks maintain their osets but are moved to a more horizontal position. The technique breaks in the North (left) where there is no further hinge to pull the beds down. Note that the faults do not warp considerably.
This cross-section appeared to have drawn all faults at the usual angle of 70 in Figure 3.10 on page 49, however once unfolded they are closer to 45.
4. RESULTS
66
4.3
Dip Data
Dip evidence points towards folding where the limbs rarely are steeper than about 50 this agrees with the BGS section in Figure 3.10 on page 49. All bedding is well distributed into a NNW-SSE orientation; Figure 4.11 shows a best-t girdle of 243.42).
Figure 4.11: Stereographical analysis of all dip data from map. From Move 2013.
The average dip of each unit across the whole area can be plotted versus maximum age. The tabulated results are shown below, and the graphed results on the next page:
Unit MSS MVL UCMS LCUL PGP LLGS LSC ULGS SYG INV KNW AUC DNV CRKV LNK MCHB
Max Age /Ma 290.0 290.0 311.0 323.9 325.4 327.0 327.0 327.0 344.0 354.0 370.0 417.0 417.0 417.0 443.0 458.0
Number 5 4 9 2 9 4 4 5 25 4 46 11 3 1 15 1
Average Dip 10.2 18.0 20.0 39.0 22.9 9.8 10.3 24.4 12.2 20.0 16.8 58.5 57.7 25.0 20.9 34.0
Standard Dev. 1.8 12.9 11.2 7.1 12.0 7.3 7.4 17.3 5.2 18.7 11.6 20.8 13.6 10.2 -
Max Dip 13 37 40 44 42 20 20 55 25 48 62 90 66 25 50 34
Min Dip 8 10 6 34 10 4 4 12 4 10 5 16 42 25 10 34
4. RESULTS 67
Figure 4.12: Average dip for each unit on the 25x20 km map, plotted against maximum age. Error bars reach the maximum and minimum dips found on the map.
4. RESULTS
68
4.4
Apparent Dip
Figure 4.13 demonstrates the eect that dip can have when exploring coal seams with vertical boreholes.
Figure 4.13: Diagram showing the eect of dip on recorded seam thickness. A vertical borehole cuts through a coal seam (grey) and the thickness is recorded (a). The true thickness is actually the length at line b.
For coal exploration, it would be worth knowing a quick conversion for determining the amount of change between the measured thickness and the true thickness. The formula is thus: true thickness = 100 Which can actually be simplied to: true thickness = 100 (cos (dip) 1) Perhaps it would useful to plot these results as a graph, Figure 4.14 shows calculated results for percentage dierences between measured and true thickness. By knowing the dip in the borehole, a simple correction can be made for the true seam thickness.
measured thicknesscos(dip)measured thickness measured thickness
4. RESULTS
69
Figure 4.14: Graph showing what eect various dip angles will have on the recorded seam thickness down a vertical borehole.
4. RESULTS
70
Applying this to the prospective coal measures, there are multiple places where care must be taken to account for the dip- with some seams appearing over 50% thicker than in reality:
Figure 4.15: 3D model showing a dip attribute map for the prospective sub-surface coal seams, pink ellipses highlight steeply dipping areas. Note that the Middle Coal Measures surface has had the area that outcrop above topography removed.
4. RESULTS
71
4.5
Fault Relationships
The relationship between the faults is likely related to their formation. Analysis of the faults across the whole area shows two main trends, with the interaction between the two having an interesting geometry. Figure 4.16 shows the original 3D model twisted so north is now top-right. This orientates the strike-slip component of the model to horizontal in the image. The Figure also contains smaller images of the two main fault orientation sets, labelled and coloured by their direction of dip. The red and blue faults are known as the Principle Displacement Faults (PDZs), with the other set known as Cross Basin Faults (CBFs), as in Dooley and Schreurs (2012).
Figure 4.16: Plan view of fault model in Move. Note that NW-SE dipping (Blue & Red) faults appear to partition o the NE-SW dipping (Yellow & Green) faults. Also note that these fault sets do not abut perpendicular, but rather at a similar acute angle across the area.
The angle that these CBFs make to the PDZs may be important. Figure 4.17 on the following page shows the result of a colour map for the angles applied across the fault nodes. Results are also shown in Table 4.1.
4. RESULTS
72
Number of Data Points Range Mean Standard Deviation
230 70 63.7 17.5
Table 4.1: The results of the fault-angle node-mapping.
Figure 4.17: By assigning a value to each node for the fault angle between the PDZ faults and the CBFs, a surface could be created to highlight areas of more acute or right-angled joins. There appears, however, to be no pattern.
4. RESULTS
73
4.6
Fault Loss
As faults modify the coal spatially, as shown in Figure 4.18 below, it is important to be able to model and perhaps predict their propagation. The rst important discovery from this project is that the faults appear to have two dominant trends, NE-SW or NW-SE. These have dierent formational histories (strike-slip and dip-slip respectively), but overall they overwhelmingly manifest themselves with normal movement.
Figure 4.18: Schematic diagram demonstrating the eect of normal faulting on coal exploration.
If a fault is found by two points, a line drawn between these will form a predictive trend for a potential fault. If NE-SW it may expected to be steeply-dipping and perhaps not a simple dip-slip displacement. If the line forms a NW-SE trend, the fault is likely to be normal. Unfortunately the down-thrown side cant be predicted, however this can be easily found by comparing depths through boreholing. Figures 4.20 and 4.21 on the following pages show fault loss maps for the important Lower and Middle Coal Measures. These are 2D plan representations of the theoretical diagram in Figure 4.18 above. An important consideration to make is that fault drag is present within the area, shown in Figure 4.23 on page 78. Figure 4.19 demonstrates the eect this will have during coal exploration- some seams may appear thinner in borehole if they are near the fault.
Figure 4.19: Schematic diagram demonstrating the eect of normal faulting on coal exploration, with a more realistic model invoking fault drag.
4. RESULTS
74
Figure 4.20: Fault loss map for Middle Coal Measures in 5x5 km area.
4. RESULTS
75
Figure 4.21: Fault loss map for Lower Coal Measures in 5x5 km area.
4. RESULTS
76
4.7
Fault Lengths: 5x5 km Move Model
The 5x5 km area where the Kier surface mines are located was analysed for fault data. Both the length of the faults, and the number of them were looked at. This only accounts for the BGS-map-scale faults, not the smaller Greenburn faulting. First, the results are presented by each fault set: NW Dip (Red) 9716.6 7 SE Dip (Blue) 15944.4 5 NE Dip (Green) 10013.4 8 SW Dip (Yellow) 10708.7 10
Av. Length /m Number
These are then combined to present results for the two major fault trends. A dominant trend is then determined: NE-SW Trend (PDZs) 25661 12 NW-SE Trend (CBFs) 20722.1 18 Dominant NE-SW (20% Greater) NW-SE (50% More)
This results can be combined to nd a single value for average fault length on each trend: NE-SW Trend (PDZs) 2138.4 NW-SE Trend (CBFs) 1152.3 Dominant NE-SW (1.85x Longer)
Av. length /m
4.8
Fault Lengths: 20x25 km Move Model
This process can be scaled up to cover the whole area. First, the results are presented by each fault set: NW Dip (Red) 2567.5 56 SE Dip (Blue) 1438.6 43 NE Dip (Green) 1438.6 132 SW Dip (Yellow) 1628.8 128
This results can be combined to nd a single value for average fault length on each trend: NE-SW Trend (PDZs) 2328.5 NW-SE Trend (CBFs) 1532.2 Dominant NE-SW (1.52x Longer)
Av. length /m
A summary of all the faults is also shown in Figure 4.22.
4. RESULTS 77
Figure 4.22: Graph showing normal fault lengths plotted by their dip-direction, against the base-10 logarithm of their overall length. The average (mean) length is also shown. note how the measures lengths have poor data resolution due to fractal scaling of the faults- the largest and smallest simply cannot be imaged as well as the medium-sized ones.
4. RESULTS
78
4.9
Greenburn Coal Seams
The results from Greenburn coal seams are displayed over the following pages. These show small-scale features in the faulting of SE-NW trending cross-basin faults (green/yellow in the project colour scheme).
Figure 4.23: Feature A in Figure 4.24. Photo taken 2013-05-28 at the Greenburn surface mine owned by Kier Mining Ltd. Fault drag is visible in the sandstone beds. Also note that the kinks in the fault are due to horizontal benches, rather than actual deviations in the plane.
Figures 4.24 and 4.23 are photographs of the Greenburn mine, taken looking toward location X on Figure 3.18 on page 56.
4. RESULTS
Figure 4.24: Photo provided by David Richardson from Kier Mining Ltd. Seams from Figure 2.11 on page 28 have been added.
79
4. RESULTS
80
Figure 4.25: Petrel model of Greenburn seams, created using the edge detection algorithm outputs such as Figure 3.20. Trace beds of 1 m thick coal are shown on the 6500 and 6600 seam, with the 6800 seam at the top.
Figure 4.26: Cross-section through the Petrel model in Figure 4.25. Some fault reactivation is visible in the South.
4.10
Petrel Isochores
4. RESULTS
Using the Greenburn seam data from Figure 3.19 on page 57, isochores could be constructed for the spaces between the three seams.
Figure 4.27: Isochores from Petrel, thickness between 6800 and 6600 seams. Hot = thick, cold = thin.
81
4. RESULTS
Figure 4.28: Isochores from Petrel, thickness between 6600 and 6500 seams. Note dierent scale from previous gure. Hot = thick, cold = thin.
82
4. RESULTS
83
4.11
Seam Depth
The methodology on page 55 demonstrated that there was a maximum possible depth from which coal can be extracted in the East Ayrshire mines. Although this is not a geophysical constraint, it is imposed by economic factors. As a rough estimate, 200 m has been chosen as the maximum possible extraction depth, shown in Figure 4.29 below:
