Report on the investigation of heavy weather

encountered by the cruise ship

Pacific Sun
200 miles north north east of North Cape, New Zealand

on 30 July 2008

resulting in injuries to 77 passengers and crew

Marine Accident Investigation Branch

Carlton House
Carlton Place
Southampton
United Kingdom
Report No 14/2009
June 2009
 The MAIB has investigated this accident pursuant to the IMO Code for the Investigation
of Marine Casualties and Incidents (Resolution A.849(20)).
MAIB gratefully acknowledges and appreciates the generous assistance provided by
New Zealand’s Transport Accident Investigation Commission during this investigation.

Extract from

The United Kingdom Merchant Shipping

(Accident Reporting and Investigation)

Regulations 2005 – Regulation 5:

“The sole objective of the investigation of an accident under the Merchant Shipping
(Accident Reporting and Investigation) Regulations 2005 shall be the prevention of
future accidents through the ascertainment of its causes and circumstances. It shall
not be the purpose of an investigation to determine liability nor, except so far as is
necessary to achieve its objective, to apportion blame.”

NOTE

This report is not written with litigation in mind and, pursuant to Regulation 13(9)
of the Merchant Shipping (Accident Reporting and Investigation) Regulations
2005, shall be inadmissible in any judicial proceedings whose purpose, or one of
whose purposes is to attribute or apportion liability or blame.

Further printed copies can be obtained via our postal address, or alternatively by:
Email: [email protected]
Tel: 023 8039 5500
Fax: 023 8023 2459
All reports can also be found on our website:
www.maib.gov.uk
 CONTENTS
Page
GLOSSARY OF ABBREVIATIONS AND ACRONYMS

SYNOPSIS 1

SECTION 1 - FACTUAL INFORMATION 2

1.1 Particulars of Pacific Sun and accident 2
1.2 Narrative 3
1.2.1 Background 3
1.2.2 Pre-departure 4
1.2.3 The cruise 4
1.2.4 The passage 5
1.2.5 The accident 7
1.2.6 Post accident events 10
1.2.7 Recovery phase 11
1.3 The three instances of heavy rolling 14
1.4 Timing of the accident 14
1.5 Injuries to passengers and crew 16
1.6 Damage 16
1.6.1 Mapping of injuries by location 16
1.6.2 Muster stations 16
1.6.3 Restaurants 22
1.6.4 Other public rooms and spaces 22
1.6.5 Crew and working areas 22
1.6.6 Open decks 25
1.6.7 Communications equipment 25
1.7 Environmental conditions 26
1.7.1 Weather forecast 26
1.7.2 Sunset and twilight 27
1.7.3 Weather hindcast from the Meteorological Service of
New Zealand 27
1.7.4 Effect of the storm on New Zealand 27
1.8 Ship motion influences 29
1.8.1 Abnormal waves 29
1.8.2 Synchronous rolling 29
1.8.3 Parametric rolling 30
1.9 Pacific Sun’s resilience to rolling 30
1.10 The stabilisers 31
1.11 Stability 33
1.12 Master and officers 33
1.13 Pacific Sun – background and management 34
1.13.1 Management 34
1.13.2 Itinerary planning 34
1.13.3 Heavy weather guidance 34
1.13.4 Emergency procedures 34
1.13.5 Mass casualty response 35
1.14 External inspection 35
1.15 IMO Regulation 35
1.16 Passenger questionnaire 36
1.17 Previous accidents 36
1.17.1 Crown Princess 36
1.17.2 Pacific Star 37

SECTION 2 - ANALYSIS 39
2.1 Aim 39
2.2 Fatigue 39
2.3 Itinerary planning 39
2.4 Decisions on passage 40
2.4.1 Mystery Island 40
2.4.2 Decision to heave to 40
2.4.3 Ability to observe the seas 41
2.5 Roll mechanism 41
2.5.1 Effect of synchronous rolling, parametric rolling and
abnormal waves 41
2.5.2 Stabilisers 41
2.5.3 Guidance to masters 42
2.6 Effect of heavy rolling 42
2.6.1 Preparations for heavy weather 42
2.6.2 Securing moveable items 43
2.6.3 Securing of fixed items 43
2.6.4 Effect on muster stations 44
2.7 Emergency response 45
2.7.1 Passenger muster 45
2.7.2 Crew alert 45
2.7.3 Communications equipment 45
2.8 Similar accidents 46

SECTION 3 - CONCLUSIONS 47
3.1 Safety issues directly contributing to the accident which have
resulted in recommendations 47
3.2 Other safety issues identified during the investigation also
leading to recommendations 48
3.3 Safety issues identified during the investigation which have not
resulted in recommendations but have been addressed 48

SECTION 4 - actions taken 49

SECTION 5 - recommendations 50
Annexes

Annex 1 Weather routing information from Weather Routing, Inc

Annex 2 Original list of injured passengers and crew, indicating the location on
board, where known

Annex 3 Medical Triage: code tags and terminology

Annex 4 MetService weather report for the accident

Annex 5 Initial study of synchronous and parametric rolling vulnerability

Annex 6 Princess Cruises Fleet Regulations – Heavy Weather (extract)

GLOSSARY OF ABBREVIATIONS AND ACRONYMS
CCTV - Closed Circuit Television
CDEM - Civil Defence and Emergency Management
CLIA - Cruise Lines International Association
ERC - Emergency Response Center (Princess Cruises)
ERO - Emergency Response Organisation
ETA - Estimated time of arrival
EU - European Union
GES - General Emergency Signal
GM - Metacentric Height
GMDSS - Global Maritime Distress and Safety System
HF - High Frequency
IACS - International Association of Classification Societies
IMO - International Maritime Organization
kW - Kilowatt
LR - Lloyd’s Register
LT - Local Time
m - metres
MCA - Maritime and Coastguard Agency
Medivac - Medical Evacuation
Metservice - Meteorological Service of New Zealand
MSL - Mean Sea Level
MTN - Marine Telecommunications Network
nm - nautical miles
NTSB - National Transportation Safety Board
OOW - Officer of the Watch
PA - Public Address System
PSA - Passenger Ship Association
PSD - Passenger Services Director
s - seconds
SERS - Ship Emergency Response Service (LR)
SMC - Safety Management Certificate
SOLAS - Safety of Life at Sea Convention 1974, and its Protocol of
1988 as amended
t - tonnes
TAIC - New Zealand Transport Accident Investigation Commission
UTC - Universal Co-ordinated Time
VSAT - Very Small Aperture Terminal
WRI - Weather Routing, Inc.
Z - Zulu Time, alternative nomenclature used by some authorities
instead of UTC

Times: Between 0200 on 26 July and 0200 on 29 July times are UTC+11 hours,
all other times are UTC+12 hours.
Pacific Sun
SYNOPSIS
During the evening of 30 July 2008 the cruise ship Pacific Sun rolled heavily in gale
force winds and high seas while returning to Auckland on the final leg of an 8-day
cruise of the South Pacific. Of the 1730 passengers and 671 crew on board, 77 were
injured, with seven sustaining major injuries.

The motion of the ship had increased during the day, and at sunset the master had
hove to into wind and swell, in doing so reducing the vessel’s speed to below that at
which the one working stabiliser was effective. Two hours later, the ship rolled heavily
three times, to an estimated angle of heel of 31º, as the master was attempting to
reduce her motion by altering course.

Many of the injuries sustained by the passengers and crew were caused by falls
and contact with unsecured furnishings and loose objects in the busy public rooms,
including those designated as passenger emergency muster stations. Following the
accident, the moving furniture and debris made many of the public rooms unusable,
and the master instructed the passengers to return to their cabins for their own safety.

Had Pacific Sun’s furnishings and fittings been sufficiently secured so as to resist
moving when she heeled, the number of injuries would have been greatly reduced.

As a consequence of this accident, Princess Cruises has taken action to: supply its
bridge teams with night vision glasses; improve deck officers’ training in the risks
associated with heavy weather; and review the securing arrangements for its vessels’
satellite communications equipment.

Princess Cruises has been recommended to: review the role of active stabilisers
in ensuring passenger safety; review the risk of injury from moving furnishings and
objects, and develop suitable means of securing such items for heavy weather;
develop a standard for securing furnishings and equipment in public spaces; and
develop its heavy weather guidance and instructions to include actions to reduce the
risk of injury to personnel.

MAIB has recommended that the Cruise Lines International Association and the
Passenger Shipping Association develop a guide on industry best practice based on
Princess Cruises’ standard for securing furnishings. The trade associations have
also been recommended to promulgate the lessons learned from this accident to their
members.

