Transportation Research Part D 36 (2015) 10–17

Contents lists available at ScienceDirect

Transportation Research Part D

journal homepage: www.elsevier.com/locate/trd

Evaluating the social cost of cruise ships air emissions in major

ports of Greece
A. Maragkogianni, S. Papaefthimiou ⇑
Technical University of Crete, School of Production Engineering & Management, 73100 Chania, Greece

a r t i c l e i n f o a b s t r a c t

Article history: Detailed NOx, SO2 and PM2.5 emissions have been estimated for cruise ships in the five busi-
Available online 3 March 2015 est Greek ports (i.e. Piraeus, Santorini, Mykonos, Corfu and Katakolo) for year 2013. The
emissions were analyzed in terms of gas species, seasonality and activity. The total in-port
Keywords: inventory of cruise shipping accounted to 2742.7 tons: with NOx being dominant
Greek ports (1887.5 tons), followed by SO2 and PM2.5 (760.9 and 94.3 tons respectively). Emissions
Cruise ships during hotelling corresponded to 88.5% of total and have significantly outweighed those
Air pollution
produced during ships’ maneuvering activities (11.5% of total). Seasonality was found to
Social cost
play a major role, as summer emissions and associated impacts were significantly
augmented. The anticipated health impacts of ship emissions can reach to €24.3 million
or to €5.3 per passenger proving the necessity of control of the emissions produced by
cruise ships in port cities or policy and measures towards a more efficient cruise industry.
Ó 2015 Elsevier Ltd. All rights reserved.

Introduction

The future of tourism development is largely dependent on the natural environment and its preservation (Hall and Lew,
1998). Thus environment is not only an important foundation for tourism sustainable development, but it can also be the
foundation for unique attractions for tourists (Zi, 2015). Tourism by its very nature is a resource dependent industry and
some commentators argue that sustainable tourism is unachievable given the industry’s ability to pollute and consume
resources (Johnson, 2002). This view has been summarized as follows: ‘‘Tourism contains the seed of its own destruction;
tourism can kill tourism, destroying the very environmental attractions which visitors come to a location to experience’’
(Glasson et al., 1995).
Air pollution can be one of the major impacts tourism can impose, causing serious health problems including lung cancer,
cardiovascular disease, and birth defects. Particulate matter (PM) emissions from marine vessels are related to increased car-
diovascular hospitalizations and have been estimated to be responsible for about 60,000 annual cardiopulmonary and lung
cancer deaths mostly along European, East Asian, and South Asian coastal areas (Corbett et al., 2007; Tian et al., 2013).
Because of the proximity of ports to urban areas their operations can influence human health, inducing serious health
problems such as premature mortality, asthma, bronchitis and heart failure symptoms (IAPH, 2007).
Cruise tourism has experienced rapid growth in recent years, and has become one of the most dynamic and fastest grow-
ing segments of touristic sector (Sun et al., 2011). From 2003 to 2013 demand for cruising worldwide has increased from 12.0
to 21.3 million passengers (+77%). Over a similar period, global land-based tourism has risen by around 57% to an estimated

⇑ Corresponding author.
E-mail address: spapaefthimiou@dpem.tuc.gr (S. Papaefthimiou).

http://dx.doi.org/10.1016/j.trd.2015.02.014
1361-9209/Ó 2015 Elsevier Ltd. All rights reserved.
 A. Maragkogianni, S. Papaefthimiou / Transportation Research Part D 36 (2015) 10–17 11

