Climate change threats to two low-lying South African coastal towns : Risks and perceptions

HOW TO CITE: Fitchett JM, Grant B, Hoogendoorn G. Climate change threats to two lowlying South African coastal towns: Risks and perceptions. S Afr J Sci. 2016;112(5/6), Art. #2015-0262, 9 pages. http://dx.doi.org/10.17159/ sajs.2016/20150262 Climate change poses a considerable threat to low-lying coastal towns. Possible risks include flooding induced by sea-level rise, increased discomfort from changes in temperature and precipitation, more frequent extreme events, biodiversity shifts, and water shortages. For coastal towns that attract many tourists, these threats can have far-reaching economic effects and may compromise the continued viability of the tourism sector. A growing number of studies are being published on the inter-relationship between climate change and tourism in the global North. As yet, little equivalent research has been conducted in developing countries with economically significant tourism sectors. This paper presents a mixed-method pilot study on two adjacent coastal towns, St Francis Bay and Cape St Francis, in the Eastern Cape Province of South Africa. We explored the climate change threats in this region, and perceptions of these threats within the tourism sector. The tourism climate index results showed that the towns are climatically well suited to tourism, but a decrease in these index scores between 1978 and 2014 suggests that climate change experienced in recent decades has detrimentally affected tourist comfort. A digital elevation model sea-level projection for the towns indicated a high risk of sea-level induced flooding by 2050, particularly for properties along the coastline. Interviews with tourism establishment respondents showed that people are aware of climate change threats, yet little adaptation is forthcoming. Rather the government is deemed responsible for adaptation, despite its limited capacity. A disjuncture therefore exists between the perceived severity of risk and the risk that is evident from scientific analyses. This gap results in poor planning for the costs associated with adaptation.


Introduction
The success of tourism in a particular location depends largely on climatic conditions. 1 Climate is commonly used as a descriptive factor when marketing destinations, and thus selection between locations often includes consideration of their relative climate suitability for preferred tourism activities. 2 Intra-annual patterns in temperature, cloud cover, wind, and rainfall influence the length and timing of peak tourist seasons. 3][6] Climate change is increasingly identified as a significant threat to tourism, because of the reliance of tourism on predictable contemporary climatic conditions and seasonality. 6,7Climate change is likely to affect the seasonality of tourism, and may threaten the infrastructure and natural environment required to accommodate and entertain tourists. 8Predictable and ameliorable weather is critical to the feasibility of outdoor activities. 9e increased variability in weather projected under climate change scenarios, together with associated risks of sea-level rise, render coastal destinations particularly vulnerable. 9It is predicted that by 2050, the global tourism sector will experience substantial climate change impacts, with fluctuations in tourist numbers as tourists opt for better-suited locations. 10The issue of climate change is therefore being integrated into sustainable tourism development initiatives and research. 11,12To attain long-term sustainable development, the tourism sector must adapt to a range of projected climate change impacts. 13 the global South, a lack of integration between climate change and tourism research is of concern because it limits future adaptation potential. 14'Adaptation' can be defined for the purposes of this study as '…the ability of a unit (e.g. a tourism operator or a community) to transform its structure, operations or organisation to survive under changes threatening its existence and success' 15 .
South Africa has a growing tourism sector, which contributes considerably to the local economy. 16The country has a wide range of tourism attractions, many of which rely on the comparatively good climate of the region, 17 characterised by warm weather and clear skies.A large proportion of the sector involves beach and nature based tourism, which rely heavily on continued ameliorable climatic conditions. 18In addition, South Africa has numerous small towns that are largely economically dependent on tourism.Although there is some variation in tourist offerings, many involve outdoor attractions -including beaches, game drives, water sports and fishing. 19uth Africa is predicted to experience temperature increases as high as 4 °C by the year 2100, as well as changes in the timing, amount, and severity of precipitation; changes in wind direction and strength; and rising sea levels. 20,21Given that tourists are highly mobile, even a few weeks of poor weather during the tourist season can affect the long-term viability of the sector in terms of destination attractiveness. 22Information on climatic conditions is easily accessible for a range of possible tourist locations, both locally and internationally, and tourists inevitably select destinations that are relatively less risky in terms of the weather. 23Climate change therefore has the potential to economically cripple tourist establishments and activities, as well as tourism-led towns -and indeed the South African tourism economy -if sufficient adaptation is not achieved.
It is insufficient to implement adaptation measures only after climate change impacts become evident; forward planning is essential. 24,25Policy documents indicate that such planning and adaptation to climate change threats are not forthcoming in the study region.We explored these climate change threats and people's perceptions within the tourist sector regarding the level of risk, to determine the levels of preparedness.This study is of value in determining the likely impacts of climate change on the tourism sector in the study areas, and broader patterns of response in terms of climate change adaptation in the tourism sector regionally.