Figure 4.29: 3D model with topograph shown in green, and a red surface 200 m below this to reect the maximum depth to which a surface mine can typically reach.
Thus by slicing this red max depth horizon through the prospective horizons it can be determined which areas are unreachable. Figure 4.31 shows plan views of these two prospective horizons, MCMS and LCMS. Figure 4.31 then shows the actual depths that could be reached with surface mining.
4. RESULTS
84
Figure 4.30: Plan view of 3D Model, showing the depths to the bases of the prospective coal measures that are below the surface. A lot of MCMS has been eroded, hence it is only really present in the South-East.
4. RESULTS
85
Figure 4.31: Plan view of 3D Model with red maximum extraction depth slice present from Figure 3.17. Note that the volcanic intrusions are still shown for reference.
4. RESULTS
86
4.12
Potential Coal Exploration Targets
By calculating the thickness between the surface topography and the seams, the Prospective Coal Measures with the greatest thickness can be located. This construction was performed vertically, as a borehole would be sunk, to avoid complications around the multiple faults and intrusions. This workow was applied to the Lower Coal Measures and the Middle Coal Measures- those with the most prospective coal seams of interest to Kier Mining. Figure 4.32 shows the calculation made to determine a colour map for the created model. It mirrors some of the theory of Figure 4.29 when the maximum mineable depth was determined. The results are shown in Figure 4.33 on the following page.
Figure 4.32: Depth-to-base for the Prospective Coal Measures. Colour map is only applied between 0 and 200 m, as that is the maximum extent of potential coal extraction.
4. RESULTS
87
Figure 4.33: Isochore thickness mapping vertically between the topography and the base of the prospective coal measures. Note that the syncline has caused the underlying to become buried too deep to extract.
4. RESULTS
88
4.13
Correcting the BGS Map
The nal step is to produce a corrected version of the solid map. This is shown in Figure 4.34 below:
Figure 4.34: New version of the BGS map created using the 3D Move model. Modied from The British Geological Survey (1980, 2000).
Interpretation
89
5. INTERPRETATION
90
5.1
Overview
The regional structure is seen in the original 20x25 km area that covers The British Geological Survey maps of Cumnock and New Cumnock. Figure 5.1 (below) shows the stereonet projection of 301 dip readings from the northern valley area of the project (it does not include those dips in the Southern Uplands). Axial traces can be interpreted on the maps that compliment the clustering on the stereonet:
Figure 5.1: Map from Figure 1.3 with the stereonet of all dips from Figure 4.11 overlain. 25 km E-W and 20 km N-S.
Also, it is noted that the Southern Uplands Fault and (and other major faults) have a similar trend to these features. If this were a normal fault, linkage between the two structural styles would be found through down-slip compression. However, as this is a strike-slip fault the two structures perhaps do not have a simple relationship. A potential interpretation is that the strike-slip faults were able to partition the strain on to their lengths, then- as they were crustal weaknesses- they acted as strain nuclei for fold propagation. Of course, perhaps it is merely coincidental that both the folds and the faults have a similar alignment.
The remainder of the interpretation will focus on the folding and faulting in turn, with linkage and relation to exploration at the end.
5. INTERPRETATION
91
5.2
Folding
The 3D models of the 5x5 km study area are important as they show clearly the main structural style in the area. Figure 5.2 below of the Middle Coal Measures horizon shows the overall structure, with the key features labelled:
Figure 5.2: Unit MCMS coloured by depth, with the colour map extending to cover only the extent of the unit, rather than the whole Z range like in the results.
The creation of this model was based mainly on the map and coal seam data. The map data was presumed to have been measured correctly, and only those areas highlighted by Kier for study (Figure 1.6 on page 17) were not trusted. The interpretation of periclinal syncline indicates a compressive stress direction from the SE (150), at right angles to the strike-slip faults. It is unlikely, therefore, that the same stress eld caused both the features simultaneously. An interpretation could be that the syncline is an earlier feature, occurring before the deposition of the sediments. Certainly there appears to be no thickening between the horizons or in the syncline (Seen in the Greenburn data and in the BGS section) so this is a viable option. This observation of no thickening also rules out the syn-strike-slip formation, as there would be package thickness changes. Perhaps then it a later feature, occurring after the Coal Measures deposition. This too would account for the lack of thickening against faults.
5. INTERPRETATION
92
Figure 4.12 on page 67 of the average, maximum and minimum dips of each unit on the map should be able to determine a timing relationship. Displayed again below, but this time with ages from Figure 2.10 on page 27. The Dinantian and Silesian sub-systems of the Carboniferous system are also shown, as these relate to the strike-slip movements in The Midland Valley.
Figure 5.3: Average dip of each unit (with maximum/minimum dip error bars) plotted by ages from Figure 2.10.
As the oldest units have the steepest dips, indicating they have been folded more. If they started to be folded earlier, and were then folded again after the deposition of the Coal Measures that show no thickening, these results would be validated.
5. INTERPRETATION
93
5.3
Faulting
The most important relationship between faults on a regional scale is the partitioning of normal faults between the main strike-slip faults:
Figure 5.4: Normal faults are partitioned between strike-slip faults.
This relationship is an indication that the green and yellow faults are cross basin faults produced as en echelon features. Importantly, this interpretation indicates that these are a later (or at least contemporaneous) feature. To back up the observation that the NE-trending faults are longer, Figure 4.22 on page 77 demonstrated that they are roughly 50% greater in average length. It also showed that the longest faults in the area are indeed in this NE-trending set. Nevertheless, there are still two possible regional structural interpretations of The Midland Valley looking purely at these major faults: an extensional graben or a strike-slip ower structure. Thus two interpretations will follow:
5. INTERPRETATION
94
5.3.1
Graben Interpretation
Figure 4.16 on page 71 demonstrates the prevalence of major NE-trending faults. Further investigation indicates that there are younger rocks on the down-dip hangingwall blocks, and horst-graben sequences are common. These have a similar trend to The Midland Valley, perhaps indicating an extensional graben formation style. A possible interpretation is shown in Figure 5.5.
Figure 5.5: A potential interpretation of a graben structure for the study area. Drawn to 6 km depth, as suggested by the LISPB survey in Figure 2.2.
5. INTERPRETATION
95
There is also evidence for a more NE-SW extension: whilst the previous interpretation used the red and blue faults to derive a NW-SE extension...
Figure 5.6: The red and blue faults from Figure 4.16 on page 71. Extension is NW-SE. Note orientation of North arrow in bottom right.
...the green and yellow faults (below) also feature structural relationships that would indicate a NE-SW extension:
Figure 5.7: The green and yellow faults from Figure 4.16 on page 71. Extension is NE-SW. Note orientation of North arrow in bottom right.
This interpretation also is validated by data from the Greenburn seams in Section 4.9 on page 78. In the Petrel model there are prominent normal-oset faults with dips that would indicate NE-SW extensionshown in Figure 4.25.
5. INTERPRETATION
96
5.3.2
Flower Interpretation
Looking at Figure 5.6 on the previous page again, the NE-trending red and blue faults suggest a pattern of spindle shapes, with transfer duplexes present. A possible interpretation is shown in Figure 5.8. Here the osets on the faults have normal oset due to dierential movement on the strike-slip faults, and rotation.
Figure 5.8: A potential interpretation of a ower structure for the study area, using the same cross-section line as Figure 5.5. Again, base is determined by LISPB from Figure 2.2.
It is dicult to determine which model for the regional formation of The Midland Valley is correct from the available data, considering all that is available is the BGS map for the area. It would appear, however, with analysis of the gross regional structure (Figure 4.16) that the red/blue faults are partitioning the green/yellow faults between them. As if it were the former that were driving the latter. During the discussion it will be possible to compare these dierent models to published theories; comparing and contrasting structures.
5. INTERPRETATION
97
All the analysed faults within the 5x5 km local scale area in Figure 4.34 have some normal down-dip sense of movement, Figure 4.26 on page 80 also shows some may be reactivated. This can be seen by looking at the relative displacement ages, and by looking at the interaction of the faults with topography. On rst inspection, this indicates an extensional formation of The Midland Valley of Scotland, as demonstrated on the regional scale interpretation in Figure 5.5 on page 94. Figure 4.25 on page 80 shows a model of the Greenburn seams from Section 4.9. The normal-oset faults here follow the NW-SE trend of the green/yellow faults. They are thus partitioned by the red/blue NE-SW trending faults, and this can be seen in the 3D modelling. Unfortunately these partitioning faults are not seen in the coal seam data as there is too much coal lost for mining to be worthwhile. Figure 5.9 shows just how close the mine gets to these major faults:
