Mine subsidence and strata control in the Newcastle district of the northern coalfield New South Wales

Transcription

1 University of Wollongong Research Online University of Wollongong Thesis Collection University of Wollongong Thesis Collections 1984 Mine subsidence and strata control in the Newcastle district of the northern coalfield New South Wales William Arthur Kapp University of Wollongong Recommended Citation Kapp, William Arthur, Mine subsidence and strata control in the Newcastle district of the northern coalfield New South Wales, Doctor of Philosophy thesis, Department of Civil and Mining Engineering, University of Wollongong, Research Online is the open access institutional repository for the University of Wollongong. For further information contact Manager Repository Services:

2

3 MINE SUBSIDENCE AND STRATA CONTROL IN THE NEWCASTLE DISTRICT OF THE NORTHERN COALFIELD NEW SOUTH WALES A thesis submitted in fulfilment of the requirements for the award of the degree of DOCTOR OF PHILOSOPHY from THE UNIVERSITY OF WOLLONGONG by WILLIAM ARTHUR KAPP, B.E. Syd. (Civil), M.E. Syd. (Mining) DEPARTMENT OF CIVIL AND MINING ENGINEERING 1984

4 STATEMENT The work submitted in this thesis has not been submitted for a degree to any other university or similar institution. W.A. KAPP r

5 Page i ABSTRACT Coal is being mined from seams which lie beneath urban areas around the City of Newcastle, New South Wales, under nearby Lake Macquarie and the Pacific Ocean. Subsidence occurs as a result of pillar extraction or longwall mining and homes and other structures or surface features can be affected. Detailed field surveys commenced in the late 1960's in the coalfields north and south of Sydney. This work was developed and extended by the author in order to provide the basic information which was used to develop a method of subsidence prediction. Because of the large quantity of results, several computer programmes were designed to handle the calculations, filing and data manipulation from the field booking sheets to the presentation of calculated results from the computer. Detailed analyses of the results of these surveys have enabled the relationship between the geometry of mine workings and the subsidence at the surface to be established for the particular geological environments. This research has shown that the massive and strong conglomerates of the Newcastle area have a significant effect on the value of the maximum subsidence. For the lower range of widths of extraction, the investigations showed that the nature of the caving of the roof strata over the mined seam influenced the magnitude of the surface subsidence. Subsidence in Newcastle was shown to be significantly less than what would be experienced in areas of mainly argillaceous strata such as in the United Kingdom because of

6 Page ii their different caving properties and strata deformation characteristics. On the basis of a theoretical subsidence prediction method developed for the Newcastle area by the author using the survey, geological and mining information, panel and pillar mining layouts are now designed for maximum coal recovery consistent with small values of maximum subsidence. With the aid of these locally established guidelines, longwall extraction is now taking place beneath the Pacific Ocean, Lake Macquarie and areas of surface development along their shorelines. The research showed that, in the Newcastle District, subsidence develops in four stages as the ratio of the width of extraction w to the depth of cover h increases for panels of critical or supercritical length. Extraction layouts with subcritical w/h values up to 0.55 result in a slight undulation of the surface and it is within this range that panel and pillar extraction layouts have been designed for protection of structures on the surface. Where the w/h ratio is greater than 0.55 and less than 0.65, a pronounced subcritical subsidence trough develops. For extraction layouts where the w/h ratio is greater than 0.65 the maximum subsidence increases rapidly as the conglomerate within the strata fails to support itself over the increasingly wide panel. The subsidence reaches its maximum possible value at a critical w/h value of around 1.3. In the supercritical range above 1.3 the maximum subsidence is 0.65 of the seam height mined. The research also revealed the factors which influenced the maximum subsidence such as the recovery of coal from the seam mined, the caving of the roof strata, whether other

7 Page iii seams have been mined, the presence of significant faulting and the stability of pillars which remain between or adjacent to extracted panels. Other features of subsidence profiles in Newcastle studied in relation to the mine geometry were the shape of the subsidence profile, the various relationships between subsidence, slope change, curvature and strain, and the time-subsidence relationships. It was also discovered that the travelling slopes and strains above an advancing face are significantly less than the final static values over the end of the panel. Strain triangles were used to investigate the magnitudes and directions of the maximum and minimum principal strains. E.v The principles developed as a result of the author's research work are unique to the Newcastle district north of Sydney where they are now used as a predictive tool in the control of mine subsidence.

8 Page iv ACKNOWLEDGEMENTS The major part of the thesis is the research based on the subsidence investigations carried out over the workings of the BHP Collieries in the Newcastle District of the Northern Coalfield. The author is grateful to The Broken Hill Proprietary Company Limited for permission to use the results of the investigations in this thesis. The author wishes to thank those of the academic staff in the Department of Civil and Mining Engineering, University of Wollongong who gave their comments and advice, especially the author's supervisors. Associate Professor R.W. Upfold and Dr. R.N. Chowdhury. Particular acknowledgement is made of the efforts and work of the Survey Department of BHP Newcastle and BHP Central Engineering Survey (Wollongong) in establishing and surveying the various subsidence grids under sometimes very difficult field conditions. The results of these surveys and the associated subsidence and strain computations formed the basis of the subsidence work in both the Northern Coalfield as described in this thesis, and in the Southern Coalfield. The following departments of the BHP Co. Ltd., also provided information used in the thesis. The author wishes to acknowledge the assistance of 1. the Manager and Survey staff of each of the several BHP Steel Division Collieries who provided details of mining in the areas of subsidence work, 2. officers of the BHP Coal Geology Department who provided relevant geological plans and associated information, and

