Impacts of jellyfish presence on tourists’ holiday destination choices and their willingness to pay for mitigation measures

Recurrent jellyfish blooms in tourist areas have negative effects on coastal and national economies. We investigate how jellyfish presence affects holiday destination choice and how the adoption of mitigation measures can reduce impacts through analyzing tourists’ preferences. A Discrete Choice Experiment approach was adopted to assess tourists’ willingness to pay for different measures. Results indicate that in scenarios of increased jellyfish numbers, 24–40% of tourists might not choose to return to a destination affected by jellyfish. Results also indicate that the adoption of measures such as jellyfish information provision, warning flags, health service presence or the deployment of jellyfish exclusion nets could reduce the proportion of non-returning tourists by 66–83%. Regarding preferences, tourists are willing to pay higher amounts for measures that provide the highest protection. We conclude that the implementation of preventive management plans should be considered by coastal administrations to promote a sense of safety among tourists and residents alike.


Introduction
The effects of current and future climatic conditions have an impact on the tourism industry and on tourists themselves. The movement of tourists, the destinations they choose or the range of activities available to them is increasingly conditioned by climatic conditions (G€ ossling et al. 2012;G€ ossling and Hall 2006). Rising temperatures and the increased frequency of extreme weather events are expected to change the capacity of the natural environment to provide the range of recreational and leisure opportunities available for society (Runting et al. 2017). Specifically, coastal areas might be particularly vulnerable to climate change due to sea level rise, ocean acidification, marine heat waves or the increase of invasive species (Canepa et al. 2013;Purcell, Uye, and Lo 2007;Richardson et al. 2009). At the same time, coastal areas are hotspots for tourism that attract increasing numbers of visitors year after year while supporting a wide range of businesses and people. In the US, in 2017, 2.4 million people were employed by marine-based tourism and recreation, contributing approximately $124 billion in GDP (NOAA 2017). In Europe, coastal and maritime tourism is the largest maritime activity and employs almost 3.2 million people, generating a total of e183 billion in gross value added and representing over one third of the maritime economy (EC 2020).
One of the potential consequences of the combination of climate change and increased anthropogenic pressure on the marine environment is the increase in jellyfish populations; however no clear conclusions have been reached on whether jellyfish populations are rising (Condon et al. 2012(Condon et al. , 2013Richardson et al. 2009). Despite the lack of a solid decision, social and scientific concerns about jellyfish blooms have significantly risen over the past decades. Whilst the number of scientific publications on jellyfish have approximately doubled each decade since 1941; news reports have dramatically increased by over 500% over the past two decades, often with alarmist headlines (Condon et al. 2012;Vandendriessche et al. 2016). Regardless of the conclusion as to whether jellyfish populations are increasing, what will become apparent is that the combination of a growing human population (UN 2019) and the rise in the use of the marine environment will lead to higher encounter rates between jellyfish and humans with all the associated impacts. The occurrence of jellyfish blooms or outbreaks, i.e. the presence of very high numbers of jellyfish in a particular area at a particular time, has a range of associated consequences for society. Among others, they are known to have negative impacts on the fishing and energy sectors (Lee et al. 2006;Patrice, L eo, and Gaby 2011), however, most noticeable are the negative effects on the recreational use of the coast, encompassing coastal tourism (Gershwin et al. 2010;Ghermandi et al. 2015; G omez and Guti errez-Hern andez 2020; Kontogianni and Emmanouilides 2014). The recurrent presence of large numbers of jellyfish around coastal areas can have detrimental impacts for the economy of those areas (Graham et al. 2014), as it has been acknowledged that coastal areas affected by jellyfish outbreaks have a lower tourist appeal (Purcell, Uye, and Lo 2007). In addition, alarmist headlines in national and international press outlets often fuel the negative perceptions around those areas (DM 2018;UH 2018aUH , 2018bVandendriessche et al. 2016), potentially multiplying the negative economic impacts. It is, therefore, crucial to design and implement management measures that mitigate potential jellyfish impacts. Importantly, the implementation of specific measures can play a fundamental role in promoting a feeling of safety and protection among tourists when enjoying coastal areas, therefore minimizing the impacts of jellyfish presence both for coastal users and on the economy of the area.
This study focuses on the Mediterranean Sea, a popular destination hotspot for European and global tourism (UNWTO 2018), which recurrently suffers from jellyfish outbreaks (Gili and Pag es 2005;Molinero et al. 2005). In general, the fishing and tourism industry are the sectors that suffer the greatest impacts from jellyfish presence (CIESM 2001;Ghermandi et al. 2015;G omez and Guti errez-Hern andez 2020;Nastasi 2010;Patrice, L eo, and Gaby 2011). In the early 1980s, jellyfish outbreaks received global attention when Pelagia noctiluca blooms, a Mediterranean jellyfish species, caused injuries to tourists and fishermen (CIESM 2001). In addition, under current climate conditions, there is a risk of an increase in the numbers of non-native jellyfish species entering the Mediterranean basin (Killi et al. 2020), potentially magnifying the consequences of jellyfish presence for particular coastal areas.
Previous studies focusing on the interactions between jellyfish and tourism have mainly concentrated on the assessment of the economic impacts from jellyfish outbreaks and whether there is a need for the implementation of mitigation strategies (Ghermandi et al. 2015;Nunes et al. 2015). In addition, the support for policies aimed at counteracting the impacts of an increase in jellyfish outbreaks (Enr ıquez and Bujosa Bestard 2020; Ghermandi et al. 2015;Remoundou et al. 2015), as well as the benefits associated with a reduction in the frequency of the outbreaks have also been assessed (Kontogianni and Emmanouilides 2014). Results from these studies indicate that there is support for general policies oriented toward the mitigation of jellyfish impacts. However, the approach of these studies was to assess the general support of the public for mitigation programs without including the assessment of coastal users' preferences for specific management measures that could contribute toward improving their coastal experiences, increase their sense of safety and therefore their overall wellbeing.
In this study, we focus on tourists' perceptions on the potential increase in jellyfish presence in the North-Western Mediterranean and their preferences for management measures aimed at mitigating associated impacts. A combination of social and econometric survey approaches using Discrete Choice Experiment was used to elicit information on the willingness to pay for specific measures and avoid the unpleasant effects derived from jellyfish-human interaction. The relevance of additional potential issues associated with beach areas, such as the presence of litter or water quality were also assessed using the same methodology. The remainder of the paper is structured as follows, Section 1.1 describes the area of study, Section 2 describes the methodology used, Section 3 presents the results from the survey and the DCE and Section 4 discusses the implications of the findings from an adaptive coastal management strategy perspective.