1.087 billion tourists in 2013, 5.0% up of 2012. Globally in 2014, 410 cruise ships (including river cruise) operated, totaling
21.6 million passengers and contributing almost $37.1 billion in revenue. In 2013 in Europe, the vast majority of cruise ships
visited ports in the Mediterranean and the Baltic, generating 31.2 million passenger visits at a total of around 250 European
port cities, an 8.7% increase over 2012. Due to cruise activities direct expenditure of €16.2 billion was generated and it was
the first annual increase following three consecutive years in which these expenditures has been decreased (CLIA, 2014). The
Mediterranean is the world’s second largest cruise shipping market: it represented 21.7% of the annual cruise capacity for
2013 while the anticipated value for 2014 is 18.9% (CLIA, 2014; MedCruise, 2014). In 2013 a total of 166 cruise ships were
active in Mediterranean waters, with a capacity of 220,352 beds and an average of 1327 beds per ship (CLIA, 2014;
MedCruise, 2014).
Cruise ships constitute one of the most energy intense forms of touristic activities (Eijgelaar et al., 2010). This is due to the
fact that cruise ships act as luxurious resort hotels throughout their journeys and this so called ‘‘hotelling’’ function is mainly
responsible for the excessive energy demand. On the other hand, significant proportion of the total energy spent is used for
the onboard activities of the crew, thus being a part of the operating cost of the vessel and significantly increasing the
emissions per passenger. IMO has estimated that the global fuel consumption in 2007, for passenger ferries and cruise ships
was 31.3 million tons (Mt), resulting to a production of 96 Mt of carbon dioxide (Buhaug et al., 2009). There has been
extremely limited research to date for calculating emissions and creating relevant inventories for individual sectors of the
maritime transport industry, such as cruise ships. Howitt et al. used data for 84 cruise ships moving in journeys to and from
New Zealand and calculated carbon emissions per passenger-kilometer (p-km), confirming that cruises emit significantly
more carbon emissions and use more fuel per p-km than economy class aviation. The operation of a cruise ship (mainly
due to the ‘‘hotelling’’ amenities included) is still about five times higher than the average energy use for the most luxurious
of hotels per visitor night, which would include many of the same comforts, such as swimming pools, casinos, gyms and
restaurants (Howitt et al., 2010).
Ports play an important role being the vital links of land and sea, acting as gateways and linking transport corridors, thus
enhancing trade and communication. The environmental effects of ports to the atmosphere and human health (due to their
proximity in densely populated areas) are extremely important and these effects are typically assessed through measured
emissions of pollutants to air. Ports are influenced by maritime emissions contributing particularly to local air pollution
(Cooper, 2003; Isakson et al., 2001). In harbor cities, maritime activity creates a problem of great acuteness for urban pollu-
tion, causing environmental problems affecting both human health and ecosystems (Miola et al., 2009).
Cruise ships can be an important parameter for deteriorating air pollution in ports. This paper presents a comprehensive
quantitative case study for cruise ships emissions (NOx, SO2 and PM2.5) during vessels’ activities (moving–maneuvering and
hotelling) in the five busiest cruise ports of Greece (i.e. Piraeus, Santorini, Mykonos, Corfu and Katakolo) for year 2013. No
such detailed previous work on cruise ship emissions has been detected in the literature, apart from a similar inventory
which has been created only for the port of Piraeus for one year, i.e. mid 2008–mid 2009 (Tzannatos, 2010). Furthermore,
an evaluation of the total external cost due to estimated air emissions in port areas is presented. Thus, the current study pro-
vides a unique inventory allowing a detailed approach in addressing the issue of air pollution generated by cruise ships at
one of the busiest touristic areas in the Mediterranean and the world.

Cruise tourism in Greece

In 2013 Greece was the third most popular destination in Europe following Italy and Spain (keeping a 14.8% share of the
total cruise passengers), while the direct annual expenditures from the cruise industry for Greece was €574 million.
5,661,867 cruise passengers visited Greek ports, while for Italy and Spain the numbers were 6,970,000 and 5,236,000 respec-
tively (CLIA, 2014). It is estimated that 11,215 workers were employed in the Greek cruise industry in 2013 (SETE, 2014). Port
of Piraeus was ranked 4th in 2013 in respect of passengers numbers in the lists of leading European cruise ports and principal
home ports, while Santorini, Corfu and Mykonos were also included in the top-10 list. For the year 2013, 139 cruise ships
(83.7% of the total cruise ship fleet in the Mediterranean) visited 42 Greek ports (related to cruise sector) which handled
4288 cruise ship calls. The current study contains data for the five busiest Greek cruise ports, which received 2565 calls from
134 cruise ships that stayed more than 26,500 h in ports and moved almost 4.2 million passengers. These values represent a
share of 59.8% and 73.8% in ship calls and total passengers respectively, for the cruise industry in Greece during 2013 (HPA,
2014). In Table 1 and Figure detailed cruise statistics (year 2013) for the studied Greek ports are presented.

