Journal ofthe Geological Society, London, Vol. 148, 1991, pp..881-891, 10 figs. Printed in Northern Ireland

A mountain front model for the Variscan deformation of the South coalfield

JULIE A. JONES Cheltenham and Gloucester College of Higher Education, Shaftesbury Hall, St Georges Place, Cheltenham, Glos. GL50 3PP,UK

Abstract: Access to theconsiderable geological database of British , together with seismic evidence of subsurface relationships, has enabled the development of a structural model for the South Walescoalfield. Detailed site investigations are used as keyexamples. The simple broad synclinal nature of the coalfield masks a more complex structural system. Thrusting of Variscan age is largely confined to the Lower and Middle Coal Measures and throughout the central and northern coalfieldis dominantlya forethrustsystem. However, along the south-crop a major backthrust system is developedwhich in places penetrates the Upper Coal Measures. The south-cropshows features analogous with a mountain front and is interpreted as a triangle zone, in which pre-existing basement faults have facilitated uplift. The lack of thrusting within the Upper Coal Measures, in contrast to the underlying Lower and Middle Coal Measures, suggests they have acted as a passive roof to the thrust system. Geographical variationin the intensity of thedeformation can be related to the proximity of, andbuttressing against, the Caledonian massif. Although broadly analogous with current models of foreland basin development and deformation, the coalfield demonstrates the importanceof pre-existing massifs and lineaments in determining the precise evolution of an area.

TheWestphalian sediments of theSouth Wales coalfield east-cropand reaching values of upto 80" inthe reveal a history of complex deformation, mainly of Variscan Carboniferous Limestone of the south-crop. age although later Mesozoic effects are also recorded. The Currentunderstanding of thestructure of theSouth results of post-depositional deformation are clearly recorded Walescoalfield stems largely from the systematicsurvey in the Coal Measures sequence andcan be studied in British work of Woodland & Evans(1964) and Squirrel1 & Coalopencast sites situated around the perimeter of the Downing (1969) whoestablished amodern correlatable coalfield and numerous deep mines located throughout the stratigraphy and documented the major structural elements. basin. Variousreviews of thestructure and basindevelopment The main extends from Pontypool havesince been undertaken (e.g. Owen & Weaver1983; in the east, westwards to Kidwelly, covering an area of 75 by Kelling 1988). An important contribution to understanding 25 km (Fig. l), and preserves a sequence of over 2000 m of the nature of the NNW-SSE faults was made by Gayer et al. Upper Carboniferous strata (Fig. 2). It is the second largest (1973) whodrew attention tothe possibleoblique slip coalfield in Britain. The basin is underlain by Millstone Grit nature and compartmentalizingeffect of the Taff gorge fault. and Carboniferous Limestone now exposed at the margins. More recently, seismic investigations (e.g. Mechie& Brooks To the north of the coalfield Devonian strata are exposed in 1984) have provided important contributions to understand- the Brecon Beacons, and Lower Palaeozoic sequences occur ing the subsurface geology of the area. in thenorth-south inlier of the Usk'anticline and the However, integrated models for basin development and east-west en echelon folds of the Cardiff area (Fig. 1). deformationhave not been presented. This and a Northwest of the coalfield the Lower Palaeozoic rocks of companion paper(Jones 1989b)presents such a model, the Welsh massif, deformed during the Caledonian orogeny, chosing key sites from a study of deep mine and opencast exhibita NE-SW structuralgrain. This area formed the sites (Jones 1989~). Thefollowing account reviews a series stable mass of St George's Land during the Carboniferous. of sites and structures from north to south across the South Atthe eastern end of the coalfield thenorth-south Usk Walescoalfield and also in theeasternmost part of the anticlineand theForest of Dean synclinemay reflect the coalfield. Key themes and issues are then drawn together in influence of a deep-seated trend, sub-parallel to the Malvern a discussion. axis (Kellaway & Hancock1983). To thesouth, in the Bristol Channelbasin, a thicksequence of Mesozoic Northern coalfield sedimentsextends onshore and onlaps the Palaeozoic sequence in the Vale of . Tower The coalfield is an elongate east-west synclinal basin and At Tower colliery (Fig. 3) the workings are concentrated in thepresent-day margins are delineated doublebya the Middle Coal Measures and extend southwards under the escarpment of CarboniferousLimestone and Upper Coal cover of the Upper Coal Measures (Pennant Measures). A Measures (Pennant sandstone) with the Lower and Middle study, based on surveyor's sections, of the Six Feet seam, CoalMeasures forming the intervening low ground. Dips indicated that the seam was repeatedly overthrust (Fig. 4). are gentle around the northern edge, steepening along the Using these sections the magnitude of thrust development 881

Downloaded from http://pubs.geoscienceworld.org/jgs/article-pdf/148/5/881/4890755/gsjgs.148.5.0881.pdf by guest on 30 September 2021 882 A. J. JONES

......

