OR/13/006 Engineering geology
|Entwisle, D C, Hobbs, P R N, Northmore, K J, Skipper*, J, Raines, M R, Self, S J, Ellison, R A, and Jones, L D. 2013. Engineering geology of British rocks and soils - Lambeth Group). British Geological Survey. (OR/13/006).|
* Geotechnical Consulting Group (GCG)
The complex and variable lithologies of the Lambeth Group give rise to variable ground conditions. This has been determined largely by the variation of depositional environments and by post-depositional processes such as erosion, weathering, leaching, pedogenic processes, soil formation processes, tectonics, and dissolution of chalk beneath the Lambeth Group. Thus the engineering geological considerations of the Lambeth Group are determined directly by this variability and complexity, and also the location and nature of engineering applications and groundwater conditions. The history of engineering within the Lambeth Group dates back to the beginnings of the industrial revolution in Britain when the pioneering engineers of the modern era, such as Brunel, Beamish, Trevithick, and Page, developed the now familiar techniques of tunnelling, excavation, shaft-sinking, drilling, and piling, much of it in east London, included within the Lambeth Group deposits. Simultaneously, these engineers shed new light on London’s geological formations. The expansion of a wide variety of large-scale engineering activity focused on central and east London during the last twenty years of the 20th century and beginning of the 21st century coincided with the distribution of deposits of the Lambeth Group, and greatly increased the knowledge of these deposits. Developments included the Jubilee Line Extension (JLE), the Limehouse Link (LL) Road, Union Rail and the Channel Tunnel Rail Link (CTRL) 2. Many such projects have evolved from London’s need to extend its transport links to the southeast.
Many of the early 19th century. projects represented firsts for engineering worldwide; for example, the Thames Tunnel (1843) was the first sub-aqueous public tunnel and the first use of a tunnelling shield (Skempton and Chrimes, 1994). These early projects proved extremely difficult because of the variable properties of the Lambeth Group and gave rise to novel technologies. Whilst the London Clay Formation has long been known as a consistent and reliable tunnelling and excavation medium, at least within the bounds of traditional techniques, this could not be said of the Lambeth Group.
Perhaps of greatest note have been the tunnels beneath the Thames, and elsewhere in London, starting with the unsuccessful Thames Driftway begun in 1806 and ending with phase 2 of the CTRL (begun 2000). Also included are the Thames, Rotherhithe, Blackwall (1 and 2), Dartford (road and rail), Victoria Underground Line, Jubilee Line Underground Extension Limehouse Link (cut and cover), Docklands, Thames-Lee Water Main (TLWM), and Thames Water Ring-Main (TWRM) tunnels. Many of the early tunnels beneath the Thames were, in places, just a few metres below the riverbed. Consequently, natural undulations or channels within the clay would result in funnelling-down of the soft or loose sediments into the face and catastrophic flooding of the works. Techniques used to combat soft ground and groundwater ingress have included compressed air (first proposed in 1826), riverbed sealing, grouting (cement and chemical), and ground freezing (Skempton and Chrimes, 1994; Clarke and Mackenzie, 1994).
The limited outcrop of the Lambeth Group and variable lithologies has meant that surface problems, such as shrink/swell, have been less prevalent and less well documented than the extensive London Clay Formation outcrop. This applies equally to coastal exposures and to landslides, the occurrence of which are low due to the limited exposure and subdued landscape. However, cuttings for railways (e.g. Park Hill railway cutting, Croydon in 1883 [Klassen, 1883]) and roads (Southwark landslide on A259 at Shoreham, W. Sussex in 1957) have resulted in engineering-induced slope failures within the Lambeth Group. Recently, wastewater schemes on the south coast have encountered the Lambeth Group (Hight et al., 2004).
Best practice for undertaking ground investigations in the Lambeth Group has been addressed by Hight et al. (2004). In particular, the report recommendations attempt to address the problems of inappropriate and incomplete sampling that has, using past conventional site investigation practice, resulted in missing key layers (such as hard bands and water-bearing sands) and in the poor quality of ‘undisturbed’ tube-samples that has resulted in almost certain underestimation of both strength and stiffness.
General considerations for ground investigation
Major projects, including tunnels, major excavation and underground structures, in which a good understanding of the Lambeth Group and its lithological variation is essential, clearly need more detailed information than routine piled foundation. For some major projects involving sensitive structures it may be necessary to have high quality, well-logged, rotary core at up to 50 m spacing. However, few projects will plan for this type of investment at the ground investigation stage and 100 m spacing are more common even in well-funded projects. If 100% rotary boreholes are considered too expensive, a mixture of rotary and cable percussion boreholes can enhance information since SPT and water strike data can be used to aid boundary definition and identifying water-bearing units. Cable percussion boreholes can also recover more accurate thicknesses of Upnor Formation gravel beds, which are rarely 100% recovered in rotary-cored boreholes. For less sensitive projects, such as housing development, site investigation may not support the expense of rotary drilling so cable percussion and pits may be used. If this is the case then 0.5 m spacing jar samples and 1 m spacing split U100’s may be adequate. Also, SPT drives should be carried on to 100 or 200 blows in the more resistive materials. As with rotary drilling, the quality of the logging is very important.
Site supervision of the ground investigation should be costed into the specification. Expensive rotary core will be of limited use if best care is not taken when drilling. Drilling may be rapidly carried out but may not recover the sand units or gravel beds precisely the layers that need to be characterised (see Lithology and stratigraphy). Runs of less than 1.5 m drilled more slowly should improve recovery.
After the quality of the driller, a most important factor is the experience and knowledge of the logger who should preferably have undertaken specific training on the Lambeth Group lithologies and lithostratigraphy. Logging needs to take place under conditions of good lighting, strictly adhering to BS EN ISO 14688-1 (BSI, 2002 et seq.) and BS 5930 (BSI, 1999 et seq.). Although core needs to be kept undisturbed and intact for laboratory testing purposes, the testing will not fully characterise the ground unless the stratigraphy of the sample is known. Most cores can be split open with palette knives or spatulas or similar tool. Alternately, scrape the surface off before logging, since the outside of the core will be contaminated with drilling fluid and disturbed sediment. Colour differences are very important and the use of colour codes such as the Munsell© colour charts (Munsell, 2009) or similar should be used. Colour photographs in suitable light with colour, monochrome and measurement scales of rotary core material are essential.
Lithology and stratigraphy
The Lambeth Group is highly variable and may therefore still surprise during ground investigations. Training in, or experience of, this stratigraphy; high quality ground investigations; and sound interpretation skills are essential to best construction practice and, hence, cost-effective decision making.
Identification of stratigraphical boundaries
Thanet Formation and Upnor Formation Boundary
Identification of the boundary between the Thanet and Upnor formations is often a problem in the London Basin area as far west as central London where the Upnor Formation directly overlies the Thanet Formation.
The boundary between the two may be clear in some boreholes as the colours or textures are different. The Upnor Formation is often darker, greener, of coarser (medium) sand and clayey, making it ‘stickier’ when reworked with water; the Thanet Formation often feels clean and the grain contacts can be easily heard when rubbed between the fingers because of the angular grains and lower fines content.
However, the boundary can be problematic for three main reasons:
- Both formations were deposited in similar sedimentary environments, occasionally leading to similar lithologies. This is particularly so in the north and west of London where only the basal part of the Thanet Formation (which is darker, contains more fines and is most similar to the Upnor Formation) is present.
- Much of the basal Upnor Formation is derived from reworked Thanet Formation materials.
- The contact between the two formations is frequently intensely burrowed, with an irregular contact (Figure 7.1) and in some cases, an indistinct, diffuse contact.
Generally, if gravel, has an appreciable clay content, with distinct clay laminae, abundant medium sand size glauconite or fossil shells are present, the sediment is Upnor Formation (Figure 7.2).
Standard Penetration Test (SPT) N-values from cable percussion boreholes can be useful in determining the boundary. N-values frequently rise sharply and stay consistently higher within the Thanet Formation, reflecting their greater density and (generally) lower clay content. In the Upnor Formation N-values are extremely variable.
Logging these sediments is best done both when freshly excavated and also, if in doubt, after a couple of hours exposure to air as the differences between the two often become more distinct during drying.
