Estimating the risk of loss of beach recreation value under climate change

Tourism Management 68 (2018) 387–400

Contents lists available at ScienceDirect

Tourism Management journal homepage: www.elsevier.com/locate/tourman

Estimating the risk of loss of beach recreation value under climate change

T

∗

Alexandra Toimil , Pedro Díaz-Simal, Inigo J. Losada, Paula Camus Environmental Hydraulics Institute “IHCantabria”, Universidad de Cantabria, Isabel Torres 15, 39005 Santander, Spain

A R T I C LE I N FO

A B S T R A C T

Keywords: Climate change Extreme events Beach recreation Tourism Erosion risk Probabilistic estimates Uncertainty

Shoreline recession due to the combined effect of waves, tides and sea level rise is increasingly becoming a major threat to beaches, one of the main assets of seaside tourist destinations. Given such an uncertain future climate and the climate-sensitive nature of many decisions that affect the long term, there is a growing need to shift current approaches towards probabilistic frameworks able to take uncertainty into account. This study contributes to climate change research by exploring the effects of erosion on the recreation value of beaches as a key indicator in the tourism sector. The new paradigm relates eroded sand to geographic and socioeconomic aspects and other physical settings, including beach type, quality and accesses, yielding monetary estimates of risk in probabilistic terms. Additionally, we look into policy implications regarding tourism management, adaptation and risk reduction. The methodology was implemented in 57 beaches in Asturias (north of Spain).

1. Introduction During the second half of the last century, residential use and industrial and service activities around many beaches worldwide intensified, giving the so-called sun and sand tourism (Aguilo et al., 2005; Sarda, Mora, Ariza, Avila, & Jimenez, 2009) a major boost. Thus, the fact that these systems are already eroding generally (Bird, 1985; Hinkel et al., 2013) poses serious risks to one of the primary resources of coastal regions, which not only maintains biodiversity and coastal protection but also provides recreation services. However, in 1994, the US Army Corps of Engineers admitted that an eroding shoreline was a serious threat to tourism and therefore to the US economy. Although there is evidence that climate change and resulting sea level rise (SLR) will cause the coastline to recede inland (Wong et al., 2014), coastal erosion is also triggered by many other factors, which raises difficulties with respect to developing appropriate management strategies (Phillips & Jones, 2006). The effect of local waves, storm surges and very large tides greatly contributes to sediment transport, and therefore to erosion. In response, while some coastal systems may be able to undergo a landward retreat, others will experience coastal squeeze (Jackson & McIlvenny, 2011), which occurs when an eroding shoreline approaches seawalls or resistant natural cliffs. This squeezing leads to adverse impacts on the environment and society, such as habitat destruction, increased exposure to flooding and loss of beach recreation value. In light of this unprecedented threat, research on climate change risks and adaptation in coastal areas with respect to tourism and recreation, a relatively unexplored field, is becoming increasingly important

∗

(Schliephack & Dickinson, 2017). However, the existing limitations of future climate and socioeconomic projections do not leave much choice but to make decisions in the context of uncertainty. For this reason, there is a growing need to shift away from widespread uncertain deterministic approaches (e.g., Bruun, 1962) and move towards probabilistic frameworks that enable effective risk-based coastal planning and management. In this setting, we aimed to bridge the gap between the scientific community and relevant stakeholders, such as tourists and local users and decision makers, who are often unaware not only of the physical processes but also of the scope, potential consequences, and implications of climate change. To this end, we developed a methodology to assess the erosion-driven loss of beach recreation value in probabilistic terms. Without limiting its generality, the procedure was applied to 57 beaches on the coast of Asturias, a region located in the northwest of Spain along a 345 km coastal shoreline stretch. Among the most common aspects of coastal management are beach sediment management and use (Ariza, Jiménez, & Sardá, 2008), requiring coastal managers and tourism businesses to fully understand users' preferences, perceptions and economic valuations based on their own experiences (Blakemore & Williams, 2008). The most extended existing asset valuation techniques include the following: i) declared preferences such as contingent valuation (based on questionnaires) and choice experiments (based on simulated experiments), which differ in the way the attributes are presented and in the structure of the willingness-to-pay question (Birdir, Ünal, Birdir, & Williams, 2013); ii) revealed preferences, such as the widely accepted travel cost method, which accounts for the asset access costs (Loomis & Santiago, 2013),

Corresponding author. E-mail addresses: [email protected] (A. Toimil), [email protected] (P. Díaz-Simal), [email protected] (I.J. Losada), [email protected] (P. Camus).

https://doi.org/10.1016/j.tourman.2018.03.024 Received 10 October 2017; Received in revised form 29 March 2018; Accepted 29 March 2018 0261-5177/ © 2018 Elsevier Ltd. All rights reserved.

Tourism Management 68 (2018) 387–400

A. Toimil et al.

towards beach recreation and discusses their policy implications in terms of tourism management, adaptation and risk reduction.

and the hedonic pricing method, which Parsons and Powell (2001) used to estimate the value of the land loss using datasets on housing sales from that time; and iii) market-based methods, such as cost-based methods (either avoided or defensive costs) and the production function, in which beaches are assumed to be environmental assets that contribute to the supply of goods or services to social agents. However, while the literature is quite diverse in terms of beach quality assessments (e.g., Morgan, 1999; Roca, Villares, & Ortego, 2009), less attention has been given thus far to the quantitative valuation of recreation. The two concepts of sun and beach tourism and beach recreation are strongly linked through cause and effect. For almost two decades, Houston (2013) focused a large part of his research on highlighting the importance of beaches to the United States economy. He noted that if California's beaches were unavailable for recreation, users would instead spend approximately $2.4 billion outside of the United States (King & Symes, 2003), highlighting the severe disadvantage of relegating recreation to a lower priority. Hence, local tourists would substitute other forms of recreation for beach recreation if beaches were eroded significantly. More recently, Alexandrakis, Manasakis, and Kampanis (2015) capitalized the value of the eroded beach in revenues from tourism businesses through hedonic pricing modeling, in which the beach value was determined by its width and the tourism businesses located there. From the beach recreation perspective, changes to the economic recreation value were estimated by King, McGregor, and Whittet (2016) through a benefit transfer approach that used information on weather, water quality, and beach facilities and services, among others. In the present paper, we used a production function–based approach to monetarily estimate the extent of potential losses of beach recreation value due to climate change–induced erosion. The analysis combines a set of environmental and social criteria along with correction factors tailored to beach-specific characteristics, quality and services. Aiming to ensure sustainable tourism and avoid significant losses of beach recreation value in the future, predictions of coastal recession need to be more reliable than ever (Ranasinghe, Callaghan, & Stive, 2012). Thus, the common practice of adopting a single value of coastal recession due to a single value of SLR is proving unsuitable for emerging risk management frameworks, which require probabilistic estimates of the combined effect of a range of climatic and non-climatic drivers and their associated uncertainties (e.g., expressed in terms of confidence levels) (Ranasinghe et al., 2012; Toimil, Losada, Camus, & Diaz-Simal, 2017a). While assessments of the impact of beach erosion are fewer than those of coastal flooding, risk and consequence valuations are even fewer. Wainwright et al. (2015) used an economic model to determine an optimal development setback location based on whether investment at a particular location is economically viable, given the amount of damage expected by coastal processes. Also focusing on the use of quantitative risk analysis for establishing setback lines, Jongejan, Ranasinghe, Wainwright, Callaghan, and Reyns (2016) combined the calculated probability density functions of annual recession extremes with property value data to obtain erosion risk estimates. Both studies yielded probabilistic estimates of coastline recession driven by combined storm erosion and SLR. While the methodology was applied only to a single beach, efficient coastal management requires the scope of the analysis to be extended across an entire region. Thus, the probabilistic erosion risk assessment we present herein not only accounts for the local tide, waves, storm surges, and SLR uncertainty but has also been implemented at regional scale (i.e., 57 beaches), resulting in robust estimates of shoreline change over the whole century for both un-interrupted and inlet-interrupted coastlines (Toimil et al., 2017a). To our knowledge, this assessment is unique in terms of coastal planning and management, as it sets out a procedure to probabilistically determine the loss of beach recreation value due to a combination of climatic drivers throughout the twenty-first century in a large area. Thus, the paper seeks to provide an insight into the risks of inaction