Study site
Our study examined two adjacent coastal towns, St Francis Bay and Cape St Francis, in the Eastern Cape Province of South Africa (Figure 1).These towns provide examples of tourism-reliant locations that face multiple climate change threats.Located on a gently sloping coastal plain, the towns are at risk of sea-level rise, which could encroach on the beach and eventually inundate much of the developed area.This area would include tourism establishments and roads and bridges, prohibiting access to the towns. 26,27Changes in temperature, rainfall and wind are also of concern.
Both towns provide predominantly outdoor tourist attractions and have distinct peak tourist seasons associated with warm temperatures and clear skies. 20Additional concerns for these towns include the Kromme River Estuary and Canal, which pose flood risks, and the position of nearby sand dunes, which could result in dust storms. 20,27 Francis Bay covers a land area of 10.04 km² at an altitude ranging from 0 masl to 34 masl. 28Cape St Francis comprises a smaller area of 4.38 km² at altitudes from 0 masl to 15 masl. 28Census 2011 data give the population of St Francis Bay as 4933 people, and 342 in Cape St Francis. 29series of canals and waterways, ideal for water sports, attracts tourists to St Francis Bay.A large number of holiday homes in both towns are rented out to tourists.Many are located along the canal system, an area of St Francis Bay commonly known as 'Little Venice'. 30A nature reserve connects the two towns and provides a popular eco-tourism destination. 31e towns are situated in a zone that receives year-round rainfall, although more rain falls during the winter months. 32,33Mean air temperatures range from 12.3 °C to 14.2 °C in winter and 18.9 °C to 20.8 °C in summer. 32,34here is marked seasonality in wind trajectories.In winter, the wind blows from west to southwest, driven by the northward trajectory of the westerly belt, and in summer it blows easterly to southeast, when atmospheric circulation is dominated by the tropical easterlies. 33,35

Methodology
We adopted an interdisciplinary mixed-method approach in this study.We wanted to determine the climate change risks to the towns, and to explore accommodation establishment proprietors' perceptions of those threats.
Climate change risks to the towns were classified on the basis of the towns' proximity to the sea and the reliance of tourism on ameliorable weather to encourage outdoor activities.The risk of sea-level rise was determined using a digital elevation model (DEM) with overlaid sealevel projections.The climatic suitability of the region for tourism was determined using the tourism climate index (TCI).
We probed the perceptions of accommodation establishment proprietors through individual face-to-face interviews in which open-ended questions were posed, and respondents were also asked to rank their concerns on a Likert-type scale.Qualitative data from these interviews were interpreted using thematic analysis.All respondents were presented with a consent letter before the interview commenced, indicating the purpose of the study and assuring them of confidentiality.The ethics requirements of the University of the Witwatersrand were adhered to prior to the commencement of this study.
Source: Authors' map drawn for the study