9 Page v 3. the staff of Collieries Research who assisted with the evaluation of results, preparation of drawings and figures and with the typing of the manuscript. The subsidence investigations in Newcastle and the application of the results require the cooperation of the Department of Industrial Relations and the Department of Mineral Resources. The encouragement of the Senior Inspector of Collieries, Newcastle, enabled early progress to be made in the use of the panel and pillar method for mining beneath residential areas. Liaison with the Department's Inspectors of Collieries assisted in establishing the subsidence work. This continuing liaison, together with the cooperation of the Department's Subsidence Engineer and officers of the Mines Subsidence Board, Newcastle, is contributing to the progress of the current investigations. Acknowledgement is made of the assistance of the Hunter District Water Board who made available the results of their surveys carried out around the sewerage treatment works at Belmont. Special acknowledgement is made of the encouragement through correspondence and personal discussion with the late Mr. R.J. Orchard who was then the Chief Surveyor and Minerals Manager with the National Coal Board (United Kingdom). His comments were very beneficial as the subsidence work progressed. The cooperation of all those associated with the subsidence work has made these investigations possible and enables the work to continue and develop in the Northern and Southern Coalfields. Their assistance is gratefully acknowledged.

10 Page vi VOLUME I CONTENTS PAGE ABSTRACT i ACKNOWLEDGEMENTS iv CONTENTS vi FIGURES xiii LIST OF SYMBOLS xvii LIST OF ABBREVIATIONS xviii CHAPTER 1 - INTRODUCTION Statement of problem and objectives l 1.2 Methodology Application of the results of the research programme 4 CHAPTER 2 - STRATA BEHAVIOUR AND SURFACE SUBSIDENCE Strata disturbances due to coal extraction General description Caving and strata movement Stability of coal pillars Effects of mining in a second seam Surface effects from coal mining Maximum subsidence Limit angle Slope, curvature and strain Aspects of strain measurement Strain triangles Effect of natural ground slope Subsidence related to time and face advance Prediction of subsidence Introduction National Coal Board method Empirical method of Bals Mathematical approaches Model studies Summary 41

11 CHAPTER 3 Page vii - SUBSIDENCE INVESTIGATIONS IN THE NEWCASTLE DISTRICT Introduction Background details * Geographical setting Geological setting Historical review Mining and associated subsidence work Development of the programme Mining in unprotected areas Panel and pillar mining Discussion of results Longwall mining in a surface sensitive area 58 CHAPTER 4 - CHARACTERISTICS OF SUBSIDENCE AND THE DEVELOPMENT OF A PREDICTION TECHNIQUE Introduction Shapes of subsidence profiles Maximum subsidence related to mine geometry Influence of various factors on maximum subsidence Length of extracted panel Overlapping subsidence profiles Nature of roof strata Geological anomalies Coal recovery Stability of pillars Mining of other seams Definition of profile shape Non dimensional profiles Location of transition point Examination of limit angle Elements of subsidence profiles Maximum developed slopes and strains Inverse curvature related to maximum strain Effect of bay length on calculated strain Strain triangles Subsidence related to time and face advance 102

12 Page viii 4.4 Prediction of subsidence Mathematical modelling of subsidence Finite element analysis Applicability to subsidence modelling and associated problems Input data Software Ill Use of finite element programmes for subsidence modelling Pennsylvania State University CANMET, Canada Summary 115 CHAPTER 5 - CONCLUSIONS Maximum subsidence related to mine geometry Influence of mining and geological factors on subsidence Profile shape and associated relationships Supplementary investigations Final comments 12 6 REFERENCES 12 8

13 Page ix VOLUME II APPENDIX A - BACKGROUND TO THE STUDY OF SUBSIDENCE IN THE NEWCASTLE DISTRICT A.l Geographical setting A Introduction A Physiography A Land use A- 3 A.2 Geological setting A History of coal formation A Structure A Stratigraphy A Conglomerate units A- 7 A.3 Coal resources A Wallarah Seam A Great Northern Seam A Seams of minor significance A Australasian Seam A Victoria Tunnel Seam A Dudley Seam A-ll 3.7 Yard Seam A-ll 3.8 Borehole Seam A Reserves and coal use A-12 A.4 Mining and subsidence A General historical review A Early subsidence A Mining methods A Current subsidence studies A Application of subsidence studies A-19 A. 5 Survey procedures A Introduction A Layout of grids A Levelling and distance measuring A Data processing A-23