2. A case study of the Asturian coast (north of Spain) Asturias is a Spanish region bordered by the Cantabrian Sea to the north, by the regions of Cantabria to the east and Galicia to the west and has 345 km of elongated, rectilinear and steep-sloped coastline. While featuring long stretches of cliffs, this area also boasts more than two hundred beaches, of which 57 were included in the assessment, particularly those that are sandy and longer than 200 m (see Fig. 1). The selected beaches are deemed pocket-type, which broadly implies that gradients in longshore sediment transport can be assumed to be negligible. Besides, 5 out of the 57 beaches are adjacent to tidal-dominated estuaries, and therefore are subject to the potential effects that these systems can have on their erosion/accretion patterns (Ranasinghe, Duong, Uhlenbrook, Roelvink, & Stive, 2013). In this regard, it is important to note that due to anthropogenic influences that include upstream damming and dredging induced by a change in land use, there is no significant fluvial sediment supply to be integrated into the sediment budget. The beaches all have macrotidal (2–5 m spring tidal range) and semi-diurnal tidal regimes, and they are mainly constituted by fine (0.2–0.3 mm) quartz sand. The most energetic waves come from the northwest to the north-northwest sectors, and are characterized by significant wave heights that may reach 10 m and peak periods of up to 20 s. When these combine with unusual levels of storm surges and high spring tides, significant damage to the waterfront may result. Between December 2013 and March 2014, a series of storm events hit the Asturian coast, causing severe coastal flooding and noteworthy beach erosion (Toimil et al., 2017a). The cumulative effects of successive storms over the beaches prevented them from recovering and providing the proper recreation services during the next summer period. The region has an Atlantic climate characterized by relatively mild and humid winters, as well as warm summers, although they are colder than in the rest of Spain. Air temperatures vary both seasonally and daily, and averages range from 9 °C (13 °C–5 °C) in January to 21 °C (23 °C–16 °C) in July. Since the ocean also has effects on precipitation, rainfall is particularly abundant, especially from October to April. Easter often marks the beginning of the beach season, and the weather is relatively warm and sunny until well into October, and even extending further into November. Two main beach seasons can thus be distinguished: from April 1 to October 31 and from November 1 to March 31, hereinafter denoted as recreation season and non-recreation season, respectively. Asturias is eminently a coastal region. Currently, 50% of the population lives within the coastal strip, where a major growth occurred during the second half of the twentieth century. The Asturian coast generates more than 48% of the economic flows and hosts approximately 55% of the major infrastructure in the region. As the Asturian tourist sector has largely been based on its natural heritage, between 8% and 11% of the GDP of the entire territory (20.7 billion1 EUR in 2013 according to FBBVA, 2017) is linked to tourism, either directly or indirectly. Although beaches are prevalent among the range of existing tourism options, the contribution of beach tourism to tourist-related GDP is an estimated value, given the geographical proximity of beaches, mountains and cultural activities within the territory, and the possibility to combine them on the same day. In 2015, Asturias received 7 million tourists, of which 82.3% were national, 14.6% international, and the remaining 3.1% intraregional. According to official touristic statistics, 47% of tourist activity was beach related, with culture and sports making up the remaining 53%. In light of these data, we assume beach-related GDP to be 971 million EUR in 2016. With respect to the 1

388

1 billion EUR = 1000 million EUR.

Tourism Management 68 (2018) 387–400

A. Toimil et al.

Fig. 1. Location of the Asturian beaches of study.

Asturian capital stock, the FBBVA (2017) estimates that it amounted to 72.31 billion EUR in 2013. From this, 1.23 billion EUR are non-residential but associated with agriculture and fishing, 14.51 billion EUR

are private supply services, and 15.22 billion EUR correspond to public infrastructures. Finally, it is important to note that Spain boasts more Blue Flag (BF) 389

Tourism Management 68 (2018) 387–400

A. Toimil et al.

Fig. 2. Flow chart of the methodology.

was specified to define the scope of the impact, and hence to obtain future shoreline changes due to coupled short- and long-term effects under RCP8.5, a very high GHG emission scenario. These changes are those that were combined with exposure and vulnerability into a damage model to provide erosion risk estimates for the tourist sector, using the loss of beach recreation value as a proxy (lower right panel in Fig. 2). Although we adopted a BaU approach to facilitate the analysis, exposure and vulnerability may also change with future socio-economic trends (e.g., increase in tourist pressure) and even with climate, as temperature directly impacts seasonal distribution, social perception and social experience. In this regard, we established the conceptual and methodological basis, but we left open the option to modify any parameter involved, or to include new ones as deemed appropriate. The hazard and impact assessments are summarized in Section 3.1 (for further details regarding the erosion modeling, please see Toimil et al., 2017a), and the assessment of exposure, vulnerability and risk is fully covered in Section 3.2. Note that this procedure may be extended to climate change risk assessments other than those arising from coastal erosion (Toimil, Losada, Diaz-Simal, Izaguirre, & Camus, 2017b).

