Quantitative Models Evaluating the Effect Climate Change

Effects on Tourism

State of the Art

Jaume Rosselló

Departament d’Economia Aplicada, Universitat de les Illes Balears,

Carretera Valldemossa km 7.5, Palma de Mallorca, Spain

Keywords: Climate Change, Climate Change Effects, Climatic Indices, Tourism Demand.

Abstract: In a context of climate change, many destinations are considering what effects can be predicted on the

tourist demand and how they should be tackled. This work analyses the most relevant perspectives

presented in the literature evaluating the effect of climate change on tourism. A review is made by showing

the results that rise from a triple point of view: the consideration of physical changes, the analysis of the

tourist attractiveness through climatic indexes and modeling tourism demand. The review suggests that,

although some methodologies are on a primary stage of development, results from the different perspectives

agree in presenting a similar map of the main affected areas (positively and negatively) in terms of tourism

demand and/or tourism attractiveness.

1 INTRODUCTION

Climate science is very certain that the Earth's

climate will change at an unprecedented rate over

the 21st century. Whether through the direct effects

of climate change, such as increased temperature, or

through ancillary effects such as sea-level rise, loss

of snow cover or impact on landscapes, the spatial

and temporal pattern of tourism demand can be

expected to adjust. Despite the apparent

overwhelming dependence of tourism on climatic

factors, it is surprising that, in spite of the economic

significance of the Travel and Tourism Industry

(9.1% of the Gross domestic product worldwide

according to the WTTC, 2012), the literature on the

sectorial implications of climate change has been

dominated by other sectors such as agriculture - with

a lower weight in economic terms (6.1% of the

Gross Domestic Product according to CIA, 2012) -

while tourism has been pushed into the background.

This circumstance is evidenced in the successive

reports of the Intergovernmental Panel on Climate

Change where tourism only recently appeared with

some strength during the last AR4 Report (IPCC,

2007).

An initial justification of the relative neglect of

tourism can be found in the uncertainties and

complexity of expected tourism demand responses.

In a recent study, Gössling et al. (2012) highlight the

complexity of understanding tourist perceptions and

reactions to the impacts of climate change in order to

anticipate the decline or increase of specific tourism

markets and seasonal shifts in tourism demand. They

argue that tourism is characterized by a large

adaptive capacity that has to be combined with other

uncertainties concerning the implementation of

future mitigation policies and its impacts on

transportation systems, the wide range of climate

change impacts on destinations, as well as broader

impacts on society and economic development.

However, the industry needs to anticipate today

the consequences of climate change on future

demand in order to plan new infrastructures

strategically and to detect business opportunities

efficiently. Despite the controversy over the

weaknesses of statistical models in predicting tourist

flows under scenarios of climate change (Gössling

and Hall, 2006; Bigano et al., 2006), results from

this literature should be contextualized under the

ceteris paribus clause entailing that all the

determinants of tourism demand remain constant,

except the climate, whose influence one wishes to

study. One of the problems arising from the

application of the clause is related to the

interpretation of results. Thus, while in the short run

the estimation of a tourism demand model and the

479

Rosselló J..

Quantitative Models Evaluating the Effect Climate Change Effects on Tourism - State of the Art.

DOI: 10.5220/0004591404790488

In Proceedings of the 3rd International Conference on Simulation and Modeling Methodologies, Technologies and Applications (MSCCEC-2013), pages

479-488

ISBN: 978-989-8565-69-3

 2013 SCITEPRESS (Science and Technology Publications, Lda.)

evaluation of one single shock can produce direct

forecasts of the tourism demand variable

(Papatheodorou, et al., 2010) when long run

determinants – such as climate – are evaluated,

forecasts have a different interpretation and have to

be taken as general projections rather than real

forecasts.

Hence, for instance, regarding the specific

market segment of winter tourism, although it is

almost impossible to provide information regarding

the change in preferences of potential tourists

visiting mountain resorts during next 50 years, it is

feasible to evaluate the physical consequences of

loss of snow coverage originated by climate change.

Extrapolation can be carried out through the

quantification of expected snowfall which implicitly

determines the availability of winter tourism

conditions. Otherwise, optimal conditions for

tourism can also be evaluated by assuming that a

certain level of favorable climate conditions is

required for main tourism activities. In this case, an

assessment of what tourists consider optimal

conditions have to be evaluated previously, in which

case the assessment of these future climatic

conditions is what later determines the loss or gain

of tourism attractiveness in terms of climate

conditions. Evidently, again, changes in climatic

preferences by people are difficult to anticipate and

only actual optimal climatic conditions can be

projected.

Within the tourism demand modeling context,

although tourism demand models had traditionally

disregarded the consideration of climatic

determinants in tourism demand modeling exercises

(Goh, 2012), since the turn of the century the

inclusion of climate variables (such as temperature,

precipitation and wind) within tourism demand

models have become more frequent. The evaluation

of climate on tourism demand is often taken as a

short-run determinant in the context of time series

models or as a push/pull factor of the destination

choice when both discrete choice models and

aggregated tourism models are considered. The

ceteris paribus clause here is clear since estimation

techniques usually entail the isolation of each of the

determinants possible in order to make simulations

concerning the effect of climatic conditions while

the rest of variables remain constant.

Therefore, a significant number of quantitative

studies have attempted to evaluate climate change

consequences on tourism over the last fifteen years.

