TECHNICAL EVALUATION OF PORT ASKAIG RECENT FERRY BERTH DEVELOPMENTS.

Information provided by
Donald Ewen Darroch
Chairperson
Islay & Jura Ferry Co Ltd
Inver Lodge
Isle of Jura


1. Standards.
The following British Standards, Lloyds Rules and Reports on Ro Ro Berths are in general use to help designers and contractors involved in the development of Ro Ro Ferry berths. They give a minimum requirement but are not comprehensive allowing any designs to be developed to best suit the circumstances.
BS. 6347. Maritime Structures. Part 1 General 2
Part 2 Quay walls and Dolphins
Part 4 Fendering
Part 8 Linkspans
Lloyds Register Rules and Regulations for Linkspans
CIRIA report on Linkspans and their safety.
LR and BS both refer to other of their Rules and Standards which should be used in conjunction with these e.g. BS 5400 Bridge Code and Ship’s Rules.
Certain information must be supplied by the client ( in this case Argyll & Bute Council ) with regards to depth of water, tidal levels and currents, exposure and wave climate and the type of vessels that are expected now and in the future to use the berths.

2. Types of Berths.
In Scottish waters with significant tidal ranges two types of RoRo berths are in use.
Type 1.The slipway, which developed as a result of the arrival of the mv “Isle of Gigha” on the West Coast in 1960’s. It was one of the first Ro Ro ferries and did not initially run on a regular route. It landed vehicles on beaches and existing stone slipways. Many of which were available and built in the 19th century under the “Congested Districts” Act or on drove routes. Vessels berthed alongside the slipway at a position appropriate for the tide level.
Calmac copied this type of vessel in their Island Class and the Western Councils built new slipways. Calmac now have 16 smaller vessels using this method and the Councils operate some ferries in a similar way. Council slipways are expensive to build because of a high proportion of tidal and underwater work in their construction. The hull form required by the vessels to use these slipways is inefficient. The transition angles between ship’s deck/ramp/slipway are often too severe and car damage occurs. Unless slipways are in well sheltered positions bow loading vessels are vulnerable to pounding and service reliability is impaired as a result.


Type 2a. Ship supported Linkspans. Orkney and Shetland County Councils led by Western Ferries followed the well established Scandinavian practice using simple linkspans supported at the outer end by the vessel during loading. Dolphins provide a berthing face for vessels. The small island ferries as well as those the size of Calmac mainline vessels, all use this type of Linkspan. There are a total of 20 vessels in Scottish waters using over 30 of this type of Linkspan. Many hundreds more use this system in Scandinavia, Holland and the Baltic. This system is also used for train ferries.

Type 2b. Mechanically Supported Linkspans. Calmac has 9 vessels which use Linkspans but they carry their own ramps even though they are open deck ships. They also have two closed deck vessels which are too large to use Port Askaig. The linkspan required for these ramped vessels is hinged on the shore above High Water mark and is supported at its outer end by a gantry with a mechanism that can adjust the outer end height under full load to suit the vessel using it.
The ship’s ramp bridges across from the car deck to the end of the Linkspan. Any movement due to swell, waves and the transit of heavy vehicles must the taken up by this short ramp. Its length limits the amount of vertical movement if vehicles are not to ground. This is made worse by the general exposure of the berths in Western Scotland which do not generally have closed sheltered ferry harbours.
The weight of these ships’ ramps, their supporting structure and mechanism reduce the carrying capacity of the vessel by about the equivalent of one commercial vehicle so reducing their earning capacity by between 15% and 40% for trucks. This additional weight, which is carried on each voyage, adds to the ship’s fuel consumption.

3. Vessels using this Port Askaig, Main berth
Calmac (presently ramped ships) currently have only 9 vessels that ought to be able to use this berth. Their beam ranges from 15.3m to 15.8m. Their freeboard varies between about 1.5m and 2.5m. These ships have ramps about 3.5m clear width, at bow and stern, centred on the centreline of the vessel. Ships of up to 16 meters beam might be expected in the future. These dimensions are breadth moulded and one should add 600mm for belting.
The maximum displacement of vessels is about 3,000 tonnes, on a draft of max 3.0m
Other Vessels (Probably non ramped ships) a private operator might use a similar but larger version of the “Sound of Jura” which operated successfully for 10 years on this route. This falls within the above parameters. Consideration should also be given to a Catamaran type of vessel such as the one under trial on the Pentland crossing. This vessel has a greater beam and freeboard than conventional singled hull ships and can be made to accept an off centre-line Linkspan.

