A Life Cycle Assessment of the Environmental Impact of Cruise Holidays

Dr Richard Farr

Senior Lecturer (Partnerships)

University of Bolton

Christine Hall

Programmes Validation Manager

University of Bolton

Abstract

Despite difficult economic conditions in recent years, the cruise industry has continued to grow strongly. Questions remain, however, as to the environmental impact of a cruise holiday. The noxious nature of the fuel employed, the sheer quantity required, and the effects of various discharges from vessels at sea are among the issues that impact upon the ‘green’ performance of the industry.

Although the media have been quick to criticise various aspects of the industry in recent years, little research has been done to quantify the actual level of environmental harm caused by the industry. In this paper the authors describe the application of Life Cycle Assessment to produce a model of contributions to climate change that is compared with the impact of a holiday spent ashore.

Keywords

Sustainable tourism, life cycle assessment, carbon footprint, cruise holiday

1. Introduction

The pursuit of sustainability requires that environmental impacts are assessed to determine the modes and levels of harm associated with activities, including tourism. Many different modes of environmental harm exist, but the one most widely recognised is the “carbon footprint”, or contribution to climate change. A variation in weather patterns over the long term, this has a number of causes including volcanic activity and changes in solar activity, but today there is a strong consensus that human activity is a major contributory factor. As the National Resource Council [2010: 2] stated, “Climate change is occurring, is caused largely by human activities, and poses significant risks for a broad range of human and natural systems.”

Whether for ethical reasons or simply to adapt to a world in which commodities including energy are more expensive, humanity is slowly becoming more ‘green’. There is evidence to suggest, however, that although an increasing number of people are prepared to consider the environment at home, in the workplace and in their daily commute, holidays are less likely to be subjected to ‘green’ scrutiny by the tourist [Leslie, 2012; Vij and Vij 2012].

Leslie [2012] describes transportation as tourism’s‘Achilles heel’since it is responsible for such a large part of the contribution to climate change. Transnational organisations such as UNWTO [1999] and UNEP [2009] concur; their efforts to calculate the carbon footprint and thereby encourage sustainability in tourism focus upon air travel, and to a lesser extent rail travel.For a holiday that largely involves staying in one place the assumption that most of the carbon dioxide (CO2) emissions will result from travel to and from the destination is reasonable, but this fails to encompass cruising, where travel continues throughout the holiday. In fact, while many cruise holidays begin and end with an element of air or rail travel, as with other holiday types, the research presented in this paper demonstrates that it is the cruise itself that makes the holiday’s largest contribution to climate change.

Ward [2010] suggested a figure for greenhouse gases resulting from cruise holidays: 960kg CO2 emitted per passenger, for a one-week cruise. For context, the World Bank [2015] estimated UK carbon emissions per capita for 2010-2014 to be 7,900kg per year, or 152kg CO2 per week – suggesting that a week spent cruising entails approximately 6.3 times as much of a contribution to climate change as a week spent at home.

Unfortunately, the process by which Ward’s [2010] figure was arrived at is not specified. It maybe derived solely from the fuel consumption of the ship – an aspect that is responsible for a formidable quantity of CO2 emissions, but which overlooks the operation of what is effectively a floating five-star hotel and leisure complex. The construction and maintenance of the ship (more difficult than fora land-based equivalent) must also be considered.

Rather than simply seeking to confirm or reject the figure of 960kg CO2 per passenger per week, the authors aimed to produce an itemised carbon footprint through the application of Life Cycle Assessment, a technique more commonly seen in manufacturing and engineering. This was achieved with a parameterised model that could be used to explore the specific problems associated with various components of a cruise holiday.