Fig. 1. General location map of South Wales and part of South West England. The study area is delineated by the inset box.

andthe three dimensional geometry of the thrust systems Theimportance of Towerand Ffyndaff sites at the was evaluated. northern margin of the coalfield is that theyclearly Where undeformed, the Six Feet seam at Tower consists demonstratethe penetration of Variscandeformation of alower leaf, 90-120cm thick,separated bya parting northwards throughout the coalfield. (oftena seatearth) from an upper leaf of approximately 60cm. Five thrusts can be recognized from the surveyor's section (Fig. 4 Tl-T5). In many cases thrusting appears only Central coalfield to affect the upper leaf, for example, in the central part of The geographical centre of the coalfield is occupied by a the sections (Tl, T2, T3). Other thrusts cut down laterally, mid-basinsyncline in whichmany of thedeep seams are (e.g. T4) presumably by lateral or oblique ramps, to include unworked due to depth limitations. To thesouth is the the lowerleaf in thedeformation. Displacement varies Maesteg anticline, an east-west-trending fold affected by a laterally andoften decreases onone thrust as movement series of NNW-SSE faults which have a dextral as well as increases onanother (e.g. in Fig. 5 T2 decreases in normal component of displacement (Fig. 3). Eastwards the displacementas T3 increases). The generalstrike and dip Maesteganticline dies outbut is replaced by the slightly directioncalculated from the map is065"/35" SE with an moresoutherly east-west-trending Pontypridd Anticline. overall shortening estimate of 31%. These two folds may represent the same structure in which case thepresent day lateral displacement is attributed to Ffyndaff opencast site (Fig. 3) dextralmovement of the NNW-SSE faultsand/or compartmentalization of the deformation. The seams here dip southwards at about 10" and thrusting occurs in severalseams, consisting largely of shortramps andlong flats which do notproduce anylarge scale St John's colliery and the Jubilee Slide repetitions of stratigraphy. At the southern end of the site A series of 16 arcuate extensional faults dissect the crest of there is intense deformation of the Nine Feet and Red Vein the Maesteg anticline. These faults are arranged en echelon, seamswith an estimated shortening of 40-5470 (Jones striking WNW-ESE and downthrow southwards by up to 1989~).All deformation is directednorthwards. Atthe 90 m. They are collectively known as the Jubilee Slide (Fig. northern end of the site bench excavations allowed a three 3). The JubileeSlide is exposedat Mynydd Llangeinor dimensional view of thrusting in theYard seam where a (88.9214.9225) and Pont y Ffald (88.8719.9260) (Woodland normalfault was visible onlybelow thethrust plane, & Evans1964), dipping southwards at 45" andexhibiting suggestinga phase of extensionalfaulting pre-dating the dip-slip slickensides. thrusting. The Jubilee Slide was exposed underground in the now

Downloaded from http://pubs.geoscienceworld.org/jgs/article-pdf/148/5/881/4890755/gsjgs.148.5.0881.pdf by guest on 30 September 2021 VA RISC AN DEFORMATION, SOUTH WALESDEFORMATION,SOUTHVARISCAN 883

important implications for mining since the lateral continuity of seams cannot be predicted. Workings above and below the driveways are restricted and the vertical extent of the upper Coal structures is unknown. Measures (Pennant) The above relationship raises the possibility for a close association of compressional andextensional structures, In perhapsinvolving re-use (inversion) of fracturea by 3 0 successive movements. The spread of fault orientations may K be a consequence of reactivation or inversion of pre-existing structures. There is less variation in the spread of fold axes 150-530 (Jones 1989~). The confinement of the extensional detachment of the Jubilee Slide within the Lower and Middle Coal Measuresis Lower Coal 210.480 Measures nodoubt related tothe ductility of thestrata. These measureshave also been the focus of compressional Milklone Grit deformationas seen in thenumerous boreholes in theSt John'sarea which exhibit no thrusting in thePennant Cabonlerous 150- 1070 Measures but several repeated sequences between the Two Limeslone FeetNine and Gellideg seams of the Lowerand Middle CoalMeasures. A similar association of anextensional

Upper ORS 70- 120 detachment and compressional zone has also been noted at _NW_ Fernhill colliery (Fig. 3) ina comparable structure, the Brownstones 120- 180 Fernhill Slide (Woodland & Evans 1964).

