The geological context and evidence for incipient inversion of the London Basin

Contexte géologique et arguments en faveur de l’inversion naissante du bassin londonien

R.C. Ghail*1, P.J. Mason2 , J.A. Skipper3 1 Department of Civil and Environmental Engineering, Imperial College London, London SW7 2AZ, UK

2 Department of Earth Sciences & Engineering, Imperial College London, London SW7 2AZ, UK 3 Geotechnical Consulting Group, 52A Cromwell Road, London SW7 5BE, UK

ABSTRACT A reappraisal of ground investigation data across London reveal that a range of unexpected ground conditions, encountered in engineering works since Victorian times, may result from the effects of ongoing inversion of the London Basin. Site investigation bore-hole data and the distribution of river terrace deposits of the Thames and its tributaries reveal a complex pattern of block movements, tilting and dextral transcurrent displacement. Significant displacements (~10 m) observed in Thames terrace gravels in borehole TQ38SE1565 at the Lower Lea Crossing, showing that movement has occurred within the last ~100 ka. Restraining bends on reactivated transcurrent faults may explain the occurrence of drift filled hollows, previously identified as fluvially scoured pingos, by faulting and upward migration of water on a flower structure under periglacial conditions. Mapping the location of these features constrains the location of active transcur-rent faults and so helps predict the likelihood of encountering hazardous ground conditions during tunnelling and ground engineering.

RÉSUMÉ Une réévaluation des résultats provenant d’études de terrain à travers Londres a révélé une diversité de phénomènes géologiques inattendus. Rencontrés lors de travaux d’ingénierie depuis l’époque victorienne ils pourraient être causés par les suites de l’inversion du basin londonien en cours. Les données provenant de forages et la répartition des sédiments de la terrasse fluviale de la Tamise et ses affluents révèlent un schéma complexe de mouvements de blocs et de failles de cisaillement inclinées ou dextres. Les mouvements significatifs (~10m) observés dans le forage TQ38SE1565 des graviers de la terrasse de la Tamise au niveau du Lower Lea Crossing indiquent un âge inférieur à ~100ka. Des courbes de retenue sur des failles de cisaillement réactivées pourraient expliquer l’existence de creux remplis de sédiments, précédemment catégorisés en tant que pingos érodés par des processus fluviaux, qui seraient plutôt créé par des failles et la migration ascendante d’eau sur des structures en fleur en conditions périglaciaires. En cartographiant ces caractéristiques, la localisation de failles de cisaillement pourra être contrainte et ainsi aider à prédire la probabilité de rencontrer des terrains hasardeux lors de percement de tunnels ou de travaux d’ingénierie.

1 INTRODUCTION

The London Basin, a wedge-shaped region ex-tending from Newbury to Rochester and Great Yar-mouth in southern England, is one of a number of post-Variscan sedimentary basins (Busby and Smith, 2001) NW-SE trending basement fractures that were reactivated as dextral transcurrent faults at various points in the Palaeogene and Neogene, causing inver-sion of the Hampshire and Weald basins (Blundell, 2002; Chadwick, 1993) the Weald, for example, has been uplifted by more than 1500 m (Jones, 1999a) since the Cretaceous. The inversion is likely a result of the combined compressive effects of the Alpine

collision and North Atlantic ridge push (Musson, 2007), but these forces appeared not to have affected the London Basin. However, a reappraisal of the data show that the geology of the London Basin, its pat-tern of drainage, and the nature of its periglacial fea-tures, all strongly indicate recent and ongoing inver-sion of the basin.

2 GEOLOGY AND TECTONICS

The London Basin (Figure 1) is one of a number of Permo-Triassic, post-Variscan, sedimentary basins (Busby and Smith, 2001) many of which were reac-

tivated during Neogene inversion (Chadwick, 1993). They opened above basement Variscan transcurrent and thrust faults (Chadwick, 1993) and while each opened at a slightly different time, all were generally depositional during the Mesozoic. However, the London Basin lies immediately north of the Variscan Front, on the margin of the Midlands microcraton, and is bounded by the chalk hills of the Chilterns and the North Downs. It formed a subaerial high throughout much of the Mesozoic but underwent complex and rapid deformation during the late Palae-ogene (Knox, 1996), and since then it has remained subaerial and quiescent (Gibbard and Lewin, 2003).

