Average expenditure on site investigation (SI) for underground construction it has been reported has reduced to around one per cent of project costs, whereas the expenditure on site instrumentation and monitoring for a project in progress, particular in urban areas, has increased. This obviously reflects concerns about the consequences of third-party damage cause by settlement, but appropriate investment in information collection before final project plans are made will reduce project risk, including that of third party damage. In any case, some degree of site investigation will be needed to select appropriate sites for monitoring instrumentation.

The main difficulty is drawing the line on how much information is necessary, and how much it will cost. Obviously a tunnelling project, even in a high-risk area such as the centre of a city, is not likely to be able to budget for the level of exploration expenditure made in the oil and gas industry, and some mining projects.

Quiet apart from the general desire to minimise general risk, some of the specific reasons for SI are:

• The characterisation of rock to be excavated so that the best method, or methods, can be selected;

• The materials and structure of the ground bordering the excavation so that the best means of primary and permanent support can then be selected and designed;

• Coupled with the above, the development of a geotechnical baseline report that can be used as a reference for informing potential contractors, and aiding design;

• An assessment of likely and acceptable degrees of settlement, including interaction with any adjacent sensitive structures in the ‘zone of influence’;

• The avoidance of ‘nasty surprises’, of whatever type, that may be hazardous or otherwise delay project progress. These could include:

a) Major changes in geology, including boulders in glacial deposits and buried fault zones that may cause of change in tunnelling method or access points;

b) Cavities such as abandoned tunnels, karstic structures, and other solution structures;

c) ‘Hidden’ water courses, especially artesian aquifers;

d) Buried steel, such as piles, abandoned pipes or debris in ‘fill’, that cannot be easily excavated;

e) Unexploded ordinance from warfare;

f) Other hazardous materials, whether naturally occurring or from old industrial processes, such as methane, oil, fibrous minerals, other carcinogenic substances, arsenic, and acid water;

g) Unexpectedly deep foundation structures, especially wooden piles that may affect the stability of surface structures even if they can be cut through. These could also include items of archaeological significance.

Methods
For their part, exponents of site investigation have adopted new technologies aimed at minimising the time and cost involved in site investigation, as well as improving accuracy. For many observers site investigation means boreholes and their sampling and instrumentation, as described in the next article, but there are many approaches that can gain valuable information.

Before investing in an active site investigation programme, it will probably be worthwhile to investigate what existing information is available on the ground conditions, structures and other data likely to be pertinent. This may include data from previous tunnelling or mining, boreholes undertaken for national geological investigations, mineral prospecting or water, oil and gas wells. The amount of data available will vary greatly from one region to another.

Even where there is substantial information available it may well be that there are major gaps in the knowledge that can indentify the main targets of a physical SI programme.

Specialist contractors usually have a wide range of geophysical methods available for site investigation. Many of these are based on identifying changes in density (including voids) or moisture that can reveal structural and material differences. According to Fugro Aperio, a specialist in engineering geophysics and other non-destructive structural testing techniques who claims to have probably the world’s biggest team of engineering geophysicists, survey techniques include:

• Ground penetrating radar (GPR),
• Ground resistivity/ERT,
• Ground conductivity,
• Magnetometry,
• Seismic methods including refraction and reflection,
• Multichannel analysis of surface waves (MASW- seismic),
• Microgravity,
• Cone penetration testing (CPT) at the surface coupled with advance positioning methods and any required sampling. These methods can present both numerical data and charts or sections that, with specialist interpretation, can display anomalies.

Ground penetrating radar found its first applications in ice fields, locating shallow non-metallic utility lines, and even in crime-scene investigation to find suspected buried bodies. In recent years it has been developed rapidly to detect all forms of subsurface anomalies, objects and cavities. It has also been adapted to detect voids behind existing tunnel linings, and has even been tried in the heads of TBMs as advance warning of problems in advance of the face. Fugro Aperio secured a 3-year Royal Society award to support research into ground penetrating radar for civil engineering applications in conjunction with Keele University, Staffordshire, England.

Penetradar Corp of Niagara Falls, New York state, has developed a number of ground penetrating radar antennae including compact devices than can be used on suitable carriers for investigating the state of existing tunnel linings including lining deterioration, subsurface moisture and its location, debonded surface tiles and lining separations.

Resistivity surveys, which, like GPR, have found widespread application in archaeological investigations, are another popular method of investigating conditions below the surface, carried out by applying an electrical test current between two dipole electrodes. Developments in improved electronics including miniaturisation has led to much more portable instruments well suited to field work, even in remote locations.