beaches than any other country in the world (BF, 2017). The BF ecolabel acknowledges beaches for their all-round high quality. High standards related to water quality, environmental management, environmental education and safety need to be not only met but also maintained. Among the 684 Spanish BFs, 14 are located along the 345 km Asturian coastal stretch, increasing its attractiveness as a major seaside tourist destination. 3. A methodology to estimate the loss of beach recreational value in a probabilistic risk framework With the aim to economically assess the loss of beach recreation value under climate change, we developed a methodology based on the Intergovernmental Panel on Climate Change (IPCC) risk framework (IPCC, 2014), in which risk results from the interaction of three components: hazard, represented by the coastal dynamics, exposure, described by the physical and socioeconomic environment, and vulnerability, associated with the susceptibility of the system to harm and its capacity to adapt. As outlined in the flow chart in Fig. 2, the methodology consists of a multi-tiered process in which the risk components are developed individually but are finally integrated to derive expected losses. To that end, erosion-driven hazards were characterized by considering that the shoreline migration responds to both long- and short-term effects (upper central panels in Fig. 2): the former caused by slow-onset SLR, and the latter by local waves, storm surges and astronomical tides. Exposure was defined by the users’ preferences (i.e., when they would rather go to the beach), expressed in terms of effective utilization rates and their purchasing power, as an indicator of how much they value the time for recreation (left panel in Fig. 2). Concerning vulnerability, a set of aspects related to the typology, the quality (i.e., water, sand and services), and the number and distribution of access points and parking facilities were defined for each beach (right panel in Fig. 2). In this work, we assumed a business as usual (BaU) scenario in which exposure and vulnerability were based on historical data, and climate change was introduced by considering the effect of the hazard for different representative concentration pathways (RCPs) (Moss et al., 2010). In this regard, a shoreline evolution model

3.1. Climate change and variability as shoreline recession triggers Any long-term coastline recession is due but not limited to the combined effects of storm erosion (i.e., large waves and local water surface elevation) and SLR (Ranasinghe et al., 2012; Toimil et al., 2017a), but also on account of the impacts that estuaries can have on the long-term evolution of adjacent beaches (Ranasinghe et al., 2013). Accordingly, the shoreline change will depend not only on cross-shore forcings as waves, storm surges, astronomical tides and SLR but also on long-shore sinks resulting from the SLR-induced sand transport that occurs from the beach into the estuary and to the ebb tidal delta. Considering the joint action of all these processes, we designed a model to predict the future shoreline evolution for each of the 57 study beaches. Aiming to meet the requirements of new emerging frameworks for coastal management, thousands of model simulations allowed us to obtain robust estimates of coastline change in probabilistic terms that also account for SLR uncertainty (the modeling framework is fully 390

Tourism Management 68 (2018) 387–400

A. Toimil et al.

Fig. 3. Flowchart of the procedure developed to obtain the number of hours per year during which beaches provide recreation services.

which a set of pilot systems for beach management services and for determining the carrying capacity of beaches was implemented. The installation of a video-monitoring and image analysis system allowed the determination of the number of users in the different areas, its space-time evolution, and its correlation with the services provided by a number of target beaches in Europe. Furthermore, bearing in mind that beaches on the Cantabrian Sea are much less congested than those of the Mediterranean, this estimate is consistent with the standard values of minimum recommended sand (m2/user) proposed by many authors in the literature (Roca, Riera, Villares, Fragell, & Junyent, 2008).

described in detail in Toimil et al., 2017a). To obtain the cumulative losses of beach recreation value by the end of the century in case no actions were taken, we estimated beach recessions between 2010 and 2100 (denoted hereinafter as R2100) by subtracting the five-year average initial position from the five-year average final position of the shoreline from each simulation. Additionally, since adaptation or risk reduction often occurs in response to unusual large recessions, we performed a second analysis in which GEV distributions were fitted to annual maxima shoreline retreats to derive a range of return periods. These allowed us to quantify losses associated with extreme retreat events within the framework of spatial planning and risk management. Finally, with the aim of computing annual damages to beach recreation value, we derived the annual beach changes (denoted hereinafter as annual beach changes) by subtracting the average shoreline position during the recreation season between two consecutive years. It is important to note that while both the R2100 and the annual beach changes indicators account for the inter-annual variability due to the combined effect of waves, storm surges, astronomical tides and SLR (they consider the accumulated net erosion by 2100 and the annual mean recession or accretion of the coastline, respectively), extreme retreat events focus particularly on unusual or exceptional recessions that are often responsible for serious damage, albeit fully or partly reversible.

3.2.2. Number of hours per year that beaches provide recreational services (H) In the second step, we calculated the average number of hours per year that beaches provide users with recreation services (see flowchart in Fig. 3). To that end, two types of data that can be derived from field surveys and/or imagery analysis need to be collected and used: geographical conditions and social behavior. Geographical conditions (i.e. latitude) provide insight into the potential number of sunny days per month (second column in Table 1) and the potential number of sunny hours per day, which, in Asturias, amounts to 8 and 10 h on average for the non-recreation and recreation seasons, respectively (expressed in potential hours per month in the third column in Table 1). Social behavior was used by us to ascertain the average number of effective days that the recreation season lasts per year on average, and how these days are distributed monthly (expressed in percentage of potential days per month in the fourth column in Table 1) according to users' habits and preferences any time of the year (e.g., jogging, soccer, swimming, and sunbathing). Here, the range of environmental conditions, such as the temperature distribution, precipitation and humidity of the study region plays a significant role. Additionally, the average number of effective hours per day needs to be considered, and they are expressed in terms of percentage of the potential hours per day in the fifth column in Table 1. Social data on users’ behavior also enabled us to derive the average occupancy rate. This rate was distributed seasonally in daily time slots according to the peak occupancy (100% occupancy) that occurs from 13:00 to 17:00 h in July and August (expressed in percentages with respect to the peak capacity in Table 2). Note that the average occupancy rates obtained in Table 2 (last row in Table 2) are displayed in the sixth column in Table 1. Finally, effective hours per month (seventh column in Table 1) are derived as a result of integrating columns three to six in Table 1, resulting in the annual number of total effective hours during which beaches provide users with recreation services as the sum of the whole column (seventh row in Table 1); in Asturias, this amounts to 498 h.