Figure 1. The Cretaceous/Tertiary transtensional basins and Var-iscan basement fractures of southern England. The region is histori-cally aseismic but nonetheless influenced by NW-directed Alpine stresses. (Developed from Musson 2007, with Geological Map Data BGS © NERC 2013 and Ordnance Survey Data © Crown copy-right/database right 2012. An Ordnance Survey/EDINA supplied service) Although its structural complexity was recognised nearly a century ago (Wooldridge, 1923). Faults have rarely been recognised in London (Aldiss, 2013) and evidence for recent fault displacement is rarer still. A very simple unfaulted synclinal model (Sherlock, 1947) persisted into the 1990s, when some faults

were tentatively added (Sumbler, 1996). The many instances of unexpected ground conditions encoun-tered in London since Victorian times (Chandler et al., 1998; Lenham et al., 2006; Mortimore et al., 2011; Newman, 2009) point to greater structural complexity than has ever been represented on geo-logical maps.

Countless site investigations have revealed brittle failures in Palaeogene sediments across London (Figure 2). A Neolithic trackway, dating from be-tween 1520 and 1100 BC, unearthed in 1993 in Beckton [TQ 427 820] was found to be cut by a fault which is infilled with clay (Greenwood and Maloney, 1994), implying a significant London earthquake within the last 3000 years. The Colchester Earth-quake of 1884, which occurred at the northern edge of the London Basin, was probably the most damag-ing in Britain in the last 400 years (Musson and Winter, 1996).

Figure 2. a) Low angle reverse faults (thrusts) in the Lambeth Group sediments beneath the Canary Wharf Crossrail station, 27 m below ground level. White box indicates the position of the detail shown in b) faults associated with periglacial features; c) Uplift of Thames terrace deposits by more than 20 m at Sheppey; and d) persistent tectonic shears in London Clay (exposed in a landslip). Scale bar is 0.5 m in each case.

Although Pleistocene (Berry, 1979; Hutchinson,

1980) and Holocene deposits (Akeroyd, 1972) mask the deeper tectonic features, a regional uplift of at least 0.07 mm a-1 across the Thames valley in the last 900 ka has been detected (Maddy et al., 2000). A higher rate in the last 400 ka has also been suggested, despite recent deglaciation (Nunn, 1983). Blundell (2002) estimates that only 50-70% of the uplift of the British Isles, Norway and France in the last 2.5 Ma years can be explained by glacier retreat and denuda-tional offloading.

Reactivation of basement transcurrent faults dur-ing the Palaeogene and Neogene has caused inver-sion of both Hampshire and Weald basins (Blundell, 2002; Chadwick, 1993). In the case of the Weald more than 1500 m of uplift has occurred since the Cretaceous (Jones, 1999a). Apart from a few scat-tered outcrops of Oligocene and Pliocene sediments, the end of the Palaeocene in the London area is marked by a gap in the stratigraphy of some 40 Ma; a significant period of subaerial conditions under a compressive regime between the Alpine orogeny and North Atlantic spreading (Musson, 2007), although it is not generally accepted that these forces affected the London Basin. However, a reappraisal of new and existing data in light of these observations shows that inversion is now affecting the London Basin (Royse et al., 2012); an interpretation in agreement with earlier inferences (Blundell, 2002; Wood and Johnson, 1978).

Evidence for fault reactivation and inversion can be found in the borehole records held at the British Geological Survey and available publically online. Approximately 2000 borehole records of more than 20 m depth were analysed for the London Basin Fo-rum Atlas Project through a number of student pro-jects. Borehole TQ38SE1565 in the Lower Lea Crossing [TQ 394 809] has a 10.3 m repeated se-quence of Kempton Park Gravel (described as Thames Ballast) and London Clay (Figure 3).

Figure 3. Cross-section in the Lea Valley/Thames confluence showing the repetition of Kempton Park Gravels and London Clay encountered in two boreholes, which has been interpreted to have been caused by thrusting.

3 DRAINAGE

The drainage network and chalk aquifer in London reflect a complex pattern of faulting (de Freitas, 2009) that appears to be controlled by the Variscan basement fractures noted above. The Thames is commonly considered to be a meandering river but closer inspection reveals a number of sharp bends and straight sections indicative of an active tectonic environment (Figure 4) with fault-controlled mean-ders.