Empirical context. Area of study
The Balearic Islands, located in the North-Western Mediterranean, are one of the major tourism destinations in Europe ( Figure 1). The archipelago is composed of four islands, Mallorca, Menorca, Ibiza and Formentera, in decreasing extension. They have a resident population of approximately 1.2 million inhabitants. Being one of the main European holiday destinations, in 2019, they were visited by over 20 million tourists (AETIB 2020). Visitors are generally attracted to the clear coastal waters of the area (ATB, 2017) and the warm temperatures throughout most of the year. Mallorca, the biggest of the islands and which concentrates the majority of the tourism (approximately 15 million tourists in 2019, AETIB 2020) was chosen as a representative location of a beach-and-sun tourism destination which is often impacted by the presence of jellyfish around its coastal waters. Jellyfish are present all year round, but their impacts are most noticeable during the summer season, when the number of coastal users increases and, hence, the interactions between humans and jellyfish are greater. In 2019, beach emergency services recorded a total of 36,810 assistances due to jellyfish stings (DGE 2020). This figure, however, underestimates the full extent of jellyfish sting numbers, as only major beaches with associated infrastructure and visited by high numbers of users, have first aid services available. The presence of jellyfish and their consequences around the island are frequently reported in both local and international newspapers and tabloid magazines, including British and German publications (MDB 2018;MM 2018;MSR 2018). These publications often adopt an alarmist tone which can have detrimental effects for the image of the island and might have an associated impact on the volume of tourists who choose Mallorca as a holiday destination.
The present study has a particular focus on two jellyfish species, Pelagia noctiluca and Physalia physalis ( Figure 2). The jellyfish Pelagia noctiluca is one of the most abundant species around the coastal waters of Mallorca. This species is known to be one of the most venomous in the Mediterranean. Its effects are not lethal, but in most cases they cause a painful sting, with local pain lasting for 1-2 weeks (Mariottini and Pane 2010). On the other hand, and although not a true jellyfish species, Physalia physalis (commonly known as Portuguese man o'war) is equipped with a particularly potent toxin that is potentially deadly to humans (Burnett 2001;Edwards and Hessinger 2000). P. physalis is not a common species in the Mediterranean; however, under particular meteorological and oceanographic conditions it can enter the Mediterranean from the Atlantic through Gibraltar's Strait. In 2010, the Mediterranean basin experienced P. physalis swarms that had dramatic consequences, including the first recorded human fatality in the region (Prieto et al. 2015). In 2018, several specimens were detected along the NW Mediterranean shores, including Mallorca, causing social alarm (DM 2018; UH 2018a, 2018b).

