Jurassic Coast - Chiswell6 Gabions

Chiswell case study: The Scheme

The Scheme index page

Gabion baskets in the process of being filled with local coarsely crushed Portland Limestone

Casting of the wave return wall on top of the existing sea wall in front of the Cove Inn
Trial length of gabions

21. In view of the innovative proposals put forward by the consulting engineers to protect the crest of the Beach with stone-filled gabion mattresses, the Weymouth and Portland Borough Council, on advice from Wessex Water and in agreement with the regional engineer of MAFF, decided that the proposals should first be tested.

22. A trial length of 400 m running northwards from the existing sea wall recommended by the consultants was reduced to an overall length of 150 m. Such a length would still allow the full range of alternative combinations of materials and construction methods to be explored. The trial, estimated at a cost of £1000/m run of Beach, was to be carried out in close collaboration with the manufacturers of the mattresses.

23. As the main purpose of the mattresses laid on the seaward slope was to slow the erosive process and thereby to protect the crest against an early reduction in height, it was essential that the construction should be flexible, strong and easily repaired. The stepped construction is intended to reduce the energy of the wave run-up of a 1978 type storm (estimated 5 year return period) and significantly to reduce the overtopping from the run-up caused by a long period wave attack as occurred in 1979 (estimated 50 year return period).

24. Two types of mattress were selected: the flexible Maccaferri with galvanised steel and plastic coated wire; the more rigid galvanised welded steel heavy gauge of the BRC cages. The Maccaferri mattresses were available in only two sizes of mesh and the smaller at 75 mm was specified. On account of the anticipated steep angles to which the filled mattresses might be subjected, an increase in the number of diaphragms normally provided in the mattresses was specified.

25. Under a sustained wave attack, the continued undermining of the mattresses on the seaward slope might lead to their being dragged off the crest unless a suitable anchorage were provided on the landward slope. Accordingly, the selected anchorage system comprised two box gabions wired together and also to the oversailing mattress. To lessen damage or displacement by heavy wave run-up, it was decided that the toe mattresses should be connected in panels of not less than 20 t in weight.

26. Under a contract let to Tilbury Construction Ltd, in August 1981, and beginning at the junction of the esplanade and the Beach, the contractor first dealt with the removal of 15 t stone blocks placed in the threatened breach at that point in 1978. A box gabion wall was then constructed on the Esplanade to retain the built-up Beach before mattress laying could begin along the raised crest

27. Following the decision by the NCC that only local quarry stone should be used for filling the stone-filled mattresses and that it should be readily identifiable as a material not normally found on the Beach, Portland Capstone was selected in spite of its brittleness. Difficulties were experienced initially in achieving satisfactory Portland stone deliveries because substantial stockpiling was not practicable on account of access difficulties.

28. To achieve the specified grading of the beach material for filling the mattresses, a portable screening plant was set up on the seaward side of the Beach and a stockpile was made. In. the event, the contractor, despite suffering a set-back when the screening plant on the Beach was severely damaged in a storm completed the works in a satisfactory manner and within a few weeks of the anticipated date.

29. The tying by hand of the stone-filled mattresses was easy in comparison with those filled with beach material where there was little clear space to thread the wires. A lesson learned on this project was that insufficient time was allowed for the labour intensive tying operation in exposed conditions which was physically tiring and tedious work. The longer time required and the added difficulty in ensuring deliveries of the correct size of Capstone were the subject of contractual claims.

30. The mattresses are expected to have a life of fifteen years. Monitoring of performance up to 1987 showed that no significant movement of the mattresses had occurred. Corrosion of the galvanised steel of the BRC mattresses is apparent, but the Maccafferi wire mattresses have been little affected. The main depreciation has occurred in the corrosion of the galvanised steel tying wire.

3 1. On 16 December 1989, the sea defences at Chiswell were severely tested by a storm that produced wave heights estimated to have a 5 year return period, and a 1 m surge over predicted tide level of a 25 year return period. The defences, including the trial gabion length, were overtopped three or four times by waves with estimated periods of up to 18 seconds and damage was caused to eight properties. The crest level of the Beach to the north of the trial gabions was lowered and large quantities of shingle were carried on to property on the landward wide.

32. Mattresses on the seaward face were undermined by wave action for the first time, and the lower layer BRC mattresses in Bays 1 and 2 settled at a steep angle. The toes of the mattresses were severely damaged where tie wires and mesh was severed, thus allowing the granular filling material to escape. Remedial works to the mattresses were carried out in June 1990 at a cost of £35 000.

33. Despite the overtopping of the crest mattresses there is little doubt that the trial length prevented major damage to the Chiswell properties in its shadow and that it performed as intended. The flexibility of the BRC mattresses is under scrutiny, but their inherent strength was amply demonstrated.

Modifications to existing sea wall

34. The existing sea wall/esplanade was built in several stages between 1958 and 1965, the northernmost length being completed in 1960. The construction of a parapet at the back of the esplanade was required to prevent over-topping in a 1978 type storm. Works were also necessary to reduce the effects of wave reflection from the existing sea wall, which was causing erosion of the beach and hence serious undermining of the existing sea wall under the same storm conditions.

35. A 1.0 m high wall, including a wave return feature, was designed to withstand the dynamic forces imposed by overtopping water. At the northern end, where the esplanade joins the Beach, the height of wave wall was increased to reduce the risk of overtopping resulting from the 'set up' of water levels caused by the trapping of water in the corner of the Esplanade. Gates were required in the wave wall to allow public access and to permit the transfer of fishing boats from the foreshore to the shelter on the landward side.

36. The protection of the existing sea wall against undermining was effected with a stepped concrete apron and steel sheet toe piles about 5 m in length.