Figure 4. Drainage patterns across London; note the sharp bends and straight sections on the Thames and Lea rivers. 1. Pre-Anglian drainage consists of NE-directed tributaries to a proto-Thames across the north of this map. 2. Modern drainage is almost perpen-dicular and ignores the path of the modern Thames. Places referred to in the text are: L=London, R=Richmond, H=Hampton, and W=Walton.

Furthermore, many tributaries to the Thames join

at a high angle, implying a regional slope independ-ent of the river floodplain. This relationship is a par-ticularly evident east of Richmond, including in the so-called lost rivers of London (Barton, 1992; de Freitas, 2009). Upstream of Richmond, the Thames takes a long detour south through Hampton, only turning north again west of Walton. These large-scale

deviations have previously been interpreted to result from river capture induced by glaciation (Gibbard and Lewin, 2003) but are consistent with a tighter structural control than that proposed by Gibbard et al. (1988), and are caused by the displacement of base-ment blocks during inversion.

Other than the Thames, the drainage pattern shows a dominantly SW—NE orientation of rivers south and north of London (indicated by dashed lines (1) in Figure 4) but a SE—NW orientation within London itself (dashed line (2) in Figure 4). Pattern type (1) is also apparent in the Chalk dry valleys south of Lon-don. These type (1) valleys predate the diversion of the Thames during the Anglian glacial and are in ex-actly the orientation expected of tributaries to the former course of the Thames.

So why are the modern tributaries almost perpen-dicular to this orientation? Indeed, why do several change course from that former direction across inner London, most notably the river Lea? Prior to the An-glian, the landscape had evolved over a very long pe-riod of stability (Gibbard and Lewin, 2003) into a mature fluvially-controlled landscape centred on the Thames flowing through St Albans and Essex. The major change to this pattern was not simply the An-glian ice advance 300–400 ka ago; if it were, the tributaries would retain their former orientation across London. Rather, it was the initiation of inver-sion some time later that caused this change in direc-tion by block tilting.

Note that the distribution of past terrace sediments lies almost exclusively to the north of the current riv-er course; this is also apparent in the sediments of the Lea, north of its abrupt change in course. Together with the change in tributary orientation, this distribu-tion implies a gradual tilting of the London basin upwards towards the NW, in contrast to the earlier tilt SE towards the Weald. Considerable uplift took place across southern England, and the Weald in par-ticular, within the last 2 to 3 Ma (Jones, 1999a, b) but apparently slowed or ceased about 400 ka ago (Maddy et al., 2000). However, strike-slip displace-ment on basement faults appears to have continued and driven uplift across London basin, probably as far north west as the Chilterns.

The faulted river terrace gravels in borehole TQ38SE1565 are inferred to belong to the Kempton Park member, dating from the early to middle

Devensian, 30 to 140 ka ago (Maddy et al., 2001). Averaged over that time, the 10.3 m displacement is consistent with fault reactivation by inversion and the uplift rates inferred earlier. Its location close to the confluence of the Lea and the Thames, the alignment the lower Lea and sediment distribution of the upper Lea are all consistent with recent, probably ongoing, dextral slip on a basement transcurrent fault under the lower Lea valley.

4 PERIGLACIAL FEATURES

While glaciation was clearly paramount in the di-version of the Thames, features associated with peri-glaciation may have had a deeper tectonic control. Throughout most of the last 400 ka, London has been under periglacial conditions, with the ground perma-nently frozen to a depth of perhaps tens of metres. The lower Lea, particularly at its confluence with the Thames, is known for its drift filled hollows: anoma-lous depressions in the London Clay infilled with al-luvium (Figure 5). These hollows are usually associ-ated with former flowing artesian areas close to the Thames and have been interpreted as fluvial scour features (Berry, 1979) in formerly open-system pin-gos (Hutchinson, 1980). However, certain aspects of these features, particularly the dyke-like diapiric in-jection of Lambeth Group sediments and even Chalk upwards into the London Clay, imply deeper driving processes.

We propose that restraining bends on the transcur-rent basement faults are the driving mechanism for producing drift filled hollows. Restraining (and re-leasing) bends are a well known feature of transcur-rent fault systems (Cunningham and Mann, 2007) and act to locally concentrate horizontal displace-ment on a transcurrent fault into vertical movement on steeply dipping faults in a flower structure.

