Offshore Defences

© Environment Agency

Sea walls are not the only way to protect the UK’s shoreline; offshore defences can be valuable too. Dominic Reeve, Vanesa Magar and Shunqi Pan, of the University of Plymouth, are at the forefront of research into sea and coast dynamics. They study the probabilities of flooding, lead research into novel coastal engineering techniques and bring together all parties interested in building surer safeguards against the erosion of the coastline.

Coastal engineering has been an active area of research at the University of Plymouth since the 1980s. The possibilities of coastal flooding, which could at any time bring the awesome power of the sea literally into our own back yards, together with the increasing threat posed by climate change, led to the establishment in 2005 of the specialist Centre for Coastal and Ocean Science and Engineering. The centre recognises the need for both innovative solutions and a joined-up approach to safeguarding UK coastlines against the potential threat from a combination of rising water levels, high tides and storm surges.

At least two of these factors were responsible for one of the UK’s biggest recent natural disasters – the 1953 floods which killed more than 300 people in England and resulted in six times that number of deaths on mainland Europe, notably in the Netherlands.

On the evening of 31 January 1953, high tides combined with an unusually large storm surge to disastrous effect. It began with extreme low pressure off the north coast of Scotland, which was funnelled down the North Sea and ravaged the coastline as it sped south. Hurricane winds were recorded at Felixstowe and throughout that night came the flooding. As well as the tragic loss of human life, in the UK alone 40,000 livestock were killed; 1,000 square kilometres of land was inundated with sea water, rendering much farmland unusable for several years; 20,000 homes were flooded and 200 miles of rail track were put out of action. At today’s prices, that disaster cost £5 billion.

Current risks

The threat of another flood on the scale of 1953 remains potent, since the combination of events generating a massive storm surge could reoccur in normal climatic timescales. The relative sea level is rising in the south-east of England and the western part of the Netherlands, with the result that those areas are effectively sinking into the sea. Add to this the likelihood of more frequent and more severe storms – another of the plausible effects of climate change – and the likelihood for flooding is increased.

To counterbalance this natural phenomenon, lessons were learned from the last floods and technology has been improved. The UK has flood barriers, improved weather forecasting, modern communications and sophisticated emergency services designed to help limit the effects of future floods. The Environment Agency, the central body responsible for flood defences, plays a pivotal role in the development of new flood defence schemes. Moreover, Shore Management Plans (SMPs) are now developed more holistically, involving several councils, and taking account of the knock-on effects of individual schemes.

The philosophy behind sea defence design has changed appreciably during the half-century since the 1953 disaster, which was the catalyst for many coastal flood defence schemes along the UK’s east and south coasts. At that time, the design approach was very much considered a ‘battle against the forces of nature’, so most of the defences constructed were mass concrete structures. The UK has experienced storms of a similar severity to those of 1953 in the last 50 years and not suffered similar consequences; that is a testament to the design of these solid defences. Nevertheless, by the 1990s, those defences had suffered significant damage from storms and the Environment Agency commissioned Halcrow, the critical infrastructure consultants, to investigate the performance of a different type of flood defence scheme, where creation and retention of a healthy beach in front of the old seawalls was paramount.

Change of culture

Since the 1950s, many new issues have impinged upon coastal engineering. These include health and safety legislation, environmental legislation leading to more environmentally sensitive engineering and the growing impact of climate change.

These have led to a more systematic, collaborative and innovative approach to coastal flood defence and protection. In the past, a bigger wall would have been the answer. Now, the effective role that beaches can play in flood defence is more widely appreciated, to the point where beach creation and retention is often a key part of newer flood defence schemes.

The risks are not going away, however, and there is now clear evidence that relative sea levels are rising around much of the UK. This, coupled with the associated change in storm patterns caused by climate change, creates a thorny problem, particularly since the natural attraction of coastal regions has seen an increase in housing developments, which has steadily increased the number of people and the value of the assets at risk of flooding.

So, what preventive actions should be undertaken to provide appropriate protection for the UK’s coastal areas? Before looking at some of the issues, the research projects and potential new techniques for coastal defence, we ought to look at some of the principles used when designing flood defences.

Risk factors

Designers do not start with the question ‘How do we prevent any flooding?’, since this clearly cannot be guaranteed. Much more likely is the query ‘What level of event would you like to defend against?’ This recognises two key points: firstly, even if it were possible to eliminate risk, it is likely to be prohibitively expensive to do so. Secondly, some general public consensus of an acceptable level of risk is required.

While there is no absolute rule dictating what this residual level of risk should be, national guidance has emerged on the basis of experience and societal consensus. In the UK, new coastal flood defences are typically designed to withstand events that might be expected to occur once every 200 years, while in The Netherlands, much of which lies below sea level, the corresponding guidance is to defend against a one-in-10,000-year event.