In St. Maughan's 450 - 610 The Moel Gilau fault I Group 0 Thesouthern limit of theJubilee Slidefault system is provided by the major southward displacement of the Moel Aa IanMarl Gilaufault (Fig. 3). Although one of thelargest faults in &,U, 330-610 South Wales it is one of the least known and has never been proved in underground workings. Its dip and throw can be deducedonly from surface exposures and near surface Fig. 2. Generalized stratigraphyof the South Wales coalfield and workings. surrounding area (Woodland& Evans 1964). In the west of the area the fault trendseast-west, but at Maesteg it bends sharply southeast before returning to an abandonedSt John's colliery near Maesteg (Fig. 3). Two east-west orientation.Eastwards the fault dies outat WNW-ESE faults displace the Victoria seam down to the approximatelyST.955.875 and is replaced by the Ty'n y south, with a combined throw of 73 m and fault inclinations Nant fault (Fig. 3). Inthe west, the MoelGilau fault of 45". Traced downwards these faults flatten at the horizon continuesinto Swansea Bay. Herethe faultpresumably of the Two Feet Nine seam. Immediately southof the shafts terminatesas it fails toemerge on the west of the bay. the interval between the Two Feet Nine and the Lower Nine Displacement is down to the south and the throw increases Feet seam is reduced from 76m to 36 m. The intervening from 61 m in the east (with Pennant Measures in both the Upperand Lower Six Feet,the Caerau, Red Vein and footwall and hangingwall) to a maximum of 1143 m in the UpperNine Feet seams are all absent.Beneath the Two west (where Pennant Measures are downthrown against the FeetNine seam north of theshafts workingsshow no Lower and MiddleCoal Measures of thefootwall). An comparableextensional faulting, suggesting that the 73 m accuratedip for the MoelGilau fault plane is difficult to displacement seen in thehigher seams is accommodated in a determine.Surface outcrop suggests a southwards inclina- detachment close to the level of the Two Feet Nine Seam tion of 55-70' in the east, shallowing to 20-30" in the west. (Fig. 6). The nearest workings in the hangingwall are in the No. 3 Two driveways to the Two Feet Nine seam, conveniently Rhondda seam 1.6 km to the south and are thought tohave aligned along the horizontal portion of the fault, suggest a terminateddue depthto (510111) ratherthan fault more complex history than simple southward extension for complication.Any change in fault dip with depth remains this fault system. Repeated sections of the Upper Six Feet unknown.However, the shallower dipthein more seam occur with occasional vertical or overturned bedding. incompetentLower and MiddleCoal Measures atthe Suchrepetition can only be accounted for by asymmetric western end of thefault suggests thatas with the much northward folding and northward directed thrusting (Fig. 7). smallerJubilee Slide, the mechanicalproperties of the Shortening is estimated at 27% (165 m). Althoughthe sedimentaryrocks may influence the angle of thefault driveways are only 50-150 m apart, correlation of structures plane. between them is often difficult. This might be explained by No NNW-SSE normalfaults displace the Moel Gilau thepresence of an offsetting structurebetween the two fault. Instead some terminate at, or are offset by, it. The driveways. However,the presence of the Upper Six Feet Pen y Castellfault (Fig. 3) forexample is displaced Seam in each driveway suggests that no major offset occurs sinistrally by some 120m. The MoelGilau fault is an and that the poor correlation is more likely to result from importanttectonic boundary in the coalfieldseparating lateral changes in the character of the structures. This has northern forethrusts from southern backthrusts.

Downloaded from http://pubs.geoscienceworld.org/jgs/article-pdf/148/5/881/4890755/gsjgs.148.5.0881.pdf by guest on 30 September 2021 884 J. A. JONES

210

200

'90

I COAL MEASURES FOLDS FAULTS Upper Pennant MeasuresTriassicPennantUpper MA Maesteg DisturbanceND Neath -0- anticline AG AbercynonGelligaer MO Mod Gilau -X- syncline Lower Pennant Measures Millstone Grit m B P Pontypridd Tyn-y-nant A thrust (177 MiddleCoalMeasures mCarboniferous Limestone c Caerphilly JS JubileeSlide - - 0 colliery L Llantwit U. Llanharan mMeasuresLower Coal mOld Red Sandstone CMMawr Cefn N Newlands B opencas K Kenfia + borehole I pc Pen-fcastell Q town I Fig. 3. Tectonic elements and stratigraphy of the study area simplified fromBGS one inch and 1 :50 OOO series maps.

Sou thern coalfieldSouthern penetration This Measures. Pennant Pennantof the Several majorthrusts are recognized,mainly occurring in Measures by thethrusting only occurs in thebackthrust two belts: the Margam and the Llanharren thrusts belts(Fig. system of the south-crop. The throw is variable, ranging up to 100 m. Surface exposure suggests no connection between 3). The Margam thrust belt consists of four major thrusts the Llanharan and Margam thrust belts. (Newlands,Kenfig andtwo unnamed thrusts) all directed southwards and striking WNW-ESE (Fig. 3) (Woodland & Evans 1964). The largest, the Newlands thrust, has a throw Park Slip Extension opencast site of 182 m, and is represented by a zone of disturbance some Thissite is located on thesouth-crop (Fig. 3) wherethe 80m wide.Within this zone, individualfaults dip Lowerand Middle Coal Measures dipnorthwards at 30°, northwardsat about 50". The termination of the thrust is notably steeperthan the south-dipping beds along the obscured in thesoutheast by the Triassiccover. The north-crop. Immediately north of Park Slip the southward displacementcontinues intoNamurian strata asdem- directed Kenfig thrust repeats the Lower and Upper Four onstrated by the Gastrioceras subcrenaturn MarineBand Feet and Two Feet Nine seams, with a throw of 68 m. which was encountered twice in the No. 1 Margam borehole Within this siteeast-west-striking southward-directed (unpublished British Coal data). thrusting is common at all levels, ranging in scale from a few The Llanharanthrust belt repeats the Middle Coal centimetresto several metres. There are two styles of Measures and is described by Woodland & Evans (1964) as thrusting seen on the extraction section (Fig. 8): long flats placingLower CoalMeasures on the Llynfi Beds of the and short ramps, and thrusts with longer ramps as seen in