Survey design and administration
To assess tourists' perceptions around potential increases in jellyfish numbers and associated consequences, as well as to answer the question of how much they are willing to pay to reduce the negative impacts associated with jellyfish presence in coastal waters, we used a combination of social science survey approaches together with the application of a discrete choice experiment method (DCE). We used DCE as they are particularly appropriate to elicit respondents' preferences among two or more multiattribute alternatives (Johnston et al. 2017). DCE are survey-based methodologies where respondents make choices among a set of hypothetical alternatives. Each alternative is defined by the same bundle of attributes; however, attribute levels differ between alternatives. The analysis of responses determines the marginal rate of substitution between any pair of attributes that differentiate the alternatives. If one of the attributes represents a monetary value, then it is possible to calculate the respondent's willingness to pay (WTP) for the other attributes (Hanley, Wright, and Adamowicz 1998;Liu et al. 2010). DCE have been extensively used to elicit preferences and estimate values for changes in many public goods (Hynes et al. 2021), human health effects (de Bekker-Grob, Ryan, and Gerard 2012) or environmental policies (Ruiz-Frau et al. 2019) among others.
A face-to-face survey was carried out in the summer of 2018 at Mallorca's airport departures' lounge. Approximately 500 tourists answered the survey. A market research company with trained interviewers was employed to administer the survey in Spanish, English and German. Respondents were randomly selected to cover a representative range of tourists. Spanish National Council Research ethics procedures were followed, and written consent was obtained from all participants. The first part of the survey contained questions aimed at characterizing respondents' visits, their perceptions around potential issues regarding beach use such as overcrowding, presence of litter, coastal water quality or jellyfish presence. Their previous experiences with jellyfish and their anticipated future behavior in hypothetical scenarios characterized by an increase in jellyfish numbers were also assessed.
The second part of the survey contained the DCE exercise. A series of five choice cards were presented to each respondent that examined their preferences for a set of attributes characterizing different beach conditions and the adopted measures regarding jellyfish management ( Figure 3). In each of the cards the respondent could choose between two beach alternatives or a third option that involved not going to the beach. Each beach alternative was characterized by five attributes, namely: presence of litter, water quality, jellyfish presence, jellyfish management measures and an associated economic cost to reach the chosen destination. The different attributes, and different levels they could adopt, are presented in Table 1. The cost attribute was calculated according to the associated costs of car and petrol use for different travel distances within Mallorca. A final section containing socio-demographic questions was included (a copy of the questionnaire can be found in Appendix A [online supplemental material]). The design of the questionnaire was tested through two rounds of focus groups (11 and 5 people, respectively). Observations from the focus groups were used to refine the design of the questionnaire. Pilot testing was conducted prior to the main survey with 100 respondents. As the structure of the DCEs did not change during the pilot phase, all pilot questionnaires were included in the final DCE analysis. The survey resulted in a total of 498 respondents completing the questionnaire, yielding a total of 2,480 observations (496 respondents Â 5 choice cards).
The open source R package support.CEs (Aizaki 2012) was used to create the DCE design, based on orthogonal main-effects arrays, and to convert the DCE design into choice cards. Twenty-five choice options were created. A blocking procedure was used to assign the options to five bundles of five choice sets; thus five versions of the choice experiment cards were produced. Each version contained a different combination of five DCE tasks and each choice task consisted of three alternatives. A choice card example is represented in Figure 3.