Table 1
Detailed cruise statistics for the studied Greek ports.

Ports Ships visiting Ship calls Revenue passengers

Piraeus 113 711 1,302,581
Santorini 62 582 778,057
Mykonos 67 485 587,501
Corfu 76 480 744,651
Katakolo 51 307 763,966
Total 369 2565 4,176,756
12 A. Maragkogianni, S. Papaefthimiou / Transportation Research Part D 36 (2015) 10–17

Fig. 1. Location of studied ports.

The studied Greek ports are Piraeus, Santorini, Mykonos, Corfu and Katakolo, and their location is depicted in Fig. 1. A
short description of these ports follows:
Piraeus is the largest port of Greece, and one of the largest in terms of passengers and freight in Europe. The port provides
services to passenger ships of any size and type including luxurious cruise vessels and high speed coastal ships. Piraeus port,
with more than 20 million passengers’ throughout every year, is an ideal hub for reaching the market center of Athens and
the famous Parthenon as well as multiple tourist attractions in the east Mediterranean including Santorini and Mykonos.
Santorini has two ports with a common anchorage. ‘‘Athinios’’ is the main port of the island for the transport of both
passengers and vehicles. In cooperation with the second port, ‘‘Fira’’, Athinios also ensures the proper disembarkation of
tourists from cruise ships. The bay of Fira is the old harbor of the island, which lies at the foot of its caldera slope. One of
the peculiarities that the port facilities of Santorini present is that cruise ships cannot anchor in a port infrastructure. For
this reason, the sea area of Fira bay has been set as port facility and it helps cruise ships to stay anchored. The current system
of four floating buoys which are anchored to 200 m depth, provides the mooring possibility. Santorini constitutes a nodal
terminal for most cruise lines around the Aegean and the wider Mediterranean region.
Mykonos has two ports in operation: the old port in the island’s down town and the new port (located 3.5 km north, in
Tourlos area) one of the largest infrastructure projects for the island in order to cover the increasing transportation needs.
Mykonos is one of the most popular cruise ship destinations in the Mediterranean today, while in 2013, 587,501 cruise pas-
sengers visited the island, 11.5% less than 2010.
Corfu is one of the most fascinating islands in Europe. Due to its strategic location (being in the entrance of the Adriatic
Sea between the mainland of Greece and Italy) it is regarded as a ‘‘must see’’ destination for cruise operators.
Katakolo is a seaside town in western Greece, situated on a headland overlooking the Ionian Sea. Due to its geographical
position and despite its small size, the port is a very popular stop for cruise ships, offering an opportunity for passengers to
visit the site of Olympia, homeplace of the Olympic Games in classical times.
In general, the busiest phase of cruise ship season in Greece extends from April to October (2378 calls or 93% of 2565 in
total). During this period, each week the majority of cruise ship visits occur between Thursday and Saturday. Furthermore, on
these days most arrivals are from 07:00 to 10:00 and most departures between 18:00 and 20:00. The average time that each
cruise ship spent in port per call varied: in Piraeus vessels stayed 12.4 h in average, while in Santorini and Mykonos this time
period was 8.8 and 11 h respectively. In ports of Corfu, cruise ships remained for 9.9 h, while in Katakolo for 7.2 h.
As regard traffic seasonality in the selected ports, passengers’ movement and ship calls reaches its peak in September,
with 704,076 passengers in total. It is worth mentioning that in case of Piraeus port (see Fig. 2) an extended touristic season
is observed with October being the busiest month (234,501 passengers). On the other hand, the cruise season for Santorini
ranges from March to end of September (during May and till end September the island is number one destination in Greece
in terms of arriving vessels) while in Mykonos from early May till end of November.
 A. Maragkogianni, S. Papaefthimiou / Transportation Research Part D 36 (2015) 10–17 13

Fig. 2. Seasonal distribution of passengers and number of ship calls in studied ports.