PORT ASKAIG NEW DUAL BERTH
These observations result from an inspection of the berth on two occasions in autumn 2009 during periods of Spring Tides. No dimensions were taken, so they may vary from those stated.
A. BERTHING FACE. The Berthing face is fendered at even spacing using V fender elements faced with UHMWPE. The outer round head has closer spaced fenders and adequately fendered. This is in line with normal practice for facing to a sheet piled quay wall. Mooring arrangements are to conventional bollards with ships’ lines. There are adequate, well positioned bollards. The gangway is man handled at a fixed position. The mooring of the vessel requires 3 men at each end of the vessel, one on the gangway plus there are 3 men on the shore, 10 in total. A modern Ferry berth should not require such a high level of manning for the mooring and unmooring of vessels. No allowance appears to have been made for any type of auto mooring.
B.LINKSPAN. The linkspan is the bridge hinged on the inshore end and supported at the outer end on each side by a single hydraulic cylinder suspended from a gantry. It was designed and supplied by Macgregor as subcontractor to the main civil works contractor. It falls short of what should have been supplied in a number of ways.

1. The linkspan does not appear to be able to accept impact from the vessels using it. The BS and LR rules require that the linkspan should be able to absorb the impact of the maximum displacement vessel likely to use the berth at a speed of 0.35 m per sec. Furthermore it should accept in an emergency up to 0.5 m/sec without being put out of order and sustaining only local damage at point of impact. The CIRIA report shows that ship impact is one of the two most common reported incidents in Linkspans.

2. The centre Line of the Linkspan. The beam of the vessels varies very little from 15.3 to15.8 and a possible 16 in future. The centre line of the Linkspan should therefore be 7.8m (+0.3 for belting) from the line of the berthing face. It is far from this, having its quayside edge almost in line with the berthing face. The active roadway for the vessels using the linkspan needs only to extend from 5.15m to 9.75m from the berthing giving an outer width of 4.60m. To allow for two lane access which ought to apply on this route the outer end should extend to seaward by 3.5m making the outer end approximately 8.1m. The actual outer end is considerably wider than this but is so off centre that 30% of the Bridge will never be used. 7.35m is considered 2 lanes within the codes any greater must be treated as 3 lanes.
The consequences of this unnecessary width are:
a. 3 lanes must be used for calculation of traffic loads unnecessarily so increasing the cost of the steelwork and support mechanism.
b. A 5 m wide strip of Linkspan adjacent to the quay that can never be used.
c. Much higher loading on the outer support cylinder due to eccentric loading on the support mechanism.
The extra steel used in the ineffective quayside of the bridge would have been better used in increasing its length. (See below)

3. Length. The roadway at the hinge should be at least 500mm above HAT (+2.4CD) but greater than this if there is either a risk of being overtopped with waves or a tidal surge. The present level is probably at +2.9CD but should be +3.2CD. Both these circumstances occur at Port Askaig as the whole concrete apron that forms the hinge for the main berth and the slipway for the Jura Ferry has already been flooded to a depth of 300mm. At HAT +2.4CD with the ship’s deck (at its threshold of the ramp) 2.5m above the water level the level of the deck is +4.9CD. This is a rise from 2.9CD to 4.9CD. i.e. 2 meters requiring a minimum length of the Linkspan of 20m. It was not measured but it appears a lot less. At LAT for the Calmac ship there is a small downward gradient of 400mm and the length is OK. However due to incorrect levels at the Jura ferry slip this vessel must use the mainline berth. In this case the ship’s deck would be at +1.0CD at LAT with a rise of 1.9m requiring a length of 19m.
There are relatively simple solutions to these problems.

4. Landing Area on the Linkspan. The codes and rules recommend a shaped profile on the deck of the outer end of Linkspan where the ship’s ramp lands. This allows the ramp to land with a safe overlap of 1.5m and to present, if necessary, an upward gradient to the Linkspan. This is especially important at lowest tides. At high water the ship’s can ramp can present down to the Linkspan so avoiding critical humps and hollows and short lengths of extreme gradients as well as the ship’s ramp tripping.
It was observed that at least one commercial articulated truck could not cross the threshold on the ship’s flap area at high water. After a number of failed attempts with wheels spinning and running backwards, the vehicle had to roll backwards onto the concrete apron and then get up speed to charge over the hump, followed by hard braking to avoid hitting the vehicles already on the deck. The driver was no novice and told me it was a frequent problem.
Such dangerous practice is avoided if the design incorporates a profiled landing area and its shape is optimised for use by the regular ferry ramp. A curved profile is the most versatile.

5. Transition Angles. The codes clearly specify an acceptable change on angle at the interfaces at the hinge end of the Linkspan as well as at its landing area. The incident above shows how the Linkspan fails at its outer end. Even at the hinge end, fully under the control of the Linkspan designer, no transition flaps are included and yet the change of gradient from level to 1:10 is critical for many vehicles and 1:8 at extremes even more severe.

6. Support Mechanism. The BS 6347 Part 8 calls for redundancy in the support mechanism of a mechanically operated Linkspan. In the event of either cylinder or its connecting pins failing there must be another duplicate system to prevent collapse. Pairs of Cylinders on each side are the normal solution to this, each independently mounted, but there is a proprietary brake system being prototyped. There appears to be no such redundancy built into this Linkspan. CIRIA cites mechanical failure of the support system as the other of the main cause of reported incidents.
There are other support systems which have natural redundancy and fail safe that could have been considered.