Choice experiment econometrics
The econometric framework for DCE lies in Lancaster's theory of consumer choices (Lancaster 1966) and McFadden's random utility theory (McFadden 1974), which assume that individuals make choices to maximize their utility (their satisfaction) and that the utility of a good can be decomposed into the utilities of the characteristics of the good. Consumers make their choices based on the utility of the characteristics (i.e. attributes), rather than on the good itself.
DCEs can be analyzed using different models. Due to its simplicity, the multinomial logit model (MNL) is often used as the starting point to subsequently test the improvements offered by more flexible models. The MNL has important limitations; specifically, it assumes independence of irrelevant alternatives (IIA) and it assumes homogeneous preferences for all respondents (Hausman and McFadden 1984). The random parameter logit (RPL) has been suggested as an alternative to overcome these shortcomings, as it allows preference heterogeneity to be accounted for across people (Train 2009).
The utility (U) of a good consists of a known or systematic component (V) and a random component (e) which is not observable by the researcher. The systematic component of utility can be further decomposed into the specific attributes of the good (bX), which in this case is a beach scenario. Thus, the utility that respondent n derives from a certain beach alternative i is given by: The RPL allows the b coefficients of observed variables to vary randomly across people rather than being fixed as in the MNL model. These unknown parameters vary accordingly to a specific function (McFadden and Train 2000). Here, cost is set as a Management measures levels: increasing levels of management measures include the measures contained in previous levels, e.g. level 5, includes jellyfish exclusion nets, first aid, warning flags and information panels fixed parameter, while the rest of the parameters are set as random following a normal distribution (Hynes et al. 2021;Revelt and Train 1998;Wielgus et al. 2009). The fixed cost parameter was adopted to avoid convergence issues and to facilitate the calculation of the implicit price of the different attributes. The RPL model also overcomes another two limitations of the MNL by allowing correlation between the error components of the different alternatives and unrestricted substitution patterns across choices (Train 2003). RPL models also provide the additional flexibility of usage on panel data. The RPL presented here has been run to allow for correlation and uses panel data.
Since the choice probability in the RPL does not have a closed form, the expression has to be approximated using simulation (Train 2009). Repeated draws of b are taken from its density f bjh ð Þ: For each draw, the product of logits is calculated, and the results are averaged across draws. In this study, Halton intelligent draws have been used for the simulation, since they have been found to provide greater accuracy than independent random draws in the estimation of RPL models (Train 2009).
Welfare estimates can be derived from the models, they are calculated in the form of marginal willingness to pay (WTP) by dividing the b parameter for an attribute by the b parameter for the cost attribute. Willingness to pay calculations allow comparison of the relative importance of changes in one attribute to changes in another attribute within the choice set design (Hynes, Tinch, and Hanley 2013). In RPL, the marginal effects need to be integrated over taste distribution in the population. This is computed by simulation from draws of the estimated distributions for the random parameters (Scarpa and Thiene 2005). All analyses were performed using the opensource R package gmnl (Sarrias and Daziano 2017).
To detect sources of heterogeneity in taste preferences while still accounting for unobserved heterogeneity, interactions of respondents' socio-demographic characteristics with choice specific attributes were explored in the model. In addition, non-linear interactions between water quality and jellyfish presence and between jellyfish presence and management options were also explored.

Results
A total of 496 respondents completed the questionnaire. Table 2 provides an overview of the socio-demographic characteristics of the sample. The sample was gender balanced (51% female; 49% male), mean age was 41.5 years (± 26.8 s.d.) and 41% held a university degree. The sample was equally split between national and international visitors.

General views and trip characteristics
Approximately, half of the sample were first time visitors (45%) while the rest had previously visited the island (55%). When visiting the island, respondents mostly do so over the summer season (48%). Regarding trip characteristics, the three main purposes for visiting Mallorca were "sun and beach" (57%), "relax" (25%) and "fun" (10%). The main factors that attracted visitors to their destination was the presence of beaches (46%) and the weather (28%) ( Table 3). On average, respondents stayed 6 (± 1.7 s.d.) nights on the island and 87% of them considered that going to the beach was an important part of their visit. When asked about the main problems perceived during their visits to the beach, respondents highlighted the presence of dirt and rubbish (36%), jellyfish (17%) and overcrowding (16%) ( Table 4). Most respondents (98%) stated they would consider coming back to the island for holidays.