Emissions calculation methodology

The environmental effects of ports to the atmosphere and human health, due to their proximity in densely populated
areas, are extremely important. An independent evaluator finds it very hard to access and elaborate relevant emissions data,
as in most cases port authorities are not obliged to measure and publicize them. The induced costs of these emissions are
practically addressed primarily to the local society, which will have to pay, in due time, the consequences.
Generally, the existing approaches for creating ship emission inventories are divided in ‘‘top-down’’ and ‘‘bottom-up’’ (or
‘‘activity-based’’) approaches. The former are fuel-based methods that estimate emitted air pollutants relying on the report-
ed amounts or marine bunker fuel sales, while for the latter fuel consumption-based or ship movements-based methods are
employed. ‘‘Bottom-up’’ approaches would generally be more accurate than top-down (Miola and Ciuffo, 2011; Buhaug et al.,
2009). In the present study a ‘‘bottom-up’’ method has been used to estimate emissions based on detailed individual activ-
ities of cruise ships in selected ports. For each studied port and for all approaching cruise vessels, activity profiles have been
created; i.e. a breakdown of a ships’ movements during modes of operation (i.e. maneuvering or at berth), with engines’
types and sizes, engines’ load factors, type of fuel consumed and time spent in each mode. In this context, information
required were: rated power of both main and auxiliary engines of each ship, load factors on both types of engines, scheduled
arrival and departure times in order to estimate the amount of time spent in different operating modes and engines’ emis-
sions factors.
For every ship call, each of the air pollutants (i.e. NOx, SO2 and PM2.5) produced during the ship’s activity in the port was
estimated through the application of the following expression:
X
Ei ¼ ðT j  Pk  LF j;k  EF i;k Þ
j;k

where E denotes the amount of ship emissions (tons); i is the specific type of emissions (NOx, SO2 or PM2.5); j is the ship’s
activity stage (i.e. moving–maneuvering or hotelling); k is the engine type, i.e. main (ME) or auxiliary (AE); P is the engine
power (kW); LF is the engine load factor during the specific activity; EF is the emissions factor (g/kW h); T is the time spent at
each of the ship’s activity stages (hours) (for maneuvering T = D/U, where D is the distance traveled by the ship in the port
before docking, U is the moving velocity of the ship during moving–maneuvering. The total emissions are calculated for each
port by summing for all cruise ships visiting during the selected time period. All necessary data regarding cruise ship calls in
Greece during 2013, i.e. vessels’ names, date and call duration (arrival and departure time), were carefully collected from
local Port authorities and compared with similar data of other sources to harmonize any discrepancies (MedCruise, 2014;
HPA, 2014).
The IHS Sea-web online database was employed to obtain various technical characteristics and data on main (ME) and
auxiliary (AE) engines for all cruise ships (IHS, 2014). Extensive work on main and auxiliary engine load factors and emission
factors of main and auxiliary engines for cruise ships during maneuvering and while at berth has been provided elsewhere
(Buhaug et al., 2009; Howitt et al., 2010; Whall et al., 2007; Murphy et al., 2013). The load and emission factors applied in
this study for the operation of main and auxiliary engines running on specific fuels and load condition, for ships maneuvering
and hotelling during summer and the rest of the year were taken from a similar survey that has been conducted for the port
of Piraeus (Tzannatos, 2010).
The least possible uncertainty in all adopted values has been maintained during the estimation of the emissions inven-
tories. Cruise ships’ hotelling duration and technical characteristics were collected from official local Port authorities and
from IHS Sea-web database respectively. The distance traveled by each vessel in the port (for the calculation moving and
14 A. Maragkogianni, S. Papaefthimiou / Transportation Research Part D 36 (2015) 10–17

