Do the developers of small hydroelectric schemes need to concern themselves with the issue of slope instability? This paper aims to answer the question by highlighting the risks to these renewable energy schemes from slope instability and how they may be avoided.

Most developers of civil engineering projects give consideration to slope stability at some stage in the development of the project. The few who choose to ignore potential problems, or do not consider them in the first place, might, with luck, complete the works with no serious problems. However, if a landslide or slope instability does occur, remedial works are invariably time consuming and expensive. They can, in some cases, lead to the abandonment of the project.

In many cases, potential slope instability can be obvious at a site if you know what to look for. In general, it is possible to assess the risk of slope instability from a desk study and preliminary walk-over survey. At the pre-construction stage, changes to the design can usually be made without incurring significant costs. However, once construction has commenced, changes to the scheme become significantly more expensive.

The key questions to be addressed are:

•Why is slope instability important?

•What is the possible impact of slope instability?

•How can potential slope instability be recognised?

•Can slope instability be avoided?

Other background questions such as ‘What are the mechanisms and relationships involved in slope failure?’ will also be discussed.

The importance of slope instability

The effects of slope instability should be considered in relation to the impact it may have on the development, the community and the environment. In this respect, all parties should have the same objective: the maintenance of stability, although there may not be agreement on how it can be achieved.

Major slope failures have recently affected the road network in Scotland1 and the construction of the Derrybrien wind farm in Ireland.2 These events have heightened the concern of the planning authorities where any form of development is proposed in ‘upland’ areas. In particular, planning authorities are now fully aware of the major environmental and economic impact that slope instability can have. The independent reports written about these events confirmed the importance of undertaking a detailed risk assessment of potential stability issues in upland areas. These issues are now the major drivers for the planning authorities. A major landslip on a hydroelectric construction or production site could affect the intake weir, the pipeline or the power house, leading to possible delay, flooding, re-design, re-construction, environmental contamination and additional cost.

All planning consents for windfarms in the UK now require a ‘Peat Stability Assessment’ (or confirmation the windfarm is not to be constructed on peat) and it is believed that similar conditions could be applied to new hydroelectric schemes. There is an existing requirement for all new development in the UK, under Policy Planning Guidance Note 14 – ‘Development on Unstable Land’ (PPG14) for slope stability to be considered as part of the development.

PPG14 sets out a range of guidance for use where developments are planned on unstable land. The guidance identifies three broad categories of ground instability caused by underground cavities, unstable slopes and ground compression of unconsolidated deposits such as peat or alluvium.

An assessment for a hydroelectric scheme or wind farm should thus consider the predisposition for instability, based on geology, slope gradients and soil type or a combination of these.

Impact of slope instability

Hydroelectric plants, by their nature, are generally constructed in mountainous areas with high annual rainfall; they are often in remote areas with no easy means of access for initial reconnoitre or construction. They are also likely to be in an area of outstanding natural beauty. These factors can combine to make it difficult to carry out a ground investigation to aid the assessment of the potential for slope instability. However, it is only by early identification of the risks that any necessary changes to the design can be made at minimal additional cost.

The consequences of landslip during construction will, at best, be a delay in the construction programme, which could result in a loss of revenue. Significant disruption to any part of the infrastructure may require re-construction or re-location and the attendant costs.

A landslip during the operational life of the plant could damage or destroy any part of the infrastructure, resulting in significant capital expense and loss of revenue.

Figure 1 illustrates the damage that can result from a broken 800mm pipe. This particular example was not the result of slope instability. There was about 700m of pipeline above the break with about 20m head of water; the erosion was the result of water held in the pipe and forebay area. There were no automatic shut vales and the pipe was not isolated for some hours; as can be seen there was a significant amount of erosion and all the debris was washed into the loch about one mile (1.6km) downstream.

Figures 2 and 3 are views of a debris flow that entered a river opposite the tail race of a new power house; the debris flow had its origin some 600m higher up the mountain and cut a swath through the commercial forest. A bridge further down stream trapped much of the timber resulting in the river backing up for a considerable distance.

