Rockfall

Prevention — much better than cure

When it comes to a quarry manager’s list of worst nightmares, few things rank higher than an uncontrolled rockfall. The potential risk to personnel and plant makes rock slope stability a key issue. Neil Tester from Carl Bro — consultants and engineers, sheds some light on the issue, using the company’s experience at Judkins Quarry in Warwickshire as an example.

Judkins Quarry operated for around 100 years producing aggregates and roadstone. It was a large quarry covering some 2.5ha and reaching a depth of 100m. Carl Bro’s involvement with the quarry has spanned two different owners, currently (WRG) Waste Recycling Group. In 2000 it became clear that the quarry had particular problems with rock slope stability which posed a potential risk to personnel, plant and the future integrity of any landfill lining system that might be adopted by the new owner.

Over the past three years a phased approach has been adopted with regard to inspection and stabilization works, including the following:

• initial inspection
• stability assessment
• risk assessment
• remedial works detailing and design
• supervision of remedial works.

The geology of the area surrounding the quarry is fairly complex and mainly comprises Precambrian and Cambrian basement to the west and much younger Triassic Mercia Mudstones to the east. These are separated by the major north-west to south-east trending Polesworth fault, which follows a line roughly parallel to the Coventry Canal just to the east of the quarry.

A detailed understanding of the structure and nature of the local geology has been built using a combination of available data, including detailed inspections. Each area requiring examination was the subject of an initial survey, including geological and discontinuity surveys. Where appropriate, the orientation, frequency and interrelationship of discontinuities was established. At Judkins, instability of the relatively intact rock was controlled by the frequency, orientation and condition of joints and fractures within the rock.

So what are the key criteria in rock slope stability assessment? Intact rock tends to fail by three main mechanisms, all of which are controlled by the orientation, frequency and nature of discontinuities within the rock.
The three mechanisms are:

• Plane Failure — this describes the sliding of a rock block or mass along a basal slip plane. Three conditions must be satisfied for instability to occur:
  (i) The azimuth (direction of dip) of the slip plane must lie within approximately 20° of the slope azimuth.
  (ii) The slip plane must daylight (outcrop) on the slope.
  (iii) The dip of the slip plane must exceed the friction angle of the plane.
• Wedge Failure — this occurs where two intersecting planes are present which dip towards each other. Movement takes place in the direction of the line of intersection. Two conditions need to be satisfied for instability to occur:
  (i) The line of intersection must daylight on the slope.
  (ii) The dip of the line of intersection must exceed the friction angle of the planes.
• Toppling Failure — this occurs where a basal plane combines with at least two sets of steeply inclined intersecting planes to produce vertically elongate blocks which overbalance and fall out of the slope. This mode of failure requires two conditions to be satisfied:
  (i) Two sets of cross-cutting discontinuities need to be present. The intersections of these discontinuity sets need to dip out of the slope.
  (ii) Basal planes dipping out of the slope at angles less than the friction angle should be present.

Often a combination of these mechanisms will operate to produce instability. In order to assess stability by these mechanisms it is normal to plot discontinuity data on a stereographic projection that allows three-dimensional data to be represented in two dimensions. Data plotted in this way is then overlain by an envelope that defines the limits of stability for the relevant mode of failure. By factoring in the angle and direction of inclination of the slope, the probability of failure can be established. The shape of the failure envelope is defined by the dip and dip direction of the design slope and the angle of friction assumed to act along the discontinuities within the rock mass.

Other failure mechanisms can occur, such as ravelling, particularly if the rock mass is heavily fractured, while failure by circular rotation or translational methods can occur in a manner akin to a soil slope if the rock is very weak and weathered. Failures of this type can be given a ‘factor of safety’. Visual inspection and, where appropriate, intrusive investigation will also often reveal any other problems that are likely to occur.

For most modes of rock slope failure, however, it is not possible to determine a factor of safety against failure. According to Richard Apted, technical director with Carl Bro’s environment team, in order to quantify risk, Carl Bro’s approach is to carry out a risk assessment based on the probability and ramifications of failure.

In the case of Judkins Quarry, having assessed the probability of instability visually and by analysis of discontinuities, each of the survey areas was individually appraised. The probability, consequences and risk of failure were established in the immediate-, short- and long-term cases, as required by the client. This resulted in the production of a risk matrix, a typical example of which is partially reproduced in figure 1.

Having identified and quantified the inherent risks at the site, work began on remedial designs. These have to be designed in cognisance of two key criteria: client budget and prioritization of the highest risk areas. In order to facilitate the implementation of cost-effective solutions within tight timescales, a proactive partnering relationship was established with the successful contractor. This involved value engineering the works as they were going on, with a commitment to amend and alter the design as necessary to make best use of available funds. This proved especially suitable for the stabilization works, as on this site it was almost impossible to economically evaluate the full extent of the works in advance.

Works carried out to date include:

• Relatively limited ‘permanent’ stabilization works (60-year design life) on the uppermost lift of the quarry face. This comprises PVC-coated steel mesh draped over the rock face (2,800m²) plus necessary top and bottom anchorages and reinforcing cables.
• Quite extensive ’temporary’ stabilization works (5–10-year design life) on selected areas of the lower lifts of the quarry face. This comprises steel mesh draped over the rock face (8,000m²) plus necessary top and bottom anchorages.
• Heavy and light hand-scaling of loose rock from seleced areas of all faces, local benching and regrading of scree (10,200m²) to reduce the risk of rockfall on less unstable areas and those areas to be covered in rock fill.
• Construction of rock catch fences, bunds and provision for rock bolts locally as required.

A key component in the successful completion of the recent phases of the project at Judkins has been effective supervision of the ongoing remedial works. This has mainly involved continual monitoring of budget costs and, where possible, modification to the design to suit actual conditions.

For example, this process resulted in revisions to the anchorage systems. Anchorage details of the rockfall protection mesh varied from area to area, making it possible, where applicable, to use anchorage trenches at the top and base of mesh areas in lieu of mechanical systems. During installation of the mesh, key areas were reinforced with additional steel cabling where the condition of the rock and stability deemed this to be required.

With regard to the other works, the most significant change was the substitution of the rock catch fence with a rock catch bund. This was considered to be an equivalent but more economic and maintenance-free solution than the initially proposed catch fence. In one area a rebound catch fence comprising a 6m tall fence, with posts set into concrete-filled 1.5m diameter rubber tyres and braced by steel cables for stability, has been used. The cables are anchored to the bench floor with 2m long steel rock bolts set into grout. Maccaferri rockfall protection netting is tensioned between the supports to provide a 3m high catch fence. The overall length of the fencing is approximately 25m.

Completed on time, to budget and to the clients satisfaction, the practical and logical approach to rock slope assessment at Judkins Quarry has been demonstrated to be a clear and concise way to appraise quarry sites and allow effective detailing and design of appropriate remedial works. The partnering approach with the main stabilization contractor proved particularly successful and allowed optimum solutions to be developed in many areas, giving best value without compromising the quality of the finished works.

Carl Bro Group Ltd, 2nd Floor, Spectrum House Edinburgh EH7 4GB; tel: (0131) 550 6300.