maneuvering times) was evaluated and a ‘‘generic’’ cruise ship path has been created and assigned to each studied port indi-
vidually. Thus the above mentioned parameters are considered to be as accurate as possible. The dominant uncertainties in
all ‘‘bottom-up’’ approaches are due to the determination of auxiliary engines’ power and the estimation of the average load
factors and emission factors of the main and auxiliary engines. In this study, and for almost 30% of the studied cruise vessels,
detailed and accurate data regarding their AE power rating were collected from the Sea-web database, while for the remain-
ing the typical auxiliary to propulsion power ratio for cruise ships (0.278) was employed. The employed load factors were
based on the most updated recent relevant studies and have taken into account the unique nature of cruise vessels, the ports
specific characteristics and local climatic conditions. Emission factors were also determined based on detailed vessel infor-
mation such as engine and fuel type, but they may also contain uncertainties.

Results and discussion

Emissions results

Emissions values (i.e. NOx, SO2 and PM2.5) for the studied Greek ports are presented in Fig. 3. Port of Piraeus leads in terms
of total emissions followed by Santorini (927 and 619 tons respectively). Emissions in Piraeus account for the one third of
total emissions, while the corresponding amount for Santorini is 23%. On the other hand, Mykonos and Corfu are very close
to each other, with 475 and 467 tons of emitted pollutants respectively. In all cases NOx emissions are dominant throughout
the year, followed by those of SO2 and PM2.5. On average the mass ratio of SO2 and PM2.5 to NOx was 40% and 5% respectively
and was kept almost constant throughout the year.
The total emissions per thousand passengers are also important as they depict the ratio of environmental burden to car-
ried passengers from arriving cruise ships. It can be of crucial importance for the induced health impacts in local inhabitants,
especially for ports with adjacent cities with high population density. In cases that passenger capacities of cruise vessels
approach their maximum, the emissions per passenger could possibly be comparable to the values of international aviation
flights. In cities as compact and densely populated as Piraeus, high doses of toxic pollutants emitted in close proximity to
people on a regular basis are imposing serious health threats to the people and substantial costs to society. For the studied
ports this ratio was between 0.33 (Katakolo) and 0.81 (Mykonos).
Fig. 4 summarizes the emissions from all studied ports under different operational modes. The estimated total emissions
during maneuvering were 316 tons, while 2426 tons of air pollutants were emitted during vessels’ hotelling (corresponding
to 88.5% of total). This is due to the fact that hotelling times for cruise ships are extended while they also operate their aux-
iliary engines at high loads throughout their stay at berth.
According to the annual inventory of Greece for green-house and other gases for the year 2013, national shipping
accounted for 30.2 and 24.6 thousand tons of NOx and SO2 respectively (MINENV, 2014; EEA, 2014). Therefore, cruise ship-
ping activity in the studied five ports contributed 6.2% and 3.1% respectively to the shipping national NOx and SO2 inventory.
For Piraeus, in 2007 the total amount of NOx and SO2 from cruise ships activity was 597 and 241 tons respectively
(Tzannatos, 2010), contributing by 0.4% and 0.83% in the total national inventory, respectively. In our study, 6 years later,
the contribution is quite similar, 0.49% for NOx and 0.74% for SO2. The amounts of three main exhaust pollutants for the stud-
ied ports would roughly constitute 0.94%, 0.72% and 0.67% of the total Mediterranean emissions from shipping, based on a
study for year 2005 in which NOx, SO2 and PM2.5 emissions during ships maneuvering or at berth were reported to be 201,
105 and 14 thousand tons, respectively (Whall et al., 2007).
Air emissions due to cruise ships traveling at sea (i.e. while approaching and leaving the studied ports) have also been
estimated, in order to evaluate their potential magnitude compared to in-port air pollution. An area of 5 nautical miles

Fig. 3. Total amount of emissions in the studied Greek ports.