The debris flows in these figures occurred in the summer flooding events of 2004 when there were many debris flows across Scotland that affected the highway network1 (figures 4 and 5).

The consequences of slope failure usually affect not only the immediate construction site but the river and many of the downstream environments associated with the river. These can include the natural ecology and commercial operations such as fish farms that rely on clean water with low suspended solids. The financial claims from a fish farm for loss of stock due to turbid water can amount to hundreds of thousands of pounds.

Avoidance of slope instability

The best way to overcome the risk of landslip is to identify and avoid those areas that may have a pre-disposition to slope failure. This requires an understanding of the complex inter-relationships that can lead to landslip.

Slope instability is largely a natural phenomena, and does not require the interference of man to trigger movement; however, when human activity is involved, the risk or likelihood of instability is significantly increased.

There are a number of well defined stages in determining whether a site is at risk of slope instability and these follow the established guidance for any development. They should include:

•Desk study – to review all available information, particularly detailed maps.

•Walkover survey of the site and adjacent areas, including access routes, pipe route, intake weirs, power house and up-slope and down-slope areas to all permanent and temporary structures.

•Review desk study and walkover to determine if there are any gaps in the information. Then design an intrusive ground investigation to confirm design assumptions.

•Ground investigation – trial pits, drilling, probing and testing followed by a factual report.

•Interpretive report and geotechnical design for cuttings and embankments, and foundations for the access routes, weirs, pipe-route and power house. A note of critical importance is that the design of temporary works must also be considered as these are often responsible for triggering landslides. A vital element in the avoidance of slope instability is an understanding of the relationships between geotechnics and geology, drainage, climate, agriculture and terrain and the mechanisms of instability.

Slope instability mechanisms

Slope instability is broadly classified according to whether the materials involved are rock or engineering soils.3 In the Scottish highland regions, instability in engineering soils is the most common cause of landslip, as natural rock falls are comparatively rare. A brief description of engineering soils instability, based on the Scottish Executive Report on landslides,4 is given below:

•Debris flow – a mixture of fine materials (clay, silt and sand) and coarse material (gravel, cobbles and boulders) with water. The water content is important as it is often the trigger to the instability and it dictates flow velocity, energy and erosive power. Debris with a high water content is described as slurry and great velocity and destruction can be wrought from a flow of such material.

•Peat instability – the main types of peat failure are: a) Bog bursts or bog flows – the emergence of a fluid form of decomposed, amorphous peat from the surface of a bog, followed by the settling of the residual peat surface, in-situ;2 b) Peat slides – the failure of the peat at or below the peat/substratum interface leading to translational sliding of detached blocks of surface vegetation together with the whole underlying peat stratum;2 c) Bog slide5 – an intermediate form of instability where failure occurs on a surface within the peat mass with rafts of surface vegetation being carried by the movement of a mass of liquid peat. Peat is a particular form of engineering soil and is described in the factfile.

Factors to be considered in a stability assessment

There is considerable observational information relating to debris and peat flows although the actual mechanisms involved in peat instability are not particularly well understood. The main influences on slope stability are geological, geotechnical, geomorphic, hydrological, topographic, climatic, agricultural and human, such as drainage and construction activity. The two soil types being considered, unconsolidated sediments and peat, are both affected to a degree by changes in any of the above list, and it is vital to appreciate that changes to the exiting equilibrium will affect the level of slope stability during construction and operation of the scheme.

Some of the issues relating to general slope, and peat instability in particular, are discussed below:

•Establish the geographical limits which could be affected by potential instability. These limits are rarely confined to the artificial boundaries imposed by land ownership; landslip occurring above a site could impact the site and property downslope or downstream of the site for several kilometres.

•Agriculture has a greater impact on peat areas that have been managed to improve grazing. Such management can include surface drainage and periodic burning, both of which can leave the surface of the peat bare for a period of time resulting in temporary desiccation of the surface. Subsequent wetting of the peat and resumption of peat accumulation results in the former desiccated surface being incorporated into the body of the peat which may introduce a discontinuity in the profile; this in turn becomes another unknown factor in the stability assessment.