1
SECTION 1 - FACTUAL INFORMATION
1.1 Particulars of Pacific Sun and accident
Vessel details

Registered owner: : Carnival PLC

(trading as P & O Cruises Australia)

Manager : Princess Cruises

Operator : P&O Australia

Port of registry : London

IMO Number : 8314122

Flag : UK

Type : Passenger cruise ship

Built: : July 1985

Construction: : Steel

Length overall: : 223.36m

Gross tonnage: : 47,546

Accident details

Date & Time: : 30 July 2008, 1945 LT UTC+12

Location of incident: : 200 miles NNE North Cape, New Zealand

Persons on board: 1730 passengers and 671 crew

Injuries/fatalities: 77 injuries to passengers and crew,

including 7 major injuries

Damage/pollution: Substantial damage to internal

furnishings. No pollution

2
1.2 Narrative
1.2.1 Background
Pacific Sun, operated by P&O Australia, was engaged in a series of cruises from
Auckland to the Pacific Islands to the north, taking the passengers to warmer
weather during the southern hemisphere winter. The accident happened during
an 8-day “Summer Daydream” cruise (Figure 1). The itinerary stated that after
departure from Auckland on 23 July 2008, 2 days were to be spent at sea
followed by port calls at Lifou, Port Vila and Mystery Island. There would then
follow a 2 day return passage to Auckland, arriving on 31 July.

Figure 1

Extract from the P&O Cruises

“South Pacific - Queensland - Asia 2007-2009” brochure

3
1.2.2 Pre-departure
At 1630 on 23 July, Pacific Sun’s master made a public address (PA)
announcement to the passengers stating that the vessel would depart late due
to strong winds on the berth. He warned that they would be heading into rough
weather overnight, which might cause the vessel to pitch and roll, and advised
that they take care when moving around the vessel.

Prior to departure, the staff captain notified the department heads by email of the
forecast heavy weather and advised them that equipment should be secured.
The deadlights1 on decks 3 and 4 were confirmed as secure before departure.

Sailing with the ship were two technicians: one employed to check the condition of
the starboard stabiliser and to identify a long-standing fault with the port stabiliser;
and one from the manufacturer of the ship’s satellite communications system.

1.2.3 The cruise

At 2210 on 23 July, Pacific Sun departed Princess Wharf, Auckland assisted by
tugs in 20 knot westerly winds. Due to the late departure, the speed required
to arrive at the first port of call of Lifou had increased from the scheduled 17.1
knots to 18.9 knots, the vessel’s maximum speed in light winds and low swell
conditions.

Early the following afternoon, with the ship making good a speed of 17 knots in
strong headwinds and heavy swell, the master advised the passengers that the
itinerary would be amended. He informed them that Pacific Sun would not call at
Lifou, and that they would head directly to Port Vila, arriving as early as possible
depending on the weather conditions. The next 2 days were spent heading
northward during which the rolling and pitching of the ship gradually moderated.

During the passage, the Global Maritime Distress and Safety System (GMDSS)
High Frequency (HF) transmitter failed. This was recorded in the GMDSS log
and a request was made for a technician to attend during the vessel’s next call in
Auckland.

Pacific Sun arrived alongside in Port Vila slightly early on the morning of 27 July.
However, the vessel’s departure was delayed for a medivac of a sick passenger
who was landed ashore at 0228 the next day, and Pacific Sun sailed from Port
Vila shortly afterwards.

At 0720 on 28 July, the master received the first weather routing information
(Annex 1) by email from Weather Routing, Inc. (WRI)2. The weather forecast
predicted that a storm would form over the South Coral Sea overnight, then drift

1 Steel covers for glass portholes

2 Weather Routing, Inc. provides meteorological consulting, including forecasts and route planning for
commercial fleets and private yachts.

4
 south easterly and develop to give gale force winds later. The master was
advised by WRI to: Modify course / speed as needed for best handling in this
unavoidable heavy weather. The winds were forecast to be force 8-9 gusting 10
from the west north west, with the sea building to 6.5m.

The late departure from Port Vila meant that, despite setting an increased
speed, Pacific Sun arrived at Mystery Island3 at 1430 on 28 July, 6½ hours later
than scheduled. At Mystery Island Pacific Sun anchored and the passengers
were ferried ashore. Because of the time this took, the master decided to delay
Pacific Sun’s departure from 1600 to 1815 so that the passengers could spend
longer ashore.

1.2.4 The passage

Pacific Sun’s passage from Mystery Island to Auckland had been planned on
the basis that the vessel would proceed at 17.3 knots. However, because of the
vessel’s later departure from Mystery Island, her speed was increased to 17.9
knots to meet the scheduled arrival time at Auckland.

WRI weather update No.2 (Annex 1) was received at 0900 on 29 July and
informed that the developing storm would gradually move south; with the wind
forecast to be force 9 from the west north west gusting force 10, with wave
heights up to 6.5m. The advice to the master was:
Route valid, as able. Adjust course / speed as needed in unavoidable
heavy weather.

At 1830 the master made a PA announcement, warning passengers of the

expected heavy weather, and the staff captain sent an email to the department
heads instructing them to check that their departmental areas were secure.

Overnight the weather deteriorated, with the wind and sea increasing from the
north west, and the rolling and pitching also increased. At 0312 on 30 July, the
master altered course to 136º and reduced the speed by 2 knots to place the
sea and swell on the starboard quarter and reduce the vessel’s rolling. At 0400
the Officer of the Watch (OOW) recorded that Pacific Sun was rolling heavily
and pitching due to rough seas and a 5m north westerly swell.

At 0700, the weather routing forecast No.3 (Annex 1) was received from WRI.
The storm centre had moved in a south easterly direction overnight rather than
the predicted southerly track (Figure 2). The master was again advised that:
As able route valid. Modify course / speed as needed for best handling
in unavoidable gale to storm force W-NW winds and large W-NW swells
associated with the aforementioned storm.

3 Mystery Island is also known as Anatom Island or, more correctly, Aneityum Island, the most southerly
island of the Vanuatu archipelago.

5
 Vessel's track

6
Vessel's proposed track

Departure
28/07 - 1800

300
miles

29/07 - 0900
Extract from Admiralty Chart BA 4602 by permission of
the Controller of HMSO and the UK Hydrographic Office

29/07 - 0900
Developing storm
moves South
30/07 - 0312 A/C
30/07 - 0700

Storm track
1945 30/07 - 1745
Accident Hove to

30/07 - 0700
Storm moves SE’ ward
Figure 2

Plot of Pacific Sun and the storm centre

 At 0900 the master advised the Director of Nautical Operations in the Carnival
Office in Sydney of his situation, and gave a revised estimated time of arrival
(ETA) of 1200 on 31 July, 5 hours later than expected.

At 1735, shortly before sunset, Pacific Sun was recorded as pitching moderately
and rolling to a maximum angle of 10º. The starboard stabiliser was confirmed
as extended and working, while the port stabiliser was housed with a known
fault. With sunset approaching, overcast skies, and virtually no moon, the
master was aware that the seas would soon become indiscernible. He therefore
decided to turn from a heading of 135º degrees to 270º to heave to into the
wind and swell.

The master announced to the passengers and crew that the vessel might roll as
she turned. The turn commenced at 1740 in hand steering and, at 1745, it was
complete. The speed was reduced to 4.5 knots over the ground, around 6 knots
through the water, and the roll severity moderated to a maximum of 5-6º at
times. At this speed, the master considered that the one stabiliser had 30-40%
the effect of both stabilisers at full speed. The master advised the OOW to alter
course if necessary to reduce rolling and to make the motion as comfortable as
possible, while he divided his time between his cabin and the bridge.

At 1826 the master made a further PA announcement to the passengers

advising them to take care as they moved around the ship during the evening,
and to use handrails wherever possible.

At 1900 the staff captain sent an email to the heads of department, stating: the
weather forecast is for strong winds and heavy swells to remain this evening.
Before close of business this evening, pls have another check. The instruction
was followed by the company’s “Heavy Weather Precautions” (Figure 3).

1.2.5 The accident

At 1935 the vessel rolled more heavily, to around 14º, on a heading of between
265º and 275º. When the master arrived on the bridge, the OOW was altering
the heading to 280º in an attempt to reduce the severity of the roll. The master
tried to, but he could not, establish the direction of the sea and swell by looking
out of the bridge windows into the darkness.

At 1937 the master used the PA system to announce to the passengers that he
was about to turn the vessel to reduce the rolling. He warned that this might
result in her rolling and pitching. The master took over the conduct of the
ship from the OOW and, at 1938, ordered 10º port rudder. As he did so, but
before the rudder had been applied, the ship started to pitch moderately and
rolled heavily to starboard to around 23º. The ship continued to roll heavily
three times; the final and largest roll heeled Pacific Sun to an estimated 31º to
starboard as the heading was altered to 260º.

7
 Figure 3

Email sent by the staff captain to department heads at 1900 on 30 July 2008

At 1938 the staff captain ordered the fore peak tank to be de-ballasted to reduce
the free surface effect and increase the metacentric height (GM) and trim the bow
up.

The heavy rolling caused passengers and crew, together with unsecured and
insufficiently secured furniture, to move across the ship. The greatest disruption
was in the upper deck large public rooms and restaurants where the majority of
passengers were located. The accident caused fear and great anxiety among
many of the passengers.

The effect of the roll is shown in the sequence from the CCTV footage in (Figure
4).