14 Page x APPENDIX B - NEWCASTLE SUBSIDENCE INVESTIGATIONS B.l Descriptions of areas studied B Introduction and location plans B Development of the programme of subsidence investigations B- 4 B.2 Details of study areas B- 7 Study 1 - Subsidence effects of static and travelling profiles B- 7 l.l Introduction B Geology and mining details B Subsidence over 2 N Panel B Subsidence related to mine geometry... B Travelling and final subsidence and strain profiles B Shape of the subsidence profile B Subsidence of the railway line B-ll 1.8 Strain triangles B Subsidence related to time and face advance B-14 Study 2 - Subsidence over two shortwalls... B Introduction B Mining and geological aspects B Elements of subsidence longitudinally over Shortwall 1 and pillar extraction B Elements of subsidence in a lateral direction across shortwalls B Features of subsidence profiles B Subsidence related to time and face advance B-22 Study 3 - Subsidence over an extensive pillar extraction area B Introduction B Mining details B Development of subsidence over lateral lines B Development of subsidence over longitudinal lines. B Subsidence along Redhead Road and railway line B Subsidence contours and strain triangles. B Subsidence profile characteristics... B Subsidence and damage at the Convent building B Influence of bay length on calculated strains B-33

15 Page xi Study 4 - Panel and pillar system using both pillar extraction and shortwalls to control subsidence 4.1 Introduction B Geographical and geological setting... B Mining procedures B Subsidence monitoring B Features of subsidence profiles B Subsidence over Q Panel and Gateshead Panel. B Increase in subsidence with time B Subsidence related to time and face position B Calculation of stability of pillars... B-48 Study 5 - Subsidence in Gateshead and related surface damages B Introduction B Subsidence over the first Belt Headings extraction B Subsidence over the Waratah and Gateshead Panels B Excessive subsidence over the second Belt Headings extraction B Evidence of pillar instability B Damages to homes and services B-54 Study 6 - Subsidence over various mine layouts in the Victoria Tunnel Seam B Introduction B Study 6A - L Panel B Study 6B - Shortwall 9 B Study 6C - Macquarie Panel B Study 6D - Subsidence damages over F Panel B Summary B-6 8 Study 7 - Subsidence along Bulls Garden Road over Dudley Seam extraction, Whitebridge... B Introduction B Subsidence in Whitebridge over NW and X Panels B Increase in subsidence in Whitebridge due to Y Panel extraction B Subsidence along Bulls Garden Road over Y, Z and 0 Panels B Subsidence in Green Valley Road, Charlestown B Summary B-7 6

16 Page xii Study 8 - Use of the panel and pillar system to control subsidence in residential and light industrial areas B Introduction B Mining details B Subsidence over 3, 4 and 5 NW Panels... B Subsidence over 6 NW and 4 NW Left pillar extraction B Summary B-82 Study 9 - Subsidence over pillar extraction in the Victoria Tunnel Seam, over a longwall in the underlying Borehole Seam and its effect on a sewerage treatment works... B Introduction B Geology and mining B Subsidence due to V.T. Seam longwalls.. B Failure of pillars and pillar remnants.. B Subsidence over the Borehole Seam longwall B Effects of early V.T. Seam workings... B Rate of subsidence development. B Summary B-92

17 Page xiii FIGURES VOLUME 1 Fig Sydney Coal Basin Fig Plan of the Newcastle-Sydney-Wollongong area Fig. 2. l Movement of strata Fig Stability of pillars Fig Multiple seam mining Fig Trough subsidence Fig Panel and pillar mining layout Fig Curvature of subsidence profile Fig Strain triangle principle Fig Effect of ground slope on strain Fig Development of subsidence Fig Travelling and final profiles Fig Relationship between subsidence and geometry of workings. Fig Bals' method of subsidence prediction Fig Analysis by Hackett Fig Part of Newcastle District showing BHP Collieries Fig. Fig. Fig. Fig. Fig Geology of Newcastle District Supercritical extraction profiles Subcritical extraction profiles Panel and pillar profiles Subsidence and contour plan, longwall and treatment works Fig Strata section Fig Subsidence and subcritical mine geometry Fig Subsidence and mine geometry, Newcastle District

18 Page xiv Fig Effect of extraction length and coal recovery on maximum subsidence Fig Geometries of long rows of coal pillars Fig Supercritical non dimensional profiles Fig Subcritical (0.5 < w/h < 0.65) non dimensional profiles Fig Subcritical (w/h = 0.28) non dimensional profiles Fig. Fig. Fig. Fig. Fig. Fig. Fig. Fig. Fig. Fig. Fig. Fig , Non-dimensional subsidence profiles Locus of transition point Increase in slope change with subsidence 'k' factors for maximum strain Location of maximum tension and compression Inverse curvature related to strain Influence of bay length on strain Principal strains Subsidence related to face position Finite element mesh Mesh showing major stratigraphic horizons Schematic representation of the finite element model Fig CANMET finite element flow diagram