3.2. A function-based approach to quantify the beach recreational value The challenge of assessing the potential loss of recreation value due to climate change and variability-induced erosion of every target beach is herein proposed to be addressed through a function-based approach. To that end, our research is mainly based on a combination of videomonitoring analysis, physical observation over the years, and stakeholders’ consultation. The methodology developed consists of a fourstep procedure. First, we estimated the area required by a user to stay comfortably at the beach. Second, we calculated the average number of hours per year that Asturian beaches provide recreation services. Third, we obtained the value of the recreation time per person. Finally, we quantified the economic and accounting valuation of the recreation value of each of the study beaches per square meter. While steps 1 to 3 refer to the characterization of exposure, step 4 incorporates the vulnerability component and designs the damage model to ultimately derive risk. Each of these steps is described in detail as follows. 3.2.1. Area required by a user (A) Within the first step, we established that the space required by a user is 15 m2 (average value of both urban and non-urban beaches). This assumption is based on the analysis developed under the CoastView EU-funded project (Contract No. EVK3-CT-2001-00054), in 391

Tourism Management 68 (2018) 387–400

A. Toimil et al.

Table 1 Effective monthly utilization rates. Month

Potential days per month

Potential hours per month

% Days of use with respect to potential days per month

% Hours of use with respect to potential hours per day

% Occupancy

Effective hours

January February March April May June July August September October November December

31 28 31 30 31 30 31 31 30 31 30 31

248 224 248 300 310 300 310 310 300 310 240 248

0 0 0 5 10 60 80 90 40 10 5 0

0 0 0 60 60 80 100 100 80 60 10 0

0 0 0 21.67 21.67 62.5 64 64 62.5 21.67 10 0

0 0 0 1.95 4.03 90 158.72 178.56 60 4.03 0.12 0

Yearly average

25%

45.83%

27.33%

Recreational season average

42.14%

77.14%

45.43%

Total effective hours per year

498

Table 2 Effective hourly utilization rates (expressed in percentages with respect to the peak occupancy reached from 13.00 to 17.00 h in July and August). Time slots

July–August (%)

June–September (%)

April-May-October (%)

November (%)

10–11 11–12 12–13 13–14 14–15 15–16 16–17 17–18 18–19 19–20

10 30 80 100 100 100 100 80 30 10

0 20 70 80 80 80 80 70 20 0

0 0 5 30 30 30 30 5 0 0

0 0 0 0 10 0 0 0 0 0

Average occupancy rate

64%

62.5%

21.67%

10%

3.2.3. Value of the recreation time per person (VRT) Within the third step, we addressed the monetary quantification of a person's recreation time. Coastal recreation generates direct economic impacts and non-market values (King et al., 2016). While direct economic impacts measure the flow of money through an economy in the form of jobs, salaries and taxes; non-market values refer to the net value a resource adds to society and can be measured in a myriad of ways. Several authors studied the opportunity cost of time by using some proportion of the individual's market wage rate or income per hour that can be determined from the sample data (ad-hoc approach) (Nichols, 1978). For instance, McConnell and Strand (1981) proposed a proportion of 60%, leading to a variation range of between 6.75 and 11.81 EUR/person/hour. Moving on to protected natural areas, Riera (2000) presented an empirical application of the travel cost approach to measure the value tourists give to the recreation services that those areas provide. On this basis, we adopted herein a value of the recreation time (VRT ) of 8.5 EUR/person/hour. Although we assumed VRT to be the same for all the beaches in order to facilitate the analysis, this value may change to account for a given casuistry (e.g., whether the beaches are crowded urban or pristine natural). Note that the VRT estimate does not depend neither on the leisure activity pursued, nor on the time of the year, as recreation is provided all year-round albeit in different forms.

Fig. 4. Flowchart of the procedure developed to obtain the beach recreation value per square meter in monetary terms.

value per square meter in monetary terms (see flowchart in Fig. 4). In this regard, we combined the data obtained in steps 1 to 3 in order to determine the economic value (EV ) of beach recreation as follows:

1 person hour € EV € year·m2 = m2 ·H year ·VRT hour·person A (1)

(

3.2.4. Economic and accounting beach recreation value per square meter (EV and AV, respectively) The fourth step involves the estimation of the beach recreation

)

(

)

(

)

(

)

where A is the area required by a user; H is the number of effective usable hours per year; and VRT is the value of the recreation time per person. 392

Tourism Management 68 (2018) 387–400

A. Toimil et al.

Table 3 Correction factor derived from location-based beach typologies. Typology 1

Typology 2

Urban beaches

Semi-urban beaches or beaches located in a summer area

1

Table 6 Accounting value per square meter (€/m2) considering per capita income to be measured in constant monetary units, and a discount rate of 4%.

Typology 3

0.66

Location-based beach typologies

Isolated beaches

0.33

Urban Semi-urban Isolated

Since recreation services provided yearly are not a direct indicator of the value of the asset itself, a capitalization of the value of future cash flow series into its present value is required. For that purpose, we obtained the accounting value (AV) of beach recreation by means of:

AV

( ) €

m2

n

=

∑

EV

t=1

(

€

year·m2

1

2

3

4

5

7055 4656 2328

5644 3725 1862

3527 2328 1164

2116 1397 698

353 233 116

Table 7 Correction factors and economic and accounting values for a sample of the Asturian beaches considering per capita income to be measured in constant monetary units, and a discount rate of 4%.

)

(1 + r )t

Beach quality categories

(2)

where EV is the economic value; r is the discount rate; n is the number of years from the present to the time horizon considered (i.e., in this case from 2010 to 2100); and t is the target year. Note that for high values of n and provided that EV is assumed constant, AV can be approximated to EV / r . The ad-hoc nature of this roadmap leads to the inclusion of a set of vulnerability parameters tailored to beach-specific characteristics and services for which additional research work is required. These parameters account for the typology of the beach, its quality, and the number and distribution of accesses (see Table 3 to Table 5). The correction factor derived from location-based beach typologies (Table 3) ranges from 0.33 for isolated beaches to 1 for urban beaches, as the latter comprises most of the year-round demand. On the other hand, five quality categories are established to account for the main beach users’ requirements (Table 4). They account for BF requirements, easy access and parking facilities, lifeguards, cleaning services and beach huts, as well as quality in terms of environment, sand and water, the latter based on the registers of the past four years. While Category 1 consists of the best services and quality indicators (i.e., combining BF, sand and services requirements), Category 5 includes inaccessible, dangerous and/or low-quality beaches. In between, a range of conditions exists, creating variation from 1 to 0.05. Finally, the number of access points is considered because users tend to settle within 300 m from the main access (Jiménez et al., 2007). In simplified terms, this is approached by considering that there is a single access located in the center of the beach that is utilized by most users (Table 5). Assuming a per capita income measured in constant monetary units and a discount rate of 4%, Table 6 shows the accounting value in accordance with location-based beach typologies and the range of quality

Beaches (from W to E)

Locationbased factor

Qualitybased factor

Lengthbased factor

EV (€/year m2)

AV (€/m2)

Navia El Moro Salinas (Luarca) Luarca Salinas San Juan Luanco La Ribera San Lorenzo Cervigon La Ñora Rodiles Borizo Palombina