So far, however, no attempt has been made to

comprehensively integrate these findings to reveal a

regular pattern, the establishment of which will

constitute general principles and cumulative

knowledge. Due to the multiple methodological

alternatives that have appeared during the last years

assessing the consequences of climate change on

tourism quantitatively, this papers aims to evaluate

them, showing how despite the different approaches

used in tackling the problem, significant agreements

in results can be found. What is more, as the distinct

alternatives are analyzed jointly for the first time, it

is possible to highlight some of the main advantages

and limitations for each one.

2 EVALUATION THROUGH

PHYSICAL CHANGES

One of the most direct consequences of climate

change is frequently illustrated by the losing of

snow-cover depth. Mountain resorts depend heavily

on tourism and snow-reliability is one of the key

elements of the offers made by winter tourism. The

financial viability of winter tourism, however,

depends on sufficient snow conditions. If climate

change occurs, the level of snow-reliability will rise

to higher altitudes over the next few decades.

Although today, adaptation strategies are

predominant in this tourism segment (e.g. artificial

snow production), to anticipate the consequences of

climate change on snow-cover depth becomes a key

element when adaptation strategies have to be

adopted. In this context, Breiling and Charamza

(1999) analyze, for all districts in Austria, the impact

of a 2°C change in temperature on seasonal snow-

cover depth. They estimate that these changes will

reduce ski season length and the usability of ski

facilities. Warming will have strong impact on low

altitude resorts, which the authors expect will

disappear first and the remaining resorts will become

more expensive. Similar studies have been carried

out for winter sports tourism in Scotland (Harrison

et al., 1999), Switzerland (Elsasser and Messerli,

2001), and Canada (Scott et al., 2006).

The methodological background of all these

studies remains in estimating accurately the amount

of precipitation that falls as snow and rain, the snow

accumulation, and the snowmelt. Historical

precipitation data can be analyzed for each alpine

resort to determine the temperatures that best predict

historical snowfall amounts. Then, using

temperature and precipitation projected data from

the different climate change scenarios (or assuming

some trend in temperatures and precipitations) it is

possible to estimate the amount of snow

SIMULTECH2013-3rdInternationalConferenceonSimulationandModelingMethodologies,Technologiesand

Applications

480

precipitation that in the meantime determines snow

accumulation. Breiling and Charamza (1999),

Harrison et al. (1999) Elsasser and Messerli, (2001),

Scott et al., (2006) and other similar studies find a

general decline in natural skiing conditions, although

this will be less of a problem at high altitude sites.

At this point it should be mentioned how the use of

snow making machines, should be considered in the

final evaluation, even though this partial solution is

also temperature dependent.

At any rate, although the specific segment of

winter tourism may be of special interest for certain

regions, within the physical evaluation perspective it

is possible to be more generalist and consider other

physical consequences which, under climate change,

could affect other segments including summer &

beach tourism. Then , in the same way that winter

tourism depends on snow, summer tourism in some

destinations rely on other physical conditions

necessary for the development of tourism activities.

Although the models under this wide perspective are

at a lower stage of development, the rise in sea level

and its consequences on beach coverage (Nicholls et

al., 2011), coral reef health (Hoegh-Guldberg, et al.

2007), proliferation of jellyfish (Purcell, 2012) and

algal blooms (Englebert et al., 2008) would be on

the agenda of future quantifications of the effects of

climate change on tourism.

In this context it is important to highlight that

although some exploratory studies have shown the

relative irrelevance of tourism opinions when faced

with a marginal loss of environmental quality

(Gössling, et al. 2006), a certain threshold level -

defined sometimes by visibility, abundance and

variety of species, occurrence of algae or physically

disappeared beaches- would exist (Gössling, et al.

2007), evidencing how tourists might respond to

climate changes in a non-linear way. Then, although

a marginal effect on environmental quality can not

be detected through simple techniques, it is

suggested that ecosystem responses to pressures are

characterized by discontinuities and thresholds

effects, resulting in difficulties for the accurate

estimation of the tourism consequences.

3 CLIMATIC INDEXES

A Climatic index refers to a set of climate variables

that are combined through a mathematical formula

in order to capture human comfort preferences.

More precisely, a tourism climatic index (TCI) can

be understood as a tool that “has evolved from more

general knowledge about the influence of climatic

conditions on the physical wellbeing of humans”

(Amelung and Viner, 2006; p.351). The potential

changes in human comfort levels suggested by

combination of the TCI with scenarios of climate

change could have profound implications for the

tourism industry. Whereas some locations are likely

to experience substantial increases in attractiveness

due to improvements in their weather conditions,

others may become significantly less appealing to

tourists, leading to shifts in the temporal patterns of

visitation and/or actual declines in the number of

visits

Mieczkowski (1985) was among the first to

apply the results of climate indices for tourism-

related activities by developing a TCI. Although 12

climate variables were initially identified from the

literature as pertinent to be included in the TCI,

meteorological data limitations reduced the number

of climate variables integrated into the TCI to seven

which were combined in five sub-indices that

comprised the most popular TCI. A standardized

rating system, ranging from 5 (optional) to -1

(extremely unfavorable), was devised to provide a

common basis of measurement for each of the sub-

indices. Then, analytically the TCI can be derived

from the following equation:

TCI=2(4Cid+Cia+2·Sun+2·Prec+Wind)

(1)

Where, Cid is the daytime thermal comfort index;

Cia is the daily thermal comfort index; Sun is an

index of the amount of sunshine; Prec is an index of

the amount of precipitation; and Wind is the index of

the appreciation of wind. Bearing in mind that TCI

scores range from -20 to 100, Mieczkowski

proposed a classification of TCI scores, with values

in excess of 60 corresponding to ‘good' conditions,

scores exceeding 70 expressing ‘very good' climatic

conditions, levels of over 80 corresponding to

‘excellent' conditions, and scores of 90 or more

standing for ‘ideal' circumstances.