Digital elevation model
To model sea-level rise for St Francis Bay and Cape St Francis, and to plot regions likely to be affected by consequent flooding, a DEM for the region was created.The input data were acquired from Viewfinder Panoramas using the DEM World Coverage Map (2014).Raw data were acquired from the Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER), an imaging instrument on board the Terra satellite. 36,37ASTER Global DEM Version 2, produced at a 3 arc-second resolution (~90 m), was used.
To perform an initial sea-level rise vulnerability assessment, we used a simple inundation model known as the 'bathtub approach'. 38This method maps land that is adjacent to the sea and is situated below a given elevation threshold to project inundated areas. 39The mapping process uses a deterministic line to indicate inundation, relative to the sea-level rise and ground elevation. 40This approach does not account for coastal dynamics including wind, waves or tide, and therefore excludes the heightened risk of storm surges. 40,41However, it facilitates the visual identification of locations at highest risk for flooding directly from sealevel rise, and provides a projection of the 'best-case scenario'.
The inundation levels were created using the raster calculator function in the spatial analyst tool in ArcGIS®.Sea-level rise scenarios were projected for the years 2050 and 2100, using sea-level rise projections of 0.4 m for 2050 and 1.6 m for 2100, [42][43][44][45] based on an average rate of change of 0.3 mm/year along the south coast of South Africa. 46This tested the Boolean statement at each cell of the DEM (interrogating whether the value within the cell was less than or equal to 0.4 m and 1.6 m, respectively).

Tourism climate index
The TCI is a numerical model that was developed to evaluate and compare tourist destinations, based on the suitability of their climates for tourists. 47,48aily climate data for the closest registered weather station, Port Elizabeth, were sourced from the South African Weather Service for the period 1978-2014 (the total period for which complete datasets were available).
To calculate the TCI, five main sub-indices were used, comprising seven climatic factors 47 (Table 1).These factors were each given a weighting (based on work by Perch-Nielsen et al. 48) and each climate variable was rated. 48We then used the standard TCI formula 47,48 : Mean annual and mean monthly TCI scores were calculated for the period 1978-2014, based on mean climate variables.This calculation determined the long-term climatic suitability of the region, whilst accounting for interannual climate variability for the southwest Indian Ocean because of El Niño Southern Oscillation, Quasi-Biennial Oscillation, and the Indian Ocean Dipole. 49In addition, changes in annual TCI scores were explored for the period 1978-2014 using linear regression, to detect progressive changes in the climatic suitability for tourism.The TCI scores were classified according to climate suitability (Table 2).

Interviews
Semi-structured interviews with respondents, who were all tourism accommodation proprietors, were conducted in St Francis Bay and Cape St Francis.During the period of fieldwork, 57 establishments existed in St Francis Bay (Figure 2) and 31 in Cape St Francis (Figure 3).
Among these 88 establishments, we interviewed 36 respondents in St Francis Bay and 17 respondents in Cape St Francis (a total of 53 people).Interviews were conducted during the off-season period of September 2014.The interview schedule included questions relating to the proprietors' understanding of climate change, their concerns regarding climate change risks to their establishments, their experiences of weather-related damage in recent years, and their climate change adaptation strategies.'Perception' is defined in this study as 'the process of interpreting sensory information received through environmental stimuli and actions taken in responding to stimuli' 50 .
Accommodation establishments were contacted telephonically prior to the interview to arrange a suitable time, and the interviews were later conducted in person.The interviews were transcribed and coded according to key themes.The frequency of the appearance of these key themes was then explored in relation to the geographic locations of each of the accommodation establishments, mapped using ArcGIS ® .