19 Page xv FIGURES VOLUME II Fig. Fig. Fig. Fig. Fig. Fig. Fig. Fig. Fig. Fig. Fig. Fig. Fig. Fig. Fig. Fig. Fig. Fig. Fig. Fig. Fig. Fig. Fig. Fig. Fig. Fig. Fig. A. 1 A. 2 A. 3 A. 4 A. 5 A. 6 A. 7 A. 8 A. 9 A.10 A.11 A.12 A.13 A.14 A.15 A.16 A.17 A.18 A.19 A.20 A.21 A.22 A.23 A.24 A.25 A.26 A.27 Coalfields of the Main Coal Province Coastal strip of the Newcastle Coal District Locations of collieries, Newcastle District Production graph of NSW coal districts Geology of Newcastle District Upper Permian and Triassic stratigraphic units Views of South Belmont and Redhead Beach Views of the Central Coast lakes and the Hawkesbury River Environment of coal formation. Borehole Seam Structures in the Northern part of the coal measures Base of the Newcastle coal measures North East part of geological cross-section South West part of geological cross-section Areas affected by Belmont and Charlestown Conglomerates Isopachs of the Charlestown Conglomerate Seam sections and areas of investigation Seam sections and areas of investigation Seam sections and areas of investigation Isoash map showing combinations of the Dudley Seam Early coal mines in Newcastle Burwood Colliery workings, 1886 Newcastle collieries, 1887 Bord and pillar mining operations Examples of recent mining layouts Computer printout of calculated subsidence Computer printout of calculated strain Computer printout of calculated strain triangle results

20 Page xvi Fig. B. 1 Part of Newcastle District showing BHP Collieries Fig. B. 2 Surface topography. Studies 1, 2 and 3 Fig. B. 3 Contoured bedrock surface, Studies 1, 2 and 3 Fig. B. 4 Coal geology plan, Studies 1, 2 and 3 Fig. B. 5 Early VT Seam workings, Studies 1, 2 and 3 Fig. B. 6 Current Dudley Seam workings, Studies 1, 2 and 3 Fig. Fig. Fig. Fig. Fig. Fig. Fig. Fig. Fig. Fig. Fig. Fig. Fig. Fig. B. 7 B.. 8 B.. 9 B..10 B..11 B..12 B..13 B,.14 B..15 B..16 B..17 B..18 B. 19 B. 20 Early Borehole Seam workings. Studies 1, 2 and 3 Surface plan. Studies 4, 5 and 6 Coal geology plan. Studies. 4, 5 and 6 Current VT Seam workings, Studies 4, 5 and 6 Surface topography. Study 7 Coal geology plan. Study 7 Early VT Seam workings. Study 7 Current Dudley Seam workings. Study 7 Early Borehole Seam workings, Study 7 Surface topography, Study 8 Contoured bedrock surface, Study 8 Coal geology plan. Study 8 Early VT Seam workings, Study 8 Current Dudley Seam workings. Study 8 There are 136 figures which apply to Studies 1 to 9. These are included with the relevant studies and are not listed here.

21 Page xvii SYMBOLS The symbols used by the National Coal Board (1975) were adopted where applicable. There are three self contained sections in Chapter 2 each of which includes a list of symbols used only in that section. a subsidence factor b bay length d horizontal distance from a point on a subsidence profile to the goaf edge f face advance distance in relation to surface point h depth of cover from the seam to the surface kj tensile strain factor k 2 compressive strain factor 1 length of an extracted area m thickness of extraction m E effective mining height r radius of curvature s subsidence of a point s observed subsidence at a station S LW subsidence due to the underlying longwall s gw subsidence due to the underlying shortwall t n distance of transition point from goaf edge w width of an extracted area E maximum strain (tensile or compressive) E distance of point of maximum strain from goaf edge G maximum slope along a subsidence trough S maximum subsidence of a point along a profile S maximum subsidence over a subcritical length panel S k maximum measured subsidence over an extraction max S L maximum subsidence due to the underlying longwall S w maximum subsidence due to the underlying shortwall W width of solid coal pillar a limit angle from the vertical

22 Page xviii LIST OF ABBREVIATIONS The following is a list of abbreviations of titles names used throughout the thesis. and AA Company ACIRL AGL AHD B BH BHP CSIRO D DIR DMR DMRes DSC HDWB JCB JD L MWSDB NCB NSW RL VT YW Australian Agricultural Company Australian Coal Industries Research Laboratories Australian Gas Light Company Australian Height Datum Burwood Colliery Borehole Seam The Broken Hill Proprietary Company Limited Commonwealth Scientific and Industrial Research Organisation Dudley Seam Department of Industrial Relations New South Wales Department of Main Roads Department of Mineral Resources Dams Safety Committee Hunter District Water Board Joint Coal Board John Darling Colliery Lambton Colliery Metropolitan Water Sewerage and Drainage Board National Coal Board (United Kingdom) New South Wales Reduced Level Victoria Tunnel Seam Young Wallsend Seam

23 CHARTER 1 INTRODUCTION

24 CHAPTER 1 INTRODUCTION 1.1 STATEMENT OF PROBLEM AND OBJECTIVES The Sydney Coal Basin is a broad flat bedded sedimentary synclinal structure. The seams outcrop near Port Stevens, 2 00 km north of Sydney and near Nowra, 150 km to the South. The Basin is defined by the outcrop of the top of the Permian strata, shown in Fig. 1.1 (Robinson and Shiels, 1975). Coal is being mined to the south of Newcastle and towards Muswellbrook, and in the Wollongong and Lithgow districts. The control of mine subsidence is important in areas where there are structures or surface features which require protection. As the Sydney Metropolitan area continues to expand and as coal mining approaches these urban areas, subsidence will continue to become an increasingly important aspect of coal mining research. Several coal seams lie beneath Newcastle, the surrounding urban and rural areas and beneath the Pacific Ocean and Lake Macquarie. Coal has been mined in Newcastle since the first colonisation in There have been many reports of subsidence damage over the years and measurements of subsidence have been made from time to time. However there had been no attempt to analyse these early results or - 1 -