Figure 5.9: The location of the Greenburn seams from Figure 3.18 on page 56 within the 5x5 km study area. Note that the seam is not mined across the major blue/red faults.
The distinction between the fault trends has been largely based on the visual interpretation of cross-cutting relationships. The data tables on page 76 however can create a more rigorous conclusion. As the NE-SW trending red/blue faults are on average 1.85x longer than the partitioned green/yellow faults, they are predicted to have come rst. This is likely the reason why the mined seams in Figure 5.9 do not cross the larger faults. As they have larger displacements they will have larger damage zones, and will alter the coal more- making it more dicult to extract.
Unfortunately, no data were available for small-scale interpretation on the major strike-slip dominated faults. As mentioned, the Greenburn seam does not cross these features. Instead only the normal faulting style can be interpreted. Figure 5.10 below shows a damage zone associated with these faults within the Coal measures in the Eastern wall of the Greenburn surface mine, similar to Figure 2.22 on page 37 after Li (2001).
5. INTERPRETATION
Figure 5.10: Large fault damage zone in the normal faults seen in the Greenburn mine. Photo taken 2013-05-29 by Ben Craven at location Y on Figure 4.9.
98
5. INTERPRETATION
99
This complements the photograph in Figure 4.23 on page 78 where a damage zone can be seen at location X on the Western wall of the mine. The interplay between the sandstone and the shale layering within the deltaic/cyclothem units such as the Coal Measures creates multiple d ecollement horizons for failure. This creates a faulting model like in Figure 4.19 (repeated below):
Figure 5.11: Schematic diagram demonstrating the eect of normal faulting on coal exploration, with a more realistic model invoking fault drag.
The isochore evidence in Figures 4.27 and 4.28 also indicate thinning on the fault planes. Again, the low competency of the shale allows the sandstone to morph around it along multiple detachment horizons. Ultimately the energy is allows to dissipate over a wider volume of rock. This has distinct implications for the mining of coal- the damage zone contains a lot of interburden that is not productive. This evidence points to gentler dipping fault planes than the usual 70 (as less energy is needed to break and slip the units) so if a new 3D Move model were to be made this should be taken into account.
5. INTERPRETATION
100
5.4
Extraction
As well as complications from fault damage zones, Figure 4.29 on page 83 demonstrated that there is a maximum depth to which coal can be mined by Kier in East Ayrshire. That Figure is repeated below:
Figure 5.12: 3D model with topograph shown in green, and a red surface 200 m below this to reect the maximum depth to which a surface mine can typically reach.
This has clear implications for the extraction of coal in the area: The seams that are most productive are demonstrably present within the Middle and Lower Coal Measures (Figure 2.11 on page 28). This is shown in the 3D Move models in Figures 4.30 and 4.31. Some observation and interpretations that can be drawn from these Figures is that the Middle Coal Measures are more prevalent in the south-east of the area (erosion has stripped them away in the north), whereas the Lower Coal Measures are more prevalent in the north-west (they are too-deep to reach within the syncline). The thickness maps in Figure 4.33 on page 87 can help pinpoint areas with maximum mining potential. My combining these with the BGS map, actual locations can be found where mining would be most productive, shown in Figure 5.13 on the next page. The thickest beds of Productive Coal Measures are found just before they go below the 200 m extraction limit, increasing the mining risk. To decrease the risk, it would be useful to locate the mine where these thickest coal sites intersect. To this end, Figure 5.14 highlights potential mine sites.
5. INTERPRETATION
101
Figure 5.13: Areas of thick productive coal measures. Maps modied from The British Geological Survey (1980, 2000).
5. INTERPRETATION 102
Figure 5.14: Green ellipses highlight areas that overlap from Figure 5.13. These will have the least risk during mining. Map modied from The British Geological Survey (1980, 2000).
Discussion
103
6. DISCUSSION
104
6.1
Folding
Interpretation of the stereonet in Figure 4.11 appears to conrm the assertions made by Coreld and Gawthorpe (1996) and Hooper (2013) described in the Introduction; because the all Carboniferous rocks are folded, the causational compressional event occurs after the deposition of the units. Conrmation for this late-stage folding also comes from the graph in Figure 4.12 on page 67 that shows a decreasing average dip through time. An interpretation for this- that there were folds already present before the deposition of the Carboniferous Coal Measures, and were later tightened by northward compression- was published by Underhill, Monaghan and Browne (2008). This model is also applied by Ritchie et al. (2003) to explain the seismic image of the Leven Syncline in Figure 2.6 on page 22. The Leven Syncline has a similar shape to that in the study area- a broadly symmetrical open synformal syncline. It is therefore possible that this has an equally similar formation history: Figure 6.1 shows that the southern end of the axial trace begins to turn southwest between two major NE-SW striking faults, similar to the faults in the study area.
Figure 6.1: Map of The Midland Valley showing major folds and faults. Study area(s) shown in red; trace of Leven-Midlothian (or Midlothian-Leven) Syncline highlighted in green. Modied after Rippon, Read and Park (1996).
An important correlational note is that the published models invoke Variscan deformation as the drive of N-S compression, as in the better-studied Welsh coal elds (Waters and Davies, 2006). Those dips that do not lie within the general trend are perhaps rotated on faults, or may be related to earlier, minor events. As there are multiple overlapping structural histories, there is bound to be scatter
6. DISCUSSION
105
in the data. For instance, folding can occur in transtensional settings (De Paola et al., 2005), producing structures that would usually only be associated with an extensional setting. It may be possible even, as this is an area that experiences a reversal in strike-slip direction (Ritchie et al., 2003; Caldwell and Young, 2013) to have folds forming like in Figure 6.2 below.
Figure 6.2: Folding in an inverted strike-slip regime. From Allen, Alsop and Zhemchuzhnikov (2001).
6. DISCUSSION
106
6.2
Faulting
The plan view of the large-scale 20x25 km area has been a very useful model to use through the project. The interplay between the principle displacement zone faults and the cross basin faults can be easily seen in the dierent highlights below:
Figure 6.3: Plan view of the 20x25 km area modelled in the project. North is the top-right.
There were two theories postulated in the Interpretation (Section 5.3) for the formational history of these faults- either a graben or a strike-slip ower structure. This is the question that has puzzled multiple researchers over the past few decades. The most modern theories, ones taking account for both eld observations and seismics, agree on a strike slip mode of formation (Ritchie et al., 2003; Hooper, 2013; Caldwell and Young, 2013). The relationship between the faults appears to follow a model of strike-slip movement rather than pure extension.
6. DISCUSSION
107
Firstly, Figure 6.4 shows a standard model for en- echelon faulting, with both compressive and extensive stresses. The Midland Valley contains extensive faulting, so attention should be paid to the 3D model on the right that shows a graben-esque opening en- echelon to the main strike-slip system:
Figure 6.4: Two styles of en- echelon faulting, compressive and extensive. Angle x is 40-45. Modied after Ramsay and Huber (1987).
A strain ellipse can be drawn to show the strain within The Midland Valley. Figure 6.5 shows the the principle displacement zone (PDZ) aligned to the red/blue NE-trending Midland Valley faults studied, and the axis of extension does appear to align with the yellow/green CBFs seen:
Figure 6.5: Strain ellipse with labelled stress tensors. Modied after Sylvester (1988) with orientations and labels from Hooper (2013).
These are also seen in real-world examples, as shown in Figure 6.6 on the next page. As a visual comparison to The Midland Valley this appears very appealing, particularly when the strain ellipse is displayed with the 3D fault model- Figure 6.7 on page 109.
6. DISCUSSION 108
Figure 6.6: Modied after Wilcox, Harding and Seely (1981). Sub-gures are as follows: B: Dasht-e Ba yaz (Iran). C: Lake Basin fault zone, Montana (USA). D: Cottage Grove, Illinois (USA).
6. DISCUSSION
Figure 6.7: Figure 6.5 overlying the 3D fault model.
109
6. DISCUSSION
110
A more statistical approach can, however, be created by determining a comparison between the internal PDZ/CBF angles in The Midland Valley and the published value of 40-45 (Ramsay and Huber, 1987). In Table 4.1 on page 72, the angles at these nodes between the fault sets were shown to demonstrate a mean value of 63.7 with a standard deviation of 17.5. This is perhaps a little too far from the published value, and equally the surface in Figure 4.17 on page 72 does not seem to represent any pattern. This may be a case of too small a data set across too large an area (only 230 nodes across 500 km2 ), or perhaps simply the area did not form as a basic wrench basin. The answer may perhaps lie in the reactivation of The Midland Valley in an opposite sense of motion (Caldwell and Young, 2013); relate to transtensional eects; or may be related to an underlying fabric.