Digital elevation model sea-level rise projection
Many accommodation establishments are situated at low elevations along the coastline.Sea-level rise is thus a particular concern for the towns we studied, both directly through the flooding of accommodation establishments and indirectly through damage to local infrastructure, which impedes access for tourists.The results of our DEM predict a considerable reduction of the beach area, with extensive coastal squeeze, by 2050, with the worst effect being predicted for the Sea Vista area of St Francis Bay (Figure 4).
Sea-level rise projected for 2050 is modelled to result in the permanent opening of the Kromme River Estuary, with a resultant increase in the salinity of the estuary and a heightened flood risk for the Kromme River.
The artificial spit which currently protects the canal system from salt water incursion is projected to be inundated by 2050, compromising freshwater supply to the towns.By 2100, the DEM projects inundation of low-lying regions of the two towns, including areas in which tourism accommodation establishments are located.The Marina Glades area is projected to be at the highest risk area of inundation by 2100.
Our DEM projections indicate that the Santareme area of St Francis Bay has the lowest threat of inundation.Of direct relevance to the tourism accommodation establishments, despite the projected high risk for Sea Vista, only one establishment is located in that region.By contrast, Marina Glades has 12 establishments that may be affected; however, only two of these are at high risk as they are located near the spit.
Cape St Francis is similarly projected to lose almost half of its beach area by 2050 owing to coastal squeeze.The town is situated relatively further inland, thus none of its accommodation establishments appear to be at risk under the 2050 sea-level rise projections.DEM sea-level projections for 2100 suggest partial inundation of the town, centred behind the Seal Point Headland.Three accommodation establishments are located behind this headland, and according to our projections are likely to be affected.Our DEM projections also indicated that by 2100 the beach will be completely inundated, which would remove the primary tourist attraction of the town.It would also heighten flood risk for the 11 beachfront accommodation establishments.
Sea-level rise additionally results in an increased risk of storm surges for low-lying areas. 25,51Furthermore, climate change projections indicate changing wind patterns, and a southward trajectory of midlatitude cyclones, both of which would further heighten the risk of storm surges. 20,49Storm surges pose the greatest threat to coastal accommodation establishments, but severe events could also result in considerable damage to town infrastructure.Flooding related to storm surges has already resulted in considerable damage to St Francis Bay during 1996 26 and 2007 52 .Hence, our DEM projections likely provide an underestimate of the flooding potential for the region and can be regarded as a 'best-case' scenario.

Tourism climate index scores
The annual TCI scores that we calculated for the region over the period 1978-2014 ranged from 73 to 86.These scores represent 'very good' to 'ideal' climatic suitability for tourism, respectively.The long-term annual average TCI score for the period 1978-2014 was 80, categorised as 'excellent' (see Table 2).An identical score was obtained for the shortterm average for the period 2009-2014.
The monthly TCI scores for the region, averaged between 1978 and 2014, peaked during the local summer months (December to February).This pattern is consistent with the accommodation establishments' recorded periods of peak tourist arrivals in the two towns (Figure 5).The pattern is notable from a climatic perspective, as the region is located within the year-round rainfall zone and is thus less likely to experience a unimodal distribution. 53The variation does, however, occur within a small range of TCI scores, all categorised as either 'very good' or 'ideal'.As evidenced by the equal TCI scores for the 35-year and 5-year calculation periods, there has been no considerable change in the climatic suitability of the region we studied.Over the period 1978-2014, the inter-annual variability in TCI score was more pronounced than any consistent long-term trend.Notably low TCI scores were evident for 1996, 2011 and 2012, and particularly high scores were evident in 1987 and 2013 (Figure 6).A weak trend towards lower TCI scores for the region, at a drop of 0.3 units per decade, was observed.The climate suitability of the region for tourism is therefore not as great a threat as sea-level rise.However, the trend of a slow drop in TCI scores indicates the importance of continual monitoring of changes in meteorological variables in the region.This would help in making adaptation plans to address these changes before they detrimentally affect the region's ability to attract tourists.