Upnor Formation and lower Reading Formation Boundary
The boundary between the Upnor Formation and lower Reading Formation is probably the most problematic of the Lambeth Group boundaries to distinguish. The strict definition of Reading Formation is of ‘continental facies’ dominantly ‘clay’, which is ‘mottled due to pedogenic processes in a humid environment’ (Ellison, 1983; Buurman, 1980). However, where the Lower Mottled Clay is thin the Upnor Formation is also pedogenically altered, which, in those areas affected, makes separating these two units extremely difficult. This is not only due to changes in colour but also the translocation of clay downwards into coarser material. A pragmatic method of classification was proposed in Page and Skipper (2000) (and is commented on in the Geology section) in which the pedogenically altered Upnor Formation and the Lower Mottled Clay are combined into the Lower Mottled Beds. Core loggers from this part of the sequence still need to make decisions about labelling units. This system used the following guidance:
- - If the sediment is primarily glauconitic, shelly and with black rounded flint gravel, or a laminated sand and clay/silt (frequently with fine but abundant glauconite), it should be labelled ‘Upnor Formation’. On the other hand if colour mottling, and fractured and coloured flint pebbles (they go brown, then white or red, then fall apart when they are weathered — see Figure 7.3 and Figure 7.4) are the primary features seen, then ‘Lower Mottled Beds’ (Lower Mottled Clay and mottled Upnor Formation) is the best assignation. If there is still an indeterminate boundary in between, then be honest and say so: ‘gradational boundary over 200 mm consisting of…’
This method allows comparison with older borehole records — if the words ‘mottling’, ‘multicoloured’, or ‘angular gravel’ are recorded, it strongly suggests the presence of the ‘Mottled Beds’.
This does not agree with the geological methods, which separate by the original depositional environment, but provides a pragmatic method that is generally easier to use than the standard geological requirement.
The Lower Mottled Beds (Page and Skipper, 2000) includes the Lower Mottled Clay and the pedogenically altered or mottled Upnor Formation (Aldiss, 2013). This classification should be used if possible. As such gravel beds are from the Upnor Formation, but where pedogenically altered, may be described as mottled Upnor Formation. However, separating the Upnor Formation from the Lower Mottled Clay can still be very difficult in some areas.
Differentiating between upper and lower mottled clay
One of the most useful ways of to differentiating between the Upper and Lower Mottled Clay when logging core is the identification of the mid-Lambeth Group Hiatus (MLH — see page 15 and Skipper, 2000), which separates the Upnor Formation and lower Reading Formation beneath from the Woolwich Formation and upper Reading Formation (which are above it). The mid-Lambeth Group Hiatus is the most useful horizon (and only true planar horizon) in the Lambeth Group, and represents the period when sedimentation ceased - between the Upnor Formation/Lower Mottled Clay deposition, and the transgression by the next depositional phase, the Woolwich Formation. Because there is such a sharp contrast in the depositional conditions between the two periods (from low sea level with much weathering to a sea level rise and drowning of former land surface), it can be recognised in core by the following:
- A sharp change down core/section from reduced (grey/black/blue) sediments to oxidised (pale, often mottled or multicoloured yellow, red or reddish brown), sediments (Figure 7.5). This is also the case in the Hampshire Basin where the Upper Mottled Clay directly overlies Lower Mottled Clay (Figure 7.5).
- In the London area, a sharp change from the Woolwich Formation shelly clays, to the first appearance of calcrete layers and Lower Mottled Clay or mottled Upnor Formation.
- In some cores from the London area, a confusing mixture of both mottled sediments AND shelly grey clay may be present just below the MLH. This is caused by crustaceans, (such as crabs), which lived in the Lower Shelly Clay sediments and dug large burrows (up to 30 mm in diameter and 2 m deep) into the underlying Lower Mottled Clay or Upnor Formation. As sea level rose, the burrows were abandoned and in filled with Lower Shelly Clay (Figure 7.7).
NB. If all else fails, the Lower Mottled Clay is generally sandier than the Upper Mottled Clay, and has a greater range of colours.
Occurrence of hard bands/layers
Hard bands or layers present a significant problem for drilling (see Hard bands). They are most common approximately a metre below the mid-Lambeth Group Hiatus (see Differentiating between upper and lower mottled clay) in the Lower Mottled Clay or the upper part of the Upnor Formation when the Lower Mottled Clay is thin or absent. This is because the MLH represents the period when little or no sedimentation took place over a wide area of SE England, and the prevailing sub-tropical climate led to the formation of a variety of duricrusts at or near to the ground surface. Over most of the Greater London area, the most commonly encountered duricrust is calcrete (calcium carbonate concretion), which can vary from just a few white nodular lumps, to a layer up 1.5 m thick. In areas of central to east London where the thickest calcrete is sometimes found, it can halt drilling progress and may damage drilling equipment (Figure 7.8). In some areas of outer northeast London to areas of Hertfordshire, Bedfordshire and Buckinghamshire, a very strong, siliceous, gravelly duricrust or silcrete, called Hertfordshire Puddingstone, (Figure 7.9) may occur. It may up to 750 mm thick and has again been known to cause damage to cable percussion drills.
In the east Kent area and around Newhaven in Sussex, a third type of duricrust — ferricrete (iron-cemented concretion) may be found (Figure 7.10) and includes the Winterbourne Ironstone. Although ferricretes are not as strong as the other two types of duricrust, they can be up to 750 mm thick and have been used locally in Kent as a building stone and mined as an iron ore.
Less commonly, hard layers occur in the Woolwich Formation. In the upper part of the Laminated Beds, a hard, grey-brown, siderite-cemented layer up to 100 mm thick may be present, especially from east London into the Essex. Drillers have reported difficulty penetrating this layer. In addition, the layer can cause confusion in the logging of cable percussion boreholes, where they have been miss-interpreted as being concretions in the London Clay Formation.
In the Upper Shelly Clay of the Woolwich Formation there are shelly limestones beds. The most well-known is the Paludina Bed, which is present between Dulwich [TQ 340 726], New Cross [TQ 363 764] and Honor Oak [TQ 355 744] in south London (Figure 7.11). This unit, is up to 300 mm thick, consists of extremely weak, thinly laminated mid to dark grey mudstone often contains fossil freshwater gastropods (the Paludina in question). It is a lacustrine deposit deposited during a freshwater incursion. Other cemented shellbeds are also found throughout the Upper Shelly Beds, but are rarely greater than 200 mm thick, and are not consistently cemented over distances more than a few tens of metres.
Presence of water-bearing beds and lenses
All the units in the Lambeth Group may contain high permeability units, but the likelihood will vary depending on the unit and location of the project.
A simplified summary is given below:
Sandy gravel beds up to five metres thick may be present in the upper part of the Upnor Formation (see Figure 7.12 for where they are most commonly encountered) where they have been deposited in large relatively steep-sided channels. They are generally water-bearing unless drained or unconfined (see Figure 7.13).
Reading Formation — Lower Mottled Clay
Permeable lenses appear to be relatively rare in central London but may be more common elsewhere. In east London and further east the Lower Mottled Clay tends to be sandy.
Woolwich Formation — Lower Shelly Clay
Occasional sandy beds (generally up to 150 mm thick but rarely, thickening to channels up to 2 m deep) (see Figure 7.14).
Woolwich Formation — Laminated Beds
Permeable units are common within the Laminated Beds, occurring as beds and channel infills of fine grey sand from 50 mm to 2 m thick, and locally thicker. The sands often contain disseminated pyrite, organic matter (lignite) and may also contain minor glauconite. The channels are steep-sided, up to 5 m wide, have very good connectivity with water/air, and widespread but can be difficult to find in site investigation boreholes, especially if sampling is not continuous (see Figure 7.15 and Figure 7.16).
Reading Formation — Upper Mottled Clay
Sand-filled channels in the Upper Mottled Clay are relatively rare in the London area, but occur within the main body of the unit or cutting down into the top of the unit. Although they have not yet been observed in open excavations in the London area they are present in excavations to the west of London, for example in the Newbury area. Here, they are commonly 2 to 3 m wide and 1 to 2 m thick within the mottled lithologies. In the Hampshire Basin they are laterally extensive, up to 100’s of metres, fluvial sand beds generally 1 to 2 m thick.
Sand-filled channels in the Upper Mottled Clay can be up to 9 m deep in London, for instance in Bermondsey excavated during the construction of the Jubilee Line Extension (Figure 7.17). The channel was in direct hydraulic continuity with local aquifers and required extensive dewatering measures. In central London a sand-filled channel measured up to 12.5 m deep and up to 200 m wide has been identified.
Note that, because of the likelihood of highly permeable lenses within the Lambeth Group, piezometer results should be interpreted in conjunction with a good understanding of the lithologies and lithostratigraphy.
Effect of sampling methods on sample quality
An important part of site investigation is the provision of suitable samples for geotechnical testing to provide relevant data for design. Sampling methods can significantly affect the quality of acquired samples and, therefore subsequent test results. Where ‘undisturbed’ samples are required then the use of sampling techniques that damage the sample may result in non-representative values, which in most cases result in data that provide conservative design parameters.