Jellyfish views and experiences
Less than a quarter of the sample (21%) had experienced the sting of a jellyfish during their lifetime. Only 5% of respondents had been stung during their time in Mallorca. When asked whether they knew the jellyfish Pelagia and Physalia, 28% claimed to know both species, 21% only knew Pelagia, 11.5% only Physalia and 39.5% did not know either.
Between 92% and 94% of respondents considered the presence of coastal measures such as, jellyfish warning flags, information panels, live information on jellyfish presence or exclusion nets as very useful tools in mitigating the impacts of jellyfish.
When asked about their potential decisions in hypothetical scenarios with increased jellyfish numbers where they would encounter jellyfish in half of beach visits, 26% of respondents considered they would not come back to Mallorca in the case of an increase in the presence of Pelagia. Conversely, a higher number of respondents (40%) would not return in the case of an increase of Physalia. However, these percentages decreased considerably in scenarios of increased jellyfish presence where mitigation measures were applied, only 6% of respondents would not come back in the case of Pelagia and 15% in the event of an increase in Physalia (Figure 4).

DCE exercise
Respondents were asked to choose between the three alternatives offered in the DCE. About 95% of respondents reported that choosing between the three options was either very easy, easy or normal. In addition, there is evidence that respondents compared the alternatives, as in 81% of the cases respondents varied their choice across the five tasks. One of the two beach alternatives was selected in 70% of the cases. Only 3% of the sample consistently chose either option A or B. When option C was chosen (30% of the choice tasks), respondents claimed they did so because they either considered that it was the best option (38%), that they were not bothered about swimming (16%) or because they considered the costs associated with options A and B too high (46%). When choosing between options A and B, 34% of respondents stated to have considered the different attributes equally, while 66% stated that they particularly focused on one of the factors. In decreasing order, 33% paid special attention to the presence of litter on the beach, 24% focused on the presence of jellyfish, 24% on water quality and 19% on the type of mitigation measures.
The levels against which the estimates were compared in all models were high water quality for the water quality attribute, the non-presence of jellyfish for the jellyfish presence attribute and the lack of management measures for the jellyfish management attribute. Presence of litter was considered a continuous attribute. The attributes and their associated levels are summarized in Table 1.
Results for the RPL show that the magnitude and the sign of the different attribute coefficients are in line with expectations (Table 5).
No significant interactions were found between respondents' sociodemographic characteristics or between pairs of attributes. Therefore, for simplicity, the RPL model without interactions is presented here.
As indicated by the coefficients, respondents experienced an increase in utility when moving away from the option of not going to the beach, as denoted by the alternative specific constant for selecting a beach (asc ¼ 5.91). They had a preference for a reduced presence of litter on the beach and for clear blue coastal waters. They were willing to pay e3.1 less for each additional piece of litter encountered on the beach and e25.9 less for going to a beach with low water quality. Similarly, respondents' disutility increased with increasing jellyfish numbers and the associated probability of being stung. Avoiding high numbers of jellyfish was the variable that elicited the highest WTP in the exercise. Respondents were willing to pay an additional e33.3 for having the possibility of enjoying jellyfish free waters, as opposed to enduring coastal waters with high numbers of jellyfish (Table 5). Medium and low jellyfish presence elicited e25.9 and e15.9, respectively. Regarding jellyfish mitigation measures, having information panels available at the beach, elicited e4.8. More active measures such as  the possibility of receiving first aid in the event of jellyfish stings (e8.9) or the presence of beach exclusion nets that would considerably reduce the presence of jellyfish at the beach (e12.4) showed higher WTP values.