Downloaded from http://pubs.geoscienceworld.org/jgs/article-pdf/148/5/881/4890755/gsjgs.148.5.0881.pdf by guest on 30 September 2021 VARISCANDEFORMATION, SOUTH WALES 885

24 OCTOBER 1985

12 SEPTEMBER 1985

EOMETRY OFTHRUSTING INLOWER LEAF UNKNOWN

T3 T4 A'

SECTIONIN TRANSPORT DIRECTION

.... KEY / / coalseam sandstone in washout / thrusts = extentandrelative elevation of face

50m SCALE 4 c Fig. 4. Interpretation of British Coal surveyors face sections from the S1 faceat Tower colliery. The strike and dipof the footwall and hangingwall cutoffs (065"/35"/SE)is oblique to the face sections (see Fig.5 for orientations). Section A-A' is reorientated perpendicularto the cutoff strike. Where cutoffs are not exposed, a minimum value has been assumed. T1, T2 etc. refers to the thrusts. N S1 FACE N203000

Hangingwall cutoff. Fie. 5. Map to show locationof the - to thrusts Tl-T4 relative the face -footwall cutoff sections at Tower colliery. Note the Ornament outlines length of ramp increase in displacementon thrust T3 as on each thrust thrust T2 dies out.

Downloaded from http://pubs.geoscienceworld.org/jgs/article-pdf/148/5/881/4890755/gsjgs.148.5.0881.pdf by guest on 30 September 2021 886 J. A. JONES

S N THE "JUBILEE SLIDE" ST JOHNSCOLLIERY MAESTEG

I See detailedsection Fig. 6. General cross sectionof the St John's area to illustrate the Jubilee Slide detaching into an ofarea intense deformation (based on the work of R. Hopkins, British Coal). Refer to Fig. 7 for detailed section.

the centre of the section, displacing the Upper Nine Feet andeast and concentrate into two zones named the seam by 45 m.This latter style of thrusting,with LlancaiachTrough andthe DowlaisTrough (Squirrel1 & considerably greater vertical displacement, is more common Downing 1969). Many faults are arranged en echelon with in thebackthrust system of thesouth-crop than in the the loss of displacement on one fault being taken up by the forethrust system north of the Moel Gilau fault. next. Some may be antithetic tomain boundary faults of the Cleavageduplexes in the Six Feetseam are possibly troughs. Any strike slip element of movement within these associated with a bedding parallel thrust flat. Many bedding troughs remains unproven. surfacesalso exhibit down-dip slickensides, furthera Thrustinginthe easternmost part of the coalfield indication of movement on thrust flats. maintains an east-west or NE-SW strike with a northwards displacement. Thrust ramps typically die out up stratigraphy beforepenetrating the Six Feetand Four Feetseams, Easternmost part of the coalfield whereas to the west these seams are complexly deformed. Three major folds cross the area, their axial traces swinging An exception to the northward directed thrusting again from northeast to east-west,subparallel to the margin of occursalong thesouth-crop of the coalfield. Herethe the coalfield (Fig. 3). All folds were active during deposition Nantgarwthrust displaces theCoal Measures 30 m of the CoalMeasures (Jones 1989a, b), with slightly southwards and places Brithdir Beds on the Hughes Beds condensedsequences onthe Pontypriddanticline in (Pennant Measures) (Fig. 3). Other unnamed thrusts in this comparison tothe flankingAbercynon-Gelligaer and areahave a similar southward displacement. The penetra- Caerphilly synclines. tion of the Pennant Measures by thrusting is again confined Changes in the three folds commonly occur across the to the backthrust system of the south-crop. NNW-SSEfaults, forexample, thedegree of fold Shorteningestimates for the easternmost coalfield are asymmetry and the appearance and disappearance of thrusts impeded by the short length of sections, but deformation is in the Abercynon-Gelligaer syncline. noticeably less intense than further west. There appears to The NNW-SSE-trending faultsdownthrow both west be a gradual eastward reduction in intensityof deformation.

- coal seam - shale or sinstone ..- - seatearth - iauir ...... sandstone

10000' -

9900' -

9800' = \ 100' 200'100' 300' 400' 500' €00' 700' 900' 900' 1OOO'1100'1200' 1300' 1400' 1500' 1600' 1700' Fig. 7. Interpretation of the structure of the Two Feet Nine driveways at St John's colliery.

Downloaded from http://pubs.geoscienceworld.org/jgs/article-pdf/148/5/881/4890755/gsjgs.148.5.0881.pdf by guest on 30 September 2021 V A RISC AN DEFORMATION,VARISCAN SOUTH WALES 887

NORTH SOUTH L 2’9” u4 6 6

0 25 metres U

- malseam - extraction limit -----.thrust -_- proposed continuation of coal seam below l extraction limit Fig. 8. A typical extraction section (along easting 86725E) across Park Slip extension opencast site (from the workof several British coal surveyors). Note that nearly all seams are deformed and thrustingis always directed southwards. Some seams,e.g. the Caerau seam, exhibit geometries which suggest the thrusting has cut down sequence. This may be due to an early ofphase folding and/or later out-of-sequence movement. I, 11, 111 and IV (Fig. 8) suggest a possible orderof thrusting in the Six Feet and Caerau seams.The Lower, Middle and Upper Nine Feet seams are productsof seam splitting rather than thrust duplication, although in addition each seamis affected by thrusting. The thrust affecting the Upper Nine Feet seam must have a steeper and longer ramp than others in the section.