0.66 0.33 0.66

0.8 0.05 0.5

1 1 1

149.00 4.66 93.13

3725.04 116.40 2328.15

1 1 0.66 1 1 1 0.66 0.66 0.66 0.66 1

0.3 1 0.8 1 0.3 0.8 0.05 0.8 1 0.8 1

1 0.5 1 1 1 0.75 1 1 0.75 1 1

84.66 141.10 180.61 282.20 84.66 169.32 9.31 149.00 139.69 148.90 282.20

2116.5 3527.50 4515.20 7055.00 2116.50 4233.00 232.82 3725.04 3492.22 3725.04 7055.00

categories deemed. As can be observed, all beaches have recreation value although variability is high. For example, among beaches that boast Category 1, those semi-urban and isolated have 34% and 67% less recreational value, respectively, than the urban type. By contrast, isolated beaches with quality rank 5 have approximately 98% less recreation value than the top-ranked ones. Taking the case of Asturias, Table 7 presents the results obtained for some of the study beaches. As shown, the clear majority meet high quality standards (Categories 1–2) and are short-type. The greatest beach recreation value amounts to 7055 EUR/m2 in terms of accounting value, and corresponds to both Luanco and Palombina, in which no correction factors are applied. They are followed by San Juan and San Lorenzo, which both have an

Table 4 Correction factor derived from beach quality categories. Category 1 Safe and easy access. Parking facilities. Existence of lifeguards, cleaning services and beach huts. High-quality sand, environment and water. 1

Category 2

Category 3

Category 4

Safe and easy access. Parking facilities. Existence of lifeguards, cleaning services and beach huts. Medium-quality sand, environment and water.

Considerably safe and easy access. Parking facilities. Existence of cleaning services and beach huts. Medium-quality sand, environment and water.

Sufficiently safe and easy access. Existence of cleaning services. Medium-quality sand, environment and water.

0.8

0.5

0.3

Category 5 Either difficult or problematic access. Lack of services. Low-quality sand, environment and water.

0.05

Table 5 Correction factor derived from length-based beach typologies. Short beaches

Medium-short beaches

Medium-long beaches

Long beaches

L ≤ 1000 m

1000 < L ≤ 2000 m

2000 < L ≤ 3000 m

L > 3000 m

1

0.75

0.5

393

0.2

Tourism Management 68 (2018) 387–400

A. Toimil et al.

Fig. 5. Spatial distribution of the cumulative losses of beach recreation value corresponding to the 75th percentile (box top mark), median (box middle mark), and 25th percentile (box bottom mark) of R2100 (full range) for each of the 57 study beaches (from W to E) by the end of the century under the RCP8.5 scenario.

Fig. 6. Spatial distribution of the differences in the accounting value of beach recreation corresponding to the 25th percentile (box top mark), median (box middle mark) and 75th percentile (box bottom mark) of R2100 (full range) for each of the 57 study beaches (from W to E) by the end of the century under the RCP8.5 scenario. The asterisk (*) in Luarca and Cervigon indicates that by 2100, these beaches will have disappeared.

accounting value beyond 4000 EUR/m2. Taking into account the 57 Asturian beaches, the average economic and accounting values are approximately 134.93 EUR/m2/year and 3375.26 EUR/m2, respectively. Considering the current beach surface for the whole region, this means a flow of 430 million EUR/year (2% of the 2013 GDP) as the annual recreation services provided, and a total recreation value of 11 billion EUR. Regarding the latter, it should be noted that the accounting value is strongly dependent on the discount rate to be applied. The 4% rate we applied in this work is based on EU policy recommendations regarding the assessment of public natural capital for long-time horizons (i.e., 80–100 year), considering intergenerational and sustainability issues (The Green Book, 2003; EC, 2014). However, the capital stock provided by FBBVA (see Section 2) matches an expected economic life of approximately 20–25 years with higher profitability requirements (private investments), leading to the use of higher discount rates. Accordingly, the 11 billion EUR of recreation value at a rate of

4% would be reduced to 6.30 billion EUR (1968.75 EUR/m2) at 7% and to 4.40 billion EUR (1375.10 EUR/m2) at 10%. The sensitivity of the results to the discount rate arises from the sensitivity of the model itself and must be interpreted accordingly. 4. Results and discussion The accounting value of recreation for the Asturian beaches obtained in Section 3.2 was combined with a set of future shoreline recessions calculated in Toimil et al. (2017a) to derive the risk of loss of beach recreation. The results we present herein do not attempt to predict direct damages to the tourism industry but to monetarily estimate the extent of potential losses of beach recreation value as a key indicator in the sector, in case no actions were taken. In this regard, Fig. 5 shows the cumulative losses of beach recreation value by 2100 under the RCP8.5 scenario. The results account for the full range of 394

Tourism Management 68 (2018) 387–400

A. Toimil et al.

Fig. 7. Losses of beach recreation value in Navia, Otur, Luarca, Salinas, San Lorenzo and Rodiles corresponding to the 75th percentile (box top mark), median (box middle mark) and 25th percentile (box bottom mark) of 5-, 10-, 25- and 50-year return period retreat events by the end of the century under the RCP8.5 scenario.

Fig. 8. Expected annual damage in San Lorenzo Beach between 2010 and 2100. The dark shaded band represents the 95% confidence levels; white circles are medians; the black dashed line is the medians slope; black dots are mean values; and black crosses are standard deviations.

current situation, we drew up an index as the quotient between the accounting recreation value in 2100 and now. This index ranges from 0 if the beach has lost all its recreation value, to 1 if the current recreation value is fully maintained. Fig. 6 shows differences in the beach recreation value corresponding to the 25th percentile, median, and 75th percentile of the full range of R2100. By 2100, according to Fig. 6, while Luarca and Cervigon will have all but disappeared, El Moro and La Ribera will exist only under recessions corresponding to the R2100 25th percentile, and their recreation value will only amount to 14% and 28% of the value they currently have, respectively. Losses of recreation value due to extreme retreat events in Navia, Luarca, Salinas, Luanco, San Lorenzo and Rodiles under the RCP8.5 scenario are provided in Fig. 7. The results correspond to the 25th

R2100 (defined in Section 3.1) at each beach, where the central mark represents the median, the lower edge of the box is the 25th percentile, and the upper one corresponds to the 75th percentile. As can be inferred, the 25th percentile, median and 75th percentile of losses will amount respectively to 2364.20, 3183.42 and 4053.68 million EUR for the whole coastline stretch, which will result in 26.26, 35.37 and 45.04 million EUR/year on average. Since losses are limited to the available foreshore width, they cannot extend beyond the disappearance of a beach. This is the case for El Moro, Luarca and Cervigon (see Fig. 5). The combination of high recreation value and large retreats rank Salinas and San Lorenzo at the top of the most affected beaches. Although it appears to be clear that, by 2100, the 57 beaches will only decline in value, in order to know how much they will be worth compared to the 395

Tourism Management 68 (2018) 387–400

A. Toimil et al.

Fig. 9. Expected cumulative damage in San Lorenzo Beach between 2010 and 2100. The dark shaded band represents the 95% confidence levels; white circles are medians; black dots are mean values; and black crosses are standard deviations.