•Forestry has a significant impact on slope stability particularly in the early stages. The creation of a forest involves disruption of the natural equilibrium and drainage of the slopes by ploughing and the installation of artificial drains. The construction of access roads further disrupts the drainage and concentrates groundwater flow into narrow, fast flowing erosive streams. The work by Winter el al1 noted that forest roads can act to retard or concentrate the downslope flow of water and thus aid its penetration into the slope below. Such a mechanism has been observed at a number of recent landslips that have affected the road network in Scotland.

•Natural drainage – a natural upland peat bog will absorb all of the precipitation that falls on it and transmit the water to the lower slopes in a controlled manner through a range of interconnections that operate at different scales and speed.2 Failure to understand this and to disrupt the transmission process for the groundwater could result in instability.

•Artificial drainage – where agricultural drainage has been used to improve the quality of the grazing or to promote forestry, the effect is to reduce the overall volume of water entering the bog and to transfer this water to the edges. This can result in ditches and streams becoming enlarged, causing increased erosion and a greater silt burden in the stream water.

Figure 6 depicts a drainage ditch that has partially closed and distorted due to slope instability and mass movement of a very shallow peat slope on a large scale.

Shear strength

In geotechnical terms, the shear strength of a soil is the physical characteristic that provides stability and coherence to a body of soil. For mineral soils such as clays or sands, such strength is variously given by an inter-particle friction value and cohesion. Depending on whether the mineral soil is predominantly cohesive (clay) or non-cohesive (sand) governs which of the components of strength control the behaviour of the soil.

For peats, where the major constituent is organic and there is likely to be little or no mineral component, the geotechnical definition of shear strength does not strictly apply. At present there is no real alternative method for defining the shear strength of peat, and the geotechnical definition is generally adopted, in the knowledge that it should be used with great caution.

It is almost impossible to predict a shear strength profile in peat and attempts to measure the shear strength using normal geotechnical methods can be misleading. Typical values of shear strength from hand shear vanes would be in the range 10-60kPa although values over 100kPa have been recorded in fibrous peat. The higher strengths are almost certainly the influence of roots or other non-decomposed material.

Stability analysis

The stability of slopes is dependent upon the shear strength of the soil to resist the disturbing forces due to the weight of the soil, the effects of the groundwater and other disturbing influencing forces.

The level of stability of a slope is normally assessed by reference to the factor of safety which is expressed, numerically, as the degree of confidence that exists for a given set of conditions, against a particular failure mechanism occurring. It is commonly expressed as the ratio of the load or action which would cause failure against the actual load or actions likely to be applied during service. This is readily determined for some types of analysis (e.g. limit equilibrium slope stability analyses).

By using geotechnical numerical methods, the factor of safety (FOS) for a given set of conditions can be determined.

The most common quantitative analysis assumes that failure will occur on a plane, approximately parallel to the ground surface and a computer program is generally used to search for the minimum value. The results of a peat stability analysis using undrained shear strengths as low as 10kPa give unrealistically high factors of safety for slopes that are known to have failed. It is therefore concluded that a standard, yet rigorous geotechnical, numerical stability analysis, when combined with shear strengths determined from vane shear tests, does not necessarily give a reliable indication of current or future stability in peat soils.

At present there is no real alternative to the measurement of shear strength as samples recovered for laboratory testing are affected by the sampling, transport and testing regime and there is no way of accounting for the effects on the true strength of the peat.

The problem with a quantitative assessment is that it requires a numerical input and the analysis cannot account for the un-quantifiable input required for a comprehensive peat stability assessment. For this reason a purely quantitative stability analysis should only be considered as a guide and a qualitative (risk) assessment of stability should be used to provide the final recommendations. It is therefore concluded that a qualitative assessment of peat stability is the preferred method with due cognisance to the quantitative assessments to allow a balanced and reasoned set of conclusions to be developed.

From the foregoing discussion, it should be apparent that the stability assessment of peat is both complex and not fully understood and that the location and impact of all the influencing factors should be taken into account.

Conclusions

Landslides are a naturally-occurring phenomena, the pre-disposition for which can often be predicted. Construction activities can exacerbate potential instability and initiate instability on otherwise stable slopes by the adoption of inappropriate design or construction methods.