Discussion Shortening estimates Thrusting in the Lower and Middle Coal Measures changes Shortening estimates are calculated at 6% for the openfolds in style and direction across the coalfield. In the centre and of the competent Pennant Measures, whereas structures in north is aforethrust systemlargely confined to discrete theincompetent Lower and MiddleCoal Measures seams. South of the Moel Gilau fault a backthrust system persistently give estimates of over30%. Even within the comprisesthrusts with steeper,longer ramps. Both lattershortening estimates vary markedly and may be a forethrustand backthrust systems show out-of-sequence function of lithology. Due to data gaps theevaluation of thrusts. In a foreland propagating thrust system, movement shorteningin a continuous seam across the basin has not in the hangingwall of a previousthrust would be termed been possible and may in anycase be invalid due to transfer out-of-sequence (Butler 1983). In a backthrust system thrust of displacement by lateral and oblique ramps. Therefore, a propagation by progressive footwall collapse would result in precise value for the overall shortening across the coalfield migration of thrusting towards the hinterland.In view of the remains speculative. It is almost certainly greater than the generalnorthward propagation of Variscanorogeny 6% determined for the Pennant Measures butmay not reach compression (Shackleton et al. 1982) such a system may not the values of35-50% calculatedfor individual structures be sustainable and propagation into the hangingwall along within the Lower and Middle Coal Measures, since vertical thesouth-crop may represent form a of foreland compressionmay have resulted in localthinning and propagation of thebackthrust system (Butler 1987) and contraction of incompetentunits. East-west extensional couldaccount for the out-of-sequence geometries seen at faults (see below) also complicate the calculation of regional Park Slip opencast site (Fig. 8). shortening. Further geographicalvariations occur eastwards as deformation is confined to lowerstratigraphic levels and decreasesin intensity. Assuming a linked fault system, a East-west faults series of lateral and/or oblique ramps is required to account Thevarious elements of theJubilee Slideoccur in the for this displacement transfer. thickened sequence in the southwest of the coalfield. Minor At all of the sitesinvestigated there wascleara syn-depositional slides havebeen recorded in thisarea difference in the extent of deformation between the often (Elliott & Ladipo198l), but it is not possible to intensely deformed Lower and Middle Coal Measures and demonstrate that thickening is directly related to the Jubilee the largelyundeformed Upper Coal Measures (Pennant Slide. Woodland & Evans (1964) recorded the extensional Measures).Only inthe backthrust systemalong the flat of theJubilee Slideas being folded along with the southcrop of the coalfield does the thrusting penetrate the bedding. The FernhillSlide shows a similar relationship Pennant Measures. (Woodland & Evans 1964). Either the extensional faulting

Downloaded from http://pubs.geoscienceworld.org/jgs/article-pdf/148/5/881/4890755/gsjgs.148.5.0881.pdf by guest on 30 September 2021 888 J. A JONES

predatesthe folding or extensionalfaults detached along Timing zonesalready weakened by the compressionaldeforma- There is no evidence for thrusting during deposition of the tion.These slides, together with the muchlarger Moel Coal Measures although early fold development had begun Gilaufault, all show anextensional displacement in the from Westphalian A onwards (Jones 1989~).In addition, no Pennant Measures, providing evidence for post-Upper Coal major hiatus or unconformity interrupts the Coal Measures Measures movement. In support of a post-Vanscan age for sequence,and the major compressive deformation is the movement Brooks et al. (1988) demonstrated large scale interpreted as late Westphalian D at the earliest. There is extensional reactivation of a major Variscan thrust in the evidence for east-westtrending extensional activity both Bristol Channel, probably during the early Cretaceous. The before and after the major compression. The later extension major extensional downthrow of the Moel Gilau fault may probablyoccurred during the Mesozoic andTertiary and well relate to this event. involved significant reactivation of east-west and NW-SE faults. NW-SSE faults Thrust transport direction Thelarge number of NNW-SSEfaults are majora Throughoutsouthwest England thedominant thrust structural component of the coalfield. Preliminary evidence direction is tothe NNW (Coward & Smallwood1984). suggests that these faults were active during sedimentation, Seago & Chapman (1988) notedchanges in the nature controlling coal seam split axes (Hartley pers. comm. 1990). of fold development across a series of NNW tear faults which Thefaults were clearlyactive during thecompressive theyinterpret contemporaneousas development and deformationas evidenced by thedextral offset of the compartmentalization of the structures. In the South Wales symmeticalMaesteg anticline axis andthe variation in coalfield east-west folds and thrusts are also compartmen- intensity and style of folding across NNW-SSE faults in the talized by NNW-SSEfaults and itseems that asin Abercynon syncline. This suggests contemporaneous activity southwestEngland these represent the overallthrust of the NNW-SSE faults with the folding and compartmen- transport direction. talization of the deformation; a similar conclusion to Gayer et al. (1973) for the NNW-SSE fault in the Taff gorge. Later extension on the NNW-SSE faults caused numerous offsets A model for the South Wales coalfield of northwarddirected thrusts. However, the termination The crosssection (Fig. 9) shows thedepth to basement anddisplacement of the NNW-SSEfaults by the Moel (probably Precambrian). Of particular note is the thickness Gilaufault indicates even younger activity of thelatter. between the base of the Carboniferous Limestone and the Currently the NNW-SSE faults show significant extensional top of thebasement. Beneath the coalfield this is 2 km downthrows to both east and west. This movement is likely (Mechie & Brooks 1984) whereas in the Vale of Glamorgan to bepre-late Triassic since although suchfaulting of a the depth to the basementis estimated at 4.2 km. Mechie& similar trend occurs in Triassicrocks of the Cardiff area, Brooks (1984) suggested thatthe basement is affected by throws do not exceed about 10 m, whereas in the coalfield faultingalong the BristolChannel fault zone,and this is the throws are commonly several hundred metres. supported by geophysical evidence (Brooks et al. 1988).