2. Methodology

In order to allow calculations of environmental impact, a mathematical model has been created in which a cruise holiday aboard a fictional cruise ship, the MS Exemplar, is specified. The vessel exists as a set of values that are referenced by formulae in a Microsoft Excel spreadsheet. The model, built according to the principles set out in the PAS2050 standard [BSI, 2008] demonstrates a full ‘cradle-to-grave’ assessment, taking account of the materials employed in the construction of the ship, and the shipbuilding process itself, with the resulting harm being amortised over the total number of passengers carried (making use of estimates of the useful life of the vessel, and the level of occupancy). Fuel, food and drink consumed are taken into account, and materials recovered for reuse at the end-of-life are discounted. The result is a measure of CO2e per passenger, per day and for a whole cruise of a specified duration. (The ‘e’ denotes CO2 equivalency, and is a standard approach among sustainability practitioners, recognising that there are a number of different greenhouse gasessuch as nitrous oxide, methane and sulphur hexafluoride,in addition to CO2, that all play a role in anthropogenic climate change [Samson, 2014]).

The authors note that cruising involves other modes of environmental harm that do not directly affect climate change, such as the emission of carcinogenic particulates, and the dumping of sewage and other wastes while at sea; these are discussed briefly in Section 4, but do not form a part of the present model; only climate change potential is quantified.

2.1Sources of Information Employed

The construction of the model drew upon a number of sources, as identified in this section. Of particular value were the DEFRA [2012] Greenhouse Gas Conversion Factors that facilitate the assessment of a wide range of operations, processes and materials in terms of their carbon footprint. Equipped with this information, once the mass and/or cost of the various things that make a cruise holiday possiblewere determined, it became possible to calculateeach element’s contribution to climate change.

To achieve this inevitably involves a degree of simplification.For example, it is assumed that all passengers bear equal responsibility, despite the fact individuals will differ in terms of their diet, the level of occupancy in each cabin, etc.Furthermore, one cruise might differ from another in terms of the itinerary, because longer sailings between ports of call would equate to more energy use, and cruise itineraries that incorporate ‘sea days’ indicate higher emissions still.A transatlantic crossingthat includessix sea days would contribute significantly more than a seven-day Western Mediterranean cruise package,with perhaps four ports of call and two sea days.Since the model is parametric in nature, many suchvariablescan be set, allowing their contribution to be explored. This may in the future allow the model to be employed in the gathering of additional information; the assumptions incorporated into the ‘base model’ are detailed in Section 2.2.

2.2Assumptions in the Model

A number of assumptions have been made in the construction of the model, any of which might later be refinedas more accurate data become available. These are reviewed in the subsections that follow.

2.2.1Embodied Material

The quantity of material used to construct the shipdirectly affects the CO2e emissions that result, but this information proved difficult to obtain in the absence of documentation from the shipbuilder. Displacement is the literal weight of the ship, but this is seldom recorded: ships are registered by their Gross Tonnage, a synthetic and highly complex assessment of enclosed volume to which no unit of measure is assigned. The approximate displacement of some vessels is revealed in promotional literature, however: it is reported that the Queen Mary 2 weighed approximately 76,000 tons [Wagner, 2006], while the Oasis-class ships weigh approximately 100,000 tons [Schnepf, 2010] – 68,946 and 90,718 metric tonnes, respectively.Since the Queen Mary 2 is a liner, not a cruise ship, her construction differs and displacement per passenger is higher: thus the figure for an Oasis-class ship was preferred. At 90,718 tonnes and carrying up to 6,296 passengers, this is 14.4 tonnes per passenger berth. If anything, this establishes a lower limit for the embodied material per passenger berth because it is unlikely that all cabinswill be at full occupancy. Also,a modern ship such as the Oasiswill tend to be more efficient in terms of its construction.Table 1 illustrates the industry trend towards larger vessels, while Figure 1 is included in order to show just how large a cruise ship can be.

Table 1: Largest Passenger Ships at the Time of Launch

Vessel / Year / Maximum passengers / Tonnage (GT)
Carnival Sunshine / 1996 / 3,400 / 101,353
Grand Princess / 1998 / 3,199 / 109,000
Carnival Triumph / 1999 / 3,470 / 101,509
Carnival Conquest / 2002 / 3,700 / 110,000
Mariner of the Seas / 2003 / 3,807 / 138,279
Queen Mary 2 / 2004 / 3,090 / 148,528
Crown Princess / 2006 / 3,858 / 113,000
Freedom of the Seas / 2009 / 4,375 / 154,407
Norwegian Epic / 2010 / 5,183 / 155,873
Oasis of the Seas / 2010 / 6,296 / 225,282