8
 Figure 4

Legends Bar, Deck 9, prior to the three large rolls

Legends Bar, Deck 9, following the three large rolls

9
1.2.6 Post accident events
The master turned the ship to starboard, and initially steadied her on a heading
of 280º with a speed over the ground of 3.5 knots.

The medical staff started to muster at the hospital on deck 3, aware that a mass
casualty incident was likely.

At 1949, the master altered course to 290º as Pacific Sun continued to pitch
moderately and roll to a maximum of 10º. At this time, the first six casualties
were reported to the bridge by radio, and the master then made a PA
announcement for injured passengers to go to the medical centre on deck 3. At
1953, following a request from the senior doctor, the master announced that the
secondary medical centre had transferred from the Outback Bar and Grill, which
had been badly affected by debris and was unusable, to the Oz Night Club, he
also requested medically trained passengers to volunteer to assist.

At 1954 the master instructed the communications officer to contact Princess

Cruises to establish the company’s Emergency Response Center (ERC)
in California. The satellite C communications system, the usual method of
communicating with the shore, was inoperative and communications were
instead established by one of the satellite B telephones.

At 1956 the Stretcher Party was mustered, and the first serious injury - of a
passenger with severe bleeding - was reported to the bridge.

At 2000, as reports of more casualties continued to be received, the master

sounded the Crew Alert signal. This required that crew return to their cabins,
collect their lifejackets and report to their crew alert stations. The master
announced that all passengers should immediately return to their cabins unless
they were in need of medical assistance. Of the four muster stations, all located
on the upper passenger decks, stations B and C were relocated due to “damage
and safety issues”. Most of the crew dealing directly with the injured passengers
remained with them initially until they felt able to carry out their emergency duty.

The staff captain contacted the Lloyd’s Register (LR) Ship Emergency Response
Service (SERS) by satellite B telephone to request assistance in calculating
the ship’s stability condition. With the satellite C system defective following the
large rolls, the information could not be sent automatically and the data was
transferred orally via the satellite B system.

At 2009, the master contacted the Princess Cruises ERC team, headed by the
Senior Vice President Marine.

At 2012 the master increased speed to 6 knots, maintaining a course of 290º to

try to further reduce the ship’s motion.

10
 Under the direction of the staff captain, the Search and Rescue party started a
thorough search of the ship to ensure that all injured passengers and crew had
been identified and were receiving care. A further announcement was made for
the passengers to return to their cabins.

At 2019, the master was informed that the starboard stabiliser fin was not
working, and he requested the Chief Technical Officer to investigate the failure.

The master was advised that several passengers continued to arrive at their
allocated muster stations, that some were uncomfortable in their cabins and
some were concerned at the sight of the crew wearing lifejackets.

By 2026 all the onboard emergency parties had mustered, accounting for all
crew on board, and the required parties were assisting with the injured.

At 2110 the crew, except the passenger muster personnel, were stood down
from Crew Alert and the operation to clear debris and clean the ship started.

The passenger muster personnel continued to account for the passengers who
were located mainly in their cabins, at one of the two medical centres, or at their
muster stations. However, several of the passengers had congregated in cabins
of friends or relatives, adding to the time taken to account for them. All of the
1730 passengers had been accounted for by 2350.

1.2.7 Recovery phase

Overnight the passengers remained in their cabins while the crew continued to
replace furnishings, remove obstructions, make temporary repairs and clean the
ship (Figures 5a, 5b, 5c and 5d). The satellite B phone line was maintained
between the bridge and the company ERC throughout the night.

At morning twilight the combined sea and swell was observed to be from the
west with a height of 8m. An occasional secondary swell from the north west
was also seen, which was noted as being a higher and steeper swell than the
predominant swell. The logbook recorded Pacific Sun as pitching and rolling
moderately at times.

The fault with the starboard stabiliser could not be identified so it was housed,
but then extended later that morning to act as a bilge keel in resisting the rolling
of the vessel. It was not possible to extend the failed port stabiliser to produce
a similar effect.

During the morning of 31 July the passengers were served breakfast in their
cabins while the ship’s officers made an assessment of the damage. At 1030,
the master announced to the passengers that they could now leave their cabins,

11
 Figure 5a

Public rooms condition following the accident - Outlook Bar and Grill,
Deck 10, Muster Station C

Figure 5b

Legends Bar, Deck 9, Muster Station B

12
 Figure 5c

Speak Easy Piano Bar, Deck 8

Figure 5d

Duty-free shop, Deck 8

13
 but advised them to take care around the ship. He also informed them that
some areas remained out of bounds while the crew continued to clear up the
debris.

Late in the morning the inoperative starboard stabiliser was housed for repair,
which was successful, and at 1347 it was extended and returned to service in
angle control mode.

During the day discussions between the ERC and the master continued in order
to confirm the status of all tanks. Particular attention was paid to the status
of ballast, black and grey water tanks to confirm the ship’s calculated stability
condition.

At 1430 the master updated the New Zealand Maritime Rescue and
Co-ordination Centre of their situation and of the status of the injured
passengers and crew.

Prior to altering Pacific Sun’s course to 148º, towards Auckland, the master
instructed all passengers to return to their cabins as a precaution. The vessel
rolled and pitched moderately throughout the turn.

At 1313 on 1 August, Pacific Sun arrived alongside in Auckland where the three
most seriously injured people were disembarked to awaiting ambulances.

1.3 The Three Instances of Heavy Rolling

The angle of heel was not recorded on board Pacific Sun; crew and passenger
accounts of the extent of the vessel’s motion vary widely. However, CCTV
recordings showed the movement of pictures in the art gallery (Figure 6), which
were suspended by light wires and able to move freely away from the bulkhead
as the ship rolled to starboard. This footage allowed estimates to be made of
the angles of heel achieved as the ship rolled, and the roll period.

The heeling angles to starboard of the three large rolls were measured as 23º,
26º and 31º respectively. The extent of the heeling angles to port is not known,
but CCTV footage of activity on board indicates the heels to port were similar in
magnitude to the heeling angles to starboard.

1.4 Timing of the accident

The accident occurred at 1940, when most of the passengers were either
at dinner or in the public rooms. The main passenger entertainment for the
evening, in the Atlantis Showlounge, had been postponed due to the heavy
weather. However, dinner was being served in the two main restaurants, and
the Outback Bar and Grill, as normal, with their galleys providing a full service.
All the bars on decks 8, 9 and 10 were open.

14
 Figure 6

31°

Art Gallery and Photo Gallery, Deck 9, at the largest recorded roll angle

How the vessel would have appeared at 31°

15
1.5 Injuries to passengers and crew
Of the 1730 passengers and 671 crew on board, 80 passengers and 11 crew
injuries were initially recorded by the medical staff. The figures were later
revised to 69 passengers and 8 crew injuries, due to the apparent duplication
of records. Most of these were bruise and laceration injuries caused by contact
with furnishings, and falls as they were thrown across the vessel. Several
other passengers and crew sustained minor injuries that did not require medical
assistance, were self-treated, or became apparent after the passengers
disembarked. The list of injuries does not include the distress and anxiety
caused to passengers, which reportedly led to several panic attacks.

Three of the injured required immediate hospital treatment on arrival in

Auckland. These major injuries included: a pelvic fracture; rib injury with splenic
injury; and extensive bruising. Seven fractures and a partial finger amputation
were also recorded. A list of the recorded injuries is attached at Annex 2.

None of the passengers or crew was considered to be in the most serious “red”
triage category. Three of the injuries to two passengers and one crew member
were considered to be “yellow” within the triage system. A summary of the triage
system is at Annex 3.

1.6 Damage
1.6.1 Mapping of injuries by location
Pacific Sun sustained no structural damage during the accident. However,
damage to internal furnishings, fixtures and equipment was extensive.

The locations where passengers sustained injuries, in so far as these could be

identified, are shown at Figure 7. The majority of injuries occurred on the busy
upper decks, in the public spaces, bars and restaurants with all the serious
injuries occurring on decks 8, 9 and 10.

1.6.2 Muster stations

Four emergency muster stations, A, B, C and D, were provided on board
(Figure 8) to congregate and account for the passengers in an emergency.
• Muster Station A (Deck 8 & 9) - Atlantis Showlounge
The deck 8 furniture consisted mainly of banquette seating and secured
tables and chairs that remained secure with the exception of one fixed
table that broke free. Only a few passengers were in this room at the
time, and no major injuries were reported.

16
 13 10 1
1 1 1 1 7
2 8 3
1
1
1
2 2 1
Figure 7

Location of injuries to passengers and crew - where known

17
18
Muster Station D Muster Station C Muster Station B Muster Station A

13 10 1
1 1 1 1 7
2 8 3
1
1
1
2 2 1
Figure 8

Location of injuries, where known, and Muster Station locations

• Muster Station B (Deck 9) - Casino and Legends Bar
This muster station combined two areas of deck 9. The effect of the
rolling in the casino area caused large secured items to move, including
a bank of gaming machines to fall from the bulkhead (Figure 9). Some
passengers only avoided injury because the adjacent chairs prevented
the gaming machines from collapsing directly on to the deck. The
electronic poker machine (Figure 10) came adrift and travelled across the
casino deck, causing several reported injuries.