To conclude the initial faulting model discussion, these en- echelon fractures are creating normal faults. These normal faults will exhibit movements that create similar structures to those modelled by Gibbs (diagram repeated below for convenience below). A modication should be made however; rather than the normal faulting driving the transfer fault, it appears to be the other way around.
Figure 6.8: Model showing possible interactions between normal faults and transfer faults, and the structures that may form at the intersections. Modied after Gibbs (1990).
6. DISCUSSION
111
6.3
6.3.1
Underlying Fabric
Stepover
The nal stage of displacement in The Midland Valley was dextral, so for comparison Figure 6.9 shows the three dierent accepted types of dextral strike-slip faulting. The nal stage of the Rhomboidal pull-apart experiment (bottom-right image) certainly appears to have similarities with the 3D Midland Valley model in Figure 4.16 on page 71. This hypothesis is shown guratively on the following page.
Figure 6.9: Line drawings of experimental sand box modelling. a) Underlapping stepover. b) Neutral stepover. c) Overlapping stepover. Modelled with a 10 cm thick layer ofbrittle overburden of silica sand. Numbers relate to master fault segments. From (Dooley and Schreurs, 2012).
By including an underlying fabric in the model, this theory can begin to account for the disparity between the values for fault set angles in Section 6.2: note that the underlying basement fabric of the rhomboidal pull-apart basin has normal fault sets at right angles to the PDZ. This also highlights the dierences between clay and sand modelling, and modelling and pure theory. The real world is often more complicated than can be modelled.
6. DISCUSSION 112
Figure 6.10: Comparison between a sandbox strike-slip fault system and the faults modelled in Move seen in Figure 4.16 on page 71. 3 boxes are highlighted on the model of The Midland valley, these are then expanded with fault throw indicators added, then related to the Figure inset from (Dooley and Schreurs, 2012).
6. DISCUSSION
113
6.3.2
Boundary Faults
Interestingly, The Midland Valley is very well dened in the North and South, not with splays and branch-line oshoots as modelled by Wu et al. (2009) in Figure 2.17. Figure 6.11 below demonstrates the eect of an underlying basement normal fault on a strike-slip zone. Model A/B appears to have a similar purely brittle initial starting conditions to The Midland Valley (rather than the salt layer modelled by the silicone), and evolves to have a similar end result; the strike-slip faults are very constrained within an area surrounding the initial fabric.
Figure 6.11: Figure 6.11 repeated for convenience: line drawings of dextral strike-slip sandbox models overlying a 90 dipping normal fault. [1]: A/B: purely brittle deformation. C/D: 5 cm of sand on 1 cm of silicone (acting as a d ecollement horizon). E/F: 4 cm of sand on 2 cm of silicone. [2]: A similar dextral system, however here the underlying fabric is unactivated. Compiled from Richard and Krantz (1991).
If The Midland Valley had therefore had an earlier period of normal faulting, it could be that this had created a fabric for the strike-slip faults to follow. In 2013, Caldwell and Young determined that The Midland Valley was indeed in a period of extension related to northward subduction of the variscan plate during the Late Devonian/Early Carboniferous (shown in Figure 6.12 on the next page). Nevertheless, Figures 6.9 and 6.10 demonstrated that this area likely has an underlying fabric trending N-S (roughly 45 to the PDZ). This may indicate that there was a previous basement structure, perhaps Caledonian, with a NE-SW trend and acted as nuclei for the Southern Uplands and Highland Boundary faults.
Thus there are at least two underlying fabrics: one that constrains the strike-slip to a graben-shaped NE-trending trough, and one that has a overlapping stepover in order to produce the shapes seen in the PDZ faults and the right-angles of the CBF nodes.
6. DISCUSSION
114
Figure 6.12: Schematic diagram showing the northerly subduction of the Variscan microplate, causing N-S back-arc subduction of The Midland Valley. The back-arc extension may have created the normal fault fabric that the strike-slip PDZ faults then follow during strike-slip motion. Modied after (Caldwell and Young, 2013).
6. DISCUSSION
115
6.4
Inversion
An important complication in the formation history of The Midland Valley is the reversal in displacement direction during the Carboniferous (Ritchie et al., 2003; Caldwell and Young, 2013), shown in Figure 6.14 on the following page. As the switch happened about 10 million years before the deposition of the mineable Coal Measures for which high-resolution data was available, evidence of the movement may not be visible in the 3D Move model. However, what may be visible are any aects of the prior fault geometries on the later structure. A simple model that results in a dextral strike-slip and extensional cross basin faults is shown in Figure 6.13 below:
Figure 6.13: A re-drawn Figure 6.4, to reect the geometry and history of The Midland Valley. Modied after (Ramsay and Huber, 1987).
However, the interpretation of Figure 6.10 determined that the underlying fabric during the nal movement of The Midland Valley (i.e. the Silesian) was orientated N-S. It may therefore be that the initial geometry of the Iapetus suture, or the rst (sinistral) strike-slip movements in The Midland Valley, set up an overlapping stepover. It was then this that was the fabric during dextral strike-slip motion in the Late Carboniferous that produced a rhomboidal pull-apart wrench basin.
6. DISCUSSION
116
Figure 6.14: Schematic diagram showing the dextral strike-slip motion of the PDZ during the Late Carboniferous. This inverts the main through-going strike-slip faults. Modied after (Caldwell and Young, 2013).
6. DISCUSSION
117
6.5
Implications for other Strike-Slip Settings
The variation on the small scale that can be achieved from the shallow-surface boreholing and coal mining is something almost impossible to match with any non-intrusive sub-surface techniques. Certainly at great depths. This makes the study of structures within The Midland Valley particularly important; so that they can be applied to similar settings. With specic application to strike-slip hydrocarbon basins, the application of the small-scale 3D seam data to wrench fault tectonics may provide statistical summaries of fault baing in sandstone reservoirs. Even the way that the fault blocks are partitioned between PDZs and CBFs is an important observation, and the categorisation of faults based on their orientation may well prove to be a useful predictive tool during exploration. Furthermore, a more detailed analysis of the sandstone/shale interbeds at and around the faults in the mines could provide more data for fault smear and fault drag modelling: Figure 5.10 on page 98 certainly represents a fault that would drastically change a reservoir potential ow. These ndings could thus be applied to the South Turgay Basin in Kazakhstan where Hydrocarbon systems are present within a reactivated strike-slip system (Allen, Alsop and Zhemchuzhnikov, 2001; Yin et al., 2012). The basin is currently constrained by seismic information, so data models from East Ayrshire could be used to estimate the small-scale structures and barriers for ow. Figure 6.15 below shows a schematic model from Yin et al. (2012) of the Aryskum Graben, note the similar geometry with well-bound basin walls and a ower structure. There is even coal present, indicating a similar deltaic/cyclothem depositional environment.
Figure 6.15: 3D schematic model of the Aryskum Graben in the South Turgay Basin. Area is roughly 30x20 km with depth to roughly 5 km (no original diagram scale). From (Yin et al., 2012).
Conclusions & Synthesis
118
7. CONCLUSIONS & SYNTHESIS The aims for the project were laid on on page 17:
Display Kiers coal seam data in 3D Produce a structural evolutionary model for the faulting around the mines... ...and relate this to the evolution of The Midland Valley Relate the faults to coal seam thickness changes Correct the BGS map in the area around Kiers current mining operations (Figure 1.6) Identify any new surface mining targets based on these new interpretations
119
These can therefore be concluded in turn:
7.1
3D Data Modelling
All the coal seams provided by Kier (and ATH Resources) were put into Midland Valleys Move and had surfaces created for them. These were then used with the borehole data to create 3D horizons to phantom against whilst building 2D cross-sections in the area. Additionally, the Greenburn coal seam data was modelled in detail with Petrel, shown in Section 3.8 on page 56. The processes for manipulating les between the various programs and formats are now known, and may provide a launch point for further research.
7.2
Midland Valley Evolution
An area in the south-west of the valley spanning 20x25 km has been studied, with a 3D model created for a 5x5 km area. The locations of these studies are shown succinctly in Figure 6.1. Analysis of the larger area in 2D began with the categorisation of the faults into 2 main sets: those trending NE-SW and those trending NW-SE. These were then categorised into principle displacement zone faults and cross basin faults: PDZs were almost twice as long and followed the general trend of The Midland Valley, also they partitioned the CBFs between them. The orientation between the two fault sets does not quite t the standard model of a wrench basin from (Ramsay and Huber, 1987), but that is presumed to be due to the complex history of formation. At least three dierent underlying fabrics have been identied, based on published literature interpretations from The Midland Valley and comparisons between strike-slip models from sand/clay box models. This are related to various stress regimes in the area during the Carboniferous, and have each inuenced in turn the following evolution of structures. By assimilating all the structures seen, analysed, and interpreted in the East Ayrshire coalelds; and comparing this to the published literature; a full model for the structural evolution of The Midland Valley can be produced. This is shown on the next page.