Figure 1: Size of anOasis-class Cruise Ship

In the model, both displacement and maximum passengers can be entered for the MS Exemplar, a warning being displayed if the resulting embodied material per passenger is implausible. The mass of the vessel is then apportioned between a simplified set of five materials; aluminium, glass, wood, paint and steel. Each has consequences for CO2e, as given in the DEFRA [2012] Conversion Factors: for example, the production of steel involves 3,100kg CO2e per tonne, while aluminium is 11,000kg CO2e per tonne. Based upon this information, and an assumption regardingthe breakdown of materials ina cruise ship (Figure 2), it is estimated that the contribution to climate change from the materials embodied by an Oasis-class ship is around 370,500 tonnes CO2e.

Figure 2: Detail from the Excel Spreadsheet

2.2.2 Construction Process

After materials have been obtained, manufacturing operations must be performed to produce thecompleted vessel, and each process has consequences for CO2e.The US Environmental Protection Agency list the energy-intensive processes in shipbuilding and ship repair as welding (particularly arc welding), forging, abrasive blasting and the application of coatings, with electricity representing 75-80% of energy costs [EPA, 2007]. This suggests that a shipyard’s stated emissions might be skewed, withemissions from generation occurring elsewhere in the supply network. The EPA have created a Shipyard Greenhouse Gas Emissions Inventory Tool [EPA, 2014] that ought tohelp a shipbuilder to measure and ultimately reduce emissions, but its use requires access to detailed information on the vessels under construction and the operations being performed.

It proved impossible to obtain any information on the sequence of shipbuilding processes employed, so a more simplistic approach had to be employed. The DEFRA [2012] Conversion Factors give a blanket figure for “other transport equipment” including ships at 0.76 kg/CO2e per pound spent; Nugent [2009] and others have reported the cost of Oasis of the Seas as being $1.4bn.Using an approximate historical exchange rate of £1 = $1.80forearly 2006, when the Oasis was ordered, ship construction equates to a little over 591,000 tonnes of CO2e. This admittedly crude estimate was the best that could be produced in the absence of detailed information on the total energy bill incurred when building, launching and fitting out a large cruise ship.(There is provision within the model for refinements to be incorporated as and when the information becomes available; this is information that the naval architect, shipbuilder or cruise line would already have.)

Obviously, the CO2e for embodied material (see Section 2.2.1) is not added to this figure, as that wouldmean double-counting the materials. Those calculations remain a part of the process because it is hoped that in the future, better information about the construction process will allow an itemised breakdown to be produced. Also, it facilitates calculations relating to the end-of-life (see Section 2.2.4).

2.2.3Vessel Operation

In addition to that arising from construction, the day-to-day operation of a cruise ship involves a contribution to climate change, with the most significant input being fuel consumption. Modern cruise ships make their own fresh water, but food and drink are another input with carbon consequences. (The authors have considered such consumables to be analogous to those in a five-star hotel.)

The fuel is typically heavy fuel oil (HFO), used to generate the power that moves the ship and runsthe on-board systems of which air conditioning is the most energy-intensive. Firm figures for consumption are a little difficult to derive because the industry measures fuel usage in tons, while information meant for the public is often expressed in gallons (imperial or US). Furthermore, some ships are equipped to use marine grade diesel or liquefied natural gas (LNG) – or a dual-fuel combination. Some will change the fuel used when in port, and some cruise lines practice “cold ironing” – getting electricity from a land-based supply while docked.

Much of the fuel usage information in the public domain is anecdotal, or incomplete. Alpeche [n.d.] reviewed this, suggesting a typical consumption of 140–150 tons (127 – 136 tonnes) of HFO per day. (Industry commentators often use a passenger capacity of around 3,000 as a typical figure.)Freedom of the Seas,a relatively large but more efficient ship, is reported to consume 2,800 gallons (9.1 tonnes) of fuel per hour when in motion [Alpeche, n.d.], which would seem to support this fuel consumption estimate.DEFRA [2012] uses a figure of 3766.5 kg CO2e per tonnefor HFO consumption, so the daily use of 127 tonnes of fuel would equate to 478.3 tonnes of CO2e emitted per day. If we assume this is shared between 3,000 passengers, this means each passenger is responsible for the emission of 1,116 kg CO2e in the course of a week-long cruise.This exceeds Ward’s [2010] figure of 960kg CO2 emitted per passenger, but the difference can be accounted for: if we confine ourselves to CO2 (not all greenhouse gases as underCO2e) and look only at direct emissions (the burning of the HFO, not includingassociated harm from its extraction, refining and transportation) the DEFRA [2012] standard is 3205.5 kg CO2emitted per tonne used, which makes a passenger’sfuel-only contribution to climate change 950 kg CO2 per passenger per week – a close match to the Ward [2010] figure.