N NEATHDISTURBANCE REACTIVATED MAESTEGBOREHOLE VALE OF GLAMORGAN DURINGVARISCAN OROGENY I I 1 MOEL GlLAU

THRUSI UNKNOWN +. ** SEMENTFAULTS UIO Lower CoalMeasures I 13 Millstone Grit [7 Carboniferous Limestone L movementPrimary on fault D OldRed Sandstone e Reactivatedmovement 13 Lower Pal;eozoic IGi Pre CambrianCrystalline Basement 01 2 345 km

Fig. 9. North-south cross section across the South Wales coalfield along easting87, based on data from the Maesteg borehole, (drilled by Cambrian Exploration) and on reflection and refraction seismicdata (Mechie & Brooks 1984; Brooks er al. 1988). Note the suggested position, orientation and reactivation of basement faults. For sequential restoration of the south-crop structure see Fig. 10.

Downloaded from http://pubs.geoscienceworld.org/jgs/article-pdf/148/5/881/4890755/gsjgs.148.5.0881.pdf by guest on 30 September 2021 VARISCANDEFORMATION, SOUTH WALES 889

~~ ~ ~ Schematic overstepping backthrust sequence in Pennant Measures- detached passive M esozoic ExtensionD. Mesozoic Gi'au roof sequence

\ / 1

C. MainVariscan Compression

rarnplng of Variscan deformation

Thinning of Coal Measures 6. Early VariSCan Activity over PontypriddAnticline Thickenina of Coal Measures

Inversion of OR.S.+Lwt: Pal.Growth Fault A. Pre-Coal Measure Template

Carb. Lst.

km Coal IDl Lwr. Measures O.R.S.+LwrFault Pal.Growth Vertical---Trajectory of nextthrust and Horizontal in sequence

Fig. 10. Sequential restorationof the south-crop structureto illustrate the proposed modelof development. StageA shows the growth sequences in the Lower Palaeozoic andORS. Stage B invokes inversionof basement faultsto account for condensed sequences in the Westphalian of the Pontypridd anticline and the southeast-crop. StageC shows the developmentof a duplex beneath the Valeof Glamorgan and the overlying backthrust system in the Westphalian sequence. Stage D illustrates Mesozoic activity on the Moel Gilau fault.

Downloaded from http://pubs.geoscienceworld.org/jgs/article-pdf/148/5/881/4890755/gsjgs.148.5.0881.pdf by guest on 30 September 2021 890 J. A. JONES