7. CONCLUSIONS & SYNTHESIS
120
Figure 7.1: Full evolutionary model, determined from interpretations during the project and literature reviews. Evidence is present in the East Ayrshire area of Scotland for all the labels present. 01: In the Late Devonian the collision and subduction of the Variscan plate due to the slow ongoing collision between the Gondwana and Laurasia continental plates (Soper et al., 1992). NE-trending normal faults form as a result of back-arc extension. 02: During the Early Carboniferous, the wrenching motion of the closing Proto-Tethys Ocean and the continental block of Laurasia set up a sinistral strike-slip zone across The Midland Valley. This activated along the fabric of the normal faults. 03: Late Carboniferous motions of the Northern European Block (Caldwell and Young, 2013) caused reactivation of the sinistral PDZ into a dextral one. This also meant that the geometry of overlapping stepovers was created by the previous CBFs. 04: Continued motion of the Variscan plate causes contraction of The Midland Valley in a NW-SE motion, tightening folds that have been created during the Carboniferous.
121
7.3
Faults and Coal Thickness Changes
The damage zones associated with the main NE-trending PDZ faults are predicted to have too great a volume of entrained interburden to make coal extraction too risky. Instead, areas between these and CBFs should be focussed on, knowing that the PDZs trend North-East and the CBFs North-West. The faulting appears to have had very little eect on the blocks between the faults, with no thickening shown in the Greenburn isochore maps. Instead, those areas with the thickest volumes of the productive Coal Measures can be looked at for exploration.
7.4
Coal Exploration Targets
Figure 7.2 shows those locations from page 101 where the productive Lower Coal Measures and Middle Coal Measures are thickest. However, this area is already intensely mined for coal, so Figure 7.3 displays this Figure with an overlay of the locations of current and past coal minings. These locations line up extremely well, so there may not be any new potential found in East Ayrshire from this project. If nothing else, this does validate the 3D modelling performed and thus if there is cause to extend mines deeper than the 200 m limit, the Move model could be used.
Figure 7.2: New BGS map with locations of potential mining sites shown. Map modied from The British Geological Survey (1980, 2000).
7. CONCLUSIONS & SYNTHESIS 122
Figure 7.3: New BGS map with locations of current and past coal minings. Taken from Kier Mining seam locations and The British Geological Survey (2013). Map modied from The British Geological Survey (1980, 2000).
123
7.5
Further Work
This project has aimed to use the data provided by Kier Mining in a variety of creative ways to complement the evolutionary study of The Midland Valley. However there is still much scope for advancement on this. The geologists at Kier are astonished that there is not more research done in the mines, as the 3D sand/shale interaction between normal faults (some of which have been reactivated) is perhaps unparalleled in other research areas. As there is a new face exposed almost every day, a normal fault could be seen in multiple slices and modelled accordingly. The damage zones and fault smears can also then be seen in three dimensions. After modelling of more seams and mines, across a wider area, this could result in an extremely accurate and well constrained view of the macro basin scale, all the way down to micro-scale fractures in the mines.
References
Allen, M. B., G. I. Alsop and V. G. Zhemchuzhnikov. 2001. Dome and basin refolding and transpressive inversion along the Karatau Fault System, southern Kazakstan. Journal of the Geological Society 158(1):8395. URL: http:// jgs.lyellcollection.org/ cgi/ doi/ 10.1144/ jgs.158.1.83 ATH Resources PLC. 2013. ATH Resources plc - in administration.. URL: http:// www.ath.co.uk/ html/ news/ archive/ 2013/ 16 05 2013.html Bamford, D. 1979. Seismic constraints on the deep geology of the Caledonides of northern Britain. Geological Society, London, Special Publications 8(1):9396. URL: http:// sp.lyellcollection.org/ cgi/ doi/ 10.1144/ GSL.SP.1979.008.01.06 BBC. 2013. Scottish Coal: Liquidators in talks with interested parties.. URL: http:// www.bbc.co.uk/ news/ uk-scotland-scotland-business-22401238 Bluck, B. J. 2002. The Midland Valley Terrane. In The Geology of Scotland, ed. N. H. Trewin. 4 ed. Geological Society Pub House chapter 5, p. 550. Bott, M. H. P. 1976. Formation of sedimentary basins of graben type by extension of the continental crust. Tectonophysics 36:7786. URL: http:// www.sciencedirect.com/ science/ article/ pii/ 004019517690007X Brewer, J. A., D. H. Matthers, M. R. Warner, J. Hall, D. K. Smythe and R. J. Whittington. 1983. BIRPS deep seismic reection studies of the British Caledonides. Nature 305(5931):206210. Browne, Mike A. E., M. Dean, I. H. S. Hall and A. D. McAdam. 1999. A lithostratigraphical framework for the Carboniferous rocks of the Midland Valley of Scotland. Version 2. Technical report British Geological Society Keyworth: . URL: http:// nora.nerc.ac.uk/ 3229/ 1/ RR99007.pdf Caldwell, W. G. E. and G. M. Young. 2013. Structural controls in the western oshore Midland Valley of Scotland: implications for Late Palaeozoic regional tectonics. Geological Magazine (c):126. URL: http:// www.journals.cambridge.org/ abstract S0016756812000878 Cameron, Ian Burnett and D. Stephenson. 1985. British Regional Geology: The Midland Valley of Scotland. Geological Journal p. 172. Carrington, Damian. 2011. Coals dirty secret: its dying.. URL: http:// www.guardian.co.uk/ environment/ damian-carrington-blog/ 2011/ nov/ 15/ coal-uk-jobs-renewables-clean-energy Cleal, C. J. and B. A. Thomas. 1996. Introduction and general background. In British Upper Carboniferous Stratigraphy. Number 11 Dordrecht: Springer Netherlands chapter 1, pp. 114. URL: http:// www.springerlink.com/ index/ 10.1007/ 978-94-011-0587-3 Coal Action Scotland. 2012. Open-Cast Coal Mining.. URL: http:// coalactionscotland.org.uk/ coal-in-scotland/ open-cast-coal-mining/ #2
124
REFERENCES CoalImp. 2013. Coal Facts - Association of UK Coal Importers.. URL: http:// www.coalimp.org.uk/ 3.html
125
Coreld, S. M. and R. L. Gawthorpe. 1996. Inversion tectonics of the Variscan foreland of the British Isles. Journal of the Geological Society 153(1987):1732. URL: http:// jgs.lyellcollection.org/ content/ 153/ 1/ 17.short Coward, M. P. 1993. The eect of Late Caledonian and Variscan continental escape tectonics on basement structure, Paleozoic basin kinematics and subsequent Mesozoic basin development. Petroleum Geology Conference series 4:10951108. URL: http:// pgc.lyellcollection.org/ content/ 4/ 1095.refs Cowie, P. A., S. Gupta and N. H. Dawers. 2000. Implications of fault array evolution for synrift depocentre development: insights from a numerical fault growth model. Basin Research 12:241261. URL: http:// doi.wiley.com/ 10.1111/ j.1365-2117.2000.00126.x Crone, John. 2000. Coal Measures Correlation in Ayrshire.. Cunningham, W. D. and P. Mann. 2007. Tectonics of strike-slip restraining and releasing bends. Geological Society, London, Special Publications 290(1):112. URL: http:// sp.lyellcollection.org/ cgi/ doi/ 10.1144/ SP290.1 De Paola, N., R. E. Holdsworth, K. J. W. McCarey and M. R. Barchi. 2005. Partitioned transtension: an alternative to basin inversion models. Journal of Structural Geology 27(4):607625. URL: http:// linkinghub.elsevier.com/ retrieve/ pii/ S0191814105000246 Dentith, M. C. and J. Hall. 1989. MAVIS-an upper crustal seismic refraction experiment in the Midland Valley of Scotland. Geophysical Journal International 99(3):627643. URL: http:// gji.oxfordjournals.org/ cgi/ doi/ 10.1111/ j.1365-246X.1989.tb02047.x Dooley, Tim P. and Guido Schreurs. 2012. Analogue modelling of intraplate strike-slip tectonics: A review and new experimental results. Tectonophysics 574-575:171. URL: http:// linkinghub.elsevier.com/ retrieve/ pii/ S0040195112004349 East Ayrshire Council. 2012. East Ayrshire Main Issues Report. Technical Report 14. URL: http:// www.east-ayrshire.gov.uk/ BusinessAndTrade/ PlanningAndBuildingStandards/ LocalAndStatutoryDevelopmentPlans/ MainIssuesReport.aspx Forsyth, Ian Hunter and P. J. Brand. 1986. Stratigraphy and stratigraphical palaeontology of Westphalian B and C in the central coaleld of Scotland. London: H.M.S.O. Fossen, Haakon. 2010. Structural Geology. Cambridge: Cambridge University Press. URL: http:// ebooks.cambridge.org/ ref/ id/ CBO9780511777806 Fraser, Douglas. 2013. Red ink and black stu.. URL: http:// www.bbc.co.uk/ news/ uk-scotland-scotland-business-21707360 George, Thomas N. 1960. The stratigraphic evolution of the Midland Valley. Transactions of the Geological Society of Glasgow 24:32107. Gibbs, Alan D. 1984. Structural evolution of extensional basin margins. Journal of the Geological Society 141(4):609620. URL: http:// jgs.lyellcollection.org/ cgi/ doi/ 10.1144/ gsjgs.141.4.0609
REFERENCES
126