In addition to fracturing and displacing the core of the flower structure upwards, water is also forced up from depth. During glacial times this water would have frozen in the near-surface permafrost, causing frost-heave that resembles ice wedge or pingo-type features and further damaging the ground. During the transition to interglacial times, flow rates in the Thames and its tributaries was much greater than, al-lowing the rapid scouring of the thawed, fractured

core of the flower structure to leave a deep oval de-pression at the surface, later infilled with alluvium of Devensian age (Hutchinson, 1980).

Figure 5. Perspective view of London Clay and Lambeth Group top surfaces inferred from ~2000 borehole records. Layer separa-tion is exaggerated for clarity. Note the SE opening depression, likely formed by Eocene tectonism exploiting earlier basement fractures, and the steep slope across the north which evidently dis-placed London Clay and may have been recently reactivated by inversion. Localised depressions (1), (2), and (3) may result from restraining or releasing bends on reactivated transcurrent faults, causing damaged ground that was exploited by periglaciation and end-glacial flood scour to form drift filled hollows, a significant engineering hazard. The drift filled hollow at (1) extends into the Lambeth Group below. The depression at (4) does not appear in the London Clay but is evident in the Chalk below; it may be a karst feature.

5 CONCLUSIONS

The straight sections and sharp bends on the Thames and its tributaries, as well as their peculiar orientation, imply and active tectonic environment that explains many of the unexpected features en-countered during tunnelling and ground engineering in London. An understanding of the geological histo-ry of the region shows that it is dominated by SE–NW oriented transcurrent basement faults. Dextral reactivation of these faults by the combined stresses of the Alpine orogeny and North Atlantic spreading has driven inversion on a number of basins in south-ern Britain such that it is reasonable to assume that the London Basin is simply the most recent example. Inversion is continuing at the present day; Mason et

al. (2015) infer current displacement rates of ~1 mm a-1 using satellite data.

Inversion offers a new insight into the origin of drift filled hollows, as scoured and periglacially al-tered restraining bend flower structures. The location of these features and mapping of the straights and sharp bends on the rivers in London constrains the location of active transcurrent faults and restraining bends (Figure 6) and so helps predict the likelihood of encountering hazardous ground conditions during tunnelling and ground engineering.

Figure 6. Summary map showing schematically the location of basement dextral transcurrent faults and blocking normal faults. The intersection of these act as restraining bends that may be re-sponsible for drift filled hollows. Notice also the alignment of trib-utaries across London parallel to transcurrent basement faults. Key features are: (1) normal faults may control straight segments on the Thames; (2) sediments on the upper Lea valley all lie to the north of the river, indicating block tilting towards the NW; (3) the unu-sual southward loop of the Thames may also result from block tilt-ing.

ACKNOWLEDGEMENT

Figures include Geological Map and Borehole Data BGS © NERC 2013 and Ordnance Survey Data © Crown copyright/database right 2012. An Ord-nance Survey/EDINA supplied service. The Lambeth

Group surface was generated in Move, supplied gra-tis by Midland Valley Exploration Ltd.

