Geotechnical Observations specialises in the measurement of pore water pressures and soil suctions. As part of our commitment to furthering the understanding of soil suctions we have prepared the items of information on soil suction, what it is, where it occurs and how it can be applied.

What is soil suction?

The water in soil voids below the water table is normally continuous. The soil may be saturated, with voids full of water or there may be occluded air bubbles present in the water. Pore pressures at depths below the water table are derived from a combination of the weight of the water lying above the given elevation and the drainage conditions below. The pore pressure normally has a positive value and can be measured using a saturated piezometer with a porous filter that is making intimate contact with the water in the soil.

If the water contained in the voids of a soil were subjected to no other force than that due to gravity, the soil lying above the water table would be completely dry. However, powerful molecular and physico-chemical forces acting at the boundary between the soil particles and the water cause the water to be either (a) drawn up into the otherwise empty void spaces or (b) held there without drainage following infiltration from the surface. The attraction that the soil exerts on the water is termed soil suction and manifests itself as a tensile hydraulic stress in a saturated piezometer with a porous filter placed in intimate contact with the water in the soil.

The magnitude of the attractive force that soil above the water table exerts on water is governed by the size of the voids in manner similar to the way that the diameter of a small bore glass tube governs the height to which water will rise inside the tube when it is immersed in water. The smaller the void, the harder it is to remove the water from the void.

The meniscus formed between adjacent particles of soil by the soil suction creates a normal force between the particles, which bonds them in a temporary way. Thus soil suction, if it can be relied upon, can enhance the stability of earth structures. However soil suction also provides an attractive force for free water, which can result in a loss of stability in loosely compacted soils or swelling in densely compacted soils.

Soil suctions can be found in all ground that lies above the water table. This may be natural level ground or slopes, fill materials and other earth structures that are constructed above the water table. Soil suctions will also be present in samples that have been recovered from a ground investigation. Laboratory measurements of suction can be very useful for assessing the quality of the samples, estimating the in situ effective stress and detecting the presence of desiccation.

The vertical and horizontal total stresses and the pore water pressure combine on an element of soil in the ground profile to give the in-situ effective stress. When the same element of soil is removed from the ground in the form of a sample the vertical and horizontal total stresses reduce to zero. However, the sample retains a pore water pressure but its value is negative because of the unloading that has taken place.
 
If the extracted sample is (a) saturated, (b) removed from the ground without allowing it to change in volume, (c) at its in-situ water content and (d) free from disturbance, the measured suction in the sample can be used to estimate the in-situ stress condition. Moreover, if a value for the vertical effective stress is assumed, it is possible to calculate the radial effective stress and hence the value of K₀.

Pore pressures and suctions in cut slopes and excavations

When clay ground is excavated to form a cut slope the reduction in effective stress can induce negative pore water pressures if the ground remains undrained. This can give the slope a temporary stability in the short term, which can be lost if the pore water pressures increase. Delayed failure of clay slopes is a common occurence in the UK.

The time to failure is frequently many decades and can be influenced by the presence of surface vegetation. The potential for instability can be assessed by measuring the insitu pore water pressures (which could still be negative) and comparing them with the long-term expected pore water pressures. Analysis will reveal the expected time to failure.

GeO flushable piezometers and suction probes are ideal for monitoring the pore water pressures in cut slopes. The figure below shows the pore water pressures recorded 40 years after the construction of a cutting formed in stiff clay and with dense surface vegetation on the upper part of the slope.

Pore pressures and suctions in embankments

Fills and compacted soils (such as those used in embankments) have an inherent suction when they are first compacted. The magnitude of this suction can be a few hundred kilopascals even under normal compaction conditions.

Infiltration into fills or compacted ground will increase the water content and reduce the suction. Frequently therefore the assumption of no suction is used for the assessment of slopes. Attempts to retain suction and hence stability by the introduction of vegetation onto slopes has had mixed success. Aside from providing natural drainage, vegetation introduces root reinforcement, it intercepts a significant amount of the infiltration that can lead to the loss of soil suction and it is often aesthetically pleasing.

However, vegetation also creates seasonal drying and wetting, which causes subsequent changes to the shallow in situ pore water pressures. Recent field observations suggest that old railway embankments constructed of dumped London clay fill deform as the shallow pore pressures vary in a seasonal manner. Finite element analyses indicate that seasonal cyclic stress changes cause a gradual outward movement, which induces strain softening and this can eventually lead to collapse through a mechanism of progressive failure (Kovacevic et al. 2001). These analyses suggest that the horizontal mid-slope movements and the number of cycles required to cause failure are linked to the amplitude of the pore pressure variation. Retaining a small suction at the boundary at the end of winter can significantly prolong the time to failure. Therefore if analyses such as these are to be used, in pro-active way, to assess the serviceability of embankments it is essential that they be fed with good data on the seasonal variation of the pore pressures, obtained from reliable field measurements.

The figure below shows data gathered from an old railway embankment in the south east of England. This was published in The Skempton Conference.

References

Kovacevic, K., Potts D.M. & Vaughan P.R. (2001) Progressive failure in clay embankments due to seasonal climate changes. Proc. 15 th Int. Conf. Soil Mech. Geotech. Engng , Istanbul , 3 , 2117-2130.

Ridley A.M., Vaughan P.R., McGinnity B. and Brady K. Pore pressure measurements in infrastructure embankments. In "Advances in Geotechnical Engineering: The Skempton Conference, 2004" , Thomas Telford, London . Vol. 2, pp 922-932.

Geotechnical Observations Limited
The Peter Vaughan Building
9 Avro Way, Brooklands, Weybridge
Surrey, KT13 0YF, United Kingdom

Tel: +44 (0)1932 352040
Fax: +44 (0)1932 356375
Email: info@geo-observations.com
Web: www.geo-observations.com