Sampling techniques are discussed in Hight et al. (2004) and are summarized in Table 7.1. However, note should be taken of the requirements of Eurocode 7 part 2 (BSI, 2007) and British Standard sampling methods (BSI, 2006).
|Block sampling||Good samples where carefully taken in appropriate situations.||Time consuming if done correctly. Limited to surface, pits and shafts.|
|Cable percussion U100 thick-wall samples||Relatively cheap, well-known technique. Standard penetration tests provide some physical data.||Likely to result in sample disturbance, particularly in stiff and ‘cemented’ material; laboratory strength and stiffness results unlikely to be representative of in situ conditions. Use for laboratory tests requiring ‘undisturbed’ samples will need to be justified.|
|Pushed thin-wall tube sampler||Provides good quality samples.||Only suitable for sampling a limited range of material, i.e. those with undrained shear strength <~250 kPa.|
|Rotary, double-tube core barrel||Generally provides good samples with some provisos.||Fairly expensive. Some disturbance of samples is likely. Drilling fluids may contaminate and/or soften the sample. Sample may be disturbed when removed from the core barrel.|
|Rotary, triple-tube core barrel||Good quality samples, Successfully used with biodegradable mud flush on a number of projects.||Expensive; not all ground investigation companies can provide this service.|
|Wireline drilling with triple tube core barrel||Good quality samples. Successfully used on a number of projects.||Expensive; only a limited number of companies provide this service. Drill bit cannot be inspected until the drill string has been withdrawn.|
Sample disturbance and core loss may occur even with careful drilling; the latter is most common in the gravels of the Upnor Formation, and sand throughout, whereas core loss in the Reading and Woolwich formations outwith coarse grained beds is generally rare, but is encountered in faulted strata.
Hight et al. (2004) note that sample disturbance is particularly likely to occur in laminated or closely interbedded units as negative pore pressures (or suctions) are likely to develop in acquired samples when taken from the ground and relieved of their in situ boundary stresses. Inevitably there will be a redistribution of water contents from more permeable silt/sand layers to less permeable clays, which will swell and increase in water content. More permeable layers will also increase the swelling process and take-up of water from drilling flush or seepage into the borehole during rotary coring or boring. Where this occurs, water content measurements in laminated units will not be representative of in situ conditions.
Other drilling-related recommendations described in Hight et al. (2004) are:
- Polymer flushes have been used successfully for rotary drilling.
- Drilling-induced sample disturbance may be reduced by the use of face discharge drill bits, and tungsten or diamond impregnated bits on ‘hard’ bands.
- Recommendations for drill bits to be used in the Lambeth Group (Beckwith et al., 1996) are:
In the Reading and Woolwich formations:
- - use of diamond saw-tooth bits,
- - switching to tungsten carbide bits when problems of core loss encountered and changing from biodegradable polymer mud flush.
In the Upnor Formation:
- - use of diamond-set, face discharge bits.
Tunnels and shafts
Tunnelling has been the engineering activity most associated with the Lambeth Group strata, particularly beneath the Thames during the early 19th C, and then throughout the 20th C, notably with the development of the London Underground system. An extensive review of tunnelling in the Lambeth Group is given in Hight et al. (2004). Much pioneering work was carried out in overcoming the difficulties posed by these strata. Preventing unacceptable entry of water ahead of, and around, a tunnel face has been a key factor in creating a workable and safe tunnelling environment within Lambeth Group strata. Hence, the absolute and relative permeability, and permeability anisotropies of the Lambeth strata, as well as water from the Thanet Formation and Chalk beneath, have been crucial factors in the design and execution of engineering remediation such as de-watering, ground-freezing, and compressed-air. Groundwater inflow, and in some cases direct river inflow, have become infamous in the Lambeth Group strata from the early days of the Thames Tunnel onward. These events have not only flooded workings, but have created voids above the face and behind tunnel linings due to the displacement of sand, silt, and occasionally clay blocks. Prior to the development of modern closed-face tunnelling techniques, the high permeability of formations within the Lambeth Group discouraged the development of the Underground system into southeast London and the Lambeth Group subcrop (Hight et al., 2004).
Early Thames tunnels
The first attempt to build a tunnel under the Thames, from Rotherhithe to Limehouse, was made by Ralph Todd between Gravesend and Tilbury in 1798. This failed due to lack of money and inflow of sand. The second attempt was the so-called Thames Driftway by Robert Vazie and Richard Trevithick. The work began in 1805, within the Upnor Formation, by hand tunnelling from Rotherhithe, but was halted in 1808 due to running sands from the Laminated Beds (Skempton and Chrimes, 1994; Hight et al., 2004).
The Thames tunnel from Rotherhithe to Wapping in East London, was begun in March 1825 with a shaft at Rotherhithe on the south side of the river, and eventually opened to the public in March, 1845 (Muir-Wood, 1994). The world’s first articulated tunnelling ‘shield’, designed by Marc Brunel, was launched in December 1825. A segmented rectangular shaped structure, made of cast iron, was settled on following rejection of a circular shape. The 366 m long tunnel was driven throughout in Lambeth Group strata, which dip slightly from south to north, in the direction of the tunnel drive.
The cross-section of the tunnel was large and unusual in that it was essentially a twin tunnel with an interconnecting arched gallery. Site investigations were begun in 1824, revealing a stratum of ‘blue alluvial earth inclining to clay of sufficient depth’, which was at the time ‘found to resist infiltration’. This is thought to refer to the Upper Mottled Clay (Muir-Wood, 1994). Tunnelling revealed this not to be the case, but more a case of ‘insufficient’ depth of clay and much of the underlying sands (Ferruginous Sand and Lower Mottled Sand). From an early stage, many inundations from the riverbed damaged the shield, made conditions extremely difficult, and culminated in the development of a swallow-hole and cessation of work between 1828 and 1835. It is believed that some of the inundations were due to the presence of natural scour hollows, up to several metres in depth, in the Lambeth Group strata, and hence critical thinning of the clay overburden (Skempton and Chrimes, 1994). In the most difficult stages of tunnelling the clay overburden (as little as 2 m thick) often cracked and subsided by as much as two or three metres due to loss of the underlying sand. However, in the later stages this clay was thick enough (>3 m) to remain stable despite sand loss due to ground water, rather than river, inflow. It is not clear to what extent these fissures were natural or induced by the water pressure. The inundations were tackled by laying clay-filled bags on the riverbed, sometimes using a primitive diving bell. The silt layers at depth within the Lambeth Group, which had been included with the clay on drill logs as ‘uniform tenacious clay’, did give engineering problems in the tunnel, but only where the sand content was high and runs occurred (Skempton and Chrimes, 1994). A major hazard, particularly in the later stages was inflow of sewage and methane from the riverbed, the latter igniting on many occasions (Muir-Wood, 1994). Currently the tunnel forms part of the East London Underground branch, and has recently been relined.
The 1st Blackwall tunnel connecting Poplar (north) with Greenwich peninsula (south), close to the present Millennium Dome, was built in 1897 and now carries northbound traffic only. The tunnel was constructed using compressed air as the strata consisted mainly of sands (Upnor and Thanet formations). As the overburden between tunnel roof and riverbed was sometimes as little as 1.7 m, the compressed air frequently blew holes in the riverbed, resulting in spectacular waterspouts. Clay blankets up to 3 m thick were routinely employed on the riverbed to counter compressed air loss and inundation of the tunnel. This was a development of the clay bags used on the Thames Tunnel, but the more generous navigation clearances downstream at Blackwall allowed a rigorous solution to a now familiar problem (Skempton and Chrimes, 1994). The eastern tunnel was opened in 1967 and carries southbound traffic. The western tunnel, narrower than the eastern, had been built by Sir Alex Binnie using James Greathead’s shield (as for Tower Subway) and opened in 1897.
Jubilee Line Extension (JLE)
The Jubilee Line Extension (JLE) extends from Charing Cross to Stratford, crossing the Thames four times. The tunnels lie mostly within Lambeth Group strata between London Bridge and Canada Water stations, and also at Canary Wharf and North Greenwich Stations (Ellison et al., 2004). The cross-section is of the line between Green Park and the given in Appendix 2 Figure A.2.9.
Within the Lambeth Group strata the tunnels were driven using a pressurized closed-face tunnel boring machine (TBM). Bermondsey Station was excavated using a combination of tunnels and deep-box. Specialised rotary trench cutting diaphragm wall tools were used in the Lambeth Group to penetrate limestone layers. Compressed air working methods (to reduce water ingress) were also used on the Central Line west of Stratford in the 1890’s, and on other tunnelling projects since. The log of Jubilee Line Extension borehole 404T contains all the units of the Lambeth Group sequence (Ellison, 1991). The information on the borehole including a generalised section, lithostratigraphy, core photographs, geologist’s and site investigation description is in Appendix 1.
Channel Tunnel Rail Link (CTRL)
The Channel Tunnel Rail Link (CTRL), which became High Speed 1 (HS1), between Folkestone and London was fully opened in November 2007. Phase 1 of the CTRL was constructed during the 1990’s to provide a high-speed rail route from the Channel Tunnel at Folkestone, via Ashford, to London. It was opened in 2003 with Eurostar services terminating at Waterloo, but with the route west of Fawkham Junction on normal (low-speed) track. Phase 2 takes the route from Dartford to St Pancras, crossing the Thames between Swanscombe and West Thurrock, and completing most of the route west of Dagenham, via Stratford, in 19 km of tunnel. With the exception of the two ends, the tunnels forming the Barking to St Pancras section of Phase 2 are driven through Lambeth Group strata. The geological cross-section along the route between Stratford and Rainham is shown in Appendix 2 Figures A.2.15 to A.2.16.
