Earthworks: Down to earth

The A465 Heads of the Valleys is a strategic trunk road linking the Midlands with south west Wales and forms a vital part of the Trans European Network. It also provides access into and out of the former industrial towns of the south Wales valleys.

The A465 was built in the 1960s. The three lane road is often congested and has a poor safety record and, as a result, the Welsh government is upgrading the road to a dual carriageway.

The overall scheme will involve upgrading 40km of the road, delivered in six sections. Sections 1 and 4 have already been finished, section 3 is under construction and sections 2, 5 and 6 are due to be completed by 2020.

Section 3, between Brynmawr and Tredegar, is well underway and is due to be completed in July 2015. This section is being delivered under an early contractor involvement (ECI) design and build contract valued at £120M, with Carillion as main contractor and Arup as the lead designer.

Work on this section comprises 7.8km of dual carriageway, 4.7km of which is being constructed offline, with four junctions, cuttings up to 18m high and embankments generally up to 10m high, although one – Carno embankment – will be 33m high. There are also eight bridges, 11 retaining structures, three underpasses and numerous small culvert and mammal crossings.

Section 3 runs close to the northern outcrop of the Lower Coal Measures, with the Brecon Beacons National Park immediately to the north. Coal and ironstone mining has affected the ground conditions throughout this section. There are historic mine entrances, including shafts and adits, the alignment crosses extensive areas of shallow workings and made ground, comprising colliery spoil and other industrial waste, is widespread.

The alignment crosses a number of steep, narrow-sided valleys on a sequence of alternating cuttings and embankments. As well as the need to excavate significant quantities of hard rock, there have been a number of geotechnical challenges to deal with, in both design and construction, including exposed coal seams in the cuttings and the frequent presence of artesian groundwater.

Early contractor involvement

The ECI phase enabled close collaboration between the contractor and designer and provided the ideal opportunity to address design and constructability challenges at the preliminary design stage. This resulted in earthworks becoming a key driver for the design, influencing the cost and programme for the scheme.

As a large amount of earthworks was required, early design predicted there would be a significant surplus of excavated soil and rock.

To reduce this surplus, it was decided early on to consider replacing a proposed viaduct over Carno valley with a 33m high embankment. Not only could an embankment be built much faster than the viaduct, there was an added benefit: it could also be used as a haul route across the valley once construction had reached about 12m high. This provided greater flexibility for subsequent earthworks, reducing haulage distances and speeding up the construction programme.

The embankment was also cheaper to build and brought environmental benefits such as minimal offsite disposal and lower visual impact.

Carbon footprint

An assessment of the carbon footprint for both the viaduct and embankment options was carried out during preliminary design. This used the CO2ST software, designed by Arup in collaboration with Cambridge University specifically for civil engineering schemes with a major earthwork component.

The carbon footprint associated with the manufacture of materials such as concrete and steel is extremely high relative to site-won earthwork materials and the use of earthworks plant. The CO2ST assessment confirmed that building the embankment would bring significant benefits compared with constructing the viaduct.

The assessment was extended to cover not just the construction element of the project, but also future operation and maintenance of the road. The proposed scheme, including the Carno embankment, was compared against the ‘do nothing’ scenario (ie not upgrading the A465).

Operational carbon emissions were shown to be significantly lower for the new scheme because of the improved alignment and changes to driving behaviour. It has been estimated the carbon footprint associated with construction will be paid back through operational savings within about 30 years. As a result, it was decided to build the embankment rather than the viaduct.

Design of the Carno embankment and underpass

A viaduct was originally proposed at Carno for a number of reasons, primarily because of the 27m high Edwardian Lower Carno Dam immediately to the north; the discharge of the embankment dam spillway into a river channel passing beneath the new road alignment; and the need to provide continued road access to the dam.

As a result, the embankment design includes a 115m long precast reinforced concrete arch underpass built at its base (piling work for the underpass was carried out in 2013 - GE October 2013). This has an 18m span and is 9.6m high, giving enough space for an access road to the dam, as well as an engineered river channel leading from the dam spillway. The size of the river channel was calculated using computational fluid dynamics to ensure it could cope with discharge flows resulting from a maximum design flood event.

To avoid the embankment encroaching on the toe of the dam and spillway, it is retained by a threetiered reinforced soil wall up to 33m high. Design of this relatively steep embankment involved detailed assessment of potential impacts on the dam, in particular the potential for slope instability of the dam itself. To limit the length of the arch underpass, the southern side of the embankment was also steepened, using reinforced soil walls and slopes.