 A. Maragkogianni, S. Papaefthimiou / Transportation Research Part D 36 (2015) 10–17 15

Fig. 4. Distribution of emissions according to maneuvering and hotelling activity.

extending outside their borders has been considered for each port (see red circles in Fig. 1) and the total emissions for all
visiting cruise ships throughout the year have been calculated. Typical values for cruise ships’ average speed and main
and auxiliary engines’ load factors while traveling at sea have been considered (Buhaug et al., 2009; Howitt et al., 2010).
The estimated total emissions to air were 295, 208, 147, 197 and 139 tons for Piraeus, Santorini, Mykonos, Corfu and
Katakolo respectively. Results indicate that the amount of these emissions compared to their in-port counterparts varies
for each port, depending mainly on the average time in port per ship call. The two extremes are Piraeus and Katakolo: in
the former vessels stay in average for 12.4 h and emissions in the 5 nautical miles area represent almost 32% of the total
in-port emissions, while for the latter ships stay for 7.2 h and the out-of-port emissions are 54% of total in-port emissions.
In any case emissions to air from cruise ships moving close to ports could have significant health impacts to adjacent areas
but can be strongly affected by local weather conditions (i.e. prevailing winds’ direction throughout year).
The seasonal distribution of ship emissions and average emissions per ship calls, are depicted in Fig. 5. During summer
period emissions of the three studied air pollutants peak (58.5% of the total), while in winter corresponds only a small
amount (0.9%). During autumn, although cruise ships operate their auxiliary engines at lower loads (due to lower ambient
temperature resulting to reduced air-conditioning needs), emissions stay high as calls by cruise ships maintain their summer
momentum. This pattern of seasonal variation of emissions, peaking during summer, is very important in terms of air quality
in port and adjacent cities areas. During summer, the augmented in-port presence of cruise ships inevitably leads to
increased shore-side demand for relevant services, traffic and congestion. Thus exhaust emissions of the various vehicles
(trucks, buses, private cars) operating in, out and around ports are added to the ships emissions, thus maximizing the detri-
mental impact upon the cities air quality. For the port of Piraeus, values of this study demonstrate an alteration in year dis-
tribution compared to a previous study, in which summer emissions were almost double than the equally placed autumn
and spring emissions (Tzannatos, 2010). For the year 2013, results show that Piraeus followed the general Greek ports emis-
sions distribution: summer, autumn, spring and winter emissions were 50.4%, 29.8%, 17.9% and 1.9% of total respectively. It is
obvious that a significant increase in cruise ships visiting port during autumn is observed, thus prolonging the touristic
season.

Fig. 5. Seasonal distribution of emissions for the studied ports.