Back to the top

The Scheme index page

Homepage

		Chiswell case study: The Scheme
The Scheme index page Gabion baskets in the process of being filled with local coarsely crushed Portland Limestone Casting of the wave return wall on top of the existing sea wall in front of the Cove Inn		Trial length of gabions 21. In view of the innovative proposals put forward by the consulting engineers to protect the crest of the Beach with stone-filled gabion mattresses, the Weymouth and Portland Borough Council, on advice from Wessex Water and in agreement with the regional engineer of MAFF, decided that the proposals should first be tested. 22. A trial length of 400 m running northwards from the existing sea wall recommended by the consultants was reduced to an overall length of 150 m. Such a length would still allow the full range of alternative combinations of materials and construction methods to be explored. The trial, estimated at a cost of £1000/m run of Beach, was to be carried out in close collaboration with the manufacturers of the mattresses. 23. As the main purpose of the mattresses laid on the seaward slope was to slow the erosive process and thereby to protect the crest against an early reduction in height, it was essential that the construction should be flexible, strong and easily repaired. The stepped construction is intended to reduce the energy of the wave run-up of a 1978 type storm (estimated 5 year return period) and significantly to reduce the overtopping from the run-up caused by a long period wave attack as occurred in 1979 (estimated 50 year return period). 24. Two types of mattress were selected: the flexible Maccaferri with galvanised steel and plastic coated wire; the more rigid galvanised welded steel heavy gauge of the BRC cages. The Maccaferri mattresses were available in only two sizes of mesh and the smaller at 75 mm was specified. On account of the anticipated steep angles to which the filled mattresses might be subjected, an increase in the number of diaphragms normally provided in the mattresses was specified. 25. Under a sustained wave attack, the continued undermining of the mattresses on the seaward slope might lead to their being dragged off the crest unless a suitable anchorage were provided on the landward slope. Accordingly, the selected anchorage system comprised two box gabions wired together and also to the oversailing mattress. To lessen damage or displacement by heavy wave run-up, it was decided that the toe mattresses should be connected in panels of not less than 20 t in weight. 26. Under a contract let to Tilbury Construction Ltd, in August 1981, and beginning at the junction of the esplanade and the Beach, the contractor first dealt with the removal of 15 t stone blocks placed in the threatened breach at that point in 1978. A box gabion wall was then constructed on the Esplanade to retain the built-up Beach before mattress laying could begin along the raised crest 27. Following the decision by the NCC that only local quarry stone should be used for filling the stone-filled mattresses and that it should be readily identifiable as a material not normally found on the Beach, Portland Capstone was selected in spite of its brittleness. Difficulties were experienced initially in achieving satisfactory Portland stone deliveries because substantial stockpiling was not practicable on account of access difficulties. 28. To achieve the specified grading of the beach material for filling the mattresses, a portable screening plant was set up on the seaward side of the Beach and a stockpile was made. In. the event, the contractor, despite suffering a set-back when the screening plant on the Beach was severely damaged in a storm completed the works in a satisfactory manner and within a few weeks of the anticipated date. 29. The tying by hand of the stone-filled mattresses was easy in comparison with those filled with beach material where there was little clear space to thread the wires. A lesson learned on this project was that insufficient time was allowed for the labour intensive tying operation in exposed conditions which was physically tiring and tedious work. The longer time required and the added difficulty in ensuring deliveries of the correct size of Capstone were the subject of contractual claims. 30. The mattresses are expected to have a life of fifteen years. Monitoring of performance up to 1987 showed that no significant movement of the mattresses had occurred. Corrosion of the galvanised steel of the BRC mattresses is apparent, but the Maccafferi wire mattresses have been little affected. The main depreciation has occurred in the corrosion of the galvanised steel tying wire. 3 1. On 16 December 1989, the sea defences at Chiswell were severely tested by a storm that produced wave heights estimated to have a 5 year return period, and a 1 m surge over predicted tide level of a 25 year return period. The defences, including the trial gabion length, were overtopped three or four times by waves with estimated periods of up to 18 seconds and damage was caused to eight properties. The crest level of the Beach to the north of the trial gabions was lowered and large quantities of shingle were carried on to property on the landward wide. 32. Mattresses on the seaward face were undermined by wave action for the first time, and the lower layer BRC mattresses in Bays 1 and 2 settled at a steep angle. The toes of the mattresses were severely damaged where tie wires and mesh was severed, thus allowing the granular filling material to escape. Remedial works to the mattresses were carried out in June 1990 at a cost of £35 000. 33. Despite the overtopping of the crest mattresses there is little doubt that the trial length prevented major damage to the Chiswell properties in its shadow and that it performed as intended. The flexibility of the BRC mattresses is under scrutiny, but their inherent strength was amply demonstrated. Modifications to existing sea wall 34. The existing sea wall/esplanade was built in several stages between 1958 and 1965, the northernmost length being completed in 1960. The construction of a parapet at the back of the esplanade was required to prevent over-topping in a 1978 type storm. Works were also necessary to reduce the effects of wave reflection from the existing sea wall, which was causing erosion of the beach and hence serious undermining of the existing sea wall under the same storm conditions. 35. A 1.0 m high wall, including a wave return feature, was designed to withstand the dynamic forces imposed by overtopping water. At the northern end, where the esplanade joins the Beach, the height of wave wall was increased to reduce the risk of overtopping resulting from the 'set up' of water levels caused by the trapping of water in the corner of the Esplanade. Gates were required in the wave wall to allow public access and to permit the transfer of fishing boats from the foreshore to the shelter on the landward side. 36. The protection of the existing sea wall against undermining was effected with a stepped concrete apron and steel sheet toe piles about 5 m in length.

		Back to the top The Scheme index page Homepage