extreme events, any annual change in the average beach position can result in erosion or accretion, denoted herein as positive and negative values, respectively. Thus, the EAD strictly corresponds to the upper side of the graph (positive values of the ordinate axis). Beyond a large inter-annual variability, between 2010 and 2100, there will be an increase in the median equal to 10 million EUR/year (average growth rate of 3.77%). When EAD accumulates over the century, the expected cumulative damage (ECD) is derived (Fig. 9). In such a case, although potential gains of recreation value due to annual average accretion net balances will be possible until 2072, the median values are nothing but losses beyond 2018. The ECD median value is projected to evolve from 76.18 million EUR in 2050 to 237.81 million EUR in 2100 (in accordance with Fig. 5), giving rise to an increase equal to 212.31%. Fig. 10 shows how San Lorenzo Beach will have its recreation value devalued over this century due to ECD. It is shown that there will be a reduction in the accounting recreation value of 15.81% (23.65% and 12.15% for the 2.5th and the 97.5th percentiles, respectively) from 2010 to 2050, and of 35.96% (63.56% and 20.68% for the 2.5th and the 97.5th percentiles, respectively) from 2050 to 2100. Moving to the regional scale, erosion risk is expected to increase rapidly during the twenty-first century. Fig. 11 shows the ECD for the study beaches between 2010 and 2100. It can be observed that any possible gain resulting from the 3000 potential shoreline evolutions can occur beyond 2042 (2019 for median values). Albeit with considerable uncertainty, ECD more than doubles from 2050 to 2100, with an increase in the median value equal to 220% (752% and 169% for the 2.5th and the 97.5th percentiles, respectively). Thus, accrued losses of beach recreation value by the end of the century will amount to 4752.54 million EUR (97.5th percentile), which represents 6.5% of the Asturian capital stock.

percentile, the median and the 75th percentile of 5-, 10-, 25- and 50year return period retreat events, the most widely used for management and planning purposes. It is important to note that return periods correspond to shoreline recession rather than to a hazard. As can be observed, if damages were accrued over the entire century, a 1-in-25-year erosion with a 0.63 ± 0.20 m projected SLR will lead to losses up to 359.46 million EUR in San Lorenzo. Similar, in Salinas, recessions corresponding to the 50th percentile that will occur at least once every 50 years with a 0.63 ± 0.20 m projected SLR will result in recreation value damages of 672.84 million EUR. By 2100, while losses in Navia will be limited by its disappearance under 10-, 25- and 50-year return period recessions, the situation will become even more critical in Luarca. Note that in contrast with what occurs with SLR, beaches might completely overcome a storm event and therefore naturally recover the lost recreation value at any time. However, as occurred in Asturias in the winter of 2014 (see Section 2), cumulative storm effects could damage beaches to the point of preventing them from any eventual recovery. Risk assessments aim to identify the areas most exposed to an impact, improving the understanding of its origin, identifying its potential extent and severity, and assisting in the elaboration of management plans and decision making. Risk must be calculated as the integral over all possible event probabilities and is commonly expressed as expected annual loss (EAL) or expected annual damage (EAD) to assets. Accordingly, we estimated the risk of loss of recreational value in EAD terms from 2010 to 2100 under the RCP8.5 scenario. On this occasion, the EAD was computed for every beach by combining the accounting beach recreation value and the annual beach change (defined in Section 3.2). As an example, Fig. 8 provides the EAD in San Lorenzo, one of the most popular tourist beaches in the region. It is worth noting that given the natural variability of the shoreline evolution process and because of 396

Tourism Management 68 (2018) 387–400

A. Toimil et al.

Fig. 10. Accounting value of beach recreation in San Lorenzo Beach between 2010 and 2100. The dark shaded band represents the 95% confidence levels; white circles are medians; black dots are mean values; black crosses are standard deviations; and the dashed line is the current accounting value of San Lorenzo's recreation.

5. Conclusions and policy implications

and the distribution of the occupancy rates among time slots), the purchasing power, and a small set of calibration parameters that include the typology of the beach, its quality, and its length are the data required for the analysis and may be updated in a timely fashion, or even effortless adapted with any desired change to any other beach in any other region worldwide. It should be noted then that the value of the recreation time per person is not only an ad-hoc parameter that varies among locations, as occurs with the area required by a user (i.e., due to the physical setting or even to the law of supply and demand), but it can also be very sensitive to the socioeconomic juncture at that time (e.g., income increase and inflation). For instance, if the amount of leisure time changes, so will the way a user invests that time. The same thing may occur with the user's perceptions in terms of comfort or even safety. Another critical aspect is the discount rate to be applied. While rates of approximately 4% usually correspond to public assets that require preservation, as they are highly valuable regardless of the time horizon considered, higher rates beyond 6%–7% are often associated with short-term commitments, private opportunity cost and private investments. But there are many more options on the way, and the number of parameters to be subjected to sensitivity analysis is large: changes in climate that may induce changes in temperature and precipitation regimes, changes in population that may lead to changes in beach occupancy rates, changes in beach maintenance that may result in changes in quality categories, etc. So, what is important is that, at the end of the day, combining the required elements within the established framework can help coastal managers to better support decision making and improve future actions. To our understanding, the loss of recreation value estimates should consider not only the coastline retreat driven by SLR (long term) but also those large recessions resulting from extreme events (short term),

Decision making for coastal planning and management requires the use of risk assessment frameworks that account for the combined effect of a range of climatic and non-climatic drivers and their associated uncertainties (e.g., SLR, changes in the likelihood of occurrence of extreme events, and socioeconomic trends difficult to predict). Within this context, we developed a methodology aiming to determine the risk of loss of beach recreation value triggered by climate change and stormerosion in probabilistic terms, enabling a robust quantification of uncertainty. Thus, we tried not only to go into the physical processes involved in coastal erosion, but also to improve the current knowledge relating climate change impacts and risks to the tourism and recreation concepts. In addition, attempts have also been made to bring scientific insights closer to decision makers and users, who must be increasingly aware of the value of what we currently have, and the losses that could be reached without implementing the appropriate climate change risk management and adaptation strategies. Thus, understanding the origin of a threat and the extent of its associated risks, namely loss of beach recreation value, may allow coastal managers to adopt a policy of adaptation or risk reduction, to anticipate, and to even provide other recreation alternatives that can help reduce any collateral impacts on the economy of the region (e.g., loss of property value and tourism revenue). Given the climate-sensitive nature of decisions that come with a long-term commitment, the new paradigm we present herein easily adapts, if deemed necessary (e.g., if more data are available or action learning is improved), enabling more flexibility in its application to other areas. Geographical conditions and social data that allow for the characterization of the recreation season (e.g., sunny days and hours