Gibbs, Alan D. 1990. Linked fault families in basin formation. Journal of Structural Geology 12(5-6):795803. URL: http:// linkinghub.elsevier.com/ retrieve/ pii/ 019181419090090L Google Inc. 2013. Google Earth.. URL: http:// www.google.com/ earth/ index.html Gosden, Emily. 2013. Coals nal chapter is not yet written.. URL: http:// www.telegraph.co.uk/ nance/ newsbysector/ energy/ 9963385/ Coals-nal-chapter-is-not-yet-written.html Harding, T P. 1974. Petroleum traps associated with wrench faults. AAPG Bulletin 58:12901304. Haszeldine, R S. 1988. Crustal lineaments in the British Isles: their relationship to Carboniferous basins. In Sedimentation in a Synorogenic Basin Complex: The Upper Carboniferous of Northwest Europe, ed. B M Besly and G. Kelling. Glasgow: Blackie & Son. Heckel, P. H. and G Clayton. 2006. The Carboniferous System. Use of the new ocial names for the subsystems, series, and stages. Geologica Acta pp. 403407. URL: http:// revistes.ub.edu/ index.php/ GEOACTA/ article/ view/ 1885 Hooper, M. D. 2013. The Carboniferous evolution of the Central Coaleld Basin, Midland Valley of Scotland: implications for basin formation and the regional tectonic setting. Phd thesis. University of Leicester. URL: https:// lra.le.ac.uk/ handle/ 2381/ 27768 Kennedy, W. Q. 1946. The Great Glen Fault. Quarterly Journal of the Geological Society 102(1-4):4176. URL: http:// jgslegacy.lyellcollection.org/ cgi/ doi/ 10.1144/ GSL.JGS.1946.102.01-04.04 Kennedy, W. Q. 1958. Tectonic evolution of the Midland Valley of Scotland. Transactions of the Geological Society of Glasgow (23):103133. Leeder, M. R. 1982. Sedimentology. Process and Product. London: George Allen & Unwin. Li, Huoyin. 2001. Major and minor structural features of a bedding shear zone along a coal seam and related gas outburst, Pingdingshan coaleld, northern China. International Journal of Coal Geology 47(2):101113. URL: http:// linkinghub.elsevier.com/ retrieve/ pii/ S0166516201000313 Macdonald, Ray and Douglas J Fettes. 2007. The tectonomagmatic evolution of Scotland. Transactions of the Royal Society of Edinburgh: Earth Sciences 97(03):213295. URL: http:// www.journals.cambridge.org/ abstract S0263593300001450 Maddox, S. J., R. A. Blow and S. R. OBrien. 1997. The geology and hydrocarbon prospectivity of the North Channel Basin. Geological Society, London, Special Publications 124(1):95111. URL: http:// sp.lyellcollection.org/ cgi/ doi/ 10.1144/ GSL.SP.1997.124.01.07 McIntosh, Bob. 2013. Westeld Opencast Coal Mine, 1977.. URL: http:// britgeoheritage.blogspot.co.uk/ 2013/ 06/ westeld-opencast-coal-mine-1977.html
REFERENCES
127
Needham, D. T. and R. J. Knipe. 1986. Accretion-and collision-related deformation in the Southern Uplands accretionary wedge, southwestern Scotland. Geology 14(4):303306. URL: http:// geology.gsapubs.org/ content/ 14/ 4/ 303.short Owen, A. W. and E. N. K. Clarkson. 1992. Trilobites from Kilbucho and Wallaces Cast and the location of the Northern Belt of the Southern Uplands during the late Ordovician. Scottish Journal of Geology 28(1):317. URL: http:// sjg.lyellcollection.org/ cgi/ doi/ 10.1144/ sjg28010003 Ramsay, John G. and Martin I. Huber. 1987. The Techniques of Modern Structural Geology: Folds and Fractures. Academic Press. Raymond, Anne C. 1991. Carboniferous rocks of the eastern and central Midland Valley of Scotland: organic petrology, organic geochemistry and eects of igneous activity PhD thesis Newcastle-upon-Tyne. URL: http:// ethos.bl.uk/ OrderDetails.do? uin=uk.bl.ethos.387371 Read, W. A. 1988. Controls on Silesian sedimentation in the Midland Valley of Scotland. In Sedimentation in a Synorogenic Basin Complex: The Upper Carboniferous of Northwest Europe, ed. B. M. Besley and G. Kelling. Glasgow: Blackie & Son pp. 222241. Read, W. A., Mike A. E. Browne, D. Stephenson and B. L. G. Upton. 2002. Carboniferous. In The Geology of Scotland, ed. N. H. Trewin. 4 ed. London: Geological Society Pub House chapter 9, p. 500. Richard, Pascal and Robert W. Krantz. 1991. Experiments on fault reactivation in strike-slip mode. Tectonophysics 188(1-2):117131. URL: http:// linkinghub.elsevier.com/ retrieve/ pii/ 004019519190318M Richardson, David. 2013. Email to Ben Craven, 31 July.. Rippon, J. H., W. A. Read and R. G. Park. 1996. The Ochil Fault and the Kincardine basin: key structures in the tectonic evolution of the Midland Valley of Scotland. Journal of the Geological Society 153(4):573587. URL: http:// jgs.lyellcollection.org/ cgi/ doi/ 10.1144/ gsjgs.153.4.0573 Ritchie, J Dereck, H Johnson, Mike A. E. Browne and Alison A Monaghan. 2003. Late Devonian-Carboniferous tectonic evolution within the Firth of Forth, Midland Valley; as revealed from 2D seismic reection data. Scottish Journal of Geology 39(2):121134. URL: http:// sjg.lyellcollection.org/ cgi/ doi/ 10.1144/ sjg39020121 Rodgers, D. A. 1980. Analysis of Pull-Apart Basin Development Produced by En Echelon Strike-Slip Faults. In Special Publication Number 4 of the International Association of Sedimentologists, ed. P. F. Ballance and H. G. Reading. Oxford: Blackwell Publishing Ltd. chapter Sedimentat. Rohde, Robert A. 2005. GeoWhen Database - Westphalian.. URL: http:// www.stratigraphy.org/ bak/ geowhen/ stages/ Westphalian.html Selley, Richard C. 2012. UK shale gas: The story so far. Marine and Petroleum Geology 31(1):100109. URL: http:// linkinghub.elsevier.com/ retrieve/ pii/ S0264817211002030
REFERENCES
128
Shelton, R. 1995. Mesozoic basin evolution of the North Channel: preliminary results. Geological Society, London, Special Publications 93(1):1720. URL: http:// sp.lyellcollection.org/ cgi/ doi/ 10.1144/ GSL.SP.1995.093.01.03 Smith, R A. 1995. The Siluro-Devonian evolution of the southern Midland Valley of Scotland. Geological Magazine 132(5):503513. URL: http:// journals.cambridge.org/ production/ action/ cjoGetFulltext? fulltextid=4464408 Smith, R A. 1999. Geology of the New Cumnock District. Nottingham: British Geological Society. Smith, R A, T. Bide, E. K. Hyslop, N. J. P. Smith, T. Coleman and A. A. McMillan. 2008. Mineral Resource map for North Ayrshire, East Ayrshire and Soth Ayrshire. OR/01/014. Technical report. Soper, N. J., R. A. Strachan, R. E. Holdsworth, R. a. Gayer and R. O. Greiling. 1992. Sinistral transpression and the Silurian closure of Iapetus. Journal of the Geological Society 149(6):871880. URL: http:// jgs.lyellcollection.org/ cgi/ doi/ 10.1144/ gsjgs.149.6.0871 Stedman, C. 1998. Namurian E1 tectonics and sedimentation in the Midland Valley of Scotland: rifting versus strike-slip inuence. In Sedimentation in a Synorogenic Basin Complex: The Upper Carboniferous of Northwest Europe, ed. B M Besley and G Kelling. Glasgow: Blackie & Son pp. 242254. Sylvester, Arthur G. 1988. Strike-slip faults. Geological Society of America Bullitin 100(11):16661703. The British Geological Survey. 1980. Geological Map 14E Cumnock. British Gelogical Survey. The British Geological Survey. 2000. Geological Map 15W New Cumncok. British Geological Society. The British Geological Survey. 2010. Mineral Planning Factsheet. Technical report. URL: http:// www.bgs.ac.uk/ mineralsuk/ planning/ mineralPlanningFactsheets.html The British Geological Survey. 2012. BGS Lexicon of Named Units.. URL: http:// www.bgs.ac.uk/ lexicon/ The British Geological Survey. 2013. GeoIndex.. URL: http:// mapapps2.bgs.ac.uk/ geoindex/ home.html? theme=boreholes The Coal Authority. 2012. Annual Report and Accounts 2011-2012. Technical report The Coal Authority London: . URL: http:// coal.decc.gov.uk/ assets/ coal/ publicationsandinformation/ 5815-annual-report-and-accounts--201112.pdf Underhill, John R, Alison A Monaghan and Mike A. E. Browne. 2008. Controls on structural styles, basin development and petroleum prospectivity in the Midland Valley of Scotland. Marine and Petroleum Geology 25(10):10001022. URL: http:// linkinghub.elsevier.com/ retrieve/ pii/ S0264817208000044 Waldron, John W.F. 2005. Extensional fault arrays in strike-slip and transtension. Journal of Structural Geology 27(1):2334. URL: http:// linkinghub.elsevier.com/ retrieve/ pii/ S019181410400152X
Walker, J. D., J. W. Geissman, S. A. Bowring and L. E. Babcock. 2012. GSA Geologic Time Scale (v. 4.0).. URL: http:// www.geosociety.org/ science/ timescale/ Waters, C N, Mike A. E. Browne, N S Jones and I D Somerville. 2012. Chapter 14 - Midland Valley of Scotland. In A revised correlation of Carboniferous rocks in the British Isles. Number 700 m. Waters, C N and S J Davies. 2006. Carboniferous: extensional basins, advancing deltas and coal swamps. In The Geology of England and Wales. London: The Geological Society chapter 9, p. 51. URL: http:// nora.nerc.ac.uk/ 1553/ Wilcox, R E, T P Harding and D R Seely. 1981. Basic wrench tectonics. The American Association of Petroleum Geologists Bulletin 57(1):7496. URL: http:// archives.datapages.com/ data/ nogs/ data/ 006/ 006028/ pdfs/ 0074.pdf Wilson, Robert W., Kenneth J. W. McCarey, Robert E. Holdsworth, Jonathan Imber, Richard R. Jones, Alastair I. F. Welbon and David Roberts. 2006. Complex fault patterns, transtension and structural segmentation of the Lofoten Ridge, Norwegian margin: Using digital mapping to link onshore and oshore geology. Tectonics 25(4):TC4018. URL: http:// doi.wiley.com/ 10.1029/ 2005TC001895 Wu, Jonathan E., Ken McClay, Paul Whitehouse and Tim P. Dooley. 2009. 4D analogue modelling of transtensional pull-apart basins. Marine and Petroleum Geology 26(8):16081623. URL: http:// linkinghub.elsevier.com/ retrieve/ pii/ S0264817208001177 Yin, W, Z Fan, J Zheng, J Yin and M Zhang. 2012. Characteristics of strike-slip inversion structures of the Karatau fault and their petroleum geological signicances in the South Turgay Basin, Kazakhstan. Petroleum Science 9:444454. URL: http:// link.springer.com/ article/ 10.1007/ s12182-012-0228-3
7.6
Software Used
MikTex 2.9 TeXcount 2.8/3.0 Mendeley Desktop 1.8.4/1.9.1/1.9.2 Move 2013.1 Petrel 2012 ArcGIS 10 Notepad++ v6.4.2 Microsoft Excel 2010 Google Earth 7.1.1.1888 Adobe Photoshop CS6 Adobe Illustrator CS6 Adobe Lightroom 4.2/5
Appendix A
Petrel Edge Detection