25 to organise a programme of subsidence investigations. Coal mining is now extending into the more outlying urban areas of Newcastle and longwall extraction is taking place in deeper mines which lie beneath the Pacific Ocean, Lake Macquarie and developed areas along their foreshores. There is a need to obtain the maximum recovery of coal in surface sensitive areas. With this in mind a research project was initiated by the author in Newcastle in order to investigate the relationship between subsidence and underground extraction and to determine the factors which would influence this relationship. This work led to the development of a method to predict mine subsidence in the Newcastle area. This method is now used to design panel and pillar mining layouts with either pillar extraction or longwalls in areas where there are surface structures or features to protect. 1.2 METHODOLOGY A survey was made of the prediction methods in use overseas and it became apparent that any approach to mine subsidence would need to be accompanied by a programme of field measurements. Although theories and prediction methods have been advanced to explain strata movement and subsidence using various mathematical approaches and physical models, they must all be assessed in the light of empirically derived relationships. It was decided to concentrate the research programme towards the development of an empirical method to investigate the relationship between maximum subsidence and - 2 -

26 mine geometry and to monitor the shape and other features of the subsidence profiles such as slope change, curvature and surface strains. This information was then to be used to test the reliability of some finite element programmes which were available for the prediction of mine subsidence. A concerted programme of subsidence research commenced in Newcastle in 1970 and involved surveys over areas of pillar extraction, shortwall mining and later, longwall mining. The 14 separate field investigations are grouped into 9 overall study areas in the Appendix which accompanies the thesis. The magnitude of each investigation varies from information in one area obtained as a simple study over several months to that in another area obtained during an extensive programme of several years duration. The studies were over the workings of several BHP collieries and were carried out under the direction and supervision of the author. Although the subsidence work in this thesis is confined to the Newcastle District of the Northern Coalfield reference is made to some of the author's published subsidence research in the Southern Coalfield. In each area where the subsidence investigations were carried out, the information which was assembled included details of the geometry of the mine layout, progressive face advance, depths of cover, topographic information and all relevant geological information. The results from the survey monitoring were analysed and related to the mining and geological information in order to develop various relationships between mining and subsidence and to examine the influence of the stratigraphy and other geological features on those relationships

27 The research programme then led on to the development of a subsidence prediction technique to enable subsidence and associated surface slopes and strains to be predicted over a variety of mining layouts, whether using pillar extraction or longwall mining methods. The various relationships were simplified to provide a practical and straight forward method for the mine operator to design mining layouts in different circumstances to take account of subsidence considerations in sensitive areas. 1.3 APPLICATION OF THE RESULTS OF THE RESEARCH PROGRAMME There have already been some direct benefits from particular applications of the author's research during the course of the subsidence investigations in the coalfields both to the north and to the south of Sydney. The author has had the opportunity to publish nine separate and distinct papers on aspects of the investigations, mainly on the work to the south of Sydney. Reference is made to these publications at various stages in this thesis and they are included in the list of references. The direct applications of the subsidence work in the Newcastle District are listed below. Reference should be made to the locality plan in Fig l. There has been pillar extraction and shortwall mining beneath residential areas in Charlestown, Whitebridge, Gateshead and beneath a light industrial area in Bennetts Green, all southern suburbs of the City of Newcastle

28 Longwall mining has taken place beneath the Pacific Ocean. This same panel extended under the shoreline and passed beneath the recently constructed Hunter District Water Board sewerage treatment works at Belmont without incurring any damage. Longwall mining has also taken place beneath Marmong Point around the shores of Lake Macquarie and is expected to be carried out beneath low-lying residential areas elsewhere under circumstances where only a small maximum subsidence can be tolerated. The effects of possible future extraction on the main northern railway line, west of Lake Macquarie, is being examined in the light of the results of the subsidence research

29 LOCALITY MAP FIG. 1.1 SYDNEY COAL BASIN

30 FIG. 1.2 PLAN OF THE NEWCASTLE - SYDNEY - WOLLONGONG AREA

31 CHARTER!2 STRATA BEHAVIOUR AND SUREACE SUBSIDENCE

32 CHAPTER 2 STRATA BEHAVIOUR AND SURFACE SUBSIDENCE 2.1 STRATA DISTURBANCES DUE TO COAL EXTRACTION General Description When a seam of coal is mined over a wide area in relation to the depth of cover, the overlying roof strata will not be self supporting. They will collapse into discreet blocks to form the goaf and the strata above will sag until they rest on the goaf material. The upper strata will deflect through to the surface and this results in the development of a subsidence trough. When any opening is made in an underground coal seam, the virgin stress conditions are disturbed. A narrow heading in the seam can be self supporting if the roof is sufficiently strong. With a weak or friable roof, there are various methods of support available. The extraction of a wider area of coal (for example a single row of pillars in a bord and pillar layout) will cause an unsupported roof to break and to fall, depending on stress conditions, the rock strength and other physical properties, and on the span of the opening. Provided that there is a sufficient depth of overlying cover, h, the roof will collapse to form the goaf and a pressure arch will eventually form as shown in Fig The type of strata above the opening and the nature 6 -