Although the Mieczkowski index was not

originally devised to explore the impacts of climate

change on tourism, since climatologists can provide

future data about the five sub-indices it is possible to

evaluate the climatic attractiveness of regions under

different climate change scenarios and to compare

them with the current situation. This has been

carried out for Europe (Rotmans, et al. 1994),

European beaches (Moreno and Amelung, 2009),

Mediterranean countries (Amelung and Viner,

2006), North America (Scott et al., 2004) and even

on a worldwide scale (Amelung et al., 2007). It

should be noted how maps depicting TCI show a

strong correlation with currently popular

destinations, suggesting that the index performs

quite well as a predictor of tourist arrivals.

QuantitativeModelsEvaluatingtheEffectClimateChangeEffectsonTourism-StateoftheArt

481

Weights used in Equation (1) give the highest

weight to the daytime comfort index to reflect the

fact that tourists are generally most active during the

day. The amount of sunshine and the amount of

precipitation are given the second-highest weights,

followed by daily thermal comfort and wind speed.

Although these weights are ultimately subjective,

they are founded on scientific knowledge owing to

the fact that they are based on the declared

preferences of tourists collected through

questionnaires. Consequently they can be adapted

for different tourist market segments. This is the

case of Morgan, Gatell, Junyent, Micallef, Özhan

and Williams (2009) who devised a specific climate

index for beach tourism that contained the same

elements as Mieczkowski’s (1985) TCI except for

the daily thermal component. The main difference

between the two indices is in the rating and

weighting schemes. While Mieczkowski (1985)

based his schemes heavily on expert judgment,

Morgan et al., (2000) based theirs on the stated

preferences of actual beach users. These preferences

were elicited from 1,354 questionnaires, filled out by

a total of north European beach users while spending

their holidays in Wales, Malta, and Turkey in 1994

and 1995.

One of the main goals of the use of climatic

indexes can be found in the analysis of seasonality.

Hence, as sub-indices composing TCI are referred to

monthly data, implications of projected climate

change on tourism seasonality can be revealed. In

this way, Amelung et al. (2007) shows how

countries in northern Europe may experience

substantial improvements in summer climatic

conditions while countries in the northern

Mediterranean that currently attract the traditional

“sun and sand” summer vacationer are likely to

become too hot for comfort in the current summer

season, which would contrast with the improvement

of climatic conditions during the non-summer

period. Thus, while northern countries are expected

to improve their tourism attractiveness under climate

change, the consequences for actual warmer

destinations remain uncertain owing to the final

balance between a loss of attractiveness during the

summer and a gain of attractiveness during the rest

of the year.

The Physiologically Equivalent Temperature has

recently emerged as alternative to the TCI.

Originally developed to assess human comfort in

general (Matzarakis et al., 1999), it has been applied

to tourism comfort by Lin and Matzarakis (2011)

who depict seasonal distribution maps of the

physiologically equivalent temperature showing that

Taiwan and Eastern China are perceived as

comfortable during spring and autumn for those

residing in temperate regions, while only the

southern region during spring and the northern

region during summer are perceived as comfortable

for those residing in sub-tropical regions.

An advantage of thermal indices is that they are

rooted in the long tradition of physiological

research; a major drawback is that they disregard

important non-thermal aspects of weather and

climate. For a proper assessment of the suitability of

climate and weather conditions for tourism purposes,

the use of composite measures is to be preferred

(Moreno and Amelung, 2009). At any rate, the main

drawback of the use of TCIs is the inability to

provide a quantitative measure of tourism impact in

economic terms or in tourism arrivals. This explains

why Amelung and Moreno (2012) recently included

the TCI as an independent variable in a tourism

demand model estimating the number of bed nights

for European countries. However, this perspective,

based on revealed preferences, brings us to the

following section.

4 TOURISM DEMAND MODELS

The neglect of tourism within the literature on

climate change came with the omission of climate

variables within tourism demand models since the

turn of the century (Goh, 2012). In the revision of

Crouch (1994), only few papers had included

climate or weather variables as determining

variables, and, on many occasions, with limited

success. A feasible explanation for this omission

would be related to the interest of researchers and

planners in income elasticities and/or price

elasticities in order to forecast tourism demand in an

accurate way - a key issue for service industries with

relatively high fixed costs – or, alternatively, in

order to evaluate the consequences of taxes or

exchange rate policies. This explains why tourism

demand literature has been dominated by time series

models and frequently linked to forecasting issues

(Song and Li, 2008). Thus, as climate is a relatively

stable variable, the climate factor does not have the

required variability and, additionally, is not

correlated with the determining variable, so no bias

in estimated elasticities is expected.