The influence and interrelationship of topography, geology, drainage, climate and construction must be studied in order to make an assessment regarding the suitability of a particular site. Having made an assessment, the geotechnical engineer should become an integral member of the design team to ensure that changes in, say, route alignment, do not result in possible instability.

The landslides that closed much of the Scottish road network occurred during a period of intense rainfall in August and there is evidence that late summer is a high risk period for slope instability in non-peat areas. One of the predictions of global warming would be wetter winters and drier summers with more frequent, high intensity rain storms.

Peat soils are a particularly difficult material to deal with as normal geotechnical strength assessments do not work particularly well with the result that stability analyses can indicate stable slopes in areas that are known to have failed. For this reason peat stability should be based on a qualitative assessment with a numerical stability assessment being one of many contributing considerations.

Potential slope instability, particularly peat instability, is believed to be critical to the successful development of a hydroelectric scheme that it should be fully addressed as part of the pre-planning design works. Then as the detail design progresses, the geotechnical engineer can assess the likely impact on stability, of design and pipe route alignment changes.


Author Info:

Ian Uglow, SLR Consulting Ltd, Treenwood House, Rowden Lane, Bradford on Avon, Wiltshire, BA15 2AU. www.slrconsulting.co.uk

This paper is based on a presentation given during the Hidroenergia 06 conference in Scotland, UK, an event organised by the British Hydropower Association and the European Small Hydropower Association


Peat

Peat is found covering extensive areas of UK, Eire, Western Europe and Canada. It is defined as the partly decomposed plant remains that have accumulated in-situ, rather than being deposited by sedimentation. When peat-forming plants die, they do not decay completely as their remains become water logged as a result of regular rainfall. The effect of water logging is to exclude air and hence limit the degree of decomposition. Consequently, instead of decaying to carbon dioxide and water, the partially decomposed material is incorporated into the underlying material and the peat ‘grows’ in-situ.
Peat is characterised by low density, high moisture content, high compressibility and low shear strength, all of which are related to the degree of decomposition and hence residual plant fabric and structure. To some extent, it is this structure that affects the retention or expulsion of water in the system and differentiates one peat from another.
Peat is an almost entirely organic material and does not have a shear strength in the normal geotechnically accepted definition. A consequence of this is that peat can and does fail on extremely shallow gradients
There are two distinct layers within a peat bog, the upper acrotelm and the lower catotelm.
The acrotelm is a fibrous surface to the peat bog, typically 0.5m thick; the acrotelm exists between the bog surface and the lowest position of the water table in dry summers. The upper surface of the acrotelm is the surface of the bog and comprises growing plants. Below this are various stages of decomposition of the vegetation as it slowly becomes assimilated into the body of the peat.
The acrotelm comprises a matrix of plant material bound by live roots, which provide a significant tensile strength. The near surface material is also highly permeable, which is probably where the misconception arises that peat as a whole is highly permeable. Deeper in the acrotelm, as a degree of decomposition occurs; the plant roots and fibres are compressed more tightly, due to the weight of the overlying material, resulting in a decrease in the permeability.
The majority of the thickness of a peat bog more than 1m thick can be catotelm which comprises completely saturated, highly decomposed material, the strength and permeability of which, are very low. The catotelm may contain a natural drainage system made up of soil pipes which may or may not be directly connected to the surface.
The shear strength of Peat is typically measured in-situ with a hand shear vane in trial pits. The instrument records the undrained shear strength of the material, and for a low permeability engineering soils, the method is believed to give a reasonable approximation to the variation and consistency of the strength. There is a caveat that should be considered in relation to peat, and that is that the hand shear vane does tend to seriously over-estimate the actual shear strength that may be available in the field. This is due to the small size of the vane and the disproportionate influence of the peat fibres on the recorded strength. Therefore, in practice, the shear strength parameters used in any stability analysis should be factored down from those recorded by the hand vane.
There are many influences on the stability of peat and observing or measuring a high shear strength should not be used to assume a high degree of stability.