The Limehouse Link is a 1.8 km dual-carriageway section adjacent to the A13 trunk road, constructed in the early 1990’s for the Docklands Development Corporation, linking Canary Wharf with the City of London. Most of the route was excavated to form a deep cut-and-cover tunnel. This necessitated massive reinforced concrete diaphragm walls up to 33 m deep and 42 m wide, constructed using a top-down method. In order to cross the Limehouse Basin a section of bottom-up construction within a cofferdam was required (Glass and Powderham, 1994). The Woolwich and Reading formations are at a depth of about 14 m and are overlain by a thin London Clay Formation, Thames Gravel, and Fill. Underdrainage to the chalk was observed both within the Thanet Formation and the Upnor Formation. Elevated pore pressures within the Laminated Beds were relieved by a series of wells, which connected it hydraulically with the under-drained Thanet Formation (Hight et al., 2004). Horizontal stresses were determined by self-boring pressuremeter tests to give values of K0 of 2.5 for the Lambeth Group clays compared with 1.5 for the London Clay Formation.
Docklands Light Railway
The Docklands Light Railway, Lewisham Extension (DLRLE) runs between Mudchute Station on the Isle of Dogs to Lewisham Station, crossing the Thames in twin tunnels from Island Garden Station on the north bank of the Thames 4.2 km to Greenwich Station on the south bank (Sugiyama et al., 1999). The tunnels were driven using a slurry shield method. More than half of the tunnel was constructed within Lambeth Group strata overlain by Terrace Gravels (Appendix 2 Figure A.2.18). A large proportion of this was within the clay formations.
Greenwich pedestrian tunnel
The Greenwich pedestrian tunnel, built in 1902, connects Greenwich Pier with the Isle of Dogs, running close to the JLE. The tunnel is 2.7 m in diameter and 366 m long. It was built to allow workers to travel from Greenwich to the docks on the north bank. The tunnel is accessed via shafts containing lifts and stairs. The north shaft is close to the Island Gardens station on the DLR, and the southern shaft close to the Cutty Sark.
Thames Water Ring Main (TWRM)
The Thames Water Ring Main (TWRM) was constructed to provide 80 km of 2.5 m diameter tunnel, several shafts, and three large underground installations linking London’s water treatment works along two principal NE–SW routes: one north and the other south of the Thames. Serious inundation took place at Tooting Bec (between Streatham and Brixton) due to high pressures within the Thanet Formation. Whilst most of the tunnels (including all of the North London section) were driven in London Clay, the 9.5 km of tunnel within the Lambeth Group were found to have only moderate water pressures (Farrow and Claye, 1994). Different tunnel lining methods were used in the London Clay and the Lambeth Group, that in the former being cheaper and quicker. Ground freezing and grouting were used to control ground water ingress (Clarke and Mackenzie, 1994).
Difficulties experienced during tunnelling
- Lithological variability
- a. Variable strength/density (sometimes associated with hard bands) leading to:
- Difficulties in controlling the alignment of the tunnelling machine,
- Variable jack pressures.
- b. Variable, inconsistent and difficult to predict distribution of water-bearing sediments.
- c. Faulting, introducing unexpected strata, such as clay suddenly changing to water bearing sand,
- d. Variable and inconsistent slurry material due to changes in lithology and material behaviour:
- May lead to clogging of slurry shield,
- Problems associated with controlling slurry conditioning,
- Problems with slurry/muck handling.
- Other reasons
- a. Face instability issues in the Laminated Beds, gravels in Upnor Formation and sands in all units.
- b. Sensitivity to compressed air (170 kPa) of fine sand and silt of the Laminated Beds.
- c. Swelling clays.
- d. Possible hydraulic continuity with water-bearing strata beneath the Lambeth Group (Thanet Formation and Chalk).
- e. Removal of oxygen from the atmosphere in dewatered sand of the Upnor Formation and potentially the Laminated Beds (Newman et al., 2013).
Difficulties experienced during shaft construction
- De-watering may need to be designed specifically for local ground conditions (may need to de-water below construction depth into Thanet Formation and Chalk),
- Hard bands or gravels may obstruct sheet steel or bored piles, or cause unbalanced caisson sinking,
- Base heave,
- Flooding of shafts in Laminated Beds and sand layers in the Lambeth Group.
|Thames Tunnel||Skempton and Chrimes, 1994||Engineering and geology|
|Rotherhithe Tunnel||Tabor, 1908||Engineering|
|Blackwall Tunnel||O’Reilly, 1997||Construction, ‘scour hollows’|
|Limehouse Link – (A13), Canary Wharf to City of London||Stevenson and DeMoor, 1994||Design and performance|
|Victoria Line||Follenfant et al., 1969|
|Jubilee Line Extension (JLE)||Linney and Page, 1996
Burland et al., 2001,
Batten et al., 1996
|Engineering geology Construction|
|Docklands Light Railway (DLR) Lewisham Extension||Sugiyama et al., 1999||Tunnelling|
|Channel Tunnel Rail (CTRL)||Beckwith et al., 1996
|Ground investigation Groundwater control|
|Crossrail||Lehane et al., 1995||Lithological variability|
Within the Lambeth Group deposits the choice of whether to found on, or to penetrate, hard layers within a weaker medium is made more difficult by the impersistence, variability in thickness and strength, and unpredictability, of such layers. These layers may be in the form of shelly limestones (Lower and Upper Shelly Clays of the Woolwich Formation), calcretes (Upnor Formation and Lower Mottled Clays) or silica-cemented gravels and sands (e.g. Upnor Formation). A 1 m thick layer of limestone nodules (calcrete) within the Lambeth Group was successfully used as a founding medium for some 2000 kN capacity 600 mm x 16 m continuous flight auger piles on the South Quay Plaza (Phase 2), Isle of Dogs (Solera, 1998). Other piles were founded in underlying Lambeth Group sands. In central London, under-reamed piles have been successfully foundered in the Upper Mottled Clay, Reading Formation.
The de-stressing of the Lambeth Group clays following excavation can result in de-structuring, swelling, and softening. Thus the relationship between strength, bearing capacity, and depth is important in the design of foundations as is timely construction after pile boring, and control of surface water. Although in central London where the water table has been lowered, the Lambeth Group is generally considered to be under-drained by the Thanet Formation and Chalk. Water bearing sand units such as sand channels in the Laminated Beds and Upnor Formation gravel should be carefully monitored during the site investigation phase of projects involving deep pile foundations. If high water pressures are encountered then actions such as bentonite support introduced towards the base of the London Clay Formation, should be considered to prevent collapse of bored piles.
Canary Wharf was one of the largest developments in Europe. Started in the late 1980’s it has expanded across many of London’s 19th and early 20th century docks. Notable infrastructure developments have included the Docklands Light Railway (DLR) and a network of roads, whilst preserving much of the waterway system. Fill, Alluvium, Terrace Gravel and Lambeth Group strata underlie the centrally situated Canary Wharf site in the West India Docks area of the Isle of Dogs. Here, the Lambeth Group is approximately 12 m thick, with typically two-thirds of this consisting of Reading Formation deposits. The Lambeth Group strata are of uniform thickness across most of the site, but limestone and gravel layers are intermittent. As a result of the heterogeneity of the strata, it was found that SPT data best characterised the engineering properties (Troughton, 1992). Considerable seepage from the gravel bed in the upper beds of the Upnor Formation, and the Thanet Formation beneath was managed by using bentonite and casing (Troughton, 1992). Major buildings were founded in the Thanet Formation, while smaller structures and roads, usually by driven piles, and cofferdams were founded within the Lambeth Group. Some driven piles reached refusal in the limestone/marl layers.
Large diameter bored piles were successfully used under dry conditions at the British Library site in Euston, London, where shaft adhesions of over 200 kPa were achieved in Lambeth Group clays and sands at between 9 and 13 m depth (O’Riordan, 1982). Design compressibility and permeability parameters for the clays and sandy clays of the Lambeth Group for the Royal Albert Dock Spine Road (RADSR) are given as mv = 0.15 m2/MN, cv = 10 m2/yr, and k = 1 x 10-8 m/s (Card and Carter, 1995). Other examples of pile design and tests are given in Hight et al. (2004).
The lack of publications on problems associated with shallow foundations on the Lambeth Group indicates that their construction generally presents no major difficulties. Those settlements that have been documented indicate that they are about half those generally found for the London Clay Formation (Morton and Au, 1974). Nevertheless, shrinkage and swelling of high plasticity Reading Formation clays, and possible instability in the Upnor Formation gravel or loose sand beds in any unit, should be considered during the site investigation and design stages.
