VLAAMS INSTITUUT VOOR DE ZEE
PLATFORM VOOR MARIEN ONDERZOEK


Persoonlijke instellingen
Naamruimten

Varianten
Handelingen


Seawalls and revetments

Uit Kust Wiki
Ga naar: navigatie, zoeken

This article introduces the hard shoreline protection structures seawalls and revetments. The article further explains how and why the application of a seawall or revetment might be used to resolve well-defined real life coastal engineering problems. It also discusses why they cannot be used as a solution for problems related to structural erosion.

Introduction

Seawalls or revetments are shore parallel structures at the transition between the low-lying (sandy) beach and the (higher) mainland or dune. The height of a seawall fills often the total height difference between beach and surface level of the mainland. In many cases adjacent at the crest of a seawall a horizontal stone covered part is present (e.g. boulevard; road; or parking places). At the initial time of construction a seawall is situated close to the position of the dune foot. In the present article with a seawall an almost vertical structure is meant. The seaward side of the seawall is thought to be rather smooth.

A revetment is, just as a seawall, a shore parallel structure. The main difference is that it is more sloping than a seawall. A revetment has a distinct slope (e.g. 1:2 or 1:4), while a seawall is often almost vertical, the surface of a revetment might be either smooth or rough (a seawall is mostly smooth) and that the height of a revetment does not necessarily fill the total height difference between beach and mainland (a seawall often covers the total height difference.

Slightly different definitions are given in the definition pages of seawall and revetment. For more information on different types, characteristics and application of revetments, see also the article Revetments. For some general information on seawalls, see also Seawall.

Solving coastal engineering problems

Clear transition beach-mainland

Figure 1 Seawall and boulevard

Especially in sandy coastal areas with a lot of human (recreational) activities, a clear and fixed distinction between beach and mainland is desirable. A seawall will serve that aim. At the sea side of the seawall a more or less normal beach is assumed to be present; at the land side a road or a boulevard is present. Staircases facilitate the access to the beach.

The coast is assumed to be stable. The beaches in front of the seawall do not erode, or in case of a structural eroding coast an essentially (time-averaged) stable situation has been achieved with e.g. regular artificial beach nourishments. So a normal beach is assumed to be present in front of the seawall (and can be used for recreational purposes). See Figure 1.

While in a situation without a seawall even a moderate storm (surge) will attack and erode the mainland, in the situation with a seawall this is prevented. Some scour in front of the seawall during a storm (surge) must be taken into account in the design. (A part of) the 'denied' erosion volume from the mainland, is now eroded just in front of the seawall. The scour hole might undermine the seawall. With e.g. the DUROSTA computation model an estimate of expected scour depths can be made [1].

Figure 2 Seawalls with measures to reduce overtopping

The design conditions for the seawall have to be properly chosen. The heavier the design conditions, the heavier the seawall must be constructed and especially the 'safe' foundation depth will increase accordingly. To build a seawall which will be safe under 'all' conditions might be an unrealistic option.

Although achieving a clear transition between beach and mainland was the primary goal in the discussion so far, automatically some protection of the (infrastructure at the) mainland is achieved. The design conditions as selected, determine the rate of provided protection.

The crest height of a seawall determines (together with the boundary conditions at sea) to a great extent the rate of overtopping (water reaching the mainland by wave run-up and breaking waves and splash water transported by landward directed wind). With an additional wall and/or a slightly curved front, rates of overtopping may be reduced; see Figure 2.

Decrease risks of valuable infrastructure and buildings

Infrastructure and buildings situated close to the edge of mainland or dunes have a chance to be destroyed during a severe storm surge (See also chances and risks).

The risk (risk = chance x consequence) is felt to be too large in an existing situation. Or for example by extension and improvements of an existing hotel the 'consequence' has been increased and so the risk would increase to a too high level. Reducing the 'chance' might result in an acceptable 'risk' level (again).

With a robust seawall or revetment the required aim can be achieved. Aspects like proper design conditions and scour holes are in this case similar to the discussion in the previous case.

Let us consider a given a stretch of sandy coast. A very severe storm surge will cause a rate of mainland erosion of say 40 m in case the stretch of coast is unprotected. With a seawall which is able to withstand these conditions the erosion of the mainland will be zero. (In front of the seawall a deep scour hole will be formed.) When the entire seawall keeps its integrity; no further problems arise. (The scour hole will be re-filled again after some time with ordinary boundary conditions.) If, however, the seawall partly collapses and locally a gap in the seawall is formed during the severe storm surge, a rather dangerous situation will occur. Large volumes of sediment from the mainland are able to disappear through the gap and will flow along the sections of the seawall which are still in good condition in both longshore directions adjacent to the gap, filling the scour hole. It is expected that the ultimate rate of erosion of the mainland behind the gap will be larger than the 40 m as mentioned for the unprotected case.

Similar phenomena will occur at the two transitions between seawall and the adjacent, unprotected parts of the coast. Especially just adjacent to an abrupt end of a seawall, relatively much erosion is expected during a severe storm surge.