Figure 9

Collapsed casino gaming machines, Casino, Deck 9

19
 Figure 10

Poker Pro machine, released from mountings, Deck 9

In the Legends Bar, where bar service was being provided, the effect
of the roll can be seen in the ‘before’ and ‘after’ pictures at Figure 4.
The settees, tables and chairs were all able to move freely, and the
CCTV footage showed passengers attempting to regain their balance by
grabbing at heavy unsecured tables and other unsecured furnishings as
they were thrown and fell across the deck. The broken glass, liquids and
fallen plant pots added to the hazards the passengers and crew faced,
and at least seven injuries were reported from this area.

• Muster Station C (Deck 10) - Outback Bar and Grill

The highest number of injuries occurred in the Outback Bar and Grill
where at least 13 passengers and crew were injured. The effect of
the rolling in this restaurant can be seen in Figure 11 where several
passengers and crew were injured either by the moving objects or cut by
the glass and crockery.

As the ship heeled, the unsecured tables and chairs slid across the full
width of the room. Passengers, unable to hold on, fell and slid across
the room with the debris, some crossing the deck several times.

• Muster Station D (Deck 9) – Terraces Lounge

The Terraces Lounge room extends the full width of the ship, and was in
use at the time of the accident, with a band playing. This room contained
a large number of heavy chairs and portable tables. One passenger was
injured in this room.

20
 Figure 11

Outlook Bar and Grill - Deck 10, Muster Station C, during the large rolls

21
1.6.3 Restaurants
Both the Burgundy and the Bordeaux main passenger restaurants on deck
8 were busy providing dinner when the accident occurred. Ten passengers
and crew were injured by falling, or sustained cuts from the broken glass and
crockery. Many passengers fell back from their chairs as the ship rolled, but
most were able to hold on to the large secured dining tables.

1.6.4 Other public rooms and spaces

A grand piano located forward of the casino in the Tivoli Square slid across the
deck and damaged an adjacent bulkhead.

One passenger in the Speak Easy Bar (deck 8) and one in the Smugglers
Lounge (deck 9) sustained serious injury.

Passengers had started to put out their luggage in the passenger cabin
alleyways in anticipation of the hotel crew taking the luggage to deck 3, ready
for offloading the following day. This luggage fell and partially blocked the
passenger alleyways.

1.6.5 Crew and working areas

In the officers, staff and crew mess rooms (Figure 12), all of which were in use
at the time of the accident, no tables or chairs were secured to the deck and all
of these moved as the ship rolled.
Figure 12

Damage to crew mess room, Deck 3

22
The stores handling area on deck 3 was a busy thoroughfare and served as
a storage area for engineering equipment and baggage handling cages. As
the ship rolled, a forklift truck which was stowed in this area, several heavy
pipes and a number of storage boxes were able to move across the full width
of the deck (Figure 13). In addition, at the extremities of the rolls the deck 3
shell doors were partly submerged and, as these were not watertight, seawater
entered the deck, making it slippery.
Figure 13

Deck 3 handling area and crew main access, before and after the large rolls

23
 In preparation for arrival in Auckland the next day, the crew had started to place
on deck 3 the garbage and other equipment for landing, and some of these
stows collapsed, blocking alleyways (Figure 14).

Figure 14

Deck 3 - main crew alleyway after accident

Under normal conditions, the preferred route to the medical centre from the
forward part of the ship for stretcher parties was via the main crew alleyway.
However, the obstructions in the store handling area and main crew alleyway
on deck 3, along with the wet decks, made passage through this area extremely
difficult.

24
1.6.6 Open decks
Many of the passengers were on deck 10 by the Lido pool bar. As the vessel
rolled, the passengers, most sitting in plastic chairs, slid across the open deck
and at least 10 were injured.

The two whirlpool spa pools on deck 11 were full with a total of 4.3t of water,
most of which emptied as the ship rolled and flowed down to deck 10, making
it slippery and hazardous for the passengers on that deck. The public address
on decks 10 and 11 was compromised by the background noise from the wind.
Consequently, some passengers could only partially hear the announcements
made by the master.

1.6.7 Communications equipment

As the vessel rolled, the MTN4 satellite communications system in the
communications office on deck 11 ripped from its securing arrangement and fell
over, tearing out connections and disabling the unit (Figure 15).

Figure 15

MTN satellite
control system

Electro Technical Office after the accident

4In May 2006, SeaMobile acquired Maritime Telecommunications Network (MTN), the industry leader in
delivering global VSAT satellite communications to the maritime industry

25
1.7 Environmental conditions
1.7.1 Weather forecast
Weather route forecast from WRI. Update #1, 27 July 2000Z (0800LT on 28
July) stated:
A ridge of high pressure from New Caledonia s’ward to 30S and e’ward
to the International dateline, will move ese’ward next 1-2 days. Low
center will form over the South Coral Sea during tonight, then drift ese-
ward and deepen to gale later tonight / early 29th, before moving se-ward
and deepening to storm later 29th thru 30th, reaching central New Zealand
late 30th.

Recommend as conditions permit direct routing to Auckland. Note that

the aforementioned Gale/storm in place, unavoidable gale to storm force
w-nw winds and large w-nw swells are likely beginning on 29th, continuing
farther south toward Auckland. Modify course / speed as needed for best
handling in this unavoidable heavy weather.

Weather route forecast from WRI. #2 sent 28 July 2100Z (0900LT on 29 July)
stated:
Developing storm near 28S / 162E will gradually move s’ward towards
41S 168E through morning of 31st with “tail end” of frontal boundary
remaining across the port Villa area…..Ridging associated with series of
high pressure systems will build around developing storm through 31st.

Route valid, as able. Adjust course / speed as needed in unavoidable

heavy weather.
29/12-00z nw-w Force 8-9 swell wnw-nnw 4.0-6.5 swells build, gusts
force 10
30/00-12z w-nw Force 8-9 wnw -nnw 6.5-5.5 gusts force 10

Weather route forecast from WRI .#3 sent 29 July 1900Z (0700LT on 30 July)
stated:
Storm centred near 33s 169e, will drift ese-se’ward through 01st with
center of system reaching /crossing central south island on 01st.
meanwhile ridge of high pressure will build e’ward in the wake of the
storm and into/across much of the s’rn Coral and n’rn Tasman seas over
next 2-3 days.

As able, route valid. Modify course/speed as needed for best handling

in unavoidable gale to storm force w-nw winds and large w-nw swells
associated with the aforementioned storm.
29/19-00z wind w-nw F8-9 swell nnw-nw 4.5-6.5 periods F10.
30/00-12 w-nw F8-9 swell wnw-nnw 6.0-8.0 periods F10.

26
1.7.2 Sunset and twilight
Sunset on 30 July was at 1741LT and civil twilight at 1806LT. There was 5%
moonlight, although the sky was overcast.

1.7.3 Weather hindcast from the Meteorological Service of New Zealand

The Meteorological Service of New Zealand (MetService) prepared a report at
the request of the New Zealand Transport Accident Investigation Commission
(TAIC). The report (Annex 4) concluded that:
From 0000 UTC on 29 July 2008 gale and storm warnings, and the
Marine Weather Bulletin for Subtropic clearly indicated that northwesterly
gales and heavy swells could be expected to affect the planned route
of the “Pacific Sun” between Aneityum in Vanuatu and Auckland. The
sequence of MSL5 analysis charts show that the gale and storm warnings
and the Marine Weather Bulletins correctly described the wind and wave
conditions that were experienced by the ship during the voyage. The
computed wind wave and swell wave product showed that the time of
highest combined wave height at the position of “Pacific Sun” coincided
approximately with the time of the heavy weather incident at 0740 UTC
30 July 2008.

MetService predicted that the expected combined sea and swell height in the
area at the time of the accident were in the region of 7m from the west with a
period of around 11s (Figure 16).

1.7.4 Effect of the storm on New Zealand

The mainland of New Zealand was buffeted by the weather system encountered
by Pacific Sun, which locally disrupted road, rail, and air travel, and electricity
and water supplies on 30 and 31 July. People were advised to avoid
unnecessary travel in the affected areas and to listen to local radio for updates
from the Ministry of Civil Defence & Emergency Management (CDEM). CDEM
reported that:
Marlborough CDEM Group declared a state of local emergency in the
Picton area at 3pm [30 June] due to flooding and the need to evacuate
some people.

There were widespread reports of landslips, flooding, and damage to trees,

buildings and roads in many regions of New Zealand during the storm.