Mobilisation of pyritic material within the lignite layers of the Lambeth Group can give rise to sulphate-rich groundwater and consequent local damage to foundation concrete, after oxidation. Deposits with a large proportion of plant remains may provide poor and variable foundation conditions.
Summary of key issues for foundation design in the Lambeth Group
- Control of groundwater (for example by site specific designed de-watering),
- Founding-on, or penetrating, strong layers or lenses (i.e. shelly limestone within the Lower and Upper Shelly Clays of the Woolwich Formation and calcrete in the Upnor Formation and Lower Mottled Clay),
- Sulphate attack of concrete foundation in the Woolwich Formation,
- De-stressing, heave and softening of clays in excavation.
- Shrink/swell in clays:
- - Seasonal moisture content changes,
- - Desiccation due to trees,
- - Heave due to removal of trees,
- - Swell due to leaking drains.
- Softening of clays in the presence of water bearing sand beds.
Embankments and use as fill materials
Due to the highly variable lateral and vertical nature and extent of the Lambeth Group lithologies and their relatively thin development, fill materials derived from these deposits are likely to be composed of more than one unit and lithology. Therefore, use of the Lambeth Group as an engineered fill will require a good knowledge of the lithologies present and available at a potential source area and their strength/compaction characteristics. During the construction of the Newbury bypass (A34) the Lambeth Group provided a good source of fill for embankments and landscaping when emplaced in the correct condition, but low plasticity clays and silty sands typically proved to be highly sensitive to changes in moisture content. Acceptable criteria for use as engineered fill should be ascertained during the planning, investigation and construction phases (Hight et al., 2004).
Data on embankments constructed from the Lambeth Group is sparse partly because it has a relatively small outcrop. A survey of the motorway networks by TRL (Perry, 1989) found failure of 7.6% of embankments constructed from Lambeth Group deposits (undertaken prior to revision of the stratigraphy, the survey reported ‘Reading Beds’ as ‘Eocene’, rather than Palaeocene in age). This was second only to the Gault Formation. Typically failure occurred on 1 in 2 slopes within 22 years of construction. Failure modes include not only slope failures but also tension and shrinkage cracks, excessive settlement, water seepage and erosion of the toe. The survey noted that drainage ditches on the slope itself contributed significantly to reducing the number of failures. The maximum allowable embankment slopes assessed during this survey are presented in Table 7.3.
|Slope height (m)||0–2.5||2.5–5.0||>5.0|
Lithological variability of the Lambeth Group may make designing subgrade on, or from, the Lambeth Group difficult. CBR values vary greatly, as do maximum dry density, and are dependent on lithology. Design of the pavement on the Newbury Bypass (A34) was based on a subgrade CBR of <2% and included all the lithologies encountered, i.e. very stiff clay and silty sand.
Cut slope failures are not very common on the Lambeth Group, partly because of its limited area of outcrop. The first well-documented case was the Park Hill cutting, Croydon, (approx. NGR 5328 1654). This was a cut and tunnel scheme a total of 1,095 m long, of which 382 m was tunnel. It was constructed by the Woodside and South Croydon Railway from the Upper Addiscombe Road (A232) to near the Combe Road (A212) between 1880 and 1885 as a route to the South Coast. The cut was up to 26 m deep and the slopes graded to 45°. The geology, described by Klaasen (1883), is shown in Table 7.4 and Figure 7.18.
|Harwich Formation||Up to 7.3||Grey SAND occasionally calcareous with clay partings and thin layers of carbonaceous matter|
|0 to ?||Brown shelly sandy GRAVEL with occasional carbonaceous fragments.|
|0 to 3||Calcareous rounded black shelly sandy flint GRAVEL generally cemented with occasional more strongly cemented concretions of flint gravel and shells.|
Lower Woolwich Formation
|0.0 to 3||White shell bed with some carbonaceous fragments. Sometimes forming tabular limestone 0.4 m long and 0.15 m thick.|
|0.9||Soft brownish grey shelly CLAY|
|2.2||Interbedded hard and soft shelly pale blue CLAY with some pyrite. Hard bands cemented by calcium carbonate from shells. Lignitic in the north of the cutting.|
Lower Mottled Clay
|2.3||Mottled red with some purple CLAY with calcium carbonate near the base in wavy bands and concretions and beds up to 5.5 m wide. Lens shaped lignite layer reported within the clay in the south of the section with associated gypsum.|
|1.2||Multicoloured red with greenish blue and purple patches. Calcium Carbonate streaks and nodules in the upper 0.6 m.|
|1.5||Vertical smooth or polished jointed or fissured greenish blue with purple and orange red CLAY|
|1.4||Purple, greenish-blue and orange red patches CLAY. At the base is a 2.5 cm thick lignite layer.|
|Upnor Formation||0.45||Purple sandy GRAVEL. Gravel is fine to coarse and black. At the top is a thin, hard orange and carbonaceous layer — an iron pan|
|0.92||Green slightly gravelly SAND. Gravel is yellow or olive green and rounded.|
|2.0||Greenish-brown slightly gravelly SAND with orange red veins. Gravel is greenish grey, rounded to subrounded flint.|
|0.6||Grey SAND with calcareous cemented sand|
|0.6||Brown, slightly gravelly fossiliferous sandy CLAY. Gravel is black, rounded, fine flint|
|Thanet Formation||Quartz-rich SAND|
During construction of the railway a section of the central cut failed on 27th August 1882 and subsequent movement in the central cutting was described as ‘inconveniently frequent’. This was followed on 6th October 1882 by slipping of a 61 m long a 10 m wide section in the north cutting after heavy rainfall. The failure occurred initially in the ‘blue clay’ of the lower Woolwich Formation. The steep slope, depth of cut and high rainfall in addition to the character of the Lambeth Group contributed to the failure of the slope.
At the time of the construction of the Park Hill cut, the Lambeth Group had a reputation for being unstable. This was also the finding of an extensive survey of the motorway networks in the UK by TRL (Perry, 1989), which reported that 2.95% of cuts in the Lambeth Group had failed or showed signs of failure. Typically the failure occurred within 22 years of construction with the most common failures being in 1 in 3 slopes.
The slopes of the cut and cut and fill section of the A259 coast road about 1.5 km north-east of Castle Hill, Newhaven, West Sussex also suffered regular slope failures. Most of the movement was in the Woolwich Formation and resulted in cracking and deformation of the road. Much of the remedial work was piecemeal patching of the road but more permanent corrective action has been carried out, including the installation of sheet pile and a bored pile wall, rock fill on the toe, road realignment, shallow and deep drainage and decreasing the slope to 1 in 4.
Landslides are rare in the Lambeth Group because of the small outcrop area and subdued landscape; although some slope failures are due to subsidence from the dissolution of the Chalk. Slope instabilities in the Lambeth Group are, however, common at coastal outcrops which, although relatively limited in extent, may be subject to continuous erosion of the toe, and hence repeated landsliding. The coastal outcrops of the Lambeth Group mainly occur at Reculver (North Kent), Newhaven (East Sussex), Shoreham, Worthing, and Felpham (W. Sussex), Portsmouth (Hampshire), Whitecliffe Bay and Alum Bay (Isle of Wight) and Studland Bay although he geological map (EW 343, Swanage published 1993) indicates that these are of the London Clay Formation.
The best examples of coastal landslides occur at sites near Newhaven, Sussex and Alum Bay and Whitecliff Bay in the Isle of Wight (Bromhead, 1979). The near vertically bedded Lambeth Group (mostly Reading Formation) form the complete slopes at Alum Bay (Figure 7.19) and Whitecliff Bay. They fail regularly, mainly as mudslides, particularly during the winter when the slopes become saturated. The mudslides slump and flow onto the beach, and are then eroded away by the sea, maintaining the steep coastal cliffs and further enhancing instability. At Newhaven, the Lambeth Group, comprising the Upnor and Woolwich Formation, is nearly horizontally bedded above the Chalk. The steep slopes in the area are associated with gulleys of weaker material. Failures in the Woolwich Formation are common and in wet winters develop large rotational slides despite attempts to re-profile the slope and to improve drainage. The underlying Upnor Formation is also commonly heavily gullied along this part of the coast.
Inland, small-scale landsliding involving the Lambeth Group and London Clay Formation was noted along the River Arun, to the south-east of Arundel, Sussex at [TQ 025 065] (Aldiss, 2002; Shephard-Thorn et al., 1982) and to the north of Angmering [TQ 5080 1060].
Whilst some clay-rich formations of the Lambeth Group are fissured, highly plastic, and of low strength, overall slope stability is frequently found to have been improved by the under-drainage effect of high permeability sand layers within the dominantly clay formations (e.g. the Lower Mottled Clays), and/or underlying sand-rich formations (e.g. the Upnor Formation beneath the Woolwich Formation). Conversely, the alternation of high and low permeability strata may result in perched water tables, resulting in small-scale, local slope failure, for example by sand runs and subsequent camber or slump.