Climate change perceptions
The interview data showed that 35 proprietors of tourism accommodation establishments believed that climate change was a cause for concern.Among all 53 respondents, 42 proprietors believed they were already experiencing the effects of climate change, and the remaining 11 felt these effects will be experienced only in 100 years' time.Respondents from establishments located closest to the coast showed more concern for climate change than those further inland.
Respondents from establishments situated along the Kromme River estuary and in the Marina Glades canal system had a medium to high level of concern regarding the threat of climate change.This finding was understandable given the risk of these water bodies flooding.Respondents whose establishments were relatively protected from possible coastal flooding reported fairly low levels of concern.They were situated at a higher altitude than the rest of the town, with large boulders along the beach which could provide defence against flooding.Thirteen respondents, with accommodation establishments scattered across the study area, did not believe that climate change would negatively affect their establishments.
More than half of the respondents (35) believed that climate change would negatively affect the towns themselves.Many of these respondents had been affected by or been within close proximity to the fires which had destroyed numerous houses on the canals in 2012.Some had also been affected by floods that occurred near the St Francis Links Golf Course in 2011.Owners of properties located close to the coast in Cape St Francis largely agreed that the town is vulnerable to the negative effects of climate change.
Fourteen respondents had experienced damage to infrastructure as a result of extreme weather events during the past 5 years.These events included a bridge in St Francis Bay being washed away four times within a 2-year period, compromising access to some of the tourism establishments.The establishments that had experienced flooding were located along the Marina Glades canal system and along the coast near the Santareme area of St Francis Bay.Cape St Francis respondents had also experienced flooding along the main road, which washed away the Sand River Bridge, hindering access to the town.Certain establishments had experienced direct problems from heavy rainfall, with water leaks causing damage to their accommodation rooms.The majority of respondents (44) felt that climate change would affect their establishments in the future.Most respondents mentioned flooding, with the threat of bridges being frequently washed away.This raised concerns about a loss of revenue.
Sea-level rise was perceived to be the largest threat for beachfront properties, with owners fearing flood damage or the infrastructure collapsing entirely if foundations become damaged.Beach erosion was mentioned, with respondents worried about coastal retreat.At the town level, concerns about infrastructure were centred on damage to roads.Overall, respondents said that the greatest impact of climate change would be a decrease in the number of guests and changes in reservation patterns.Concern was raised by respondents that climate change would lessen the ability of the towns to offer tourism activities, including fishing and hiking, which would reduce their competitive advantage relative to other coastal towns.
When asked who should be responsible for mitigating the negative effects of climate change, many respondents felt that government should be responsible for the future provision of climate change adaptation plans.They felt these plans should include the provision of information about climate change adaptation and covering the costs involved.They believed that the lack of information from local government was an indication of the minimal extent of climate change risks.A number of respondents were concerned about hotter, more humid weather, and had installed airconditioning systems to make their establishments more comfortable for tourists under such conditions.The promotion of tourism during winter months or during adverse weather conditions was being pursued by some establishments, through increasing the range of indoor activities available, such as lounge areas with digital satellite television, DVD players, Wi-Fi, and a selection of books and board games.
Although respondents demonstrated a good understanding of climate change threats, they appeared to underestimate the severity.None of the respondents mentioned relocating their business in response to sealevel rise.Rather, sea-level rise was classified as a town-level issue, to be addressed by government through the use of sea defences, including dolosse (interlocking concrete blocks) in the port of St Francis Bay.Some respondents believed that the effects of climate change are distant, and that they would no longer be alive by the time such change occurs.Others believed that adaptations would only become necessary when climate change directly affects their businesses.This view was interesting to us, because of the regular flooding which participants said they have already experienced, with considerable damage to individual establishments.However, such opinions might be influenced by the high cost of infrastructural change, or a belief that insurance will pay for flood damage.