C. JURA FERRY BERTH.
1. Berthing Face. The former open structure used by the original “Sound of Gigha” has been replaced by a sheet pile wall. The original open “Fence” was designed as a wave screen and allowed waves to be dissipated as they passed through it. With the solid wall, reflected waves create a “jabble” under relatively light conditions and cause the Jura Ferry to roll excessively. Floating fenders are necessary to reduce damage but they cannot be a permanent solution.
This problem would have been eliminated if the double sheet pile wall had been topped off at low water level and the last 3m made up of precast COBS (1mx1mx1m) which are hollow blocks used to face breakwaters. The small boats on the inside would have been less subjected to overtopping spray filling them in bad weather, and the Jura ferry would be able to lie at the berth as the wave climate would be substantially reduced. Without the floating fenders the “Eilean Dhiura” would be able to berth 1m to the south.

2. Slip. The stub ended slip worked satisfactorily with the old “Sound of Gigha” but less so with the new “Eilean Dhiura”. This type of slip in tidal ranges of 2 meters can be very satisfactory, cheap and better able to cope with bad weather as the bow is well clear of the ground. If the tidal rise is over 2m it is limited at extreme water levels. This could be solved by making two adjacent slips one with the landing point where the ship’s ramp lands 1 meter higher than the other. Performance will also be enhanced if the slip is curved in profile so that at highest tides the transition angles can be improved by increasing the overlap. At the moment the “Eilean Dhiura” lies in the mainline berth and has to use its own berth with a very poor Ramp/Linkspan interface when the mainland ferry is present.

D.TRAFFIC MANAGEMENT. The regulations require that foot passengers are physically separated from vehicular traffic. Sight Lines for drivers are also important as is good signage.
The car marshalling area is excellent considering the topography. Sight Lines from the Calmac ship’s are poor and vehicles must turn immediately on exiting the vessel. If the road had been widened to seaward in front of the slipway by the hotel, the approach line to the ship would have been much improved and safer. Only a meter or two would have made a great difference.
It is inevitable that there will be crossing traffic and pedestrians in this area. The raised pavement between shop and terminal acts as a speed bump but there are no other measures. This speed bump also slows the loading procedure for the ship. There is no actual control of passengers once they are on land, nor is there anything to stop local traffic moving across the ferry traffic.
During the 6-8 minutes that the vessel is discharging, a simple traffic light system stopping all traffic in the marshalling area and local traffic bound for the Hotel or shop would greatly reduce the risk of accident. The lights would allow pedestrians to cross between terminal and shop during this period. It would also show GO for the ferry traffic and give it a clear run out of Port Askaig. To start loading the ship, the pedestrian crossing would be at RED and all local traffic held from passing through boarding traffic. Such signal sequences could be controlled by the harbour master and would reduce risk to pedestrians.

CONCLUSION.
A successful Ferry Berth requires a careful liaison between: The client in this case Argyll & Bute Council, the appointed consulting Engineer who must be experienced not only in Marine works but in SHIPFIT*. The regular ship operators should of course be consulted and indeed asked to sign off the design where it will be effected, namely, mooring points, gangways, services, Linkspan interfaces and passenger and car control.
It is quite obvious that A&B Council were let down by their consultant whose design and specification fell short in these vital areas. Calmac have a history of changing ship’s and expecting new terminals to be built for them, but with experienced foresight and little extra cost, terminals built in the 70s and 80’s should be able to accommodate current vessels. There are many precedents of this. Other ships of the type and size for this route should be taken into account, (there are not many of them), otherwise taxpayers money is being used to bolster a monopoly. Indeed we would hope that A&B Council have not allowed themselves to be in such a position, that a challenge could make them vulnerable to EU demanding their funds back if the terminals could be shown to be sustaining – favouring a monopoly and precluding the use of terminals by others.
Another reason for this poor interface is that the contractors have no experience of understanding of SHIPFIT so that when they are preparing their tender they ask for a price from a Linkspan supplier. The main contractor is only interested in the lowest price and will not submit a better alternative even it is a penny more. Indeed if this “penny more” better linkspan saves thousands in Civil Works it is the last thing that the main Contractor wants to know as it reduces his work and therefore profit.
Not one of the parties alone can be blamed for these faults but the arrangements for designing, specifying, tendering and implementation used create these results. Most independent and many local authority ports adopt a different approach with better results.
The recent record of construction of Ro Ro berths in Argyll does not look well when one considers the number of rebuilds, major modifications and unused berths that result in lack of foresight and understanding of marine matters. Perhaps within the public sphere there is another organisation who would be better placed to manage the Argyll piers and harbours than the present incumbents?



*SHIPFIT : Is a recent term to describe the interface between the ship and the quay. It is rather specialised in RoRo operations and requires extensive and detailed knowledge of the bow and stern openings of Ferries and Ro Ro Ships.