397

Tourism Management 68 (2018) 387–400

A. Toimil et al.

Fig. 11. Expected cumulative damage for the 57 beaches under study between 2010 and 2100. The dark shaded band represents the 95% confidence levels; white circles are medians; black dots are mean values; and black crosses are standard deviations.

euros worth of beach recreation value, the challenge now is appropriately managing this risk. Although many beaches might naturally adapt, those where inland migration is not possible for geomorphological conditions or those experiencing coastal squeeze due to coastal urbanization could have this natural response constrained. In this regard, there are a myriad of approaches that range from cost-intensive construction of hard defenses to soft sand nourishments, which appear to be one of the preferred adaptation options in response to shoreline retreat (Eurosion, 2004). The efficacy and appropriate timing of one option or any combination thereof will vary among contexts and should be thoroughly analyzed. Thus, where irreversible loss is evident and popular measures become unsustainable, planned strategies that give some scope to natural processes might be worth exploring.

as cumulative storm effects steadily damage beaches and can prevent them from recovery. Besides, it is important to point out that although the 57 beaches under study are assumed to be pocket-type, and thus with consequently negligible gradients in longshore sediment transport, this configuration is not necessarily found in other regions, but the coast may be open or even subject to anthropogenic action such as severe urban pressure and groin construction, which must also be accounted for if it was the case. This is occurring on the Spanish Mediterranean coast, where these structures disrupt the longshore drift sediment transport, depriving down-drift beaches of sediment supply and rapidly accelerating coastal erosion (Eurosion, 2004). While indisputable for any long-term assessment, probabilistic modeling that includes SLR uncertainty is a major pillar in this work, underpinning the risk estimates of loss of beach recreation value. The results give insight into the serious risks a 50-year return period retreat event would have by the end of the century, leading to recreation value losses in Salinas Beach up to 672.84 million EUR. However, the most striking is that also by 2100, even a 5-year return period retreat will lead to the disappearance of beaches (e.g., Luarca). In that event, users would have no choice but to spend their recreation time elsewhere. Moreover, if no actions were taken, regional ECD will increase by approximately 220% between 2050 and 2100, which means that losses could amount up to 6.5% of the Asturian capital stock. Particularly in the case of San Lorenzo, the increase in EAD median value will equal 10 million EUR between 2010 and 2100, leading, by the end of the century, to a median value of ECD amounting up to 237.81 million EUR. In light of such an unprecedented erosion threatening millions of

Author contribution A.T., P. D-S., and I.J.L. developed the concept and the methodology for this research. A.T. performed the analysis and prepared the original manuscript. A.T. and P.C. designed the figures and comments on the results. P. D-S. and I.J.L supervised the whole work. A.T., P. D-S., I.J.L., and P.C. discussed and commented on the manuscript. Acknowledgements The authors would like to acknowledge comments provided by Prof. Roshanka Ranasinghe to an earlier version of this manuscript. This work has been supported by the Spanish Ministry of Agriculture and

398

Tourism Management 68 (2018) 387–400

A. Toimil et al.

Fishery, Food and Environment (MAPAMA). The authors are also grateful to the Government of Asturias for the technical assistance and the data provided.

Ranasinghe, R., Callaghan, D., & Stive, M. J. F. (2012). Estimating coastal recession due to sea level rise: Beyond the Bruun rule. Climatic Change, 110, 561–574. http://dx.doi. org/10.1007/s10584-011-0107-8. Ranasinghe, R., Duong, T. M., Uhlenbrook, S., Roelvink, D., & Stive, M. (2013). Climatechange impact assessment for inlet-interrupted coastlines. Nature Climate Change, 3, 83–87. Riera, A. (2000). Mass tourism and the demand for protected natural areas: A travel cost approach. Journal of Environmental Economics and Management, 39, 97–116. Roca, E., Riera, C., Villares, M., Fragell, R., & Junyent, R. (2008). A combined assessment of beach occupancy and public perceptions of beach quality: A case study in the costa brava, Spain. Ocean & Coastal Management, 51, 839–846. Roca, E., Villares, M., & Ortego, M. I. (2009). Assessing public perceptions on beach quality according to beach users' profile: A case study in the costa brava (Spain). Tourism Management, 30, 598–607. Sarda, R., Mora, J., Ariza, E., Avila, C., & Jimenez, J. A. (2009). Decadal shifts in beach user sand avaiilability on the costa brava (northwestern mediterranean coast). Tourism Management, 30, 158–168. http://dx.doi.org/10.1016/j.tourman.2008.05. 011. Schliephack, J., & Dickinson, J. E. (2017). Tourists' representations of coastal managed realignment as a climate change adaptation strategy. Tourism Management, 59, 182–192. http://dx.doi.org/10.1016/j.tourman.2016.08.004. The Green Book, Appraisal and evaluation in Central Government (2003). Treasury guidance. London: HM Treasury. Toimil, A., Losada, I. J., Camus, P., & Diaz-Simal, P. (2017a). Managing coastal erosion under climate change at the regional scale. Coastal Engineering, 128, 106–122. http:// dx.doi.org/10.1016/j.coastaleng.2017.08.004. Toimil, A., Losada, I. J., Diaz-Simal, P., Izaguirre, C., & Camus, P. (2017b). Multi-sectoral high-resolution assessment of climate change consequences of coastal flooding. Climatic Change, 145(3–4), 431–444. http://dx.doi.org/10.1007/s10584-017-2104-z. US Army Corps of Engineers (1994). Shoreline protection and beach erosion control study. Phase 1: Cost comparison of shoreline protection projects of the US Corps of Engineers. Washington, DC: Water Resources Support Centre. Download link: http://www.iwr. usace.army.mil/Portals/70/docs/iwrreports/94-PS-1.pdf. Wainwright, D. J., Ranasinghe, R., Callaghan, D. P., Woodroffe, C. D., Jongejan, R., Dougherty, A. J., et al. (2015). Moving from deterministic towards probabilistic coastal hazard and risk assessment: Development of a modelling framework and application to Narrabeen Beach, New South Wales, Australia. Coastal Engineering, 96, 92–99. Wong, P. P., Losada, I. J., Gattuso, J.-P., Hinkel, J., Khattabi, A., McInnes, K. L., et al. (2014). Coastal systems and low-lying areas. In: Climate change 2014: Impacts, adaptation, and vulnerability. In C. B. Field, V. R. Barros, D. J. Dokken, K. J. Mach, M. D. Mastrandrea, T. E. Bilir, M. Chatterjee, K. L. Ebi, Y. O. Estrada, R. C. Genova, B. Girma, E. S. Kissel, A. N. Levy, S. MacCracken, P. R. Mastrandrea, & L. L. White (Eds.). Part A: Global and sectoral aspects. Contribution of working group II to the fifth assessment Report of the intergovernmental panel on climate change (pp. 361–409). Cambridge, United Kingdom and New York, NY, USA: Cambridge University Press.