5 Mean Sea Level

27
28
Taken from MetService Weather Report (Annex 4)
Figure 16

MetService wave analysis for 0600Z (1800 ship’s time) 30 July 2008
1.8 Ship motion influences
1.8.1 Abnormal waves
The mariner’s handbook notes that:
A well found ship properly handled is designed to withstand the longest
and highest waves she is likely to encounter as long as they retain their
original shape. But when waves are distorted by meeting shoal water,
a strong opposing tidal stream or current, or another wave system,
abnormal steep fronted waves must be expected. Abnormal waves
may occur anywhere in the world where appropriate conditions arise. In
places where waves are normally large, abnormal waves may be massive
and capable of wreaking severe structural damage on the largest of
ships, or even causing them to founder.

The Maxwave project6, undertaken by the European Union, showed that

abnormal waves occur much more frequently than had been previously believed.

MetService notes in its report (Annex 4) that:

Individual waves may be much higher than this representative [significant
wave height] measure; approximately 1 in 100 waves will be twice the
significant wave height.

It is possible, therefore, that the waves experienced by Pacific Sun at the time of
the three large rolls could have been exacerbated and steepened into abnormal
waves due to the presence of the secondary wave pattern, which was observed
by the master at daybreak the following morning.

1.8.2 Synchronous rolling

Synchronous rolling is an example of resonance. Large rolling motions may be
excited when the natural roll period of a ship coincides with the encountered
wave period.

MAIB studied Pacific Sun’s vulnerability to synchronous rolling (Annex 5).

The calculated encounter periods were not coincident with Pacific Sun’s
predicted natural roll period, therefore synchronous rolling is thought an unlikely
contributor in this case.

6 Maxwave, full title ‘Rogue Waves – Forecast and impact on marine structures’ Dec 2000 to Dec 2003
was a sub programme of EU Programme FP5-EESD-1999, Sustainable marine ecosystems, Operational
forecasting of environmental constraints of offshore activities.

29
1.8.3 Parametric rolling
Parametric rolling can be a particular problem on ships designed or modified
with a flat transom and large bow flare. This is a common design, found
particularly on many modern container ships, which maximises deck cargo
capacity while minimising hull resistance with fine hull lines. Cruise ship hull
forms like Pacific Sun can be considered not dissimilar to that of a container
vessel.

The parametric rolling phenomenon occurs as a result of the transverse

stability alternating between being ”stiff”7 and ”tender”8 as a wave passes along
the length of the hull, leading to a rhythmical increase of any slight angle of
heel that may have been initially present. Ships are particularly vulnerable to
parametric rolling when the encountered wave period is half the vessel’s natural
roll period, and the wave length is roughly equal to the ship’s length.

The study at Annex 5 examined different wave periods at various ship

headings, corresponding to a ship speed of 6 knots through the water to
establish if the criteria for parametric rolling were met. It was found that in
head seas and near head seas with a wave period of 11s, the wave length was
roughly equal to Pacific Sun’s hull length, and the encountered wave period was
half that of the ship’s predicted natural roll period of 17.3s.

The wave period from analysis and prediction from the MetService report
(Annex 4) shows a significant wave height of 7m, from the north west with a
wave period of around 11s, less than 2 hours before the accident (Figure 16).

Parametric rolling cannot, therefore, be discounted as a contributing factor to

the accident. However this cannot be confirmed without much greater study
and analysis to establish the susceptibility of Pacific Sun’s hull form to this
phenomenon.

1.9  Pacific Sun’s resilience to rolling

In compliance with SOLAS requirements, Pacific Sun’s main propulsion
machinery and all auxiliary machinery was designed to operate when rolling up
to an angle of 22.5º to port and starboard while simultaneously pitching up to
7.5º by the head or stern.

7 Said of a vessel when she offers exceptional resistance to forces tending to list her.
8Said of a vessel having a small righting moment; so being easily moved from her position of equilibrium,
and slow returning to it.

Dictionary of Nautical Words and Terms, C.W.T Layton.

30
1.10 The stabilisers
Pacific Sun was fitted with a pair of Sperry Marine Gyrofin roll stabilisers. In
calm conditions the stabilisers were housed within the ship’s hull. When
required one, or both, is extended hydraulically depending on the extent of
the rolling (Figure 17). Stabilisers have no effect in reducing a ship’s pitching
motion.

Figure 17

Typical stabiliser machinery unit as fitted to Pacific Sun

The control surfaces on stabilising fins work in the same way as aircraft wing
ailerons. When tilted down, they deflect the water flow downwards and, in doing
so generate lift. Similarly an upwards tilt produces a downward pressure. A
gyro sensor will detect rolling torque and so move the fin control surface to
oppose a roll before the ship has started to heel over. Active gyro fin stabilisers
are operated in either the angle control or lift control modes, the lift control mode
producing maximum fin efficiency. The operation of the stabilisers could be seen
from a panel located on the bridge (Figure 18), but no audible alarm was fitted
to indicate that an operating stabiliser had failed.

31
 Figure 18

stabiliser fin angle

Bridge Gyrofin control unit

Pacific Sun’s port stabiliser was inoperative because parts of the system were
heavily worn and could not be replaced until the next period in dry dock. Her
starboard stabiliser was operating in angle control mode due to a failure of
the lift transducers required for the lift control mode. This was caused by salt
water entering the transducer compartment. An overhaul of both the stabilisers
had been scheduled for the vessel’s previous dry dock maintenance period, in
Brisbane in 2004. However, Brisbane’s dry dock was insufficiently wide for the
stabilisers to be withdrawn, and the overhaul was postponed to the forthcoming
2008 dry dock.

A single stabiliser will produce around 60% of the effect of two. Active
stabilisers become ineffective at speeds of around less than half their design
speed, in Pacific Sun’s case at speeds through the water of less than 10 knots.
The speed through the water at the time of the accident was around 6 knots,

32
 and this would have produced no lift effect from the stabiliser. However, the
extended starboard fin would have generated some resistance to the vessel’s
rolling, producing an effect similar to that provided by a bilge keel.

There is no statutory requirement for stabilisers to be fitted to seagoing vessels.

However, stabilisers have become a standard item on cruise vessels for
passenger comfort. Their condition is monitored by Classification Societies only
to ensure the watertight integrity of the units.

1.11 Stability
At 0700 on 30 July, routine calculation of Pacific Sun’s stability showed a
metacentric height (GM) of 1.45m when corrected for free surface effect. The
vessel was trimmed 0.37m by the stern. At the time of the accident, Pacific
Sun comfortably complied with the statutory ‘intact’ and ‘damaged’ stability
requirements.

1.12 Master and officers

Master
The master had been employed mainly with Princess Cruises since 1993, and
he was promoted to the position of master in April 2007. He joined Pacific Sun
on 18 April 2008 for the first time, and was appropriately qualified including an
STCW II/2 certificate.

Staff Captain
The staff captain had been at sea for 20 years, the last 10 years of which had
been with P&O Australia. He held an STCW II/ 2 certificate and had been a staff
captain for 4 years. He had been on board Pacific Sun for 5 weeks.

Officer of the watch

The on-watch first officer had been employed on Princess Cruises’ vessels for
several years. He held an STCW II/2 certificate, and had been on board for 1
month prior to the accident.

Passenger Services Director

The passenger services director (PSD), in charge of the passenger muster
personnel, had been at sea for 21 years. She had first joined Pacific Sun 18
months before the accident, for the first of her 4 month appointments. She was
promoted to PSD in May 2008.

Senior Doctor
The senior of two doctors on board had been qualified for 20 years, initially
as an anaesthetist, and she had subsequently worked as an accident and
emergency doctor trained in triage procedures. She had worked on several
company vessels for 2 years.

33
1.13  Pacific Sun – background and management
1.13.1 Management
Pacific Sun, originally named Jubilee, was one of three “Holiday Class” ships
constructed for Carnival Cruise Lines by Kockums Varv, Malmo, Sweden. She
commenced operation in 1986. At build, her two Sulzer 11755kW main engines
provided a maximum speed of 21.5 knots, however Pacific Sun’s achievable
speed was about 19 knots at the time of the accident.

In 2004, Jubilee was renamed Pacific Sun, flagged into the British Registry,
and traded under the banner of P&O Cruises Australia, managed by Princess
Cruises. The vessel has remained with Lloyd’s Register Classification Society
throughout her service life.

As part of an internal realignment of management responsibilities, management

of Pacific Sun transferred to P&O Cruises UK in September 2008.

1.13.2 Itinerary planning

P&O Australia has operated cruise ships from Australia and New Zealand for
several decades, the cruise routes being developed with ships originally built as
passenger liners.

Pacific Sun was engaged on a series of winter cruises from Auckland to the
Pacific islands, mainly Fiji, New Caledonia, and the Vanuatu archipelago with
occasional longer cruises. The “Summer Daydream” cruise was a regularly
repeated itinerary, which was operated at various times of the year by P&O
Australia’s fleet. As this was a tried and tested itinerary, it did not require
additional management consideration on this occasion.

1.13.3 Heavy weather guidance

The heavy weather checklist was reviewed throughout the voyage; it was
referred to frequently as the conditions worsened and also following the
accident. Guidance to the master about the required actions in heavy weather
was provided within the fleet regulations and is shown at (Annex 6).