Discussion
Our research identified climate change threats and risks to coastal tourism, and compared this reality with the perceptions of proprietors of tourism accommodation establishments.We examined the vulnerability of the coastal area to climate change in the context of the broader tourism sector.Our DEM sea-level rise projections indicated considerable risk to beach attractions in both towns by 2050, and risk to the infrastructure and to accommodation establishments in the towns by 2100.Accommodation establishments located closest to the coastline or inland water are at the highest risk, although a combination of natural and man-made barriers protects certain properties within these areas.
The TCI results confirmed that the climate of the region is presently very well suited to tourism, particularly during the summer months.The small range of scores stems from the use of the international TCI model, which does not account for the relatively ideal climates throughout South Africa compared with much of the northern hemisphere.The greatest detriment to the TCI scores are rainfall and cloud cover, both of which predominate in the winter months.Notably, temperatures during winter do not detract from the TCI score.However, a weak downward trend in scores from 1978 to 2014 suggests that climate change might already be altering the climate suitability of the region for tourism.This trend is predominantly driven by changes in precipitation in the region over the 35-year period.
Although summer temperatures are currently within the bracket of climate suitability, continued increase would further detriment the TCI scores for the region if the comfort threshold is exceeded. 53Because of the towns' proximity to the ocean, as well as warm water from the Agulhas current retroflection 54 and a large area of inland water, 55 such temperature increase would be associated with a proportional increase in humidity.This would further decrease the region's TCI scores.
Comparing the perceptions of respondents to the results of our TCI and DEM analyses, a low level of concern amongst respondents regarding sea-level rise is notable.Based on our findings and other studies, 24 we strongly recommend that accommodation establishment proprietors in St Francis Bay and Cape St Francis should consider implementing their own mitigation measures as a matter of urgency.
Little information on climate change adaptation plans for the study sites is available from the local municipality. 56,57This lack of concern bodes ill for future adaption and mitigation, particularly given the perception by respondents that government is responsible for and will provide adaptation that is sufficient to protect the local tourism sector from the threat of climate change. 58,59spondents were particularly concerned about changes to the environmental comfort of their patrons.This finding is interesting, as the DEM and TCI results indicate that sea-level poses a greater threat in a shorter time than changes in the climatic suitability of the region for tourism.Respondents acknowledged that sea-level rise may pose a threat, but they did not perceive the immediacy or extent of the problem -namely that their establishments could be severely damaged or entirely destroyed within the next 50 to 100 years. 60,61Furthermore, many respondents said that in 50 to 100 years they will no longer be active in the tourism sector and therefore this is not an issue they will personally face.
Little consideration is made for the long-term sustainability of the tourism sector in the region.Respondents appear to be making cost-benefit assessments regarding the implementation of adaptation measures, by comparing the cost of specific forms of mitigation against the probability of a climate threat occurring, whilst considering additional benefits or detriments of such measures should the climate threat not be realised.They demonstrate willingness to adapt to changing meteorological conditions by installing air-conditioning to improve comfort during hot days, and satellite television and Wi-Fi to provide alternative activities on increasingly frequent rainy days.These climate events are more tangible than sea-level rise, given the recent occurrence of heat waves, cold fronts, and rainy days.Such adaptation measures also improve the amenities of establishments under any climatic conditions, which will afford visitors greater relaxation and business facilities.Finally, they are relatively low-cost adaptations.Therefore, the proprietor is likely to recoup the costs of such adaptation measures over a short period, with no harm should climate change not occur in the projected period.
Our study showed that the respondents were not yet investing in infrastructural changes to address flooding induced by sea-level rise.
Although floods have occurred in the recent past, they were discrete events that are often assumed to be freak accidents rather than indicative of progressive change.Owing to substantial fluctuations in the return rates of flood events and the heightened variability under climate change, there is no certainty that a flood could occur the next year -or even in the next 10 years.Therefore, the cost outlay to make infrastructure changes might not be offset by a direct or immediate benefit. 25Infrastructural changes, such as building retaining walls, are considerably more expensive than installing satellite television, with potentially no payback for the current proprietor.Furthermore, unlike satellite television and Wi-Fi, which contribute positively to the establishment under current conditions, infrastructural changes could negatively affect the amenities of an establishment by blocking sea views.Finally, as discussed above, many respondents said the government is responsible for infrastructural changes to prevent flooding induced by rising sea levels.Such action by the state would render personal retaining walls and other infrastructural adaptations unnecessary and would mean poor strategic spending by individuals.
Although such decisions can be understood from the perspective of proprietors' perceptions of the contemporary climate, they are not consistent with modelled climate change risks for the region.The results of our DEM sea-level rise projections indicate that sea water incursion, with further flood risk from storm surges, is likely to occur by 2050.This requires urgent action to improve infrastructure to adapt to the rising sea levels.Climate models for the region 20,21 and the results of our TCI time-trend analysis suggest that decreases in climatic suitability of the region will take longer to be detected by tourists, and thus require less urgent adaptation.Although the adaptation measures that the respondents are already implementing are relatively cheap and provide immediate additional benefits, they are not a substitute for infrastructural adaptation.They would be insufficient if an establishment is flooded, if the beach retreats, or if the arterial roads are flooded and this prevents tourist access. 24