2.2.4 End-of-Life

At the end of a product’s useful life, additional harm may result – or additional benefit may be derived, for example through reuse or recycling. Barring vast improvements in efficiency, longer-lived products are generally more ‘green’, and ships perform well in this regard with a lifespan that can be measured in decades.

Ultimately, a ship will typically be broken up, such that any valuable materials can be reclaimed. This process of reclamation can introduce additional environmental problems of contamination and perhaps health hazards from materials such as asbestos, but these do not impact directly upon climate change, and so are not addressed in this study. (Taylan [2013] describes ship breaking in considerable detail.)

The expected useful lifeof the MS Exemplar is specified as a variable within the Excel model, as is the proportion of materials recovered. The DEFRA [2012] Conversion Factors detail savings for recycling including 1,300 kg CO2e for ferrous and9,000 kg CO2e for nonferrous metals, and these have been incorporated into the model, significantly reducing environmental impact from embodied material.

2.3 Calculations Performed in Summary

The model described addresses four phases of a product lifecycle; sourcing raw materials, construction, the use phase, and the end-of-life. The CO2e associated with materials and construction, and the ‘discount’ received by recycling at the end-of-life, are one-off entries, and are apportioned over the entire life of the cruise ship. Thus, if MS Exemplar(a ship of 50,000 tonnes displacement, carrying 3,000 passengers on average) operates a week-long cruise, fifty weeks a year for twenty-five years, the contribution to climate change from the ship itself must be divided by 3,750,000 (3000 x 50 x 25) to find each passenger’s personal share. Naturally, this amortisation across the whole lifespan makes the impact from the ship itself relatively small, at something like 63.5kg CO2e per passenger per week – assuming a 95% recycling rate for the vessel’s major materials.

To this is added the personal share of CO2e resulting from ship operations. The major element is that of fuel usage, suggested in Section 2.2.3 to be 1,116 kg CO2e per week; other necessary additions to arrive at a final figure include the contribution to climate change from hotel operations and services performed on board, the CO2e footprint resulting from transport to and from the cruise ship, and that of any excursions undertaken. These are reviewed in Section 3.

3. Model Elements Derived from Land-Based Holidays

The land-based equivalent contribution to climate change for a hotel guest is not merely useful for the purposes of comparison; it provides information that can be used to complete the cruise model, since the harm resulting from the provision of food and drink, etc. are broadly similar for both types of holiday.

A working group was established in 2011 to enhance understanding of the impacts of the hotel industry.Underpinned by the Green House Gas Protocol Standards [World Resources Institute, 2011], the Hotel Carbon Measurement Initiative (HCMI) [WTTC, 2012] was piloted in a variety of different styles and sizes of hotel in several destinations. It was designed to be applicable to any accommodation in any location, and in many ways it matches the parameters of a stay in a cruise ship cabin.

Several of the large hotel chains have gathered information on energy consumption, with the aim of informing their sustainability strategies [Jones et al, 2014]. Some of those hotels employed the HCMItool for this purpose – including the market-leading Accor Group, selected for a comparison between cruising and a land-based holiday. The Pullman Barcelona Skipper hotel, situated in Barcelona, Spain is described as a high-class contemporary style hotel, comprising 241 rooms with two pools, a gym and spa, plus meetings rooms capable of providing space for up to 800 delegates.This type of land-based hotel and its level of facilities equate, in terms of the star rating, to a cruise ship of the kind typified by the MS Exemplar, matching Mintel’s [2014] classification of a cruise ship as hotel-at-sea.