1.13.4 Emergency procedures

In an emergency, the Crew Alert Signal was usually sounded prior to the
General Emergency Signal (GES) to ensure that the stairwells, evacuation
routes and muster stations were manned in anticipation of the passengers going
to their muster stations and therefore prevent the congestion of passengers
and crew in stairwells and alleyways. Once the crew were in position the
GES would be sounded if required and passengers, having first collected
their lifejackets and warm clothing from their cabins, would go to their muster
stations.

On this occasion the passengers were instructed by the master to return to their
cabins at the same time as the crew were ordered to their Crew Alert positions.

34
1.13.5 Mass casualty response
Two mass casualty exercises had been held recently on board Pacific Sun. The
first, held in December 2007 at the company’s behest, had been to develop
the company-wide triage and mass casualty response system. The second, a
ship-board drill in April 2008, had identified the Oz Nightclub as the preferred
secondary medical centre and this arrangement was expected to be formally
adopted once approval had been received from the vessel’s managers.

1.14 External inspection

The Maritime and Coastguard Agency (MCA) issued Pacific Sun’s Safety
Management Certificate (SMC) on 12 December 2006 and the Passenger Ship
Safety Certificate in September 2007. The last Port state control was carried out
in May 2008.

1.15 IMO Regulation

The Safety of Life at Sea Convention (SOLAS), regulation II-2 Part B Reg 5. 3.3
Furniture in stairway enclosures of passenger ships, states:
Furniture in stairway enclosures shall be limited to seating. It shall be
fixed, limited to six seats on each deck in each stairway enclosure,
be of restricted fire risk determined in accordance with the Fire Test
Procedure Code, and shall not restrict the passenger escape route.
The Administration may permit additional seating in the main reception
area within a stairway enclosure if it is fixed, non-combustible and
does not restrict the passenger escape route. Furniture shall not be
permitted in passenger and crew corridors forming escape routes in cabin
areas. In addition to the above, lockers of non-combustible material,
providing storage for non-hazardous safety equipment required by these
regulations, may be permitted. Drinking water dispensers and ice cube
machines may be permitted in corridors provided they are fixed and
do not restrict the width of the escape routes. This applies as well to
decorative flower or plant arrangements, statues or other objects of art
such as paintings and tapestries in corridors and stairways.

SOLAS also states:

Muster stations shall be provided close to the embarkation stations. Each
muster station shall have sufficient clear deck space to accommodate all
persons assigned to that muster station, but at least 0.35m² per person.

There are no requirements for furnishing within muster stations to be secured,

as there are for escape routes.

35
1.16 Passenger questionnaire
MAIB passenger questionnaires9 were sent to the home addresses of all the
passengers on board who were older than 18. A total of 310 responses were
received.

1.17 Previous accidents

1.17.1  Crown Princess
The following is an extract from The National Transportation Safety Board
(NTSB) report on its investigation into a heeling accident on M/V Crown
Princess, Atlantic Ocean off Port Canaveral, Florida July 18, 2006.
On July 18, 2006, the cruise ship Crown Princess, which had been in
service about a month, departed Port Canaveral, Florida, for Brooklyn,
New York, its last port on a 10 day round-trip voyage to the Caribbean.
Slightly more than an hour after departing, while on a heading to intersect
its track to Brooklyn, the vessel’s automatic steering system began a turn
to port. In an effort to counter the effects of a perceived high rate of turn,
the second officer, the senior watch officer on the bridge, disengaged the
automatic steering mode of the vessel’s integrated navigation system and
took manual control of the steering. The second officer turned the wheel
first to port and then from port to starboard several times, eventually
causing the vessel to heel at a maximum angle of about 24º to starboard.

The heeling caused people to be thrown about or struck by unsecured

objects, resulting in 14 serious and 284 minor injuries to passengers and
crewmembers.

The vessel incurred no damage to its structure but sustained

considerable damage to unsecured interior components and to cabinets
and their contents.

The National Transportation Safety Board determines that the probable

cause of the Crown Princess accident was the second officer’s incorrect
wheel commands, executed first to counter an unanticipated high rate of
turn and then to counter the vessel’s heeling.

The Safety Board’s investigation of the accident identified the following

safety issues: [not inclusive]
• Reporting of heeling incidents and accidents.
• Emergency response following severe incidents.

As a result of its investigation, the Safety Board makes recommendations

to the U.S. Coast Guard, to the Cruise Lines International Association,
and to SAM Electronics and Sperry Marine (manufacturers of integrated
navigation systems).

9 Facilitated by New Zealand Transport Accident Investigation Commission (TAIC).

36
 The report commented on the effect of the heeling on the passenger areas,
stating:
Further, given the debris on the vessel after the accident, including water
from the pools, broken glass, and displaced and overturned furniture,
directing passengers to their muster stations could have exposed them
to hazards of slips, falls, and blunt or lacerating injury. In addition, the
number of injured passengers and crew and the severity of their injuries
were uncertain. Having passengers report to muster stations could have
delayed treatment of the injured. Therefore, the Safety Board concludes
that the captain’s decision not to order passengers to their muster stations
after the accident was appropriate.

Following the accident, in August 2006, the Senior Vice President of Princess
Cruises sent an instruction to the masters of each vessel, including Pacific
Sun instructing them to review the securing arrangements to prevent items,
especially heavy objects, from moving in heavy weather or if the vessel is
subject to large angles of heel (Figure 19).

1.17.2  Pacific Star

Pacific Star sailed from Auckland, New Zealand, in the late afternoon of 10 July
2007 on Summer Daydream cruise, the same cruise itinerary as Pacific Sun.
Overnight the ship experienced easterly winds in excess of 90 knots, and a swell
height of more than 5m. As a result, she suffered damage to internal fittings
and to a number of windows in cabins and public spaces, as well as structural
damage to the forecastle bulwark and the loss of a satellite communications
dome. Significantly, the rough weather also exacerbated pre-existing corrosion
damage causing the ship to take water through small holes in her side.
Following an underwater survey at her next port of call, Port Vila, Vanuatu,
she was instructed by the MCA to proceed directly to a dry dock repair facility,
without passengers.

The subsequent MAIB investigation focussed on the shell plating weakened by

corrosion, and why this had not been detected during survey.

37
 Figure 19

[2006]

Email sent to Princess Cruises Masters following the Crown Prince accident in 2006

38
SECTION 2 - ANALYSIS
2.1 Aim
The purpose of the analysis is to determine the contributory causes and
circumstances of the accident as a basis for making recommendations to
prevent similar accidents occurring in the future.

2.2 Fatigue
All personnel directly involved in this accident were well rested, and there is no
evidence that fatigue contributed to this accident.

2.3 Itinerary planning

Pacific Sun was on a routine “Summer Daydream” cruise to the Pacific islands
to the north of New Zealand that formed part of the company’s programme
during the summer and winter months. P&O Australia accepted that
modifications would be required to several cruises each year, particularly during
the winter months. The cruise planning process did not address this cruise
particularly as it had been completed on numerous occasions in the past.

The cruise started and finished with two lengthy ocean passages, and the
vessel’s itinerary required these to be made at between 17 and 18 knots. At the
time of the accident Pacific Sun had a maximum speed of 19 knots, rather than
the 21 knots she could achieve after each period in dry dock. The schedule
therefore allowed the master very little flexibility to make up time lost due to the
effects of bad weather or if the ship’s departure was delayed.

The passengers’ cruise experience was initially diminished when the first port of
Lifou was missed due to the late departure from Auckland and poor weather en
route, and Pacific Sun headed directly to Port Vila. The delayed departure from
Port Vila led to a greatly reduced call at Mystery Island, which reduced their
enjoyment of the cruise further.

While the master delayed his departure from Mystery Island for as long as
possible, his decision meant that the speed required on the return leg, of almost
18 knots into forecast bad weather, made it highly improbable that the ship
would arrive in Auckland on time.

The master was fully aware of the consequences of arriving late at the
turnaround port. Revised travel arrangements could be required for passengers
leaving the ship; those joining for the following cruise could have had to be
accommodated until it arrived; and storing and bunkering operations had to be
re-scheduled. Although the master was not put under any pressure to arrive
in Auckland sooner than was safe, the schedule had placed him in a difficult
situation and it would have been natural for him to make every effort to arrive at
the turnaround port on time or to limit the delay to a minimum.

39
 Had allowances been made during the itinerary planning stage for the reductions
in the ship’s achievable speed and the greater probability of rough weather in
the winter months, the master would have had greater flexibility to achieve the
outward and return passages on schedule.

2.4 Decisions on passage

2.4.1 Mystery Island
As Pacific Sun sailed late from Mystery Island, her master was aware that it
was unlikely that she would arrive in Auckland on schedule given the worsening
weather forecast. With the benefit of hindsight, had he decided to miss out the
visit to Mystery Island and instead proceed directly from Port Vila for Auckland,
it is possible that he would have passed ahead of the forecast storm and
arrived alongside early. This would have had an impact on only one cruise; the
subsequent cruise could potentially then have departed on time.