Conclusion
This research quantified the nature and extent of two predominant climate change threats and risks to the tourism sector in St Francis Bay and Cape St Francis.These main threats are sea-level rise and climatic suitability for tourism.Climatic suitability for tourism was explored over the historical period of the three most recent decades, using the TCI, with future trends inferred on the basis of climate forecasts.For sealevel rise, the threats were predicted using DEMs.The results showed that sea-level rise poses a serious threat to sustained tourism in the region, whereas a drop in the climatic suitability of the region is a less immediate concern.
Tourism accommodation establishments in the region have made smallscale adaptations to these threats, but these predominantly focus on climatic suitability.The reasons for this approach are the relatively low cost of such adaptations and the perception that government is responsible for larger-scale infrastructural changes to mitigate the negative effects of sea-level rise.Many respondents said they will not witness in person the future negative effects of climate change in the region.In addition, none appeared to be considering relocating because of any current risks or threats.Should mitigatory actions not be undertaken sufficiently by government and tourism accommodation establishments, the tourism sector in the region will likely be negatively affected.The current adaptation measures are inadequate to protect the beaches and prevent infrastructural damage.Over time, this will result in individual establishments losing their competitive advantage to regions with a lower flood risk or those with more robust and proactive adaptation plans.
South Africa has a varied climate and natural environment, attracting a range of tourists with a large spectrum of interests.Copyright of South African Journal of Science is the property of Academy of Science of South Africa and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission.However, users may print, download, or email articles for individual use.

Figure 1 :
Figure 1: Map indicating the location of study sites.

Figure 2 :
Figure 2: Map indicating the locations and types of tourism accomm odation establishments in St Francis Bay.

Figure 3 :
Figure 3: Map indicating the locations and types of accommodation establishments in Cape St Francis.

Figure 4 :
Figure 4: Digital elevation model illustrating predicted sea-level rise for both towns for the years 2050 and 2100.(a) Terrestrial altudinal gradient with contemporary sea level (0 masl) and (b) sea-level rise (SLR) induced inundation relative to terrestrial area.

Figure 5 :
Figure 5: Mean monthly tourism climate index (TCI) scores for 1978-2014 for the region (St Francis Bay and Cape St Francis).

Table 2 :
Rating categories of the tourism climate index

Table 1 :
Components of tourism climate index (TCI) http://www.sajs.co.za Volume 112 | Number 5/6 May/June 2016 Source: Authors' map drawn for the study May/June 2016 of this pilot study provide initial insight into climate change risks and perceptions in low-lying coastal towns that rely heavily on beach and outdoor attractions.Although inland tourism destinations experience different climate change threats, the ambivalence of respondents to climate change threats in St Francis Bay and Cape St Francis might well be shared.In general, the sustainability of the tourism sector in South Africa relies on adequate and timely adaptation to climate change threats, so as not to lose tourists to destinations that become better climatically suited.Such adaptation requires an understanding of the nature and extent of climate change threats, and appropriate measures to protect individual establishments and towns.There is an urgent need for further studies in this line, especially given the dominance of the tourism sector in the South African economy.Our pilot study provides a valuable methodological framework for continued research.