33 and extent of jointing are two of the main factors which influence the formation and the geometry of the arch (Price, Malkin and Knill, 1969 and Lama et. al., 1984). Movement is transferred through the overlying strata to the surface when the ratio of the width of the extracted area to the depth of cover, w/h, exceeds a certain value. The value of w/h at which subsidence first becomes evident depends on the nature of the strata, particularly on their caving properties immediately above the seam and the deformation characteristics of the upper strata. Its value is greatest where the overburden contains thick strong sandstone beds. As the width of an extracted area increases, so does the width of the pressure arch (Fig. 2.1). According to Wardell and Eynon, 1968, a study of the initial movement of the surface in relation to the area of the extraction can indicate the maximum width of the pressure arch. Movement of the surface is gradual and it does not necessarily fracture or fissure except where the seam is close to the surface Caving And Strata Movement When coal is mined over a wide enough area, the roof strata breaks into discreet blocks of varying size and angularity. The collapsed roof fills the space evacuated by the coal and occupies a greater volume than the roof strata before mining. This is known as the primary caving zone. Above the primary caving zone, the strata are deflected and broken to occupy the space vacated by the goaf material. This is known as the secondary caving zone in which the stratigraphic horizons maintain their identity but are significantly deformed and have extensive fracturing

34 The analogy made between the zone of strata in bending and the deformation of a beam would only be applicable before the strata zone which lies above the secondary caving zone reaches the elastic limit. This can be illustrated with two examples from Chapter 3 of the thesis. Elastic deformation would occur for instance when the minimum radius of curvature is 500 km, associated with a maximum subsidence of 45 mm when there is a competent conglomerate bed 100 m thick, high in the strata. The conglomerate would deflect without fracturing under these circumstances. On the other hand with a subsidence of 1100 mm and an associated minimum radius of curvature of 3.8 km the deformation is less likely to be in the elastic region of the stress strain curve and the strata is more likely to have exceeded the elastic limit.

35 The overlying strata then deflect down without fracturing. In the lower portion of this deflected zone, bed separation and microfractures can occur although the major breaks associated with the secondary caving zone are absent. On a regional scale, the zone in bending could be analagous to a bending beam. With mining at shallow depths, any failure which would occur would originate from a zone of induced tensile stresses at localised points on the extreme fibres of the beam and is more likely to occur in the lower strata. However, it is considered that any induced tensile stresses would be too small to cause the rock in a triaxially confined state to fracture more than on a localised scale. This zone of strata bending remains impermeable when mining takes place beneath bodies of surface water (Orchard, 1969). Cracking can occur at the surface because of the tensile strains induced by mining even when such strains are small. However, where there are no major geological or topographic anomalies, the depths of such cracks over total extraction at moderate or great depths are minimal (Hellwig, 1950) and could be of the order of 15 m (Orchard, 1969). Cracks will generally occur along a plane of weakness as provided by surface jointing or the surface expression of a dyke or fault. Three zones of strata movement have been defined - the primary zone; the secondary caving zone; and the region of strata bending. These zones are not distinct but there is a gradual transition from one to another. The heights at which the transitions can be considered to occur depend on factors related to the stratigraphy, other geological features, the geometry of the opening and the mining practice. The height of the primary caving zone, - 8 -

36 The height of the caving zone of at least half! the extraction width above seam level would apply to I the initial period of excavation or to a subcritical! width area of extraction. Fbr supercritical width! extractions the height of the caving zone will depend ; primarily on the bulking factor of the overlying i strata which is different for different rock types as shown in Table 2.1.

37 in particular, depends principally on the nature of the roof rocks. Argillaceous strata cave well resulting in a small increase in volume of the caved rocks and give better support conditions to the overlying beds which slowly bend over (Ropski and Lama, 1973). Massive beds of arenaceous strata cave at large intervals in big blocks resulting in a greater increase in volume of the caved rock, and ultimately less surface subsidence than over caved argillaceous strata. The heights of the different zones are strongly influenced by the lithology of the overlying strata. A comparison of values from different sources allows some estimates to be made, as shown in Table 2.1 (Wardell, 1975). In general terms the height of the primary caving zone is usually taken to be five times the seam thickness and the secondary caving zone extends to 8-10 times the mining height above the seam. More particularly the maximum caving height (primary caving zone) can vary from twice the seam height mined where the roof is made up of weak, well laminated shaley beds, up to 5 times the mining height where the roof is a massive sandstone. The height of this zone is at least half the extraction ' width above the seam level (Farmer and Altounyon, 1980). The height of the caving zone was estimated by Kenny (1969) and Wilson (1975) to be between 2 and 4 times the mining height. Model studies carried out by ACIRL indicated that the height of the primary caving zone in arenaceous strata in the Sydney Basin is less than 15 m (ACIRL, 1969). Ropski and Lama (1973) consider that in general terms the height of the primary caving zone is usually taken to be five times the seam thickness and that the secondary caving zone extends to 8-10 times the mining height above the seam. Investigations were carried out in a mine in Poland above longwall workings in a 1 m thick seam at a depth of cover of - 9 -