Existing row of dunes does not meet safety requirements

Figure 3 Scour in front of a revetment

A row of dunes is apparently too weak (too slender) to guarantee the safety requirements. Under design conditions a break-through is expected; the low-lying hinterland behind the slender row of dunes will be flooded. A seawall or revetment might be chosen as a solution, provided that the seawall or revetment will keep its integrity during the design conditions. A risky alternative would be that the structure is 'allowed' to collapse in a latter stage of the storm surge. The time left to the end of the storm surge (with lower water levels) is then thought to be too short to cause yet a break-through.

If a revetment is selected as protection tool, two aspects call for some remarks, viz.:

  • slope of revetment in relation to depth of scour hole;
  • level of upper end of revetment.

At least for rather smooth revetments it has (also experimentally) been proven that the depth of the scour hole depends on the slope characteristics. During tests in the Delta Flume of WL|Delft Hydraulics it turned out that with a slope of 1:3.6 a deeper scour hole was found than with a slope of 1:1.8 (other test conditions the same). The deepest point of a scour hole is not always found at the intersection point between revetment and cross-shore profile (see Figure 3 for a sketch). It is conceivable that the depth of a scour hole is less for a rough slope than for a smooth slope.

Figure 4 Scour in front of a revetment; Delta Flume test

Figure 4 shows scouring in front of (smooth) revetment after a test in the Delta Flume of Delft Hydraulics. Most of the water has already released from the flume. The still water level during the test was much higher.

The level of the upper end of a revetment (above that level the normal front slope of the dunes is assumed to be present up to the top of the dunes) determines the still occurring (remaining) dune erosion above that level, but also to some extent the depth of the scour hole. The higher that level the less remaining erosion, but also the deeper the scour hole. It is remarkable that if the level of the upper end of a revetment is equal to the storm surge level (or lower than that level), no reduction in dune erosion is found compared to a situation without any protection.

Structural erosion problems

The former section showed why and how seawalls or revetments may be used to resolve a coastal engineering problem or to achieve to goal. It is important to realize that seawalls and revetments are no solution for structural erosion problems. We pay attention to this issue in this article, because in coastal engineering practice too often this principally 'wrong' combination is applied. Many bad examples can be found all over the world.

Structural erosion caused by a gradient in the longshore sediment transport, means that volumes of sediment are lost out of the control volume area in a cross-shore profile (for a more detailed article on this, see hard structures and structural erosion). This loss process takes mainly place under ordinary conditions; the contribution of storm conditions to this loss process is often rather small. The initial losses of sediments out of a cross-shore profile take place where water and waves are; where actual longshore sediment transports do occur; so in the 'wet' part of a cross-shore profile. The 'dry' parts of a cross-shore profile are not involved in the longshore sediment transports; it looks like that the 'dry' parts do not form an integral part of the cross-shore profile during ordinary conditions.


Figure 5 Seawall and gradient in longshore transport

During high tides and/or modest storms (storm surges) all parts of a cross-shore profile participate in the coastal processes. By offshore directed cross-shore sediment transports, sediments from the higher parts of the profile ('dry' beach; even mainland under the more severe conditions) are transported to deeper water, filling the 'gap' that has been developed because of the gradient in the longshore sediment transport (see Figure 5). This sequence of processes causes a permanent loss of material out of the upper parts of a cross-shore profile.

By 'protecting' the mainland in this case with a seawall, one indeed prevents that sediments from the mainland are transported in seaward direction (less filling of the 'gap'). The losses, however, continue; the 'dry' beach disappears; it becomes deeper and deeper in front of the seawall. Initially, right after the construction of the seawall, still a more or less normal beach was present. The beach did 'protect' the seawall to some extent; only moderate storms could reach the seawall. When the beach had disappeared, much more frequent wave attacks directly to the seawall will occur. (Most likely in the design of the seawall this was not taken into account.) Damage occurs; reinforcements have to take place.

A somewhat confusing element is the time-scale of the developments as have been discussed so far. Local people (their houses are at stake) have noticed in the past that every storm surge has taken some square metres of their gardens. The edge of the mainland is coming closer and closer to their houses. Not seldom the responsible coastal zone manager is 'forced' by the local people 'to do something'. Building locally a seawall or revetment (e.g. in front of the properties which are situated closest to the sea) indeed seems to resolve the problem. During the next storm surge, the just 'protected' parts of the coast do not show any further erosion; in the un-protected parts the erosion of the mainland continued. Local people believe that this solution 'works' (own experience). The coastal zone manager is forced to build seawalls along the other parts of the coast. However, when time elapses, it will be quite clear that a quite wrong solution has been chosen. Only with huge costs the situation can be redressed.

See also

References

  1. Steetzel, H.J. (1993). Cross-shore Transport during Storm Surges. Ph.D. Thesis Delft University of Technology.
The main author of this article is Jan van de Graaff
Please note that others may also have edited the contents of this article.