Geoscientist, February 2018
Chris Eccles, Director and Dr Simon Ferley, Technical Director, TerraConsult Ltd
Geological conditions in mid-Cheshire, combined with the anthropogenic legacy, will have significant effects on the proposed HS2 Phase 2B high speed railway line, say Chris Eccles and Simon Ferley
This 20 km long section of the HS2 Phase 2B route passes through gently undulating farm land crossing over a number of A-roads, canals and two minor railway lines which would in many areas of lowland UK have routine ground engineering risks. However, this part of the HS2 route will be technically challenging for design and construction due to the presence of deep salt karst and several hundred years of human impact on the ground including below ground infrastructure. In November 2016, HS2 announced a change in both the horizontal and vertical alignments through Cheshire due to the ground related risks and the presence of below ground infrastructure.
HS2 Alignment in Mid Cheshire
The HS2 alignment through Cheshire was first published in January 2013 and a revised alignment published in November 2016. This revised route involved moving the track in an area southwest of Northwich up to 800 m west. The reason for this change was the geology and subsurface anthropogenic features.
The route leaves the Crewe to Preston railway line north of Crewe, passes between Winsford and Middlewich, crosses the A530 King Street south east of Northwich and then runs to the east of Northwich passing entirely through farmland and passing close to a number of villages.
The 20.1 km length of the 2016 route covered here includes 2 Rail Crossings, 11 River/Canal crossings, 10 Road Crossings & re-routing of 2.3km of dual carriageway, Embankments up to 26 m high and a large rolling stock depot. There are no cuttings.
Superficial deposits in this part of Cheshire are up to 30 m thick. Drift deposits at surface are nearly all glacial till, with alluvium in the river valleys (Figure 2). Almost no glacial sand and gravel or fluvioglacial deposits are mapped below this part of the route.
The bedrock consists of the Mercia Mudstone Group (formerly Keuper Series, see Table 1) over 1.5 km thick. The subdivision underlying the whole of this area is the Sidmouth Mudstone Formation, which includes major deposits of halite (rock salt) of economic importance – especially the Northwich Halite. This is a series of halite strata interbedded with marl (75% salt 25% marl). This member is 200 to 285 m thick and thins east to west.
The route runs through the north of the Cheshire Basin Syncline with strata dipping south. There is a single major fault called the King Street fault (crossed by HS2 at Chainage 22.5 km) forming a curved line running roughly north-south through Cheshire. Bedrock to the east of the fault has dropped by more than 500 m relative to the west.
Mining and Storage in the Halite in Cheshire
Salt extraction has a long history in Cheshire stretching almost continuously over 2000 years. In 1670, John Jackson, digging for coal in Marbury (3 miles north of Northwich), found salt. From then until 1928 mining continued in Northwich in Witton, Dunkirk, Marbury & Marston areas. The oldest mine still working is Compass Mineral’s Winsford salt mine, operating since 1844.
Pumping of brine from wells sunk into wet rockhead (‘wild brine pumping’) started in the 1800s and continued until 2006. This often caused subsidence many miles from wells: due to hydrogeological conditions, groundwater was not drawn into wells uniformly with preferential flow following so-called ‘brine runs.’ The most active subsidence often took place where relatively salt free water was drawn towards a well and met bedrock halite.
Pumping from flooded conventional mines continued until 1938 when the practice was banned. This was called ‘bastard brine pumping’ and created extensive subsidence, particularly to the north of Northwich where large lakes or ‘flashes’ formed.
The first controlled solution mining in Cheshire commenced in 1928 and it is via this method that salt is currently mined for the chemical industry in Cheshire.
Materials of many kinds are also stored in the salt caverns, including documents (Winsford), hazardous waste (Winsford) solvent waste, strategic oil reserves (currently being decommissioned) and gas. Gas storage began in 2006, and became operational in 2011, four lying close to the proposed route.
Where the rock salt bearing rocks of the Northwich Halite are present at rockhead, it is called “wet rockhead.” This term indicates where saturated brine is present from dissolution of the upper section of the halite and is also where the original brine extraction took place. In contrast, the traditional name for the area where the salt has not dissolved (because it is sealed beneath a cover of the Byley/Wych Mudstone) is “dry rock head” (“dry” here meaning simply that water is not dissolving salt). Wet rockhead is present below about 57 % of the length of the route covered here.