All Greenburn seams from page 56 had an edge-detection algorithm applied in Petrel.
130
APPENDIX A. PETREL EDGE DETECTION 131
Figure A.1: RDRs edge-detection algorithm applied to a 2 m spaced convergent interpolation surface on the 6800 seam. Hotter colours indicate edges- e.g. curvier surfaces with steeper dips.
Appendix B
Full Word Count

A full summary of the project word count
134
APPENDIX B. FULL WORD COUNT
135
B.1
Full Word Count
File: Project.tex Encoding: ascii Sum count: 10000 Words in text: 9621 Words in headers: 187 Words outside text (captions, etc.): 3109 Number of headers: 66 Number of floats/tables/figures: 124 Number of math inlines: 2 Number of math displayed: 0 Subcounts: text+headers+captions (#headers/#floats/#inlines/#displayed) 423+2+0 (2/2/0/0) Chapter 0+1+0 (1/0/0/0) Chapter: Introduction 535+1+51 (1/3/0/0) Section: Introduction 123+2+10 (1/1/0/0) Section: UK Coal 0+2+0 (1/0/0/0) Section: Hydrocarbon Exploration 69+1+0 (1/0/0/0) Subsection: Conventionals 74+1+13 (1/1/0/0) Subsection: Unconventionals 71+2+16 (1/1/0/0) Section: Project Aims 0+2+0 (1/0/0/0) Chapter: Geological Background 726+7+326 (1/7/0/0) Section: Research on The Midland Valley of Scotland 462+4+40 (1/2/0/0) Section: Ayrshire in the Carboniferous 19+6+26 (1/1/0/0) Subsection: Stratigraphy and Tectonic History- Literature Compilation 42+6+93 (1/1/0/0) Subsection: Coal Seam Correlation in East Ayrshire 0+3+0 (1/0/0/0) Section: Expected Structural Styles 189+2+74 (1/4/0/0) Subsection: Normal Faulting 268+2+189 (1/6/0/0) Subsection: Strike-Slip Faulting 100+2+55 (1/1/0/0) Subsection: Inter-Coal Faulting 0+1+0 (1/0/0/0) Chapter: Methodology 132+5+0 (1/0/0/0) Section: Data Obtained for the Project 45+2+13 (1/1/0/0) Section: Modelling Workflow 549+4+186 (1/7/0/0) Section: Modelling Workflow- Large-Scale Data 47+4+30 (1/1/0/0) Section: Modelling Workflow- Small-Scale Data 56+6+66 (1/2/0/0) Section: BGS Cross Section - New Cumnock 14E 381+4+77 (1/5/0/0) Section: Completing the 3D Model 178+4+28 (1/1/0/0) Section: Additional Modelling: Seam Depth 176+3+80 (1/3/0/0) Section: Additional Modelling: \emph{Petrel} 0+1+0 (1/0/0/0) Chapter: Results 110+26+162 (5/8/0/0) Section: \emph{Move} Models 80+5+88 (1/2/0/0) Section: Unfolding the BGS Cross Section 70+2+37 (1/3/0/0) Section: Dip Data 122+2+88 (1/3/2/0) Section: Apparent Dip 142+2+89 (1/3/0/0) Section: Fault Relationships 203+2+52 (1/4/0/0) Section: Fault Loss 87+6+0 (1/3/0/0) Section: Fault Lengths: 5x5 km \emph{Move} Model 48+6+54 (1/3/0/0) Section: Fault Lengths: 20x25 km \emph{Move} Model 48+3+114 (1/4/0/0) Section: Greenburn Coal Seams 20+2+32 (1/2/0/0) Section: \emph{Petrel} Isochores 100+2+89 (1/3/0/0) Section: Seam Depth 105+4+56 (1/2/0/0) Section: Potential Coal Exploration Targets 19+4+14 (1/1/0/0) Section: Correcting the BGS Map 0+1+0 (1/0/0/0) Chapter: Interpretation 191+1+19 (1/1/0/0) Section: Overview 304+1+44 (1/2/0/0) Section: Folding 127+1+7 (1/1/0/0) Section: Faulting}\label{sec:faultint 124+2+64 (1/3/0/0) Subsection: Graben Interpretation 616+2+98 (1/4/0/0) Subsection: Flower Interpretation 216+1+56 (1/4/0/0) Section: Extraction 0+1+0 (1/0/0/0) Chapter: Discussion 273+1+34 (1/2/0/0) Section: Folding 456+1+91 (1/6/0/0) Section: Faulting 0+2+0 (1/0/0/0) Section: Underlying Fabric 130+1+81 (1/2/0/0) Subsection: Stepover 236+2+106 (1/2/0/0) Subsection: Boundary Faults 170+1+38 (1/2/0/0) Section: Inversion 247+5+28 (1/1/0/0) Section: Implications for other Strike-Slip Settings 77+2+0 (1/0/0/0) Chapter: Conclusions \& Synthesis 84+3+0 (1/0/0/0) Section: 3D Data Modelling 218+3+161 (1/1/0/0) Section: Midland Valley Evolution 89+5+0 (1/0/0/0) Section: Faults and Coal Thickness Changes 96+3+34 (1/2/0/0) Section: Coal Exploration Targets 148+2+0 (1/0/0/0) Section: Further Work
Appendix C
Digital Data (CD)