However, with increasing interest in climate

issues and, more precisely, in evaluating the

consequences of climate change on tourism, part of

the literature on tourism demand modeling has been

reoriented in order to integrate climate and weather

factors in the estimation of tourism demand. Hence,

consumers are assumed to be revealing their climate

preferences through purchasing habits. Knowing

their climate preferences and, ceteris paribus, it is

SIMULTECH2013-3rdInternationalConferenceonSimulationandModelingMethodologies,Technologiesand

Applications

482

possible to project future tendencies through

projected climatic conditions. Note how within this

framework the focus of the analysis is the tourists

not the physical conditions. Within the tourism

demand framework, three main perspectives have

been adopted: time series analysis, discrete choice

models, and aggregated tourism models.

4.1 Time Series Analysis

In the context of time series analysis, it is preferable

to talk about weather (the short term atmospheric

conditions) rather than climate (the mean

atmospheric conditions of a region) because under

this perspective the most popular proposal is to

attempt to capture some kind of short term

relationship between tourism demand and an

extreme weather event. Thus, Subak, et al. (2000)

assess impacts on tourism of the anomalously hot

summer of 1995 and the warm period from

November 1994 through October 1995 in the U.K

finding how the tourism sector showed clear

differences in its response to winter and summer

warm anomalies. Agnew and Palutikof (2006) look

into the sensitivity of UK tourism to weather

conditions using monthly data for domestic tourism

and annual data for trips abroad, showing that

outbound flows of tourists are responsive to weather

variability of the preceding year, whereas domestic

tourism is responsive to variability within the year of

the trip. Using the anomalously warm year of 1995

in the UK, the potential impact of climate change is

evaluated suggesting that the generally warmer and

drier conditions of 1995 benefited the UK domestic

tourist industry but wetter and duller-than-average

conditions in the previous year seemed to encourage

more trips abroad.

In a more general framework, it is proposed that,

using a monthly time series model, on the one hand,

the cyclical-trend component can be captured

through an ARIMA model (Rosselló et al., 2011) or

even including prices, and other economic

determining variables (Álvarez and Rosselló, 2010;

Rosselló, 2011). On the other hand, because

meteorological variables can present a high

variability and are not present in the long-run, it is

hypothesized that affect the short run of the time

series and consequently can not be captured though

ARIMA or Economic factors, and remain in the

error term. Then, the hypothesis to be tested is

whether short term extreme weather episodes are

related to this residual term. Analytically, the

problem can be summarized in terms of a Transfer

Function Model:

  

tbtptp

dLaLYL









(2)

where Y

is the number of tourism flows a month t; a

is the innovation or moving average term; d

is a

weather variable (or a set of weather variables) that

could influence the number of tourism flows;







and







are the lag operator polynomials for both

and a

, respectively capturing the cyclical-trend

component (the long-term component) of Y

as is

common practice in ARIMA modeling; and







are the lag operator polynomial (or transfer function)

for the weather determining variables, thus assuming

that some lag between the observation of weather

variables and tourist flow data occurs.

The estimation of equation (2) makes the

prediction of within-sample values possible, which

can be compared with simulated predictions using

scenarios of climate change thus affecting the d

variables. For instance, Rosselló et al. (2011) using

the transfer function methodology found a

significant relationship between British tourists

abroad and different British weather variables such

as temperature, heat waves, air frost days and

sunshine duration. Using different simulations in the

context of average temperature warming they found

that an additional 1◦C to the UK average

temperature will provoke an annual decrease of

1.73% of British outbound flows, a percentage that

is not homogeneous throughout the year because of

the expected higher impact during the winter time.

This result suggests, again, the presence of non-

linear relationships between temperatures and tourist

flows.

The use of more complex structures has

dominated the most recent literature. Then,

Kulendran and Dwyer (2012) use the

Autoregressive Conditional Heteroskedasticity

modeling approach for identifying the relationship

between climate variables such as maximum

temperature, relative humidity, and hours of

sunshine and seasonal variation, defined as the

repetitive and predictable movement around the

trend line in holiday tourism demand in the context

of seasonal variation in holiday tourism demand to

Australia from the US, UK, Japan, and New

Zealand. Otero-Giráldez et al. (2012) found also a

significant positive connection between the North

Atlantic Oscillation –as a meteorological indicator-

and tourism demand in Galicia (Spain) using

autoregressive distributed lag model. Goh (2012)

built an error correction tourism model for tourism

demand which also looks into the presence of

structural changes in the estimations using the TCI

as a determining variable, showing how the climatic

index is found to have a significant and positive

relationship for all the tourism demand series

analyzed.

Summarizing, the analysis of the relationship

between climate (or weather) and tourism through

QuantitativeModelsEvaluatingtheEffectClimateChangeEffectsonTourism-StateoftheArt

483

time series models has evidenced a current, real

relationship between tourism and climatology. Like

TCI, one of the main goals of the use of time series

can be found in the analysis of seasonality.

However, it should be noted how, within this

framework, only short term relationships are suitable

to be captured despite climate change issues belong

to the long term.