At ‘wet’ rockhead, groundwater has circulated and dissolved salt from the strata and only the interbedded insoluble marl rocks remain. When the salt dissolves, the marls collapse and are broken up (brecciated) and softened. Large voids up to 6 to 10 m high can be found in the brecciated/collapsed strata, but large ‘rafts’ of intact marl also occur. The overall thickness of brecciated ground typically varies from 15 to 120 m below drift. It can be difficult to distinguish between glacial deposits and brecciated bedrock because both have similar composition.
The development of the collapsed strata was strongly influenced by groundwater flow and the modification of the groundwater regime during the last, and possibly earlier, glaciations. The last (Devensian) glaciation buried Cheshire in a thick ice-sheet which had water at its base. This raised groundwater pressures and flushed out saline waters from the ground, forcing fresh salt-free water to the dissolution surface of the salt – thus cutting deeper into the sequence. When the ice melted, further dissolution of the salt occurred as new regional groundwater patterns formed. Because of glaciations, the depth of salt dissolution is up to 120 m deep.
Man-Made Cavities in Areas of Dry Rockhead
In some areas of dry rockhead there has been, (or there is still) solution-mining of the Northwich Halite at depth The 2013 route passes through 1.46 km of the Holford Brinefield east of Lostock Green, passing over 14 brine cavities. Here, 20 m of drift lie over Byley and Wych Mudstones, with solution mining in the underlying Northwich Halite. Cavities may be up to 170 m diameter and are as little as 30 m or less apart, with the shallowest cavities being only 60 m below surface in the north of the brinefield. Some cavities at Holford are being used to store solvent waste.
In a number of locations at Holford there has been break-through between adjacent cavities. In Cheshire no solution mined cavities have collapsed; but near Preesall in Lancashire similar solution mining was carried out of the Northwich Halite to shallower depth, resulting in a series of collapses that created several lakes.
Due to the nature of rock salt to creep over time, the ground above the Holford Brinefield is slowly settling by three or four millimetres a year. The settlement bowl extends to a wider area than the actual plan extent of the cavities.
The presence of gas storage caverns below ground presents a lower risk than the Holford Brinefield because the roofs of gas storage caverns are usually deeper than 600 m. Individual storage caverns may be 100 m diameter and more than 100 m high. However, locating HS2 directly over such a cavern is impossible because the UK Government classifies these gas storage facilities in Cheshire as “nationally strategic.”
Both 2013 and 2016 routes avoid the EDF gas storage facility of Warmingham between Crewe & Middlewich. The 2013 route crossed both Storenergy’s Stublach Gas Storage facility (southwest of Northwich) and the King Street Energy facility, which has planning permission for a new storage field.
Between 2013 and 2016 HS2 carried out a review of the risk from constructing the route across the solution-mined brine caverns at Holford Brinefield and reassessed the importance of the gas storage caverns to the south of Holford Brinefield. By moving the 2016 route to the west it avoids both Holford Brinefield and the gas storage caverns. However, it crosses surface facilities for the planned King Street Energy Storage Facility and intersects the edge of the settlement bowl of the Holford Brinefield.
Ground Affected By Natural Salt Subsidence
Areas with “wet rockhead” comprise 54 % of the 2016 route in this 20.1 km length. Natural subsidence can occur here in a number of ways.
Groundwater flowing at rockhead in depressions causes slow dissolution. This occurs faster along the river valleys and where groundwater flows along preferential pathways. Areas of “wet rockhead” can often appear stable for long periods, but if conditions change the rate of solution may accelerate. Where there has been historic uncontrolled wild brine pumping, subsidence features may still be active many years after the activity ceased.
One area where there has been recent reactivation occurs near Lea House Farm (Chainage 15.75 km). Here the farmer has reported that the field, level in 1990, now has two parallel settlement troughs up to 2 m deep. Subsidence appears to be accelerating. Not far to the south of this at Yew Tree Farm close to chainage 15.45 km there are new subsidence hollows (picture).
Another subsidence feature in open farmland is located to the northeast of Northwich, where a pond has formed since 2004 in a new hollow. The cause of subsidence features in areas of wet rockhead remote from historic wild brine pumping is usually unknown. They could be due to salt dissolution at depth at rockhead (often more than 60 m down) and migration of the void. New subsidence features in open farmland usually go uninvestigated. The Cheshire Brine Board has made an assessment of farmland hollows near Plumley, (some 1.0 km east of the HS2 route). Findings did not indicate deep-seated solution of rock salt, but concluded that uneven drainage and consolidation of heterogeneous drift deposits were the most likely culprits. The presence of subsidence features caused by shallow movement in drift still presents a significant potential risk to a high speed railway line.