Dip readings, fault angles and lengths, Move and Petrel Models
136

Wrench Tectonics in The Carboniferous Coals of The Midland Valley of Scotland

Enviado por

Dados do documento

Título original

Direitos autorais

Compartilhar este documento

Compartilhar ou incorporar documento

Opções de compartilhamento

Você considera este documento útil?

Este conteúdo é inapropriado?

Direitos autorais:

Wrench Tectonics in The Carboniferous Coals of The Midland Valley of Scotland

Enviado por

Direitos autorais:

Masters Thesis

Ben Craven BSc (Hons)

Wrench Tectonics in the Carboniferous Coals of The Midland Valley of Scotland

August 14, 2013 Word Count: 10000

University of Leeds Department of Earth Sciences

Declaration of Academic Integrity

4.6 4.7 4.8 4.9 4.10 4.11 4.12 4.13

1.1 1.2 1.3 1.4 1.5 1.6

2.13 2.14 2.15

3.5 3.6 3.7

3.14 3.15 3.16

4.11 4.12 4.13

6.2 6.3 6.4 6.5 6.6 6.7 6.8 6.9

Figure 1.3: Simplied version of the geological map. 25 km E-W, 20 km N-S.

This project will focus on the specic problems listed below:

Research on The Midland Valley of Scotland

Ayrshire in the Carboniferous

Stratigraphy and Tectonic History- Literature Compilation

Coal Seam Correlation in East Ayrshire

Expected Structural Styles

Data Obtained for the Project

ATH Resources PLC:

Midland Valley Ltd.:

The British Geological Society:

Modelling Workow- Large-Scale Data

Modelling Workow- Small-Scale Data

BGS Cross Section - New Cumnock 14E

Completing the 3D Model

Figure 3.14: 3D Model Bounds. 5000x5000x1100 m.

Figure 3.15: 3D model of faults in the 5x5 km study area.

Additional Modelling: Seam Depth

Additional Modelling: Petrel

4. RESULTS 3D Modelling - Base Passage Group Formation

4. RESULTS 3D Modelling - Base Lower Coal Measures

4. RESULTS 3D Modelling - Base Middle Coal Measures

4. RESULTS 3D Modelling - Base Upper Coal Measures

Unfolding the BGS Cross Section

This technique is modied to produce Figure 4.10:

Number of Data Points Range Mean Standard Deviation

230 70 63.7 17.5

Table 4.1: The results of the fault-angle node-mapping.

Fault Lengths: 5x5 km Move Model

Av. Length /m Number

Av. Length /m Number

Fault Lengths: 20x25 km Move Model

Av. Length /m Number

A summary of all the faults is also shown in Figure 4.22.

Greenburn Coal Seams

Potential Coal Exploration Targets

Correcting the BGS Map

Figure 5.4: Normal faults are partitioned between strike-slip faults.

Figure 6.7: Figure 6.5 overlying the 3D fault model.

Implications for other Strike-Slip Settings

Conclusions & Synthesis

These can therefore be concluded in turn:

Midland Valley Evolution

7. CONCLUSIONS & SYNTHESIS

7. CONCLUSIONS & SYNTHESIS

Faults and Coal Thickness Changes

Coal Exploration Targets

7. CONCLUSIONS & SYNTHESIS 122

7. CONCLUSIONS & SYNTHESIS

Petrel Edge Detection

APPENDIX A. PETREL EDGE DETECTION 131

APPENDIX A. PETREL EDGE DETECTION 132

APPENDIX A. PETREL EDGE DETECTION 133