Statistics regarding maximum sea levels – themselves dependent on a combination of factors – can be recorded over time and fitted into a probability function (see box alongside). Such statistics inform the insurance industry on the risk factors for people and their properties in particular areas. People are sometimes understandably mystified and angry as they view their flooded homes and ask their insurer ‘How come the one-in-50-year event has already flooded my house two years running?’ Probabilities are just that – probable, not guarantees.

Tide levels rise

Elements of climate change are predictable since levels of carbon dioxide can be measured, as can rising temperatures. Its effects on precipitation are less certain but it is estimated that sea levels will be between 0.5 and 3metres higher in 2080 than they are today. So, the Centre for Coastal and Ocean Science and Engineering’s research role is becoming more urgent.

Coastal flooding has several causes: ‘overflow’ happens when the sea level exceeds the crest of the sea wall; ‘wave-overtopping’ is when waves, often exacerbated by storm conditions, run up over the crest of existing defences, and variations of this result in wave damage to the defence, either as breach of the sea wall or ‘toe failure’ at the base of the defence by extreme (and sometimes rapid) seabed erosion.

A storm event is defined by a continuum of combinations of water levels and wave conditions, corresponding to a contour line of fixed probability on the joint distribution function. The combinations providing the worst case scenario in relation to selected criteria can then be identified.

Empirical formulae are available for this purpose but depend on a large number of variables since wave-breaking is a complex phenomenon. Wave height depends on water depth (one empirical rule states that wave height is limited to a maximum of 0.78 of water depth), so waves propagating in shallow water begin to shoal and will eventually break, dissipating energy and reducing the wave height.

Figure 1: the positioning of the offshore reefs along the coast of Sea Palling © Mike Page

Limiting wave impact

The crucial point is to find a means of limiting the water-depth. This will serve as a natural regulator on the wave height and consequently the impact of waves. Water depth depends on two factors: firstly the water level controlled by tides plus surge, over which we can have no influence and, secondly, the level of the beach, over which we do – to a degree.

A new approach to constructing coastal defences has developed over recent years. Instead of placing structures at the top of the beach, they have been positioned further offshore in order to influence the wave patterns and so retain a ‘healthy beach’ in front of the stretch of coast to be protected. This dissipates storm waves by ensuring they break before reaching the upper beach.

An example of offshore defence structures is the Sea Palling detached breakwater scheme, developed by Halcrow. It is especially apt that this area was chosen because in 1953, an hour-and-a-half after the floods hit the coastal areas around The Wash, the sea broke through the defences at Sea Palling, where seven people drowned.

By the late 1980s the beach in front of the existing seawalls on the north Norfolk coast had become lowered to levels considered critical, so the construction of nine offshore reefs was sanctioned (see Figure1). The reefs, also known as ‘shore–parallel breakwaters’, were constructed in two stages: StageOne, laying down Reefs 5-8, was completed between 1993 and 1995 and Reefs 9-13 were added in 1997 during StageTwo, when the beach was also recharged.

The reefs constructed in Stage Two were modified from those of the Stage One reef design, which had lower crest levels that were shorter in length and more closely spaced. Simultaneously, some intermediate works were undertaken to ensure the integrity of existing defences where beach loss occurred in areas not defended by the reefs.

The originally planned Reefs 1-4 were reserved for a subsequent stage but have yet to be constructed, leaving Reef 5 as the northernmost reef.

Measuring outcomes

Academic researchers have taken the opportunity to work with the Sea Palling scheme designers in the LEACOAST and LEACOAST2 projects, funded by the EPSRC. This has involved teams from the universities of Liverpool, East Anglia, and Plymouth, as well as the Proudman Oceanographic Laboratory. A combination of advanced measuring, remote-sensing monitoring techniques with radar and video cameras, plus modern computer models was used to study the impacts of the offshore reefs on beach morphology under both wave and tide conditions. The quality of data was further enhanced by regular bathymetric and topographic surveys.

In parallel, a linked project was also funded by the Department for Environment, Food and Rural Affairs and the Environment Agency to enable the leading coastal engineering companies, HR Wallingford and Halcrow, to utilise and extend the research results to improve guidelines for the design of such coastal defence structures in the future.

The outcomes of the Sea Palling scheme have ensured protection to the beach and land behind it from erosion, while being less intrusive than traditional seawalls, thus offering additional benefits in terms of tourism. It is classified as a flood defence as it continues to successfully defend a key section of the East Anglian coast, behind which sits the Norfolk Broads, whose delicate eco-structure is characterised by its freshwater environment. Such a classification would surprise engineers involved in the previous design as the prime effect of the breakwaters is to stabilise the beach levels, and ensure the incident wave energy is dissipated, rather than to reflect incident wave energy.