The access road and river channel are skewed relative to the alignment of the embankment, so significant asymmetric loading of the arch was predicted, particularly at the portal ends. Finite element analyses were carried out, allowing for different construction heights and earthworks loading corresponding to the range of temporary and permanent conditions along the 115m length of the arch.

Construction of earthworks

Earthworks design was modelled in 3D for incorporation into a BIM system for use by the wider project team. This enabled detailed programming of the works; allowed the team to look at early visualisations; and also meant clash detection could be managed as the different aspects of the design were completed.

The 3D earthworks models were particularly useful to earthworks contractor Walters, as the models could be used by its GPS-enabled plant for profiling earthworks. The plant had a range of features, including an automated cut off system to prevent over-dig.

Throughout construction, the 3D BIM model was used to identify areas where temporary works could prove particularly challenging. As a result, the design of a number of structures was refined, including the reinforced soil retaining walls at Carno, which were amended to minimise excavations, to enable simpler temporary works during poor weather conditions and to speed up construction.

Close collaboration and updating using actual ground information has also enabled the design to be refined throughout construction. This has been particularly beneficial to the programme, for example where zoning of materials in the Carno embankment has been redesigned at various stages in response to changes to planned earthworks.

Mining risk treatment and mitigation

Mine entrances and shallow workings of the coal seams and intervening ironstone horizons presented hazards to construction and future operation of the road along most of the alignment.

Treatment and mitigation measures were developed, based on ground information gathered during construction, which allowed them to be optimised and avoided unnecessary treatment. All of these measures were developed in close liaison with Environment Agency Wales (now Natural Resources Wales).

Many of the shallow workings below the alignment form groundwater connections with the Sirhowy, Ebbw and Clydach Rivers. The design of mitigation and treatment measures was therefore largely influenced by potential effects on the hydrogeological regime, the watercourses and wider environment.

Where entrances such as shafts were present, the mining risk posed to construction and operation was generally considered too high for the shaft to remain in place untreated. Measures typically comprised locating the shaft using trial pitting or probing and, if not already treated, backfilling with gravel or grout and placement of a reinforced concrete cap.

The risk of subsidence as a result of shallow workings was assessed based on a literature review of past subsidence events, including the South Wales Mining Desk Study completed by Arup on behalf of the Welsh Office in 1986.

Subsidence risk zones were defined along the full length of the alignment, based on aspects such as the presence of potentially worked seams, the likely seam thicknesses and depth of cover. It was concluded that, even for areas within the subsidence risk zones, the risk of shallow workings leading to subsidence was likely to be very low. However, because of the extent of the area underlain by shallow seams, it is statistically likely that some subsidence events may occur during the design life of the new road.

Obviously, it would be uneconomic and impractical to treat all shallow workings in an attempt to eradicate the potential for future subsidence, high strength geogrids have been used within the road structure above potentially worked shallow seams. While the geogrids will not prevent future subsidence, or negate the need for road repairs if subsidence occurs, they will provide some temporary spanning support to the road and mitigate the immediate danger to road users.

However, when it comes to structures, mining-related subsidence could have far greater effects in terms of safety, repair costs and disruption. As a result, a preventative approach was considered essential for those structures within a subsidence risk zone, typically involving traditional drilling and grouting.

Sirhowy Bridge

The 50m span Sirhowy Bridge is a case in point. It is built in an area underlain by workings from two shallow ironstone seams. These underdrain the site, lowering the groundwater table and providing a connection to the Sirhowy River.

The Sirhowy Bridge was designed to be supported by raft foundations over stiff glacial deposits and bedrock. The plan was to build the bridge in two halves, with the northern half constructed first and traffic transferred onto it. The current road bridge to the south was then to be demolished, enabling the southern half of the new bridge to be built in its place.

However, during construction, the northern half of the bridge was going to be supported by a raft foundation under significant loading, with one edge over shallow ironstone workings.

Traditional drill and grout works to infill the workings would not have been permitted by Natural Resources Wales, because of the potential environmental impact to the Sirhowy River. However, because the foundation was relatively small, it could not be designed to economically span or cantilever over the ironstone workings, without significant risk of instability and tilting. Instead, it was decided to infill the workings with gravel.

Using gravel rather than grout eliminated the risk of grout migrating into the river and the grade of gravel used allows continued groundwater flow through the workings. Close spacing of drill holes, at 0.75m centres, was specified to ensure voids were filled adequately.

The many innovative solutions adopted on section 3 of the upgrading of the A465, has been greatly influenced by the early input of the contractor during the preliminary design phases, plus the continued input of the designer and adaptation of the design during construction. This has highlighted the benefits that can be gained by ECI and collaboration to efficiently manage ground related risks and uncertainty.

Mark Cooper is a senior geotechnical engineer at Arup