REFERENCES

Akeroyd, A.V., 1972. Archaeological and Historical Evidence for Subsidence in Southern Britain. Philosophical Transactions of the Royal Society of London. Series A, Mathematical and Physical Sciences 272, 151-169. Aldiss, D.T., 2013. Under-representation of faults on geological maps of the London region: reasons, consequences and solutions. Proceedings of the Geologists' Association. Barton, N., 1992. The Lost Rivers of London: A Study of Their Effects Upon London and Londoners, and the Effects of London and Londoners on Them, third ed. Historical Publications Ltd, Whitstable. Berry, F.G., 1979. Late Quaternary scour-hollows and related features in central London. Quarterly Journal of Engineering Geology and Hydrogeology 12, 9-29. Blundell, D.J., 2002. Cenozoic inversion and uplift of southern Britain. Geological Society, London, Special Publications 196, 85-101. Busby, J.P., Smith, N.J.P., 2001. The nature of the Variscan basement in southeast England: evidence from integrated potential field modelling. Geological Magazine 138, 669-685. Chadwick, R.A., 1993. Aspects of basin inversion in southern Britain. Journal of the Geological Society 150, 311-322. Chandler, R.J., Willis, M.R., Hamilton, P.S., Andreou, I., 1998. Tectonic shear zones in the London Clay Formation, Géotechnique, pp. 257-270. Cunningham, W.D., Mann, P., 2007. Tectonics of strike-slip restraining and releasing bends. Geological Society, London, Special Publications 290, 1-12. de Freitas, M.H., 2009. Geology; its principles, practice and potential for Geotechnics. Quarterly Journal of Engineering Geology and Hydrogeology 42, 397-441. Gibbard, P.L., Lewin, J., 2003. The history of the major rivers of southern Britain during the Tertiary. Journal of the Geological Society 160, 829-845. Gibbard, P.L., Rose, J., Bridgland, D.R., 1988. The History of the Great Northwest European Rivers During the Past Three Million Years [and Discussion]. Philosophical Transactions of the Royal Society of London. B, Biological Sciences 318, 559-602. Greenwood, P., Maloney, C., 1994. London Fieldwork and Publication Round-up 1993, London Archaeologist. Archaeology Data Service, York, p. 210. Hutchinson, J.N., 1980. Possible late Quaternary pingo remnants in central London. Nature 284, 253-255. Jones, D.K.C., 1999a. Evolving models of the Tertiary evolutionary geomorphology of southern England, with special reference to the Chalklands. Geological Society, London, Special Publications 162, 1-23. Jones, D.K.C., 1999b. On the uplift and denudation of the Weald. Geological Society, London, Special Publications 162, 25-43. Knox, R.W.O.B., 1996. Tectonic controls on sequence development in the Palaeocene and earliest Eocene of southeast England: implications for North Sea stratigraphy. Geological Society, London, Special Publications 103, 209-230.

Lenham, J., Meyer, V., Edmonds, H., Harris, D., Mortimore, R., Reynolds, J., Black, M., 2006. What lies beneath: surveying the Thames at Woolwich. Proceedings of the ICE - Civil Engineering 159, 32-41. Maddy, D., Bridgland, D., Westaway, R., 2001. Uplift-driven valley incision and climate-controlled river terrace development in the Thames Valley, UK. Quaternary International 79, 23-36. Maddy, D., Bridgland, D.R., Green, C.P., 2000. Crustal uplift in southern England: evidence from the river terrace records. Geomorphology 33, 167-181. Mason, P.J., Ghail, R.C., Bischoff, C., Skipper, J.A., 2015. Detecting and monitoring small-scale discrete ground movements across London, using PSInSAR. Proceedings XVI ECSMGE, 0221. Mortimore, R., Newman, T.G., Royse, K., Scholes, H., Lawrence, U., 2011. Chalk: its stratigraphy, structure and engineering geology in east London and the Thames Gateway. Quarterly Journal of Engineering Geology and Hydrogeology 44, 419-444. Musson, R.M.W., 2007. British earthquakes. Proceedings of the Geologists' Association 118, 305-337. Musson, R.M.W., Winter, P.W., 1996. Seismic hazard maps for the U.K. Natural Hazards 14, 141-154. Newman, T., 2009. The impact of adverse geological conditions on the design and construction of the Thames Water Ring Main in Greater London, UK. Quarterly Journal of Engineering Geology and Hydrogeology 42, 5-20. Nunn, P.D., 1983. The Development of the River Thames in Central London during the Flandrian. Transactions of the Institute of British Geographers 8, 187-213. Royse, K.R., de Freitas, M., Burgess, W.G., Cosgrove, J., Ghail, R.C., Gibbard, P., King, C., Lawrence, U., Mortimore, R.N., Owen, H., Skipper, J., 2012. Geology of London, UK. Proceedings of the Geologists' Association 123, 22-45. Sherlock, R.L., 1947. London and the Thames Valley, 2nd Edition ed. His Majesty's Stationary Office, London. Skipper, J.A., 2015. The Lambeth Group in the Crossrail Project of London, UK - the geological model. Proceedings XVI ECSMGE, 0445. Sumbler, M.G., 1996. London and the Thames Valley, Fourth Edition ed, London, p. 173. Wood, W.R., Johnson, D.L., 1978. A Survey of Disturbance Processes in Archaeological Site Formation. Advances in Archaeological Method and Theory 1, 315-381. Wooldridge, S.W., 1923. The minor structures of the London Basin. Proceedings of the Geologists' Association 34, 175-IN171.