It would have been a very robust decision for the master to recommend to the
company that the visit to Mystery Island be cancelled. The visit to one of the
three ports on the itinerary had already been cancelled due to poor weather,
and to cancel a second would have heavily impacted on the overall cruise.
Further, he was looking at the 48 hours ahead weather forecast, which indicated
nothing substantially worse than he and Pacific Sun had encountered many
times before. Therefore while, in hindsight, it would have been better to have
cancelled the visit to Mystery Island, the master’s decision to continue with the
visit was understandable.

2.4.2 Decision to heave to

Having left Mystery Island, the weather routing information from WRI advised
the master to set a direct route to Auckland and modify courses and speeds as
necessary in the prevailing conditions. At the start of the passage the storm
was forecast to move south. However, it moved south east over the following
24 hours, converging on Pacific Sun’s route. As the weather conditions
deteriorated, the master altered course and speed to make the movement of the
ship as comfortable as possible by placing the sea and swell on her starboard
quarter.

The master did not consider heaving to until just before sunset, when he
realised that he would be unable to see the sea and swell in the ensuing
darkness. In hindsight, had he decided to heave to earlier in the day, the
depression would have passed further to the south of Pacific Sun, and she
would have encountered lesser winds and smaller waves.

By deciding to run parallel to the predicted path of the storm for as long as
possible, and only to heave to at sunset, the master inadvertently placed Pacific
Sun in the area of the worst sea conditions, where she was most vulnerable to
the wind and seas because her one operational stabiliser was ineffective at slow
speed.

40
2.4.3 Ability to observe the seas
On 30 July it was a dark, overcast night, with virtually no moon, and neither the
OOW nor the master was able to identify the direction of the sea and swell. A
better view of the sea could have enabled them to identify the best heading for
the ship when hove to, to reduce the rolling, and to time any course alterations
so as to avoid particularly large waves.

Portable night vision glasses are now commercially available and are not
prohibitively expensive. In this case, had the bridge team had access to night
vision glasses they would have gained a better understanding of the sea
conditions they faced, and might have been provided with some warning of the
approach of any abnormally large waves.

2.5 Roll mechanism

2.5.1 Effect of synchronous rolling, parametric rolling and abnormal waves
Whether Pacific Sun rolled to large angles due to the presence of an abnormal
wave or waves, or was subjected to the effects of parametric rolling cannot be
stated with any certainty.

MAIB’s calculations into the ship’s vulnerability to synchronous rolling or

parametric rolling (Annex 5) have indicated that while synchronous rolling
is unlikely, parametric rolling was possible if Pacific Sun’s hull form was
susceptible to this behaviour. However, the lack of a steady build up to the
three large roll angles experienced does conflict with normal parametric rolling
behaviour.

It is more likely that the large sea and swell, possibly combined with the second
more northerly swell identified the following morning, produced a series of large
steep sided abnormal waves. These abnormal waves, unseen in the dark,
would have most likely been the cause for Pacific Sun rolling so heavily.

2.5.2 Stabilisers
Both stabilisers were due to be extensively overhauled during the dry dock
maintenance period scheduled for a few weeks after the accident. With only
one stabiliser working, Pacific Sun’s ability to dampen her roll was reduced,
and the potential for the remaining stabiliser to fail, leaving the vessel without
active stabilisers, increased as the time since the system’s last major overhaul
increased.

Shore managers and the ship’s officers understood the critical nature of the
stabilisers, particularly for schedules in the South Pacific winter, and remedial
maintenance was provided immediately prior to the cruise. However it was not
possible, or expected, to bring the defective port stabiliser back into service
before the next dry dock period, and it was considered acceptable for the vessel
to continue to operate with only one working stabiliser.

41
 Had both stabilisers been maintained, as originally intended, during the vessel’s
previous dry dock period in 2004, it is likely that both would have been available
for the cruise. By deferring that maintenance, the likelihood increased that
Pacific Sun would, at some point, be left without working stabilisers during a
winter storm, as occurred on this occasion.

The malfunctioning port stabiliser did not directly contribute to the accident.
However, Pacific Sun’s master would have been better able to handle the rolling
of the ship throughout the cruise if both stabilisers had been working.

It is not possible to determine whether the remaining stabiliser failed prior to,
during, or after the large rolls because no alarm sounded (or was required
to sound), and the failure was not noticed until around 30 minutes after the
accident. The master and officers were therefore unaware that the stabiliser had
failed and was not providing the roll reduction they anticipated.

When the ship was hove to, she was already vulnerable to the effect of the
forecast seas and swell. At 10 knots or more, had one, or both stabilisers been
operational, they would have moderated Pacific Sun’s heavy rolling. However
this speed would also have resulted in increased pitching. To avoid heavy
pitching, the master reduced Pacific Sun’s speed to 6 knots, 4.5 knots through
the water. At this speed, he believed that the starboard stabiliser would be
partially effective in the active mode. In reality, at that speed the stabiliser was
only effective as a bilge keel.

2.5.3 Guidance to masters

Pacific Sun’s master was not aware that at speeds of less than 10 knots the
active stabilising effect was lost completely and the stabilisers, if extended,
would only be acting as bilge keels. Neither did he have any information on
Pacific Sun’s vulnerability to either synchronous or parametric rolling in certain
conditions, particularly when operating at speeds when the active stabilisers
were ineffective. No such ship specific guidance or training in these risks is
provided to the masters of ships within the Carnival group of companies.

Carnival Group should consider making an assessment for each of its ships
of their vulnerability to synchronous and parametric rolling and other handling
characteristics, for example stabiliser effectiveness, to provide its masters with
vessel specific guidance on ship handling in rough weather.

2.6 Effect of heavy rolling

2.6.1 Preparations for heavy weather
Before encountering and during the heavy weather, the staff captain sent routine
e-mails to all heads of department instructing them to secure their areas, and
referred them to the relevant fleet instruction. All departments secured their
equipment in their normal manner. While this level of securing was effective

42
 during the passage north from Auckland, it was ineffective for the conditions
experienced on the return leg when many items were insufficiently secured to
withstand the severe rolling.

Had the severity of the weather conditions experienced by Pacific Sun after
sailing from Mystery Island been anticipated, the senior officers could have
taken a number of actions to mitigate the effects of extreme ship motion. These
could, for example, have included:
• A further review of the securing arrangements for all moveable objects.
• Postponing the preparations for arrival, including delaying the collection
of passenger baggage and the pre-positioning of stores, garbage and
equipment in alleyways and handling areas ready for landing.
• Restricting passenger movement and activity.
• Other actions, such as emptying the spa baths and modifying menus to
reduce the hazards present in the galleys during cooking.

These actions could impact on the passengers’ enjoyment of the cruise, and
therefore would not be adopted lightly. To guide ship’s staff, a methodology is
needed to identify when such steps are appropriate.

2.6.2 Securing moveable items

The officers and crew believed that they were prepared for heavy weather,
but were hampered by physical limitations. For example, few items of heavy
furniture had securing points as part of their design, and the decks did not have
suitable fittings for these heavy items to be secured to. In the large public
areas, heavy items were able to move unrestricted during the rolling, posing a
significant hazard to passengers and crew and causing many of the reported
injuries.

Had the vessel been provided with additional means of securing for bad
weather, then ship’s staff could have secured these heavy objects more
thoroughly.

2.6.3 Securing of fixed items

A number of items that were, in theory, permanently fixed broke free of their
securing arrangements. These included such items as a grand piano, a
bank of gaming machines in the casino, and heavy office equipment such
as photocopiers, many of which were being used by passengers to steady
themselves against the motion. It was pure good fortune that some passengers
and crew were not more seriously injured or killed by the unexpected movement
of supposedly fixed heavy items.

43
 Attempts to identify and secure items, and especially heavy objects, on Princess
Cruises’ vessels following the Crown Princess accident in July 2006 were not
successful in preventing similar items from breaking free on Pacific Sun. In the
absence of an industry standard, Princess Cruises should develop a company
standard for securing fixed items on board its vessels and apply it across its fleet
as soon as practicable.

2.6.4 Effect on muster stations

Muster stations are allocated to provide protection for passengers and to
congregate them away from potential danger in an emergency. The muster
stations on board Pacific Sun, all in public areas, could have been required to
assemble the passengers had the situation deteriorated further.

On this occasion, due to the quantity of loose items and debris, and the
continued risk of further heavy rolling, two and possibly three of the four muster
stations would have been unable to safely accommodate the numbers of
passengers they were required to hold. This would have left the master and
PSD with the difficult task of accounting for, and managing, all the passengers
on board without using the nominated or alternative muster stations.

In the fourth muster station, Muster Station Alpha, the Atlantis Showlounge, all
furniture, including rotating chairs, was secured. Although one table released
from its mounting, the impact of the rolling on this room was minimal and no
injuries were reported there. This is in striking contrast to the condition of the
three other muster stations with mainly unsecured tables and chairs where
significant numbers of injuries occurred.