4.2 Discrete Choice Modeling

Within the framework of revealed preferences, a

second perspective is the use of Discrete Choice

Models. In this context the relevant question is why

people choose a particular destination, and the

theoretical background is found in Lancaster (1966)

where the source of utility is proposed to be the

characteristics of the commodities and services not

the commodities or services by themselves. Then,

taking the utility theory into account within the

context of tourism decisions, as formally described

for the first time in Morley (1992), a new framework

is introduced that allows different perspectives of

tourism decisions and a larger set of explanatory

variables to be considered. Analytically, the utility

that a tourist n derives from choosing to visit

destination i is assumed to take the following form:

nininni







(3)

where β

’x

is the deterministic portion of the utility

received if destination i is visited. Therefore, x

are

the observed attributes characterizing the

alternatives available to tourists and β

is the vector

of estimated coefficients for tourist n representing

his/her tastes. Finally, the error term ε

captures the

variation in preferences between tourists in the

population. As the individual is assumed to visit the

destination yielding the greatest utility, the

probability π

of him choosing the i-th alternative is:





ijxx

njnjnnininni







(4)

Thus, individuals or households with exactly the

same socioeconomic and demographic

characteristics might choose very different

destinations. However, over and above the

consideration of the utility theory, through the use of

random utility models it is generally recognized that

tourists have different tastes and that choosing a

final destination is not an independent decision, but

the final decision of a set of choices. In this sense, it

is argued that once tourists have decided to go on

holiday and have established a budget and mode of

transport, they choose a destination conditional upon

their preferences and the attributes characterizing the

alternatives in the choice set (Eugenio-Martin,

2003).

This framework for modeling tourism demand

from a microeconomic perspective has become of

interest to different tourism stakeholders, such as

tourism marketing analysts, because of the high

potential for identifying the determinants of

destination choice decisions. It is important to

highlight that choosing a destination is considered to

be one of the most complex stages in the decision

process by tourists, with a wide number of variables

(dependent on the aim of the study) that are likely to

influence such decisions (Marcussen, 2011).

In the context of climate change and tourism,

Eugenio-Martín and Campos-Soria (2010), using a

Discrete Choice Model for European households,

focus their analysis on the relationship between

climate in the home area and the choice of taking

holidays in the region of origin or abroad, showing

that the climate in the region of residence is a strong

determinant of holiday destination choice. They

show that residents in regions with better climate

indices have a higher probability of travelling

domestically and a lower probability of travelling

abroad while colder regions trend to travel abroad

with a higher frequency than warmer ones.

Thanks to the estimation of the β

vector of

equation (4) they project how individuals’ choice

changes when input climatic data from Europe

change, evaluating the probability of travel abroad

and/or domestically. Thus, under a scenario of

climate change, a not very strong relationship is

found when the evaluation of both travelling abroad

and travelling domestically is considered. However,

it seems that a rise in temperature increases the

likelihood of traveling domestically only and lowers

the likelihood of traveling abroad, a result that was

also obtained within the time series framework.

Using the same methodology, Bujosa and

Rosselló (2013) investigate the impact of climate

change on destination choice decisions in the context

of summer domestic coastal tourism in Spain. Once

destinations are characterized in terms of travel cost

and coastal ‘attractors’ (temperature and beach-

related attributes) the observed pattern of

interprovincial domestic trips is modeled, showing

trade-offs between temperature and attractiveness in

the likelihood of a particular destination being

chosen. Using A1FI and B1 climate change

scenarios they show how Spain’s colder Northern

provinces would benefit from rising temperatures

while provinces in the south would experience a

decrease in the frequency of trips. In this application

it is important to highlight how a squared term for

the temperature was used, thus providing an

estimation of maximum or minimum comfort

temperatures and evidencing the non-linear

relationship between tourism and climate.

SIMULTECH2013-3rdInternationalConferenceonSimulationandModelingMethodologies,Technologiesand

Applications

484

Although models evaluating the effects of

climate change on tourism within the discrete choice

modeling framework have just emerged, it should be

noted how the assumptions about the utility function

should be deeper investigated. Then, the use of

additive sum of the individual utilities would be that

the tourist is assumed to be risk-neutral, an

assumption sometimes difficult to justify within the

framework of tourism choices.

4.3 Aggregated Tourism Demand

Models

Aggregated tourism demand modeling continues to

be a popular issue in tourism literature. Reviews by

Lim (1999), Li et al. (2005) and Song and Li (2008)

show that tourism demand estimation mainly

focuses on time series models, thus including

variables with significant short term variability, such

as prices and income, and neglecting structural

determinants, such as climate, which are expected to

be collected by the constant term. This practice can

be justified by the interests of tourism planners, who

have preferably been interested in income elasticities

and/or price elasticities in order to forecast tourism

demand in an accurate way (a key issue for service

industries with relatively high fixed costs), or

alternatively in order to evaluate the consequences

of taxes or exchange rate policies.

However, with the growing interest of climate

issues, a set of aggregated tourism demand models

have emerged focusing their interest on climate

variables. Hence, the pioneering study of Maddison

(2001) sets out a cross-sectional model for chosen

destinations of British tourists using classical price

determinants of tourism demand and incorporating

climate variables in terms of attractors. The

estimation of the model permits the quantification of

the trade-off between climate and holiday

expenditure and, because of the introduction of non-

linear effects (through a 4

order polynomial) the

identification of ‘optimal’ climate for British

generating tourism. The findings are used to predict

the impact of several climate change scenarios on

different tourist destinations. In a similar way, Lise

and Tol (2002), using aggregated data and

regression analysis, find optimal temperatures at

travel destinations for different tourists and tourist

activities, showing that OECD tourists prefer an

average of the hottest month of the year temperature

of 21 ◦C indicating that, under a scenario of gradual

warming, tourists will spend their holidays in

different places than they currently do. With a global

perspective, Hamilton et al. (2005a and 2005b) set

out what is known as the Hamburg Tourism Model

(HTM), consisting of the estimation of two

equations for international tourist departures and

arrivals for a specific year. Analytically:

dddddd

YCTTGA lnln

3210





(5)

ooooo

GYBTT

lnln

543

210





(6)

where A refers to the total number of arrivals in a

country d; D the total number of departures from s

country O; P the population in thousands; G the area

in squared kilometers; T the country’s mean yearly

temperature for the period 1961-1990 in degrees

centigrade; C the length of the coastline in

kilometers; Y the country’s per-capita income; B the

number of countries bordering a particular country;

and



and



parameters to be estimated.