Subsidence in the Lambeth Group is usually due to the dissolution or mining of the Chalk. Karst features, formed by dissolution of the Chalk, are likely to be concentrated at the feather-edge of the Lambeth Group where surface run off is concentrated by clay beds (Edmonds, 1995). The result is an irregular undulating junction between the Chalk and Lambeth Group commonly with dissolution pits (dolines) or pipes. The Lambeth Group subsides into the karstic features.
Subsidence is more likely to occur in sand and silt deposits, and catastrophic subsidence may occur as a result of very heavy rainfall or water main failures. Soakaways should not be sited where large quantities of water may cause dissolution of the Chalk or where sands or fine-grained materials may be washed into karstic features. Fossil irregularities in the Chalk subcrop resulting from these karstic processes are up to 10 m deep in West Sussex and elsewhere in southern England.
Large-scale subsidence features involving the Lambeth Group, indicated as ‘Foundered strata’ on geological maps, are shown on the BGS Brighton-Worthing 1:50 000 sheet (318/333), particularly at West Blatchington, Brighton, and probably also in the Hove and Worthing areas (Young and Lake, 1988). These are thin relic outliers of Lambeth Group overlying the Chalk, where preferential saturation and dissolution of the Chalk, at the feather-edge of present and former Lambeth Group outcrops, has led to subsidence. Much of the outcrop is mantled by thin superficial deposits, which are often rich in Lambeth Group materials resulting from various periglacial processes, including solifluction. One possible mechanism is that sulphate-rich acidic groundwater formed by oxidation of pyrite from lignite beds within the Woolwich Formation overlying the Chalk have partially converted chalk to gypsum; this being more prone to dissolution and hence subsidence (Young and Lake, 1988). Similar conditions occur in the Chichester area (Aldiss, 2002).
Where founding on Lambeth Group strata underlain by the Chalk in southern England, consideration should be given to the likelihood of solution features and other karst/periglacial features affecting the Chalk/Lambeth Group junction, particularly at the feather-edge of the Lambeth Group outcrop. These will have serious implications for foundation design, in particular the risks associated with re-mobilising Lambeth Group material infilling solution features, for example as a result of water pipe fracture.
Karst features may be identified beneath the Lambeth Group using a number of methods including probing and geophysics (Bell et al., 2004). If it is essential to find the dissolution features then removal of the Lambeth Group cover may be required. For example, the karstic feature may be infilled with Lambeth Group materials as seen during the construction of the A34 at Chieveley (Rhodes and Marychurch, 1998). The difference in behaviour between chalk and the infill meant that standard road construction was not suitable. Alternatives included digging out the Lambeth Group and replacing with a suitable material or constructing a reinforced road. The size of the dissolution features meant that the reinforced road was the best option.
Man-induced subsidence occurs where chalk was mined below the Lambeth Group. Chalk, usually taken from surface, is added to clay to reduce shrinkage and as a flux in brick making and flint mining. The presence of former shallow chalk mines adjacent to and beneath the Lambeth Group outcrop represent a potential foundation collapse hazard (for example in the Coley district of Reading, where former shallow chalk mines resulted in the need for remedial foundation work to houses (Edmonds, 2008). Small vertical Chalk mines, dene holes, occur in Kent and Essex, and sand mines in the Thanet Formation are known in the Blackheath area, south east London.
Deoxygenated air in tunnels and deep excavations
Deoxygenated, pressurised air primarily in the Upnor Formation has caused health and safety risks during tunnelling or deep shaft excavation operations in London and has caused deaths (Lewis and Harris, 1998; Newman et al., 2013). The hazardous conditions occur where the deoxygenated, pressurised air is intercepted by tunnels or deep shafts as follows:
- Groundwater lowering by pumping of the Chalk during the 19th century and the first part of the 20th century for potable and industrial water,
- Ingress of air into mostly coarse-grained deposits such as the Upnor Formation now above the water table,
- Rising groundwater during the late 20th century and early 21st century as the industrial need for borehole water is reduced,
- Trapped of pressurised air due to rising ground water within parts of the Upnor Formation beneath clay beds including the Lower Mottled Clay,
- Deoxygenation of the trapped air in the Upnor Formation probably by ‘green rust’, a mixed Fe(II) and Fe(III) layered double hydroxide (Newman et al., 2013),
- Interception of the deoxygenated, pressurised air in the Upnor Formation by tunnelling or construction of deep shafts.
When carrying out a desk study for construction concerning the Lambeth Group it is important to consider whether any old extractive industry workings may be present. These workings may be poorly documented and include deep, steep-sided pits, which may be infilled. The fill may have markedly different geotechnical characteristics to the undisturbed materials. Buildings overlapping the fill and in situ material may be liable to differential settlement unless foundations are designed accordingly.
Materials extracted from the Lambeth Group include sand, ochre and lignite, and clays for the manufacture of brick, pipes and tiles and as a fuller’s earth for absorbing lanolin, oils and other greasy impurities as part of the finishing process for cloth. Although extraction is becoming more limited, during the 19th century and early part of the 20th century numerous small pits may have been opened for local purposes. The extent and depth of many of these pits is generally unknown or poorly documented although relevant information can often be found in:
- memoirs of the Geological Survey, especially those published during the 19th century and first half of the 20th century, when most of this activity occurred,
- the British Geological Survey quarries database, which includes historical data,
- historical O.S. maps,
- County mineral records.
During the 19th century and first half of the 20th century, quarries and pits provided much of the geological information on the geology on the Lambeth Group and are often well described in British Geological Survey memoirs, reports and published scientific papers.
Brick and tile manufacture
The Reading Formation clays have been used for brick and tile manufacture throughout much of the outcrop and are described in the Memoirs of the British Geological Survey as the borrow pits often provide the best exposures for geological descriptions. In recent times, activity has been restricted to small-scale operations near Chesham [SP 984 018], Buckinghamshire; Knowl Hill [SU 819 797], near Maidenhead, Berkshire (Sumbler et al., 1996); and Michelmersh [SU 343 260], near Romsey, Hampshire. Similar workings have been carried out elsewhere within the outcrop, for example at Arundel [TQ 000 073] and Westhampnett [SU 881 065], Sussex (Aldiss, 2002). Modern robotic brick making methods, however, do not favour heterogeneous or smectitic clays, and use of Reading Formation clays is now very limited (Bloodworth et al., 2001).
Sand and gravel
Sand and gravel has been taken from the Lambeth Group for local use, often with clays for brick and tile manufacture. Few of the pits were very large but two areas, at Orsett [e.g. TQ 565 810] in south Essex and Upnor [e.g. TQ 765 695] in north Kent, have a number of large sand and gravel pits. These pits provide the best examples of the Upnor Formation sequence. The pit at Shelford [TR 166 614], to the north of Canterbury, ceased extraction of sand gravel from the Upnor Formation in 2007–2008 and is now a landfill site for non-hazardous waste.
Fuller’s earth is a clay-rich material containing high proportion of smectite (usually calcium smectite) which imparts extremely high plasticity and shrink-swell potential, and low residual strength. An investigation into the fuller’s earth resources in England and Wales (Moorlock and Highley, 1992) found that some parts of the Lambeth Group, notably the Lower Mottled Clay, contained high proportions of smectite The Lambeth Group is not currently considered to be a viable economic source of fuller’s earth but a 1 m thick clay bed beneath clean white sand was extracted during the 19th century by ‘clothiers’ (Blake and Munckton, 1903).
Lignite was extracted from a seam up to 4 m thick along a drift mine near Cobham, Surrey. It was quarried for domestic purposes at Cobham Hall by Lord Darnley. A drift mine was started in 1947 and produced 80 tons (81.2 Mg) per week and expansion planned were formulated to increase output to 150 tons (152.4 Mg) a week. Difficulties were encountered including flooding, which was controlled by pumping. The proposed expansion did not happen due to the water problems, methane encountered in a gallery driven deeper into the hillside and difficulty in selling the product. In 1953 the mine was closed and the entrances blown up. Deep depressions in the woods nearby are probably due to the collapse of the workings (Kent Underground Research Group, 1991). The site was rediscovered during site investigation for the Channel Tunnel Rail Link, where collapsed adits were also found. Excavations in the lignite were recorded during embankment construction (Collinson et al., 2003).