38 TABLE 2.1 Heights of Primary Caving Zone Assumed max.caving Type of Lithological height caving characteristics (x seam height) Regular shaped Sandstone, with closely 10 blocks. spaced vertical or near vertical joint planes or induced fractures. Irregular Strong, poorly laminated 5 shaped slabs. and jointed sandy beds. Smaller, regular Strong, not so well 3.3 blocks and slabs. laminated possibly jointed silty or sandy beds. Coarsely frag- Laminated silty or sandy 2.5 mented roof beds. beds. Finer, highly Weak, well laminated shaley 2 fragmented beds. roof beds

39 220 m. Up to 16 m above the seam, the strata consisted of shales and fine grained sandstone. The height of the primary caving zone above the seam was found to be 1.5 to twice the seam thickness, and the secondary caving zone extended to 3 to 3.5 times the seam thickness above the seam. On the other hand, a study in the U.S.A. (Webster, Haycocks and Karmis, 1984) showed that strata are broken and fractured up to a height of from 10 to 12 times the thickness of the extracted seam. This height would include the primary and secondary caving zones. Seam gas studies and associated gas investigations have given an indication of the heights of the various zones of strata deformation. Curves have been produced relating to the co-efficient of degassing to the distance above and below the seam being mined (L. Lunarzewski - personal communication). Without specifying the nature of the strata, a consensus of Continental European experience indicates that the primary caving zone, in which the co-efficient of degassing is 100%, extends to a distance of six times the mining height above the seam. The top of the secondary caving zone was taken to be at a distance of 20 times the mining height above the seam and the limit of the effects of degassing was at around 120 m above a 3m high extraction, or at 40 times the mining height. The distance below the seam for gas permeation is generally considered to be half the distance for gas permeation above the seam. Orchard (1975) also considers experiences with the heights above and below an extraction from which methane can be drained. Strata movement occurs laterally beyond the goaf edge, as well as vertically above and below the extraction. The concept of a "limit angle" or "angle of draw" is discussed later with respect to surface subsidence. However an

40 imaginary line joining the goaf edge with the limit of detectable surface movement does not represent the limit of movement at intermediate depths. The deflection of the strata over the extracted area results in a cantilevering effect over the edges of the solid coal. Investigations in the massive sandstone strata of the Southern Coalfield, New South Wales, showed that over longwall face, the cantilever length of the sandstone in the goaf reached distances as much as 70 to 105 m before failure occurred (Lama et.al, 1984). It follows that the limit of movement near to the seam is of greater lateral extent than indicated by a line at 35. in this region the beds maintain their continuity although some degree of cracking develops. Extensometers were installed in vertical boreholes drilled down from a heading above a retreating longwall face in order to determine relative vertical settlements above the seam and to study strata fracturing and bed separation (Farmer and Altounyan, 1980 and Gupta and Farmer, 1983) Stability Of Coal Pillars Where coal is extracted over a wide area and the roof caves, abutment stresses are imposed on the solid coal on both sides of the mined opening. Where in a series of extracted coal panels, these panels are separated by rows of coal pillars, additional loading is applied to the pillars due to the extracted areas on either side. Whether a pillar fails or remains stable is important to the effects on the deformation of the overlying strata and on the surface subsidence

41 Several theoretical and empirical methods have been used to examine the stability of coal pillars as summarised by Hulstruid (1976). The method now discussed is widely used and has been adopted by the National Coal Board in the Subsidence Engineers' Handbook (N.C.B., 1975). The stability of the rows of pillars which remained between the longwall panels was investigated by Wilson and Ashwin (1972). This paper examined the relationship between confining pressure and the stress at failure of British coal measure rocks, the width of the yield zone around the pillar where the coal is in a fractured condition, and the load taken by the goaf areas adjacent to the pillar. The work indicated that the stress distribution across the pillars and the surrounding goaf areas could be shown by Fig. 2.2a. The coal around the edges of the pillar is fractured and Wilson and Ashwin (1972) found that the width of this fractured zone was m h, where m is the thickness of extraction and h is the depth of cover. Although this coal does not provide support to the superincumbent strata it does contain the pillar core to some degree causing it to be in a triaxially confined state of stress rather than in a uniaxial condition. Using basic soil mechanics principles and by assuming hydrostatic virgin stress conditions in the sedimentary rocks of the British coal measure strata, the vertical peak abutment stress was approximated as being 4rt h where n is the average density of the superincumbent strata. The constant value of 4 was obtained from the triaxial testing of British coals and coal measure rocks and represents the slope of the straight line joining the strength values at confining pressures of 0 and 20 MPa. The slope of this straight line can also be obtained by considering the angle

42 of internal friction of the material which represents its internal resistance to shear. It is given by tan p = 1 + sin <f> 1 - sin 0 where in this case d is the angle of internal friction. Experience from several British collieries led to the consideration that the distance from the goaf edge to the point where the rising stress in the waste reaches the cover load was 0.3 h. The pillar supports the strata which are directly above it and as the distance from the goaf edge increases over the goaf, it takes an increasing amount of the cover load until at a distance of about 0.3 h, the goaf supports the full weight of the superincumbent strata (Fig. 2.2b). The hypothesis by Wilson and Ashwin (1972) is based on conditions in the collieries of the United Kingdom and by using the method a family of curves was developed for U.K. conditions in the Subsidence Engineers' Handbook to enable the stability of pillars between extracted panels to be examined. The method is used in some of the field study areas in this thesis to compare the results of the stability calculations of pillars in a panel and pillar layout with pillar stability as determined from the results of the subsidence investigations. General rules of thumb have been developed in the United Kingdom over the years and continue to be used. As discussed by Whittaker and Singh (1978) the width of a pillar, W, is related to the depth of cover, h, by the relationship