As an example, Hyundai Engineering & Construction Co carried out site investigation for mountain tunnel construction for a motorway in South Korea using the Sting/Swift automatic resistivity imaging system from AGI (Advanced Geosciences, In) of Austin, US. As can be seen from the accompanying illustration of one of the survey lines, a longitudinal section along the 2km highway tunnel alignment was covered using 27 electrodes at 20-m spacing for a total length of 520m. The dipole-dipole electrode array was used, and the recorded data of apparent resistivity was converted to ‘true’ resistivity, with corrections for topography, using the DIPRO software developed by Dr Kim of the Korean Institute of Geology, Mining & Minerals. High resistivity zones, shown in red (Figure 1, opposite), indicate solid rock, whereas low resistive areas (yellow, green and blue) indicate fractured zones with water, making the zone more conductive.

Several zones of low resistivity could be identified. During construction progress stopped at 380-350m at the weak zone for probe drilling at the face to better locate the weak zone, which was confirmed at 350-
340m. The resistivity indicated fracture zone at 80-140m corresponds to the findings of a geological survey.

As outlined in the next article, several investigative methods have been developed into modules suitable for borehole use. In addition to obtaining quality samples from its boreholes, both onshore and offshore, for the Lake Mead Intake Number Three project, Crux Subsurface carried out additional surveys through its Crux Oriented Borehole Logging (COBL) service. In situ testing included optical and acoustical televiewer surveys, natural gamma radiation detection, a temperature profile, permeability testing and dilatometer testing. Crux was working with MWH and CH2M-Hill for the Southern Nevada Water Authority.

The importance of detecting historic sites in advance of proposed tunnel routes is illustrated by the case of the Bosporus crossings. As part of the Istanbul Strait Highway Transit Project, the ATAS Company has prepared a plan to preserve historical artefacts on the European side from the Cankrtaran shoreline to Haydarpasa. This is to avoid the delay of about 3.5 years to the previous Marmaray Project. GPR was used to scan the tunnel route to detect historical remnants. This and other geophysical methods have been used especially at underpasses and cut-and-cover sections.

Pilot
In some circumstances tunnelling project promoters go to the opposite extreme in site investigation by driving a pilot bore on the alignment. This is not likely to be an economic option unless the bore were, in some way, to form part of the ultimate structure, such as by enlargement or perhaps by forming a service or access tunnel. The major advantage is that a pilot bore can demonstrate the behaviour of excavated ground on a more lifelike scale. There could also be a political element to the decision to make a pilot bore such as to create a near fait accomplis, or to demonstrate progress in full construction when the promoters are still being cautious or do not have full funding available.

It is likely that a pilot bore would still be preceded by a more conventional site investigation programme such as to check for structural anomalies that could make a pilot bore a waste of time if it were planned for a poor alignment.

Project monitoring
Although not strictly site investigation, the process of monitoring for any movements which may be a result of underground construction activity has important links with original site investigation and its findings. The information found during site investigation, especially related to the location and condition of existing structures, and any other major features such as caves and waterways, may have an important bearing on planning the instrumentation stations for monitoring.

As part of the upgrading of Victoria Underground metro station in London, Itmsoil installed, commissioned and managed a range of instrumentation including five Leica TM30 total stations to monitor movement of buildings within the predicted zone of influence.

As part of its R&D programme Itmsoil has been developing special low-profile reference prism reflectors used as surveying targets. Of particular relevance to Victoria is a miniature version that is deployed on several listed and other ‘sensitive’ buildings such as two theatres, the major surface railway terminus and office blocks. Whilst functioning perfectly for surveying purposes they are difficult for the casual observer to note on the buildings, and so do not detract from the aesthetic values of the listed structures. Eight reference prisms are laced outside of the ‘zone of influence’ of the works. In addition to the five total stations on the surface itmsoil has deployed 13 more Leica TM30 total stations underground.

As an indication of the huge investment likely to be required for the monitoring of major underground projects now, on its Crossrail monitoring contract, Itmsoil is installing 300 motorised theodolites associated with many times more reference prisms throughout the scheme.

Within the Soletanche Bachy group, SolData conducts chiefly instrumentation and monitoring activities. Geophysical surveys are mainly the activity of sister company EDG (Europeene de Geophysique). Since 1 January it has become part of SolData and renamed SolData Geophysic. It has a staff of 30 specialists conducting seismic, electrical resistivity and radar tomography surveys amongst other services.