Hamilton et al. (2005a and 2005b) use the HTM

to analyze how climate change alters the relative

appeal of countries, studying the redistribution of

tourist arrivals and departures due to changes in

population, per-capita income and climate change.

The results show how, in the medium to long term,

tourism will grow in absolute terms but this increase

will be smaller than population and income changes

and not homogeneously distributed, with it being

higher for colder countries and lower for warmer

ones. Climate change would also imply that the

currently dominant group of international tourists -

sun and beach lovers from Western Europe - would

stay closer to home, implying a relatively small fall

in total international tourist numbers and total

distance travelled. However, it should be noted how

the authors highlight that changes induced by

climate change are generally much smaller than

those resulting from population and economic

growth.

The HTM, as an aggregated model, has been

criticized and extended in many ways, although the

improvement of the methodology has often implied

a loss of generalization. Thereby, Bigano et. al.

(2006) extend the HTM by considering substitution

between domestic and international tourism, while

also analyzing tourist expenditure. However, the

consideration of these two issues implies the

limitation of the sample of countries to be included

in the model due to data restrictions, and only 45

origin-countries travelling to 200 destination-

countries are considered. Hamilton & Tol (2007)

assess the impact of climate change on tourism from

a regional perspective in Germany, the UK and

Ireland, based on different scenarios of climate

change for the regions under analysis. They suggest

that non-uniform warming within countries might

lead to tourist behavior patterns that are regionally

different, pointing to the need to develop HTM

methods on a lower scale than a national one.

Recent updates of the HTM can be found in

Rosselló and Santana (2012) and Tol and Walsh

QuantitativeModelsEvaluatingtheEffectClimateChangeEffectsonTourism-StateoftheArt

485

(2012), who use bilateral tourism flows and consider

the nature of the data dynamics by testing the change

in tourism preferences in reference to climate

conditions. Despite the relatively high level of

complexity in the specification and estimation, using

specific projected climatic, population and economic

data related to A2, B1 and B2 scenarios, Rosselló

and Santana (2012) forecast tourist arrivals for 2080

finding similar results to previous works, thus

providing more evidence that climate change would

imply that the currently dominant international

tourism flow from North to South would be weaker.

5 CONCLUSIONS

Many tourist destinations around the world are

considering what kind of effects climate change

would lead to, with the consequences on tourist

demands as one of the most recurrent ones. Despite

the uncertainties and complexity of evaluating

expected tourism demand responses in a context of

climate change the literature has just started to

answer this question assuming the inability to

control the complexity nature of tourism demand

decisions by turning to the ceteris paribus clause. In

this work, the most relevant methodologies

presented in the literature quantitatively evaluating

the effect of climate change on tourism and

projecting this relationship to the future are

summarized in three categories: the consideration of

physical changes, the analysis of tourist

attractiveness through climatic indexes, and tourism

demand modeling based on revealed preferences.

One of the common results that the different

methodologies show is that climate change is, on the

whole, bad news for warm destinations whatever the

methodology used. Then, the search for a more

comfortable climate is found to be one of the main

motivations determining global tourism flows and,

as such, climate change will imply a loss of

attractiveness for traditional winter resorts and

traditional warmer destinations around the world.

However, it seems that climate change, on the one

hand, would increase domestic trips, especially in

colder countries, and, on the other hand, could be

good news for seasonality. This is what the revealed

preference methods show us. What is more, this

would be an opportunity for the industry to capture

the increasing segment of short breaks during non-

summer seasons. At any rate, further research in this

respect is encouraged. Another agreement that is

shown from the different perspectives is the non-

linear relationship that apparently exists between

tourism and climate. More precisely, an inverted u-

shape in the relationship between temperature and

tourism demand has been found using different

perspectives, thus revealing the existence of optimal

climatic conditions for the practice of tourism.

Nevertheless, some questions remain. Will some

destinations be too hot? The inverted u-shape can be

explained by both the existence of a turning point

(destination will be too hot) or by the increase of

competitors. What are the most sensitive marked

segments to climate change? What will the induced

effects of climate change be on biodiversity loss, dry

episodes, beach transformations, etc.? Thus, it seems

clear that more quantitative studies are needed

Given the aim of this paper, the integration of

research findings is limited to studies that report

quantitative evidence of the relationship between

tourism and climate with the aim of extrapolate the

relationship in the context of different climate

change scenarios. Other non-empirical studies,

however, may be applicable insofar as they provide

relevant theories which can contribute to the

knowledge of future reaction of tourism to climate

change. This study does not attempt to integrate the

findings of all the relevant empirical studies. For

instance, there is bound to be a number of

unpublished studies. Nevertheless, the set of studies

examined is reasonably comprehensive and the

resulting large set of findings is fairly representative.