Engineering geological summary
The engineering geological descriptions and characteristics of the Reading, Woolwich and Upnor formations are summarised in Table 7.5, Table 7.6 and Table 7.7, respectively. Note that in all cases where the Lambeth Group overlies Chalk strata, the potential problem of subsidence or collapse of Lambeth Group materials into voids created by dissolution of the underlying Chalk should be considered and assessed.
|Formation||Unit||Lithology||Engineering Description||Foundations||Excavation||Engineered Fill||Site Investigation|
|Reading Formation||Upper and Lower Mottled Clays||Clay||Stiff to very stiff often closely or very closely fissured, brown, grey, red or purple (Lower Mottled Clay only) mottled or multicoloured CLAY. Numerous fissures are often listric and polished (grey) giving rise to a ‘blocky’ texture. Upper Mottled Clays of high to extremely high plasticity. Lower Mottled Clays of intermediate to very high plasticity.||Shallow Foundations:
Clay lithologies may be prone to shrink/swell movements that can be exacerbated by presence of trees, leaking drains and high water tables.
Presence of water-bearing sand bodies, beds or laminae may make foundation construction difficult. Water ingress may lead to reduced bearing capacity of clays.
|Diggable. Fissuring likely to give rise to instability in excavations and provide potential pathways for water ingress.||
Variability and relatively thin nature of each unit mean fill materials are likely to be composed of more than one unit and lithology. Acceptance criteria should be taken into account at planning, investigation and construction stages.
May prove to be a good source of fill, material for embankments and landscaping if in an acceptable condition.
|Important to determine groundwater conditions, thickness of clay sequence and lithological variability (e.g. sand-filled channels). Samples required to ascertain strengths and shrink- swell potential.|
|Sand||Dense to very dense, orange, brown or grey sometimes red or mottled, sometimes slightly clayey or silty fine to medium coarse SAND. Sometimes weakly cemented. Generally occurs as impersistent layers or lenses but may be more extensive in some places.||Diggable. Likely to be water-bearing and unstable in excavations, requiring immediate support. Their presence within mostly clay will affect tunnelling methods and operations.||Important to determine position, extent and thickness of sand and associated groundwater conditions.|
|Variable, clay with channel sands||See description of clays and sands above. Contact between sand and infilled clay channel is usually sharp.||As above.||Important to determine position, extent and thickness of sand-filled channels and associated groundwater conditions.|
|Lower Mottled Clay||Cemented sands||Very weak to occasionally strong orange, brown or grey sometimes red or mottled, generally iron-cemented sand (SANDSTONE). Generally thin
(<1 m thickness) and often impersistent.
|May require hard digging locally; variable strength leads to variable stability in excavations, particularly below the water table.||Important to determine elevation, thickness, extent and strength of cemented sand layers prior to construction.|
|Very weak/powdery to strong carbonate concretions (CALCRETE) ranging from gravel-size up to 0.5 m diameter. Exceptionally, in east London, concretions coalesce to form a strong to very strong bluish-grey or grey, sometimes nodular, fine-grained crystalline LIMESTONE, up to 1.6 m thick. Inconsistent in lateral and vertical extent and strength.||Digging, ripping or pneumatic tools may be required due to variable strengths. May enhance stability in excavation but dependant on hard-band thickness, strength of surrounding strata and potential water ingress.||Important to determine elevation, thickness, extent and strength of hard bands prior to construction.|
|Formation||Unit||Lithology||Engineering Description||Foundations||Excavation||Engineered Fill||Site Investigation|
|Woolwich Formation||Upper Shelly Clay/Lower Shelly Clay||Shelly clay||Firm to very stiff, often closely to extremely closely fissured, sometimes thinly to thickly bedded, generally dark grey sometimes mottled brownish grey shelly CLAY. Some beds, up to 1 m thick, are almost entirely shells, locally weakly cemented (see limestone below).||Shallow Foundations: Clay lithologies may be prone to shrink/swell movements that can be exacerbated by presence of trees, leaking drains and high water tables.
Presence of water-bearing sand bodies, beds or laminae may make foundation construction difficult. Water ingress may lead to reduced bearing capacity of clays.
|Diggable. Strength contrasts between clay-dominant and shell-dominant lithologies may lead to instability in excavations.||
Variability and relatively thin nature of each unit mean fill materials are likely to be composed of more than one unit and lithology. Acceptance criteria should be taken into account at planning, investigation and construction stages.
|Important to determine groundwater conditions and lithological variability, particularly thickness and extent of shell bands.|
|Laminated Beds||Clay, silts and sands||Variable, thinly interbedded succession of CLAY, SILT and SAND. Beds usually <50 mm thick and typically laminated on a millimetre scale. Localised sand bodies (probable channels) up to about 4 m thick occur, particularly in SE London.||Diggable. Usually water-bearing, giving rise to perched water tables and instability in excavations.||Important to determine presence of water-bearing Laminated Beds of sand and silt and associated perched water tables; also presence and extent of possible water-bearing sand-filled channels.|
|Upper Shelly Clay||Shelly sand (Generally in the east of London)||Medium dense to very dense, sometimes laminated, grey sometimes brown, occasionally with organic remains, silty, fine to medium, occasionally coarse SAND (representing infilled channels). Generally high sulphate and organic contents.||Diggable. Impersistent and often water-bearing, leading to unexpected water strikes and instability in excavations. Immediate support required.||Important to determine position, extent and thickness of sand-filled channels and associated groundwater conditions.|
|Upper Shelly Clay/Lower Shelly Clay||Shelly Mudstone and LIMESTONE (Limited to south and east London)||Weak generally thin but up to 300 mm thick beds of shelly MUDSTONE and strong dark grey LIMESTONE (Paludina limestone, Upper Shelly Clay).||Digging, ripping or pneumatic tools may be required due to variable strengths. May be stable in excavation but dependant on hard-band thickness, strength of surrounding strata and potential water ingress.||Important to determine elevation, thickness, extent and strength of hard bands prior to construction.|
|Lower Shelly Clay||Lignite (Mainly to south and east of London)||Firm to weak, sometimes thickly to thinly laminated, sometimes with extremely closely spaced fissures/fractures, dark brown or black, sometimes clayey or sandy LIGNITE. Sometimes with interbeds or thick laminations of black coal.||Diggable, but trees and large roots preserved in situ may cause difficulties locally. Variable thickness, strength and close fracturing/jointing may result in instability in excavations. May be stable in short-term.||Unsuitable||Important to determine presence and extent of lignite bands associated with variable thicknesses and strengths.|
|Formation||Lithology||Engineering Description||Foundations||Excavation||Engineered Fill||Site Investigation|
|Upnor Formation||Glauconitic sand||Medium dense to very dense greenish grey or green becoming orange or brown, occasionally gravely, sometimes shelly, clayey or silty fine to medium, sometimes coarse SAND. Gravel often rounded fine to coarse. Thin seams of grey clay are also present. Clay-dominated units of firm to stiff CLAY up to 0.3 m thick with minor sand laminae may also occur. Clays have high smectite content.||Shallow Foundations: Clay lithologies within dominantly sand units may be prone to shrink/swell movements that can be exacerbated by presence of trees, leaking drains and high water tables.
Presence of water-bearing sand bodies, beds or laminae may make foundation construction difficult.
|Diggable. Generally water-bearing with possible artesian conditions if in hydraulic continuity with underlying Thanet Formation. Interbedded clay bands may give rise to perched water tables. Generally unstable in excavation with immediate support required.||
Variability and relatively thin nature of each unit mean fill materials are likely to be composed of more than one unit and lithology. Acceptance criteria should be taken into account at planning, investigation and construction stages.
|Important to determine presence depth and thickness of sands and associated clay bands (often highly plastic smectite-rich) and associated groundwater conditions, particularly potential artesian pressures.|
|Gravel||Dense to very dense, usually well-rounded flint GRAVEL. Gravel generally less than 30 mm diameter with occasional cobbles (up to 200 mm).||Diggable. Highly permeable and possibly water-bearing. Immediate support required in excavations. Flint gravel will increase wear on cutting and excavation machinery.||Important to determine presence and extent of potentially abrasive gravel beds, and associated groundwater conditions. May pose drilling/sampling difficulties.|
|Hard bands||Weak to moderately strong, irregular-shaped carbonate concretions (CALCRETE) which locally may be 0.5 m diameter.
Strong to extremely strong silcrete nodules which may be up to 3 m long and 1 m thick.
|Generally diggable but may require ripping or pneumatic tools locally. Variable size, strength and extent of concretions may cause problems in excavation. Depending on thickness, may enhance stability of excavations.||Important to determine elevation, thickness, extent and strength of hard bands prior to construction.|
- SKEMPTON, A W, and CHRIMES, M M. 1994. Thames Tunnel: geology, site investigation and geotechnical problems. Geotechnique, 44, 191–216.
- CLARKE, R P J, and MACKENZIE, C N P. 1994. Overcoming ground difficulties at Tooting Bec. Proceedings of the Institute Civil Engineering, 102, 60–75.
- HIGHT, D W, ELLISON, R A., AND PAGE, D P. 2004. The engineering properties of the Lambeth Group. Report RP576 Construction Industry Research and Information Association (CIRIA), London.
- MUNSELL. 2009. Munsell soil colour charts. Gretag Macbeth, New York, USA.