43 W = O.lh + 15 (metres) Also Holland (1967) considers that if the pillar width W is greater than 10 to 12 times the pillar height h, it is unlikely to fail. It was shown by model pillar tests that in a panel and pillar layout, the ratio of the pillar height to the pillar width ratio is a major factor in influencing the pillar strength (Whittaker, 1984). Where the roof or the floor of a coal seam is of a similar or weaker strength than the coal, it is likely that the overburden pressures applied through the pillar will result in its penetration of the roof or floor (Cook and Hood, 1978). in headings, the phenomenon of floor heave results from strata beneath the pillar being pushed down and exuding out in the roadway as floor heave. Similarly, even with large pillars with sufficient overburden pressures, the strata above pillars may move outwards and down to occupy some of the void vacated by the mined coal as appears to have occurred in at least one example in the Southern Coalfield, New South Wales (Kapp, 1982a) Effects Of Mining In A Second Seam Extraction from a seam of coal will affect the strata both above and below it. The amount of disturbance to overlying or underlying seams depends on the method of extraction in the first seam, on the distances between the seams and their relative locations above or below each other, the nature of the interburden strata and other geometrical and geological factors. The effects of these

44 factors on the mining of upper and lower seams were discussed in an empirical manner by Webster, Haycocks and Karmis (1984) in relation to the Appalachian (USA) Coalfields. It is usual to extract the uppermost seam first, so minimising the disturbance to other seams, although it is sometimes unavoidable that the lower seam is extracted first. Extraction of pillars in a lower seam over a wide area in relation to the depth will result in caving of the roof and deformation of the upper strata. The sizes of pillars which have remained in the top seam and the seam separation distance will affect the stability of those pillars. For successful mining in a second seam, overseas experience varies. It has been suggested that the seam to be mined should be separated by a distance of at least 60 times the seam height from an underlying seam which has been totally extracted although Lazer (1965) considers that if the extracted area is large enough the conditions encountered in a seam 15 m above can be similar to those in seams as high as 250 m above the earlier extraction (Webster, et.al., 1984). If the top seam has been extracted so that the immediately overlying strata has caved and surface subsidence has occurred, experience both in the Newcastle District and in the Southern Coalfield shows that extraction in a lower seam over a wide area will result in a greater subsidence and a greater time rate of subsidence than if the top seam had not been extracted (Kapp, 1982b). The amount of increase in the subsidence and the time rate of subsidence will depend on the seam separation, mining heights in both seams, the percentage of coal recovered from the top seam and the sizes or the slimness of pillars or pillar remnants (stooks) which remain in the upper seam

45 It was recommended in the Inquiry into Mining Beneath Stored Waters (some of which was based on British experience), that for better subsidence and strata control pillars should be in vertical alignment (Reynolds, 1976). Experiences of multi-seam workings in the United Kingdom are discussed by Hunter (1980). In a single seam working, the maximum strata abutment pressure of four times the normal cover load is located at around 5 m from the ribside, depending on the seam height and the depth of cover. Fig. 2.3a shows that the pressure reduces to a value less than the cover load over the goaf area and this zone is known as the relaxed zone. As the height above the workings increases, the maximum abutment pressure decreases and its location and the location of the relaxed zone move along lines at 11 to the vertical. Where there are minable seams less than 90 m apart, the face lengths of panels should be reduced (Fig. 2.3b) so that the workings do not enter the zone of increased strata pressures from the first seam mined. In this zone, severe crushing effects have been experienced. At this stage there do not appear to be agreed guidelines with regard to strata deformation and multi seam mining. The effects of the many variables have yet to be reliably assessed in relation to the degree of disturbance to seams overlying earlier mined seams and in any case, this would depend on the geology of the particular coalfield under consideration. Mathematical modelling, discussed later, will become increasingly important in the study of multi seam mining

46 2.2 SURFACE EFFECTS FROM COAL MINING Subsidence is a three dimensional phenomenon which is usually simplified in a two dimensional section across the area of extraction where the maximum subsidence occurs. The main elements of subsidence are the vertical displacement (subsidence), the change in ground slope, and the curvature of the subsidence profile which gives rise to the strains at the surface. The various profiles which represent surface movements over coal extraction from a horizontal or near horizontal seam, are shown in Fig The values for the different elements of subsidence vary widely according to the extraction geometry, geology and mining practice. A basic relationship common to all subsidence work is that for extracted panels of critical length, the maximum subsidence increases over panels of increasing ratios of width of extraction w to depth of cover h until the critical extraction width is reached. The maximum subsidence is influenced by several factors associated with the geology, geometry and mining practice Maximum Subsidence The maximum subsidence S is the maximum vertical displacement along a subsidence profile and occurs over the centre of the extraction in a flat or gently sloping coal seam. As the extracted area becomes wider, the amount of subsidence increases until an upper limit of the maximum subsidence is reached when the area of extraction becomes critical (Fig. 2.4b). This occurs when the ratio of the