The results provided in this article would be a

useful guide for other researchers and practitioners

interested in carrying out similar studies

investigating the effects of climate change on

tourism. The selection of the most suitable approach,

however, will depend upon the circumstances and

objectives of the study being planned. It would be

wrong to blindly adopt any one approach without

first judging its limitations and assumptions. In this

regard, this work provides a convenient reference.

REFERENCES

Agnew, M., & Palutikof, J. (2006). Impacts of short-term

climate variability in the UK on demand for domestic

and international tourism. Climate Research, 21: 109-

120.

Álvarez, M., & Rosselló, J. (2010). Forecasting British

tourist arrivals to Balearic Islands using

meteorological variables. Tourism Economics, 16:

153-168.

Amelung, B., & Moreno, A. (2012). Costing the impact of

climate change on tourism in Europe: results of the

PESETA project. Climatic Change 112: 83-100.

SIMULTECH2013-3rdInternationalConferenceonSimulationandModelingMethodologies,Technologiesand

Applications

486

Amelung, B., & Viner, D. (2006). Mediterranean Tourism:

Exploring the Future with the Tourism Climatic Index.

Journal of Sustainable Tourism, 14: 349-366.

Amelung, B., Nicholls, S., & Viner, D. (2007).

Implications of Global Climate Change for Tourism

Flows and Seasonality. Journal of Travel Research 45:

285-296.

Bigano, A., Hamilton, J. M., & Tol, R. S. J. (2006). The

impact of climate holiday destination choice. Climatic

Change, 76, 389-406.

Bigano. A, Hamilton, J. M., Maddison, D. J., & R. S. J.

Tol (2006). Predicting tourism flows under climate

change an editorial comment on Gössling and Hall

(2006). Climatic Change, 79:175-180.

Breiling, M., & Charamza, P. (1999). The impact of global

warming on winter tourism and skiing: a regionalised

model for Austrian snow conditions. Regional

Environmental Change, 1: 4-14.

Bujosa, A., & Rosselló, J. (2013). Climate change and

summer mass tourism: the case of Spanish domestic

tourism. Climatic Change 117: 363 - 375.

CIA, Central Intelligence Agency (2012). The World

Factbook. Accessed 19th June 2012 at: https://

www.cia.gov/library/publications /the-world-fact

book/fields/2012.html

Crouch, G. I. (1994). The Study of International Tourism

Demand: A Survey of Practice Journal of Travel

Research, 32: 41-55.

Elsasser, H., & Messerli, P. (2001). The vulnerability of

the snow industry in the Swiss Alps. Journal of

Mountain Research and Development, 21: 335-339.

Englebert, E. T., McDermott, C., & Kleinheinz, G. T.

(2008). Effects of the nuisance algae, Cladophora, on

Escherichia coli at recreational beaches in Wisconsin.

Science of the Total Environment, 404, 10-17.

Eugenio-Martín, J. L., & Campos-Soria, J. A. (2010).

Climate in the region of origin and destination choice

in outbound tourism demand. Tourism Management,

31: 744 -751

Eugenio-Martín, J. L., (2003). Modelling determinants of

tourism demand as a five-stage process: A discrete

choice methodological approach. Tourism and

Hospitality Research, 4(4):341-354.

Goh, C. (2012). Exploring impact of climate on tourism

demand. Annals of Tourism Research 39: 1859-1883.

Gössling, S., & Hall, C. (2006). Uncertainties in predicting

tourist flows under scenarios of climate change.

Climatic Change, 79: 163-173.

Gössling, S., Bredberg, M., Randow, M., Sandström, E., &

Svensson, P. (2006) Tourist Perceptions of Climate

Change: A Study of International Tourists in Zanzibar.

Current Issues in Tourism, 9: 419-435

Gössling, S., Lindén, O., Helmersson, J., Liljenberg, J., &

Quarm, S. (2007). Diving and global environmental

change: a Mauritius case study. In B. Garrod & S.

Gössling (Eds.), New frontiers in marine tourism.

Amsterdam: Elsevier.

Gössling, S., Scott, D., Hall, C. M., Ceron, J.P., & Dubois,

G. (2012). Uncertainties in predicting tourist flows

under scenarios of climate change. Annals of Tourism

Research, 39: 36-58.

Hamilton, J., & Tol, R.S.J. (2007). The impact of climate

change on tourism in Germany, the UK and Ireland: A

simulation study. Regional Environmental Change, 7:

161-172.

Hamilton, J. M., Maddison, D., & Tol, R.S.J. (2005a).

Climate change and international tourism: A

simulation study. Global Environmental Change, 15:

253-266.

Hamilton, J. M., Maddison, D., and Tol, R.S.J. (2005b).

The effects on climate change on international

tourism. Climate Research, 29: 245-254.

Harrison, S. J., Winterbottom, S.J., and Shephard, C.

(1999). The potential effects of climate change on the

Scottish tourist industry. Tourism Management, 20:

203-211.

Hoegh-Guldberg, O., Mumby, P. J., Hooten, A. J.,

Steneck, R. S., Greenfield, P., Gomez, E., Harvell, D.

R., Sale, P. F., Edwards, A. J., Caldeira, K. Knowlton,

N., Eakin, C. M., Iglesias Prieto, R., Muthiga, N.,

Bradbury, R. H., Dubi, A. & Hatziolos, M. E. (2007).