- ELLISON, R A. 1983. Facies distribution in the Woolwich and Reading Beds of the London Basin, England. Proceedings of the Geological Association, 94, 311–319.
- BUURMAN, P. 1980. Palaeosols in the Reading Beds (Palaeocene) of Alum Bay, Isle of Wight, UK. Sedimentology. 27, 593–606.
- PAGE, D, and SKIPPER, J. 2000. Lithological Characteristics of the Lambeth Group. Ground Engineering, 33, 38–43.
- BSI, 2007. BS EN 1997-2 Eurocode 7. Geotechnical design. Ground investigation and testing. British Standard Institute, London, UK. 196pp.
- BSI. 2006. BS EN ISO 22475-1. Geotechnical Investigation and testing. Sampling methods and groundwater measurements. Technical principals for execution. BSI, London, UK. 134pp.
- MUIR-WOOD, A M. 1994. The Thames Tunnel 1825-43: Where shield tunnelling began. Proceeding of the Institution of Civil Engineers, Civil Engineering, 102, 130–130.
- ELLISON, R A, WOODS, M A, ALLEN, D J, FORSTER, A, PHAROAH, T C, and KING, C. 2004. Geology of London. Special Memoir for 1:50 000 geological sheets 256 (North London), 257 (Romford), 270 (South London) and 271 (Dartford) (England and Wales). British Geological Survey, Keyworth, Nottingham, UK.
- ELLISON, R A. 1991. Lithostratigraphy of the Woolwich and Reading Beds along the proposed Jubilee Line Extension, south-east, London. British Geological Survey Technical Report WA/91/5C. British Geological Survey, Keyworth, Nottingham, UK.
- GLASS, P R, and POWDERHAM, A J. 1994. Application of the observational method at the Limehouse Link. Geotechnique, 44, 665–679.
- SUGIYAMA, T, HAGIWARA, T, NOMOTO, T, NOMOTO, M, YUTAKA, A, MAIR, R J, BOLTON, M D, and SOGA, K. 1999.Observations of ground movements during tunnel construction by slurry shield method at the Docklands Light Railway Lewisham Extension – East London. Soils and Foundations, Japanese Geotechnical Society, 39, 99–112.
- FARROW, J P, and CLAYE, P M. 1994. Civil engineering and tunnel design. Proceedings of Institute of Engineering, Civil Engineering, 102, 23–33.
- NEWMAN, T G, GHAIL, R C, and SKIPPER, J A. 2013. Deoxygenated gas occurrences in the Lambeth Group of central London, UK. Quarterly Journal of Engineering Geology and Hydrogeology, 46, 167–177. DOI: 10.1144/qjegh2012-013
- TABOR, E H. 1908. The Rotherhithe Tunnel. Minutes of the Proceedings, Proceedings of the Civil Engineers. 175, (1), session 1908–1909, 190–208.
- O’REILLY, M P. 1997. The first Blackwall Tunnel. Tunnels and Tunnelling, 29 (5), 42–45.
- STEVENSON, M C, and DE MOOR, E K. 1994. Limehouse Link cut and cover tunnel: design and performance. Proceeding of the 13th International Conference on Soil Mechanics, New Delhi, 2, 882–890. Rotterdam: Balkema.
- FOLLENFANT, H G, CUTHBERT, E W, MORGAN, H D, BARTLETT, J V, CLARK, J A M, HOOK, G S, LEE, J J, MASON, P L, THOMAS, D G, and BUBBERS, B L. 1969. The Victoria Line. Proceeding of the Institution of Civil Engineers, supplement 1970, paper 7270 S. 337–475.
- LINNEY, L F, and PAGE, D P. 1992. The engineering geology of the Woolwich and Reading Beds and its implications in the design and construction of the Jubilee Line Extension. Preprints Volume of The 28th Annual Conference of the Engineering Group of the Geological Society. In The Engineering Geology of Construction, Manchester. UK.
- BURLAND, J B, STANDING, J R, and JARDINE, F M. 2001. Case studies from the construction of the Jubilee Line Extension. CIRIA, London.
- BATTEN, M, POWRIE, W, BOORMAN, R, and YU, H-Y. 1996. Measurement of prop loads in a large braced excavation during the construction of the JLE station at Canada Water, East London. In: Geotechnical Aspects of underground construction in Soft Ground — Proceedings of the International Symposium at City University, London, 15–17 April 1996 (R J Mair and RN Taylors, editors), Balkema, Rotterdam, 57–62.
- WHITTAKER, D. 2004. Groundwater control for the Stratford CTRL station box. Proceedings of the Institute of Civil Engineers, Geotechnical Engineering, 157, 183–191.
- LEHANE, B M, CHAPMAN, J P, PAUL, S P, and JOHNSON J G A. 1995. The apparent variability of the Woolwich and Reading Beds. Proceedings of the XI European conference on soil mechanics and foundation Engineering, Copenhagen, 8, 95–102.
- SOLERA, S A. 1998. Continuous flight auger piles in the Woolwich and Reading Beds in the Isle of Dogs, London. Proceedings of the 7th International Conference on Piling and Deep Foundations. Vienna, DFI Paper 1.9.
- TROUGHTON, V M. 1992. The design and performance of foundations for the Canary Wharf development in London's Docklands. Geotechnique, 42, 381–393.
- O’RIORDAN, N J. 1982. The mobilization of shaft adhesion down a bored, cast in-situ pile in the Woolwich and Reading beds. Ground Engineering, 15, 17–26.
- CARD, G B, and CARTER, G R. 1995. Case history of a piled embankment in London's Dockland. In: Engineering Geology of construction. Eddleston M., Walthall, S., Cripps J.C., and Culshaw, M.G. (Editors). Manchester 1995. Geological Society Engineering Geology Special Publication, 10, 79–84.
- MORTON, K, and AU, E. 1974. Settlement observation on eight structures in London. In: ‘Proceedings British Biotechnical Society Conference ‘Settlement of Structures’ Cambridge, UK. Pentach Press. 183–203.
- PERRY, J. 1989. A survey of slope condition on motorway earthworks in England and Wales. Department of Transport, TRRL Research Report RR 199, Transport and Road Research Laboratory, Crowthorne, UK.
- BROMHEAD, E N. 1979 Factors affecting the transition between the various types of mass movement in coastal cliffs consisting largely of over-consolidated clay with special reference to southern England. Quarterly Journal Engineering Geology, 12, 291–300.
- ALDISS, D. 2002. Geology of the Chichester and Bognor district. Sheet description of the British Geological Survey, Sheet 317, 332 (England and Wales). British Geological Survey, Keyworth, Nottingham, UK.
- SHEPHARD-THORN, E R, BERRY, F G, and WYATT, R J. 1982. Geological notes and local details for 1:10000 sheets SU80NW, NE, SW and SE, SU90NW, NE, SW and SE, TQ00NW, SW, West Sussex Coastal Plain between Chichester and Littlehampton (West Sussex Coastal Plain between Chichester and Littlehampton). British Geological Survey Technical Report WA/VG/82/02. British Geological Survey, Keyworth, Nottingham, UK.
- YOUNG, B, and LAKE, R D. 1988. Geology of the country around Brighton and Worthing. Memoir of the British Geological Survey, (England and Wales), sheet 318 and 333. British Geological Survey, Keyworth, Nottingham, UK.
- RHODES, S J, and MARYCHURCH, I M. 1998. Chalk solution features at three sites in south-east England: their formation and treatment. In: J G Maund and M Eddleston (eds). Geohazards in engineering geology. Engineering Geology Special Publication No.15, Geological Society, London, 277–289.
- EDMONDS, C N. 2008. Karst and mining geohazards with particular reference to the chalk outcrop of England. Quarterly Journal of Engineering Geology and Hydrogeology, 41, 261–278.
- SUMBLER, M G, PHAROAH, T C, BARON, A J M, COX, B M, OWEN, H G, WOOD, C J, ELLISON, R A, ZALASIEWICZ, J A, BRIDGLAND, D R, WYMER, J J, BALSO, P S, HIGHLY, D E, GREY, D R C, MONKHOUSE, R A, STRANGE, P J, ROBINSON, J E, IVIMEY-COOK, H C, and SHEPHARD-THORN, E R. 1996. London and the Thames Valley. British Regional Geology 13. HMSO, London.
- BLAKE, J H, and MONCKTON, H W. 1903. The geology of the country around Reading. Memoir of the British Geological Survey, Sheet 268 (England and Wales). British Geological Survey, Keyworth, Nottingham, UK.
- KENT UNDERGROUND RESEARCH GROUP, 1991. Kent and East Sussex Underground. Meresborough Books, Rainham, Kent. UK. 128p.
- COLLINSON, M E, HOOKER, J J, and GROCKE, D R. 2003. Cogham lignite bed and penecontemporaneous macrofloras of southern England. A record of vegetation and fire across the Paleocene-Eocene Thermal maximum. Special Papers of the Geological Society of America, 369, 333–350.