Most earthworks structures are designed to be in place for a long time. The typical design life of an embankment or cutting can be between 60 and 120 years.

Many structures in use today are even older. But, as Dr Tom Dijkstra, senior lecturer in engineering geology at Loughborough University, notes: “The weather now is very different from what it was 120 years ago and it will be very different in another 120 years.”

He adds: “I can’t overemphasise the importance of changing conditions.”

As a result of climate change, we’re experiencing harsher weather, including heavier rainfall, more extreme temperatures (high and low), stronger winds and higher sea levels. All of these can change how the the material in an embankment or cutting behaves. They may affect its strength or stability, resulting in damage, a loss of integrity or even a catastrophic collapse (see Table 1).

Earthworks structures have been used for a wide range of purposes for many years. They include embankments and cuttings for railways and roads; and flood barriers; dams and other water-retaining structures.

In countries that experienced significant industrial growth in the 18th and 19th centuries, many of these structures were built quickly and without following standards governing even the most basic engineering criteria, such as the minimum slope angle. Meanwhile, some structures built in the past half-century that would meet current standards are at or beyond their design life and have been subjected to decades of action by various weather effects.

As a result, failures occur. Some are catastrophic, such as the landslip that caused a fatal derailment in Aberdeenshire in August 2020; others are relatively minor but disruptive, such as the rock falls that occur regularly in railway cuttings in the south of England. Each time a failure happens, it seems to be unexpected – a structure that has withstood wet weather or high temperatures for decades suddenly loses strength or integrity.

Until recently, earthworks have been treated as if they’ll continue to perform as well as they did on day one, when in fact they are deteriorating.

Evaluating the Deterioration of Geotechnical Infrastructure Assets Using Performance Curves is a paper given at the 2019 International Conference on Smart Infrastructure and Construction. It notes that deterioration is caused not only by the loading imposed on a structure (by traffic, for instance) but also by environmental actions such as seasonal pore water pressure cycles as well as changes in peak pore water pressures caused by extreme weather conditions. 

These actions change the properties of the structure both at the soil scale (e.g. a change in strength, stiffness, permeability or structure) and at the asset scale (e.g. changes in slope geometry or structural integrity owing to deformation). 

The paper explains that the performance, behaviour and deterioration of an earthworks structure over its lifetime can be described using an inverted ‘bathtub’ curve, by plotting performance on one axis and time (or asset age) on the other (see diagram below).

Every earthworks structure will have a slightly different deterioration curve, depending on how well it was constructed, the materials used and the loading and environmental conditions it is subjected to. For organisations and individuals who own and operate such structures, the key question is, “Where is my structure on this deterioration curve?” – as an effective maintenance intervention during stage 3 could restore performance and extend its lifespan.

A substantial body of research

The ACHILLES programme, comprising the British Geological Survey and the universities of Newcastle, Loughborough, Durham, Southampton, Bath and Leeds, has been collecting data on the performance of earthworks for 20 years. It is using that data to develop predictive models to identify the rates of deterioration of different materials and structures.

This information will help asset owners and engineers to plan interventions such as monitoring, maintenance and renewal activities and to forecast their costs, timing and benefits.

The programme has modelled the impact of future climate scenarios on a representative section of cuttings in London clay in southern England. 

The model simulates the processes that affect slope pore water pressures – including meteorological information, infiltration, evapotranspiration and overland flow – to understand how pore water pressures change and how slopes deform as a result.

The researchers found that the magnitude of annual fluctuations in pore water pressure plays a significant role in the rate of progressive failure, and this annual variation is higher in future climate scenarios, leading to a more rapid deterioration in strength. 

The value of saturated conductivity also plays a role in time to failure, with the stand-up time of the cut slope being longer when the hydraulic conductivity of the material is lower.

In the UK, different asset types will have different deterioration profiles because of their history. The first embankments on the rail network were constructed almost 200 years ago, and were designed and built empirically. Many failed soon after construction, leaving the structures that are still used now with variable geometrical and historic weaknesses. 

Construction of the strategic road network began in the late 1950s and continued into the 1990s, with highway embankments being built using design standards based on saturated soil mechanics theory and informed by research into the properties of the fill materials. They are younger than those on the rail network, are less likely to include zones of weakness and are more homogenous in their geometry and material composition. However, many are reaching the end of their planned life of 60 years.

The embankments supporting the UK’s high-speed railway lines are much younger than those on the rail or strategic road networks: the first section of High Speed 1 opened in 2003 and the earthworks structures may still be in their bedding-in period, while structures for HS2 are currently being built. Other earthworks structures, including flood defences and dams, have been built throughout this period.

The lifespans of these different assets and their place on the deterioration curve can be represented graphically (see diagram below).

Some of the oldest earthworks structures in the UK are owned and managed by Network Rail. Its asset inventory includes 70,000 soil cuttings, 20,000 rock cuttings and 100,000 embankments, formed from a variety of materials.