References Aguilo, E., Alegre, J., & Sard, M. (2005). The persistence of the sun and sand tourism model. Tourism Management, 26, 219–231. http://dx.doi.org/10.1016/j.tourman. 2003.11.004. Alexandrakis, G., Manasakis, C., & Kampanis, N. A. (2015). Valuating the effects of beach erosion to tourism revenue. A management perspective. Ocean & Coastal Management, 11, 1–11. Ariza, E., Jiménez, J. A., & Sardá, R. (2008). A critical assessment of beach management on the Catalan Coast. Ocean & Coastal Management, 51, 141–160. BF (2017). Blue Flag's criteria. Download link: http://www.blueflag.global/criteria/ (last accessed: July 2017) . Bird, E. C. F. (1985). Coast changes. New York: Wiley&Sons. Birdir, S., Ünal, Ö., Birdir, K., & Williams, A. T. (2013). Willingness to pay as an economic instrument for coastal tourism management: Cases from Mersin, Turkey. Tourism Management, 36, 279–283. http://dx.doi.org/10.1016/j.tourman.2012.10.020. Blakemore, F., & Williams, A. (2008). British tourists' valuation of a Turkish beach using contingent valuation and travel cost methods. Journal of Coastal Research, 24, 1469–1480. http://dx.doi.org/10.2112/06-0813.1. Bruun, P. (1962). Sea level rise as a cause of shore erosion. Journal of the Waterways and Harbors Division, 88, 117–130. EC (2014). Guide to cost-benefit analysis of investment projects. Economic appraisal too for Cohesion Policy 2014-2020European Commission. Eurosion (2004). Living with coastal eorison in Europe: Sediment and space for sustainability. Part-1 Major Findings and Policy Recommendationsof the EUROSION ProjectDirectorate General Environment European Commission. FBBVA (2017). Acumulación y productividad del capital en España y sus comunidades autónomas en el siglo XXI. Informes 2017 Economía y Sociedad. ISBN: 978-84-9293770-7. (Download link: https://w3.grupobbva.com/TLFU/dat/DE_2017_Ivie_Inf_ Stock_1964-2014.pdf. Hinkel, J., Nicholls, R. J., Tol, R. S. J., Wabg, Z. B., Hamilton, J. M., Boot, G., et al. (2013). A global analysis of erosion of sandy beaches and sea-level rise: An application of DIVA. Global and Planetary Change, 111, 150–158. http://dx.doi.org/10.1016/j. gloplacha.2013.09.002. Houston, J. R. (2013). The economic value of beaches – a 2013 update. Shore and Beach, 81, 3–10. IPCC (2014). Summary for policymakers. In C. B. Field, (Ed.). Climate change 2014: Impacts, adaptation and vulnerability. Part A: Global and sectoral aspects. Contribution of working group II to the fifth assessment Report of the intergovernmental panel on climate change (pp. 1–32). Cambridge, United Kingdom and New York, NY, USA: Cambridge University Press 2014. Jackson, A. C., & McIlvenny, J. (2011). Coastal squeeze on rocky shores in northern Scotland and some possible ecological impacts. Journal of Experimental Marine Biology and Ecology, 400, 314–321. Jiménez, J. A., Osorio, A., Marino-Tapia, I., Davidson, M., Medina, R., Kroon, A., et al. (2007). Beach recreation planning using video-derived coastal state indicators. Coastal Engineering, 54, 507–521. Jongejan, R., Ranasinghe, R., Wainwright, D., Callaghan, D. P., & Reyns, J. (2016). Drawing the line on coastline recession risk. Ocean & Coastal Management, 122, 87–94. King, P. G., McGregor, A. R., & Whittet, J. D. (2016). Can California coastal managers plan for sea-level rise in a cost effective way? Journal of Environmental Planning and Management, 59(1), 98–119. King, P., & Symes, D. (2003). The potential loss in gross national product and gross state product from a failure to maintain California's beaches. San Francisco State University. Loomis, J., & Santiago, L. (2013). Economic valuation of beach quality improvements: Comparing incremental attribute values estimated from two stated preference valuation methods. Coastal Management, 41, 75–86. http://dx.doi.org/10.1080/ 08920753.2012.749754. McConnell, K. E., & Strand, I. (1981). Measuring the cost of time in recreation demand analysis: An application to sportfishing. American Journal of Agricultural Economics, 63, 153–156. Morgan, R. (1999). A novel, user-based rating system for tourist beaches. Tourism Management, 20, 393–410. Moss, R. H., Edmonds, J. A., Hibbard, K. A., Manning, M. R., Rose, S. K., van Vuuren, D. P., et al. (2010). The next generation of scenarios for climate change research and assessment. Nature, 463, 747–756. Nichols, L. M., Bowes, M., & Dweyer, J. F. (1978). Reflective travel time in travel-costbased estimates of recreation use and value. Forestry Research Reports 78–12, (AES, University of Illinois). Parsons, G. R., & Powell, M. (2001). Measuring the cost of beach retreat. Coastal Management, 29, 91–103. Phillips, M. R., & Jones, A. L. (2006). Erosion and tourism infrastructure in the coastal zone: Problems, consequences and management. Tourism Management, 27, 517–524. http://dx.doi.org/10.1016/j.tourman.2005.10.019.

Alexandra Toimil is Postdoctoral Research Associate at the Environmental Hydraulics Institute “IHCantabria”. She is working in the field of climate change, adaptation and risk reduction in coastal areas. Her research interests focus on impact assessment, risk assessment and risk management, for both the natural and the socio-economic systems, and including the most concerning sectors such as ecosystems, agriculture, housing, industry and tourism among others.

Pedro Díaz Simal is Research Associate at the Environmental Hydraulics Institute “IHCantabria” and Associate Professor at the Universidad de Cantabria. His working fields include risk assessment, applied economics to transport and tourism, and environmental economics, the latter mainly focused on environmental project assessment and ecosystem services valuation.

399

Tourism Management 68 (2018) 387–400

A. Toimil et al. Iñigo J. Losada is the Research's Director of the Environmental Hydraulics Institute “IHCantabria” and Full Professor of Hydraulic Engineering at the Universidad de Cantabria. He is past coordinating leading author of the IPCC's Fifth Assessment Report, leading the chapter on impacts, vulnerability, risks and adaptation in coastal areas and a contributing author to the chapter on South and Central America. His interests and expertise include marine climate, coastal dynamics, coastal structures and coastal protection modeling, climate change impacts assessment and adaptation, as well as extreme risk assessment and mitigation.

Paula Camus is Postdoctoral Research Associate at the Environmental Hydraulics Institute “IHCantabria”. Her research activities are mainly related to the application of data mining techniques to characterize wave climate, develop downscaling methodologies for marine climate, and assess climate change coastal impacts.

400