Known Salt Subsidence from Wild Brine Mining
The best known case along the HS2 alignment is at Billinge, between Northwich and Middlewich. The route crosses a 1.5 x 0.5 km subsidence trough with a series of six flashes. Here a viaduct will cross both the Northwich to Middlewich railway line and the Trent and Mersey Canal, in an area of wet rockhead with intact salt at about 60 m below ground, with collapsed bedrock and glacial deposits above. The subsidence feature at Billinge is mentioned in both of the geological memoirs for Geological Map Sheets 109 and 110. These suggest that it is not known whether the subsidence was due to pumping at Northwich (2.5 km away) or Middlewich (4 km). The location of the subsidence is thought to be partially controlled by the King Street Fault, running parallel to the trough’s eastern edge. The brine pumping will have drawn groundwater principally from just above the intact bedrock at the base of previously collapsed strata. Therefore, the already weakened and collapsed strata (formed in glacial periods) underwent a further period of subsidence as wild brine pumping reduced strength and increased compressibility, creating a c. 60 m thickness of highly compressible strata above rockhead.
Subsidence in this area has required the railway embankment to be rebuilt requiring raising in stages; the canal towpath to be reconstructed; the railway bridge over Whatcroft Lane (Figure 9) to be raised by over 2.7 m and strengthened in stages. Other pictures of the historic effects of wild brine pumping on roads, canals and railways are in Figure 13 to 15.
Section of Route Over Winsford Salt Mine
The salt mine is located northeast of Winsford and is owned & operated by Compass Minerals Ltd. There is an extensive network of caverns about 4 x 3 km in plan, typically 140 to 220 m below ground. However, at the northern end of the mine the roof is less than 100 m deep. The mining is ‘room and pillar’ with caverns up to 20 m wide and 8 m high. The pillars of salt are usually 20 m square with a 68 to 75% extraction ratio. Historically a single bed of salt was mined but in the last few years Compass have started to mine from a second, shallower level, in the southern part of the mine directly below the HS2 route. Overall the HS2 route passes over a 3.1 km length of the mine.
Above the mine runs the 1.16 km long River Dane viaduct, which is up to 26 m high. Further to the north a 160 m long viaduct crosses Puddinglake Brook and the Trent & Mersey Canal. This is up to 13.4 m high. The HS2 route also crosses hazardous waste landfill within the salt mine. The mine sits within in an area of wet rockhead where depth to intact rock decreases from 100 m in the south (below the Dane viaduct) to 60 m (below the Puddinglake viaduct). The Dane viaduct also runs over an area where there is mining at two levels; but at the Puddinglake viaduct the depth to the mine below intact rockhead is only about 30 m and is even shallower than in the south.
Due to the thickness of the drift deposits and the collapsed breccia beneath (known to have voids above the mine) foundations for the HS2 viaducts are likely to be piled – potentially requiring 120 m deep piles below Dane Viaduct. These will probably be the deepest on-shore piles in the UK and will be particularly difficult to construct. Whilst shorter piles will be required below the Puddinglake viaduct, here it is likely that substantial backfilling and stabilisation work will be required due to the depth of mining below bedrock.
Drainage and Salt Karst
The 2013 proposed alignment of HS2 required embankments up to 18 m high and cuttings up to 10.2 m deep. The November 2016 HS2 alignment has no cuttings and the embankments are up to 26 m high. Regarding these changes HS2 stated “the route has been elevated across the area …… to allow for careful management of drainage and geological risks.” The 2016 alignment recognised the potential for drainage to be a significant risk in areas of wet rockhead, particularly deep cuttings which would act as deep drains and could trigger subsidence. The drainage from embankments located in areas of wet rockhead will have to be carefully managed with ditches and attenuation features requiring to be lined to prevent localised infiltration into the ground which in the long term could lead to subsidence. It should be noted that in his Glossop Lecture, Dr Tony Waltham indicated that 95 % of all subsidence collapses in karst areas are the result of engineering structures interfering with natural drainage.
Ground problems have led to many innovative solutions in the Northwich area. The use of timber ring-beams for properties enabled them to overcome large scale settlements and then be jacked level again (Figure 10). Many conventional bridges in the area have jacking points (Figure 11) so they can be re-levelled. Hayhurst (1898) & Town Bridge (1899) in Northwich were constructed with the turntable on floating pontoons (Figure 12) on the Weaver Navigation so that as long as the water level can be maintained in the river the bridges will continue to remain level and operational.
The construction of HS2 in Cheshire will doubtless require its own innovative solution to stand alongside these historic achievements.