Manos Tsoukalas, senior asset engineer at Network Rail, says: “Pick any weather extreme – high wind, rainfall, temperature extremes – and you’ll see an effect on earthworks structures, either from an individual weather effect or a combination.

These include:

• Heavier rainfall. Climate change is resulting in heavier rainfall, which increases the pore water pressure of the material in a cutting or embankment. It can also cause damage through indirect impacts – for example, a drain or ditch at the base of an embankment may overtop and wash out the toe of the earthwork. “It could be something that has been fine for 200 years but fails because now we get more water,” Tsoukalas says.
• Hotter temperatures combined with higher rainfall. In southern England, climate change is resulting in higher temperatures alongside prolonged periods of rain. A clay embankment subject to these conditions will expand when it is wet and shrink when it is dry and hot. If this is combined with overgrown vegetation and thirsty trees, the soil will lose even more moisture, leading to the ground above losing shape and affecting the quality of the track above. “Because we are getting a lot more heat in summer, desiccation is an issue,” Tsoukalas says. “We have plans in place with our track maintenance engineers to identify the thirsty trees on embankments and remove them individually. Taking out trees with large canopies makes a huge difference.”
• Wind and tall trees. The load from high winds on tall trees can cause the ground to move, which can result in a failure even when there is no rainfall effect.
• Colder temperatures. The year 2021 was the worst on record for frost at Folkestone in Kent, causing the faces of chalk cliffs to become brittle and creating small fissures that produced rubble. “In the past we would have designed for the danger of large boulders forming, but now we have to take different actions because we are getting this ‘rubbling’, where it is raining small rocks,” Tsoukalas explains.

Network Rail is installing remote condition monitoring (RCM) and cameras on its high-risk earthworks structures to help manage the network in adverse weather events. “The technology has helped us to identify trends in those high-risk areas,” Tsoukalas says. “They are used to stop the operation of trains in a landslip situation.”

The RCM consists of sensors fitted with tilt meters, which register movement. Network Rail has set allowable movement thresholds above which the line would be closed automatically. Cameras linked to the sensors can relay images to show engineers exactly what is happening.

RCM was initially installed to give advance warning of a failure – particularly of a landslip in a cutting – but Network Rail is now using it to help plan maintenance interventions. For example, following two failures of a section of cutting at High Brooms in Kent, the organisation planned immediate remedial measures. Instead, RCM was installed, and the cutting was monitored until the sensors started showing that there was movement. “By installing the RCM we were able to keep the line open for another year,” Tsoukalas says. “The RCM told us when we needed to intervene, which meant we could get the maximum life out of the asset.”

Network Rail is also trialling products that measure water using sensors in watercourses to provide information about what is happening in rivers, pipes and drainage systems near its earthworks structures.

TfL has studied the effects of inclement weather on its earthworks structures. It has identified four main modes of failure:

• Flow failure of granular material resulting from high-intensity short-duration rainfall
• Embankment failure caused by pore water pressure increase as a result of prolonged rainfall on structures with low soil moisture deficit
• Scour erosion at the toes of embankments from watercourses running through, or parallel with, the structure
• Shattering of rock in chalk cuttings caused by cold temperatures

TfL has identified all of the assets that could be affected by one or more of these failure modes and categorised them as high, medium or low risk. It has also instigated a network-wide heavy rain and flooding plan and installed rain gauges across London to provide live data. If rainfall reaches a trigger level, speed restrictions are applied automatically.

Much of TfL’s network is built on embankments or in cuttings formed of London clay, most of which are too steep by today’s standards. The material is prone to shrinking in high temperatures, especially where there is mature vegetation. This can cause track deflection.

Under design codes for earthworks structures, design engineers must consider the worst case, but there is no requirement to look at the cumulative effects of changes in the long term.

It may be possible to anticipate some of the impacts of climate change, such as higher pore water pressure or cracking caused by extreme cold, and design these in – e.g. by reducing the slope angle of a cutting or installing extra drainage. This does not acknowledge that the structure will deteriorate over time, though.

Road pavements are designed for a certain lifespan (often 40 years), but there’s an assumption that they will be maintained regularly throughout that time and there will be interventions – e.g. resurfacing – because some elements are expected to deteriorate.

This is not the case for most earthworks structures. They are also designed for a specific lifespan – 120 years in some cases – but there has been an assumption that they will perform to the same level as they did on day one. There is no acknowledgement that a structure will deteriorate or that it will need interventions to maintain its integrity.

But, if the structure is designed from the outset with monitoring built in and a clear maintenance strategy, there is less chance that it will fail and every chance that it can reach or exceed its lifespan. Modelling by the ACHILLES programme has shown the economic benefits of early intervention. 

Acknowledging that earthworks structures do deteriorate, and that the effects of climate change contribute to that deterioration, will affect how they’re designed and maintained to maximise lifespan and minimise failure.