Coral Reefs Under Rapid Climate Change and Ocean

Acidification. Science, 318: 1737-1742.

IPCC, Intergovernamental Panel on Climate Change

(2007). Fourth Assessment Report: Climate Change

2007 (AR4). Working Group II Report "Impacts,

Adaptation and Vulnerability".

Kulendran, N., & Dwyer, L. (2012). Modeling seasonal

variation in tourism flows with climate variables.

Tourism Analysis, 17: 121-137.

Lancaster, K.L. (1966). A New Approach to Consumer

Theory. Journal of Political Economy, 74(2): 132-157.

Li, G., Song, H., & Witt, S. F. (2005). Recent

developments in econometric modeling and

forecasting. Journal of Travel Research, 44, 82-99.

Lim, C. (1999). A Meta analysis review of International

tourism demand. Journal of Travel Research, 37, 273-

284.

Lin, T.P., & Matzarakis, A. (2011). Tourism climate

.information based on human thermal perception in

Taiwan and Eastern China. Tourism Management, 32:

492-500.

Maddison, D. (2001). In search of warmer climates? The

impact of climate change on flows of British tourist.

Climatic Change, 49: 193-208.

Marcussen, C.H. (2011). Understanding Destination

Choices of German Travelers. Tourism Analysis, 16:

649-662.

Matzarakis, A., Mayer, H. & Iziomon, M. G. (1999).

Applications of a universal thermal index:

Physiological equivalent temperature. International

Journal of Biometeorology 43(2):76–84.

Mieczkowski, Z. (1985). The tourism climatic index: A

method of evaluating world climates for tourism.

Canadian Geographer, 29 : 220-233.

Moreno, A. & Amelung, B. (2009). Climate Change and

Tourist Comfort on Europe's Beaches in Summer: A

Reassessment. Coastal Management, 37: 550-568.

Morgan, R., Gatell, E., Junyent, R., Micallef, A., Ozhan,

E., & Williams, A. T. (2000). An improved user-based

QuantitativeModelsEvaluatingtheEffectClimateChangeEffectsonTourism-StateoftheArt

487

beach climate index. Journal of Coastal Conservation,

6: 41-50.

Morley, C. L. (1992). A Microeconomic Theory of

International Tourism Demand. Annals of Tourism

Research, 19: 250-267.

Nicholls, R. J., Marinova, N., Lowe, J. A., Brown, S.,

Vellinga, P., Gusmão, D., Hinkel, J. & Tol, R. S. J.

(2011) Sea-level rise and its possible impacts given a

'beyond 4°C world' in the twenty-first century.

Philosophical Transactions of the Royal Society A,

369: 161-181.

Otero-Giráldez, M.S., Álvarez-Díaz, M., & González-

Gómez M. (2012) Estimating the long-run effects of

socioeconomic and meteorological factors on the

domestic tourism demand for Galicia (Spain). Tourism

Management, 33: 1301-1308.

Papatheodorou, A., Rosselló, J., & Xiao, H. (2010).

Global Economic Crisis and Tourism: Consequences

and Perspectives. Journal of Travel Research, 49: 39-

45.

Purcell, J. E. (2012). Jellyfish and ctenophore blooms

coincide with human proliferations and environmental

perturbations. Annual Review of Marine Science, 4:

209-235.

Rosselló, J.; Riera, A., & Cardenas, V (2011). The impact

of weather variability on British outbound flows.

Climatic Change, 105: 281-292.

Rosselló, J., and Santana, M. (2012). Climate change and

global international tourism: An evaluation for

different scenarios. Working paper: Departament

d’Economia Aplicada. Universitat de les Illes Balears,

Palma.

Rosselló, J. (2011). The impact of the NAO index on

European Airline Transit. Transportation Research

Part D: Transport and Environment, 16: 183-187.

Rotmans, J., Hulme, M. & Downing, T. E. (1994). Climate

change implications for Europe. Global

Environmental Change, 4: 97-124.

Scott, D., McBoyle, G., & Schwartzentruber, M. (2004).

Climate change and the distribution of climatic

resources for tourism in North America. Climate

Research, 27: 105-117.

Scott, D., McBoyle, G., Minogue, A., & Mills, B. (2006).

Climate Change and the Sustainability of Ski-based

Tourism in Eastern North America: A Reassessment.

Journal of Sustainable Tourism, 14: 376-398.

Song, H., & Li, G. (2008). Tourism demand modelling

and forecasting -A review of recent research. Tourism

Management, 29: 203-220.

Subak, S., Palutikof, J., Agnew, M., Watson, S., Bentham,

C., Cannel, M., Hulme, M. McNally, S., Thornes, J.,

Waughray, D., & Woods, J. (2000). The impact of the

anomalous weather of 1995 on the UK Economy.

Climatic Change, 44: 1-26.

Tol, R.S.J., & Walsh, S. (2012) The Impact of Climate on

Tourist Destination Choice. Working paper: ERSI.

Economic and Social Research Institute, Dublin.

WTTC, World Travel and Tourism Council (2012) Travel

& Tourism Economic Impact 2012: WORLD. WTTC:

London.

SIMULTECH2013-3rdInternationalConferenceonSimulationandModelingMethodologies,Technologiesand

Applications

488