The simplestmodel involves northward propagating et al. (1986) amajor backthrust at the mountain front is Variscan deformation migrating largely within the Westpha- most appropriate in view of the lack of asub-Pennant lian sequence across the present day Vale of Glamorgan. unconformity and the presence of considerable thrusting in Later uplift of the south-crop is then required to produce the coalfield foreland of the position of any tip fold. The thepresent day geometry. However, this model cannot south-crop may represent part of a triangle zone (Bossart explain the reduced subsidence along the south-crop during 1957) with the Pennant Measures forming a passive roof to theWestphalian (Jones 1989a) or the well-developed the deformation in the Lower and Middle Coal Measures. backthrustsystem, and also does not agree with recent The change in lithology fromthe incompetent Lower and seismic evidence (see below). Middle Coal Measures to the competent sandstones of the An alternativemodel transfers displacement into the Pennant Measures may have assisted the developmentof the CoalMeasures from the Lower Palaeozoic and Devonian upperdetachment with anabrupt vertical change in sequencebeneath the Vale of Glamorganand the Bristol deformation (Morley 1987). Channel (Fig. 10). The necessary ramp musthave been The triggermechanism forthe development of a situated north of the backthrust system and could have been mountain front geometry may be the sticking of a foreland the precursor of the Moel Gilau fault. From this position a propagating sequence, perhaps as a result of a facies change forethrustsystem propagated northward and acontem- along the line of thrust detachment e.g. the termination of poraneousbackthrust system developed against the coal horizons (a possible plane of easy slippage) at the limit south-crop in response to uplift of the area and pinning of of thecoal basin. In addition, the presence of basement the system in front of the leading branch line. This uplift, faults may have resulted in buttressing of the deformation initiated during sedimentation, may have been achieved by facilitating uplift of the south-crop and development of the inversion of Devonian and other Lower Palaeozoic growth backthrustsystem. Continued deformation following lock- faultsas described in southwest Dyfed by Powell (1989). up of the system would accentuate the structure and may Indeed an eastward projection of the Ritec fault along strike account for the piercement of the upper detachment along from southwest Dyfed would take that structure beneath the the south-crop by thrusts penetrating the Pennant Measures. southernpart of the coalfield. Furtherevidence for such The westwards increase in intensity of the deformation in faults is based on condensed sequences in the Westphalian theSouth Wales coalfield may bea consequence of the along the Pontypriddanticline and southeast-crop which arcuate form of the pre-existing Caledonian massif to the couldbe either footwall or hangingwallcondensed north of the basin.Unless the coalbasin significantly sequencesresulting respectively fromextensional or onlappedthe massif, sticking and buttressing probably compressional activity on pre-existingbasement faults. occurred earlier in the west than in the east. Owen (1954) considered the sub-Devonianbasement Althoughprecise geometries remain speculative, this exterteda strong influence onlater deformation of the model (Fig. 10) of majorramp development perhaps SouthWales area.Hartley & Warr (1990) takeup this accentuated by reactivationof, and buttressing against, theme and propose that pre-existingbasement faults have earlier faults is preferred since it more adequately explains beena major influence on basindevelopment throughout thestructural and stratigraphic relationships seen at the southwest England. Similarly, Price & Todd (1988), working surface and in the subsurface in South Wales. in the Irish Variscides, see a clear correlation between late Palaeozoicextensional growth faults and Variscan thrusts Conclusions and folds, the latter often an inversionof the former. Detailedinvestigations of BritishCoal opencast and A shortening estimate for the deformation beneath the deepmine sites has illustrated the complexity of deformation Vale ofGlamorgan can only atpresent be an educated of the Lowerand Middle Coal measures, with local guess. However, based on values of 40% in the Bristol area shortening estimates of up to 50%. Overall shortening for (Williams & Chapman 1986) and in SW Dyfed of 45% the coalfield may be in the order of 30%, which contrasts (Coward & Smallwood 1984) shorteningin the order of withestimates from southwest Dyfed of 45% (Coward & 30-50% is anticipated in the Bristol Channel and the Vale Smallwood 1984) and for the Bristol Mendips area of 40% of Glamorganareas. The thickenedsequence beneath the (Williams & Chapman 1986). Vale of Glamorganmight result partly or entirelyfrom Deformation has taken the form of a forethrust system thrust repetition of the Lower Palaeozoic sequence, perhaps throughout the central and northern coalfield and a major involvingdevelopment of aduplex structure. Support for backthrust system along the southcrop. The extent and style the modelcomes from published and unpublished seismic of deformation in theSouth Walescoalfield varies both reflection data e.g., Brooks et al. (1988). stratigraphically and geographically. The controlling factors Thesouth-crop of theSouth Walescoalfield displays are thought to be asfollows. geometriescomparable with those of mountainfronts or (a)The variation in mechanicalproperties of the frontalculminations (Butler 1986), viz: elevation of the sediments. Thecompetent sand-dominated Upper Coal CarboniferousLimestone above its regionallevel, steeply Measures remained little deformed and acted as a passive inclinedbedding (aforeland dipping panel) and a well roof to the thrust system. developed backthrust system. Vann et al. (1986) commented (b) Thepresence of pre-existingfaults. It is proposed that mountain fronts were commonly overlain by ‘molasse’ that pre-existing faults beneath the south-crop facilitated the which,although conformable, frequently exhibited con- formation of a triangle zone and mountain front structure. siderably less deformation than the underlying sequence. In (c) The proximity tothe Caledonian massif.Some the SouthWales coalfield thePennant Measures may geographicalvariations in deformation areattributed to represent such a deposit, conformable with, but much less buttressing the Caledonian massif. The closer proximity of deformedthan, the underlying Lower and Middle Coal the coalbasin tothe massif in the west hasresulted in Measures. Of the four possible solutions proposed by Vann increased intensity of deformation.