Had all the muster stations been secured to the same standard as Muster
Station Alpha, the number of injuries sustained due to moving furniture would
have been greatly reduced.

Pacific Sun temporarily heeled to an estimated angle of 31º, yet her machinery
was only required to operate to an angle of 22.5º. In the worst case, the vessel
could have lost engine power and then lain beam to the seas and rolling heavily
until power was restored. In these circumstances, the master might well have
sounded the GES, sending the passengers to their muster stations as part
of his emergency response. In such an event, passenger safety would have
depended on the tenability of the muster stations.

There are no SOLAS regulations to ensure that furnishings and equipment on

board passenger ships’ muster stations are secured, and that no extraneous
objects are placed there. The onus therefore lies with the operating company
to decide on the appropriate degree to which furniture and other objects are
secured in public areas on passenger vessels which have been designated as
muster stations.

44
2.7 Emergency response
2.7.1 Passenger muster
It took the crew almost 4 hours to account for all the passengers following the
master’s instruction for them to return to their cabins. With no established
system to account for the passengers in their cabins, the existing system of
mustering passengers had to be adapted. Accounting for passengers was
made more difficult because some passengers were in the Medical Centre and
Secondary Medical Centre, and a number had chosen to go to their muster
stations or had joined friends in other cabins.

In previous accidents on Princess Cruises’ vessels the passengers had

been instructed to return to their cabins as part of the emergency response.
However, the opportunity had not been taken to formalise the process, and
Pacific Sun’s ERO did not provide for a suitable system of accounting for
passengers once they had returned to their cabins. Had such a system
been established, and the crew trained in its use, the time taken to ensure all
passengers were accounted for would have been much reduced.

2.7.2 Crew alert

The sight of crew in lifejackets caused concern to many passengers. The
system of Crew Alert normally preceding General Emergency meant that the
crew were in possession of their lifejackets and the passengers were not.

Had the master not sounded the Crew Alert Signal, he would have been unable
quickly to account for the crew and would not have had emergency teams, such
as the fire parties and the passenger assistance party, in place had the situation
deteriorated. The master therefore made the most appropriate decision in
responding to his current problems and anticipating the possibility that the
situation could deteriorate further.

Sounding of the Crew Alert Signal during the initial confusion of an actual
emergency will always have a negative impact on some passengers, regardless
of announcements that may be made to reassure them. This situation exists
extensively throughout the cruise industry and better solutions have yet to be
found.

2.7.3 Communications equipment

The collapse of the main satellite C communications system reduced the
master’s ability to communicate effectively with external organisations and with
the Princess Cruises ERC.

The failure of the GMDSS HF and MF system prior to the accident also limited
the vessel’s ability to communicate effectively and might have had a detrimental
effect had the situation deteriorated and further external communications been
necessary.

45
 Although not part of the GMDSS equipment, the loss of the satellite C system
meant the vessel was not able to link into the company’s ERC’s computerised
emergency management system. Had consideration been given to the system’s
importance, the satellite C equipment could have been more effectively
secured, or additional redundancy provided in the form of an “emergency only”
satellite communication system.

2.8 Similar accidents

The NTSB report into the heeling accident on board Crown Princess, and the
MAIB’s report into the Pacific Star accident, have shown that large passenger
vessels heel and roll for numerous reasons. Such reasons can include heavy
weather, mechanical fault, such as a steering gear failure, or human error, and
the resultant heeling might, or might not, be predictable. Regardless of the
cause of the large angles of heel, cruise vessels should be able to withstand
such accidents without endangering the passengers and crew through
unsecured or insufficiently secured furnishings and equipment breaking loose.

This accident demonstrates that the actions taken by Princess Cruises had
not ensured that procedures for securing furnishings on board its vessels were
sufficiently robust to remove the risk of injury to its passengers and crew in the
event of future heeling accidents, however caused.

46
SECTION 3 - CONCLUSIONS
3.1 Safety issues directly contributing to the accident
which have resulted in recommendations
1. Had both stabilisers been maintained, as originally intended, during the
vessel’s previous dry dock period in 2004, it is likely that both would have
been available for the cruise. By deferring the maintenance, the likelihood
increased that Pacific Sun would, at some point, be left without working
stabilisers during a winter storm, as occurred on this occasion. [2.5.2]

2. It was not possible to determine whether the one working stabiliser failed
prior to, during, or after the large rolls because no alarm sounded, and the
failure was not noticed until 30 minutes after the accident. [2.5.2]

3. Had the severity of the conditions been anticipated, the senior officers could
have taken a number of actions to mitigate the effects of extreme ship
motion. These could have included, for example: a further review of the
securing arrangements; postponing the preparations for arrival; restricting
passenger activity; and modifying menus to reduce the hazards present in
the galleys during cooking. [2.6.1]

4. Had the ship’s staff been provided with additional means of securing for bad
weather they could have secured moveable objects more thoroughly. [2.6.2]

5. A number of items that were, in theory, permanently fixed broke free of their
securing arrangements. [2.6.3]

6. Attempts to identify and secure items, and especially heavy objects, on

Princess Cruises’ vessels following the Crown Princess accident in July 2006
were not successful in preventing similar items from breaking free on Pacific
Sun. [2.6.3, 2.8]

7. There are no IMO SOLAS regulations to ensure that furnishings and

equipment are secured, and that no extraneous objects are placed in the
muster stations of passenger ships. [2.6.4]

8. The collapse of the main satellite C communications system reduced the

officer’s ability to communicate effectively with external organisations and
with Princess Cruises. [2.7.3]

47
3.2 Other safety issues identified during the investigation
also leading to recommendations
1. Passenger vessels can experience large angles of heel for a number of
reasons including: heavy weather; mechanical fault, such as steering gear
failure; or human error, and the resulting heel angles can be unpredictable.
Regardless of the cause of the large angles of heel, cruise vessels should
be able to withstand such incidents without endangering passengers through
unsecured and insufficiently secured furnishings and equipment breaking
loose. [2.8]

3.3 Safety issues identified during the investigation which

have not resulted in recommendations but have been
addressed
1. Had allowances been made during the itinerary planning stage for the
reductions in the ship’s achievable speed and the greater probability of rough
weather in the winter months, the master would have had greater flexibility to
achieve the outward and return passages on schedule. [2.3]

2. Had the bridge team had access to night vision glasses, they would have
better understood the sea conditions they faced, and might have had some
warning of any abnormal waves approaching. [2.4.3]

3. The ship’s emergency response organisation did not provide for a suitable
system for accounting for passengers once they had returned to their cabins.
Had one been established, and the crew trained in its use, the time taken to
ensure all passengers were accounted for would have been much reduced.
[2.7.1]

4. The sight of crew in lifejackets caused concern to many passengers. The

dual system of Crew Alert and General Emergency meant that the crew were
in possession of their lifejackets before the passengers. [2.7.2]

5.  Pacific Sun’s master was not aware that at speeds of less than 10 knots the
active stabilising effect was lost completely and the stabilisers, if extended,
would only be acting as bilge keels. Neither did he have any information
on Pacific Sun’s vulnerability to either synchronous or parametric rolling at
speeds when the active stabilisers were ineffective. [2.5.3]

48
SECTION 4 - actions taken
Princess Cruises has:
• Introduced night vision equipment to assist bridge teams.
• Provided additional training and guidance to masters and deck officers in the
effective handling of its vessels in heavy weather.
• Undertaken a technical review of critical emergency and communications
equipment, and the provision of additional redundancy in portable satellite
communications to ensure effective emergency response.
• Reviewed its itinerary planning process.

Carnival PLC has:

• Carried out a trial muster of passengers in their cabins, instead of their
muster stations, and will amend the fleet’s onboard emergency response
plans accordingly.

Cruise Lines International Association (CLIA) has:

• Agreed to discuss at the CLIA Committee on Technical and Safety Matters
the potential negative impact on passengers of crew mustering in lifejackets
prior to any formal announcement of emergency stations and explore
possible alternative arrangements.

49
SECTION 5 - recommendations
2009/138 Princess Cruises is recommended to:
• Review the role of active stabilisers in ensuring passenger safety. The
review should include system redundancy, planned maintenance, and
the need to fit suitable alarms to warn of system degradation or failure.
• Carry out a fleet-wide assessment of the risk of injury from moving
furnishings and objects following large angles of heel. This should
include the calculation of the forces involved and identify suitable
means of securing objects both routinely, and for heavy weather.
• Develop a standard for securing furnishings and equipment in public
spaces, particularly in muster stations and their access routes.
• Develop its heavy weather guidance and instructions to include
actions to reduce the risk of injury.

2009/139 Cruise Lines International Association and the Passenger Shipping

Association are recommended to:
• Use the Princess Cruises’ standard for securing furnishings and
equipment in passenger vessels (once completed) as a basis for
developing a guide on industry best practice.

• Disseminate the safety issues, actions taken and lessons learnt

following this accident to their members.

Marine Accident Investigation Branch

June 2009

Safety recommendations shall in no case create a presumption of blame or liability

50