16 A. Maragkogianni, S. Papaefthimiou / Transportation Research Part D 36 (2015) 10–17

Externalities in ports

For an accurate estimation of total external cost due to emissions to air in studied ports, results from two methodologies
have been used: Clean Air for Europe (CAFE) and New Energy Externalities Development for Sustainability (NEEDS). Both
CAFE and NEEDS model PM2.5 factors employing the Developing Harmonised European Approaches for Transport Costing
(HEATCO). Values are expressed as damages per ton of emission of PM, SO2 and NOx. It includes four different sensitivity
scenarios for each EU-25 Member States and surrounding seas. HEATCO, focuses on cost-benefit analysis proposing guide-
lines in order to value changes in travel time, accident risks and environmental costs. NEEDS is the most updated method-
ology, covering all major pollutants and all EU Member States. It includes all European sea territories (thus being appropriate
for correctly calculating the external costs of maritime transport related emissions), and quantifies not only health effects
(that correspond to over 90% of the total external effects) but also the side effects of emitted NOx and SO2 on materials
(i.e. buildings), biodiversity and crops. For maritime transport, in NEEDS, specific damage cost values for all major pollutants
have been calculated for all European sea regions using the EcoSense model (Korzhenevych et al., 2013; Holland et al., 2005).
The average external cost in million euro, has been estimated for each pollutant in studied ports and aggregated costs are
presented in Table 2, and their detailed variation in Fig. 6. The lowest estimates result from the application of CAFE (i.e.
€12.4 million) while in the case of NEEDS the anticipated total external cost reaches €24.3 million. The average cost for all
ports per cruise passenger is €5.3 and €2.5 for CAFE and NEEDS respectively. Our cost estimates can be compared to results
found in other studies. Tzannatos estimated for Piraeus in 2008–2009 the external costs from in-port cruise ships activity to
€16.5 million or €10.4 euro per cruise passenger, while in the current study the per passenger values are €9.1 and €6.0 for
CAFE and NEEDS respectively (Tzannatos, 2010). McArthur and Osland for the port of Bergen-Norway indicated a total social
cost between €10 and €21.5 million or ranging between €6.79 and €14.63 per cruise passenger (McArthur and Osland, 2013).
Berechman and Tseng found that the cost of emissions from ships at berth in Kaohsiung-Taiwan, was $119.2 million in 2010
(Berechman and Tseng, 2012), while Song estimated the total social cost of air emissions for the Yangshan port of Shanghai
equal to $287 million (Song, 2014).
The methodology for the estimation of external costs might lead to underestimation since the population daily exposed to
the ship-generated pollutants should also include daily visitors and employees which produce a significantly higher number
than that corresponding to permanent inhabitants. Furthermore, although the applied external cost factors are widely used
in many research studies, it is acknowledged that there is adequate margin for improvement through the broadening of the
range of incurred damages and the refinement of their impact evaluation. More specifically, it is important to note that the
applied external factors do not include structures of cultural value (such as antiquities), for which Greek regions (such as
Piraeus-Athens, Santorini and Katakolo) would have particular interest.

Table 2
Estimation of total external cost.

Ports NEEDS (million €) CAFE (million €)

Piraeus 11.88 7.88
Santorini 4.61 1.94
Mykonos 2.84 0.80
Corfu 3.35 1.34
Katakolo 1.57 0.46
Total 24.25 12.42

Fig. 6. Distribution of externalities (for both methodologies) and emissions in the studied Greek ports.
 A. Maragkogianni, S. Papaefthimiou / Transportation Research Part D 36 (2015) 10–17 17

Conclusions

An activity based approach has been employed to create NOx, SO2 and PM2.5 emissions inventory for cruise ships
approaching ports of Piraeus, Santorini, Mykonos, Corfu and Katakolo, for year 2013. The total in-port inventory of cruise
shipping accounted to 2742.7 tons: with NOx being dominant (1887.5 tons), followed by SO2 and PM2.5 (760.9 and 94.3 tons
respectively). Emissions during hotelling corresponded to 88.5% of total and have significantly outweighed those produced
during ships’ maneuvering activities (11.5% of total). Emissions during summer prevailed due to the augmented in-port pres-
ence of cruise ships. The obtained inventory based on cruise shipping activity within the five busiest Greek ports, contributed
6.2% and 3.1% respectively to the relevant national NOx and SO2 inventory, while it would roughly constitute 0.94%, 0.72%
and 0.67% of the total emissions from shipping for the three main exhaust pollutants in the Mediterranean. The anticipated
health impacts of ship emissions have been expressed in monetary values, ranging in total from €12.4 to €24.3 million (or
€2.5 to €5.3 per passenger). The future annual costs will depend largely on growth and efficiency of cruise industry and
the traffic in the ports.