Downloaded from http://pubs.geoscienceworld.org/jgs/article-pdf/148/5/881/4890755/gsjgs.148.5.0881.pdf by guest on 30 September 2021 V A R IS C A N DEFORMATION, SOUTH WALESDEFORMATION,VARISCANSOUTH 891

1 am indebted to R. A.Gayer and British Coal Staff for their Isles. Specialpublication of the YorkshireGeological Society, Wigley, support during this NERC funded project. British coal are thanked 243-253. for permission to publish data (Figs4,6,7, 8). The comments of R. KELLAWAY,G. A. & HANCOCK,P. L. 1983. Structure of the Bristol District, A. Gayer,A. Hartley and referees have greatly improved the the Forest of Dean and Malvern fault zone. In: HANCWK,P. L. (ed.) The Variscan fold belt in the British Isles. Hilger, Bristol, 88-107. manuscript. KELLING,G. 1988. Silesian sedimentation and tectonics in the SouthWales basin: a brief review. In: BESLY,B. & KELLING,G. (eds) Sedimentation in a synorogenic basin complex, the Upper Carboniferous of northwest References Europe. Blackie, London. 38-42. MECHIE,J. & BROOKS,M. 1984. A seismic study of deep geological structure BOSSART,D. 0.1957. Relationships of the Porcupine Hills to early Laramide in the Bristol Channel area, S.W. Britain. Geophysical Journal of the movements.Alberta Society of Petroleum Geologistr, 7th annual field Royal Astronomical Society, 78, 66-689. conference guidebook, 46-51. MORLEY,C. K. 1987. Lateral and vertical changes of deformation style in the BROOKS,M., TRAYNER,P. M. & TRIMEILE,T. J. 1988. Mesozoic reactivation Osen-Roa thrust sheet, Oslo region. Journal of Structural Geology, 9, of Variscan thrusting in the Bristol Channel area, UK. Journal ofthe 331-343. Geological Society, London, 145. 439-444. OWEN, T.R. 1954. The structure of the Neath disturbance between Bryniau BUTLER,R. W. H. 1983. Balanced cross sections and their implications for Gleisonand Glynneath, SouthWales. Quarterly Journal theof the deep structure of the northwest Alps. Journal of Structural Geology, Geological Society of London, 109, 333-65. 5, 125-37. OWEN, T.R. & WEAVER,J. D. 1983. The structure of the Main South Wales -1986. Thrust tectonics, deep structure and crustal subduction in the Alps Coalfieldand its margins. In: HANCWK,P. L. (ed.) The Vararircan fold andHimalayas. Journal ofthe Geological Society, London, 143, belt in the British Isles. Hilger, Bristol, 74-87. 857-875. PRICE, C.A. & TODD,S. P. 1988. A model for the development of the Irish -1987. Thrust sequences. Journal of the Geological Society, London, 144, Variscides. Journal of the Geological Society, London, 145,935-939. 619-634. POWELL,C. M. 1989. Structural controls on Palaeozoic basin evolution and COWARD,M. & SMALLWOOD,S. 1984. An interpretation of the Variscan inversion in southwest Wales. Journal of the Geological Society, London, tectonics of S.W. Britain. In: HU~TON,D. H. W. & SANDERSON,J. D. 146,439-446. (eds) VariscanTectonics of the North Atlantic Region.Geological SEAGO,R. D. & CHAPMAN,T. D. 1988. The confrontation of styles and the Society, London, Special Publication, 14, 89-102. evolution of a foreland basin in central SW England. Journal ofthe ELLIOTT,T. & LADIPO,K. 0.1981. Synsedimentarygravity slides (growth Geological Society, London, 145,789-800. faults) in the Coal Measures of South Wales. Nature, 291, 220-222. SHACKLETON,R. M., REIS, A. C. & COWARD,M. P. 1982. An interpretation GAYER,R. A., ALLEN,K. C., BASSE~T,M. G. & EDWARDS,D. 1973. The of Variscan structures in southwest England. Journal of the Geological structure of the Taff Gorge area, Glamorgan, and the stratigraphy of the Society, London, 139, 553-541. OldRed Sandstone-Carboniferous Limestone transition. Geological SOUIRRELL,H. C. & DOWNING,R. A. 1969. Geology ofthe South Wales Journal, 8, 345-374. coa@eldPart 1 The Country aroundNewport (Mon). 3rd edition. HARTLEY,A. J. & WARR,L. N. 1990. Upper Carboniferous basin evolution Memoir of the Geological Survey of Great Britain. in SW Britain. Proceedings of rhe Ussher Society, 7, 212-216. VANN,I. R., GRAHAM,R. H. & HAYWARD,A. B. 1986. The structure of JONES,J. A. 1989a. Sedimentation and tectonics in the eastern part of the South Mountain Fronts. Journal of Structural Geology, 8, 215-227. Wales Coalfield. PhD thesis, University of Wales, Cardiff. WILLIAMS,G. D. & CHAPMAN,T. 1986. The Bristol Mendip Foreland thrust - 19896. The influence of contemporaneous tectonic activity on belt. Journal of the Geological Society, London, 143, 63-73. Westphalian sedimentation in the South WalesCoalfield. In: WOODLAND,A. W. & EVANS,W. B. 1964. Thegeology of the South Wales ARTHURTON,R. S., GUTTERIDGE,P. & NOLAN,S. C. (eds) Therole of Coa@eld (Part IV). The country around Pontypridd and Maesteg (3rd tectonicsin Deuonian and Carboniferous sedimentationin the Britkh edition). Memoir of the Geological Survey of Great Britain.

Received 18 July 1990; revised typescript accepted 30 April 1991.

Downloaded from http://pubs.geoscienceworld.org/jgs/article-pdf/148/5/881/4890755/gsjgs.148.5.0881.pdf by guest on 30 September 2021