Practical research

The other key success factor of the LEACOAST projects has been in helping to develop a new approach to designing and monitoring flood defence schemes. It has proved the benefits of extensive field measurements, essential for understanding the complex coastal processes interacting with the nearshore structures and validating mathematical models. Both process-based and statistical modelling approaches developed within the projects are extremely useful to assist in quantifying the level of uncertainty associated with the more traditional deterministic modelling.

Facilitating the involvement of practising engineering organisations as well as government agencies throughout the projects has permitted them to provide advice and, latterly and directly, the researchers to disseminate their scientific findings to industry to ensure the most updated research results of the projects can be turned into practical use and applied more widely and consistently by the coastal engineers.

The Sea Palling scheme has shown the effectiveness of maintaining a healthy beach to provide protection from flooding and erosion. It also has additional benefits: it provides a solution that is less intrusive to the eye and provides a sheltered and extended beach behind the reefs which can be used for walking, swimming and other water sports. It additionally helps restore the beach vegetation and wildlife in the areas sheltered by the reefs.

Other beach protection schemes are in place in the area such as the groyne systems on the north and south and the sea wall on the south of the reefs, and it is clear that such defences do not provide the added benefits of beach protection schemes. It is therefore likely that the benefit of reefs as a flood defence measure outgrows the added costs, at least in areas at great risk of coastal flooding and erosion.

Offshore breakwater schemes have been used for beach erosion control in many countries and in great lakes as well as in coastal zones. The threat posed by increasingly turbulent seas and the rise in water levels is likely to lead to a significant increase in the number of offshore breakwater schemes in the future.

The authors would like to thank Gary Atkins for his help in drafting this article.

BOX: Calculating the probability of flooding

Risk analysis is key to the work of authorities like the Environment Agency – responsible for coastal defences – and research centres such as the Centre of Coastal and Ocean Science and Engineering at the University of Plymouth. They know that flooding will occur if the sea level is higher than the crest of the seawall, and that sea level variations comprise a predictable tidal component and a fairly random ‘surge’ component arising from changes in atmospheric pressure and wind and waves at the sea surface. Sea level changes, combined with high tides and storm surges, can present significant potential for flooding.

Over a period of many years, the sequence of these annual sea level highs can be collated and used to fit a probability function, which may be extrapolated to find the probability of water levels exceeding a particular value in any year or, conversely, finding the level that corresponds to a given probability of ‘return period’ (the interval of time before an event may be expected to recur).

Implicit in this process is the assumption that individual measurements are independent and identically distributed. In symbols, the probability that water level w is less than a particular value W is described by the function Fw(W), known as the cumulative distribution function:

Prob(w < W) = Fw(W)

For very low values of W, the cumulative probability distribution function takes values close to 0, while for large values it approaches 1. Thus, if we build a sea wall with a crest level of H then the probability in any year that the maximum sea level exceeds this, and that there will therefore be flooding, is given by:

1 – Fw(H)

If we set 1-Fw(W) p then the probability that the annual maximum water level exceeds W is p. The probability p is sometimes written as 1/T where T is the ‘return period’ and is the average time, in years, between successive events exceeding the level W.

Confusion begins for the lay person when they hear the statement that ‘a defence will protect against storms with a return period of 50 years’, the assumption is that there will be one flooding event every 50 years, or that because it flooded last year there will be no floods for another 49 years. The confusion lies between the statistic ‘return period’ and the probability that, in any year, the chance of flooding is 1/50.

Flood defences will usually be designed to give a specified level of protection for a defined period of time – their design life. For coastal flood defences this might be 200 years. We can work out the probability that a defence designed to protect against the one-in-200-year storm is flooded at least once during the course of its design life of 200 years as:

1 – (1 – Fw(W))²⁰⁰= 1 – (1 – 1/200)²⁰⁰≈ 0.63

That is, the probability of flooding occurring during a period of 200 years is 63%, not 1% or even a 10% probability, but 63%. This means there is almost a two-in-three chance that this flood defence will fail during its design life of 200 years. To someone who is protected by such flood defence, this figure can come as an unpleasant shock.

In practice, the situation is more complicated because waves are a significant cause of flooding as well as damage to defences. In this case, extreme sea states will have to be determined through joint probability analyses of waves and water levels.

BIOGRAPHY – Dominic Reeve, Vanesa Magar and Shunqi Pan

Dominic Reeve is Professor of Coastal Dynamics at the University of Plymouth. He has worked in industry and academia, published over 150 research papers and is the author of two textbooks on Coastal Engineering and on Risk.

Dr Vanesa Magar is currently RCUK Academic Fellow at the University of Plymouth. Her research has focused on mathematical modelling applied to long-term morphodynamics, to transport processes and fluid dynamics (in Coastal Engineering, Geophysics and Biology).

Dr Shunqi Pan is Reader in Coastal Engineering at the University of Plymouth and has more than 15 years experience in physical and numerical modelling of coastal hydrodynamics and morphodynamics.