References

Berechman, J., Tseng, P., 2012. Estimating the environmental costs of port related emissions: the case of Kaohsiung. Transp. Res. Part D 17 (1), 35–38.
Buhaug, Ø. et al, 2009. Second IMO GHG Study. International Maritime Organization (IMO), London, UK.
CLIA, 2014. The Cruise Industry: Contribution of Cruise Tourism to the Economies of Europe. Cruise Lines International Association (2014 Edition).
Cooper, D., 2003. Exhaust emissions from ships at berth. Atmos. Environ. 37 (27), 3817–3830.
Corbett, J. et al, 2007. Mortality from ship emissions:a global assessment. Environ. Sci. Technol. 41 (24), 8512–8518.
EEA, 2014. Annual European Union Greenhouse Gas Inventory 1990–2012 and Inventory Report 2014. Brussels: European Environment Agency, European
Commission, DG Climate Action; May 2014.
Eijgelaar, E., Thaper, C., Peeters, P., 2010. Antarctic cruise tourism: the paradoxes of ambassadorship, ‘‘last chance tourism’’ and GHG emissions. J. Sustain.
Tour. 18 (3), 337–354.
Glasson, J., Godfrey, K., Goodey, B., 1995. Towards Visitor Impact Management: Visitor Impacts, Carrying Capacity and Management Responses in Europe’s
Historic Towns and Cities (Urban and Regional Planning and Development). Avebury, Surrey, UK.
Hall, C., Lew, A., 1998. Sustainable Tourism: A Geographical Perspective, first ed. Prentice-Hall, London.
Holland, M. et al., 2005. Damages per tonne emission of PM2.5, NH3, SO2, NOx and VOCs from each EU25 Member State (excluding Cyprus) and Surrounding
Seas, for European Commission. AEA: Technology Environment.
Howitt, O.J.A., Revol, V.G., Smith, I.J., Rodger, C.J., 2010. Carbon emissions from international cruise ship passengers’ travel to and from New Zealand. Energy
Policy 38 (5), 2552–2560.
HPA, 2014. Greek cruise statistics 2012–2013, Hellenic Ports Association: 2014.
IAPH, 2007. Tool Box for port Clean Air Programs. Improving Air Quality While Promoting Business Development. International Association of Ports and
Harbors, Netherlands.
IHS, 2014. Sea-Web, www.sea-web.com: Database (assessed May–June 2014).
Isakson, J., Persson, T., Selin Lindgren, E., 2001. Identification and assessment of ship emissions and their effects in the harbour of Goteborg. Atmos. Environ.
35 (21), 3659–3666.
Johnson, D., 2002. Environmentally sustainable cruise tourism: a reality check. Marine Policy 26 (4), 261–270.
Korzhenevych, A. et al, 2013. Update of the Handbook on External Costs of Transport, for European Commission. Ricardo-AEA, London, UK.
McArthur, D., Osland, L., 2013. Ships in a city harbor: an economic valuation of atmospheric emissions. Transp. Res. Part D 21, 47–52.
MedCruise, 2014. Cruise activities in Medcruise Ports (Statistics 2013). MedCruise, Piraeus.
MINENV, 2014. Annual Inventory Submission under the Convention and the Kyoto Protocol for Greenhouse and Other Gases for the Years 1990–2012.
Ministry of Environment, Energy and Climate Change, Athens, April 2014.
Miola, A., Ciuffo, B., 2011. Estimating air emissions from ships: meta-analysis of modelling approaches and available data sources. Atmos. Environ. 45 (13),
2242–2251.
Miola, A. et al, 2009. External Cost of Transportation-Case Study: Maritime Transport. JRC, European Commission, Brussels.
Murphy, A., Landamore, M., Pazouki, K., Gibson, M., 2013. Modelling ship emission factors and emission indices. In: Low Carbon Shipping Conference,
London, 9–10 September 2013.
SETE, 2014. Association of Greek Tourism Enterprises, SETE (www.sete.gr). Athens.
Song, S., 2014. Ship emissions inventory, social cost and eco-efficiency in Shanghai Yangshan port. Atmos. Environ. 82, 288–297.
Sun, X., Jiao, Y., Tian, P., 2011. Marketing research and revenue optimization for the cruise industry: a concise review. Int. J. Hospit. Manage. 30 (3), 746–755.
Tian, L. et al, 2013. Shipping emissions associated with increased cardiovascular hospitalizations. Atmos. Environ. 74, 320–325.
Tzannatos, E., 2010. Ship emissions and their externalities for the port of Piraeus-Greece. Atmos. Environ. 44 (3), 400–407.
Whall, C. et al, 2007. Ship Emissions Inventory: Mediterranean Sea. Entec UK Limited. April 2007.
Zi, T., 2015. An integrated approach to evaluating the coupling coordination between tourism and the environment. Tour. Manage. 46, 11–19.