Discussion
The focus of this study was to assess changes in tourists' behavior under potential future scenarios of increased jellyfish presence and their preferences and willingness to pay for specific jellyfish mitigation measures.
In general terms, the study indicates that without the adoption of suitable management measures to mitigate jellyfish associated impacts, sun and beach destinations in the Mediterranean such as Mallorca, could suffer significant negative economic impacts under scenarios of increased jellyfish numbers. Our survey indicates that the lack of additional management measures could translate into a decrease between 24% and almost 40% of tourists depending on the jellyfish species under consideration. Such a reduction could have significant effects on coastal economies heavily reliant on the tourism industry (Purcell, Uye, and Lo 2007;Richardson et al. 2009;Tomlinson et al. 2018).
Although tourists identified the presence of litter on beaches as the most bothering problem in comparison to other issues such as jellyfish presence, overcrowding or water quality during the survey, the DCE analysis clearly indicated that the presence of high numbers of jellyfish is the factor that has a greater impact on beach users' wellbeing. The avoidance of high jellyfish presence was the variable that elicited the highest willingness to pay from tourists (e33.3). The magnitude of this result was higher than estimates in a similar study performed along the Catalan coast, also in the NW Mediterranean area, which ascertained that beach users were willing to pay lower amounts (e3.20) to reduce the risk of encountering jellyfish in their visits to the coast .
Personal negative experiences with jellyfish or more importantly, negative media reports about coastal areas affected by jellyfish outbreaks can persuade tourists to seek alternative holiday destinations (Lucas, Gelcich, and Uye 2014;Richardson et al. 2009). The implementation of additional safety measures, however, could potentially minimize the change in tourists' destination choice.
According to results, the adoption of mitigation measures could significantly reduce the loss of tourists by 66% in the case of Physalia and 83% in the case of Pelagia. Therefore, the adoption of measures which increase the sense of the beach user's safety is fundamental to promote an image of safe enjoyment of the coast (Bordehore et al. 2016;Gershwin et al. 2010;Lucas, Gelcich, and Uye 2014). Care has to be taken, however, when considering the number of returning and non-returning visitors, as DCEs often suffer from hypothetical bias, meaning that stated intentions might not always reflect real behavior (Ajzen, Brown, and Carvajal 2004;Hensher 2010). Several factors are known to influence tourists' revisit intentions for coastal and marine areas, among others accessibility, facilities or the quality of the marine environment have an influence on shaping their intentions (Ariesta, Sukotjo, and Suleman 2020;Schuhmann et al. 2019). Another aspect to consider relates to the fact the survey was administered among departing visitors, who might have different preferences to arriving visitors; therefore results might not be directly transferable.
Previous studies have pointed toward coastal users' general support for policies aimed at counteracting the effects of jellyfish outbreaks in the Mediterranean, in countries such as Spain, Israel or France (Ghermandi et al. 2015;Kontogianni and Emmanouilides 2014;Nunes et al. 2015). Although the support for such policies has been elicited, there is no information on the type of mitigation measures and preferences that coastal users have for specific measures that increase their sense of safety and therefore, wellbeing. The more detailed nature of the DCE presented here can help local administrations and coastal managers understand which mitigation measures are preferred by beach users. The survey indicated that respondents had a clear preference for those measures that maximized their safety, as they were willing to pay e12.4 per visit to the beach for measures that included jellyfish exclusion nets, the presence of first aid assistance, jellyfish warning flags and information panels on jellyfish species and the procedure to follow in the event of a sting. The installation of exclusion nets is not a widespread practice in the Mediterranean where jellyfish stings rarely have fatal consequences (Mariottini and Pane 2010), when compared to other parts of the world such as Australia, where exclusion nets are commonly used to prevent the entrance of potentially deadly box jellyfish species (Gershwin et al. 2010). The implementation of measures without exclusion nets, that is, the presence of first aid assistance, jellyfish warning flags and information panels, still elicited e8.96 per beach visit, meaning that the utility of the beach user is still higher in a scenario with mitigation measures than in a situation without any type of measures. In this study, we chose not to extrapolate the WTP estimates to the entire volume of incoming tourists as our sample suffered from a nationality bias. While the number of international vs. domestic tourists is approximately 70% vs. 30% (AETIB 2018), our sample was biased toward domestic tourists (50% international vs. 50% domestic). Due to the inability to access further visitors' data, representativeness for the rest of sociodemographic characteristics could not be evaluated.

Management measures
The different outputs of the study highlight the increasing need for adaptation that coastal holiday destinations have in the face of potential ecosystem shifts (De Donno et al. 2014;Hattam et al. 2015;Kontogianni and Emmanouilides 2014). This is particularly important for the Mediterranean region which is more regularly impacted by the increased frequency of P. noctiluca outbreaks (Gili and Pag es 2005;Molinero et al. 2005) and is at risk of receiving non-native jellyfish species, some of them with potentially negative impacts for the human use of the coast (Killi et al. 2020). It is therefore fundamental to develop management strategies that enable the coexistence of coastal users with jellyfish.
Effective management plans need coordinated research efforts that allow a better understanding of underlaying ecological mechanisms of jellyfish breakouts, the adoption of effective prevention policies and mitigation strategies, and appropriate planning of health services in areas highly frequented by tourists (Canepa et al. 2013;De Donno et al. 2014). Specific elements that could contribute to the effectiveness of management plans are the provision of information, the availability of life guard or first aid assistance and a monitoring system that allows for continuous information on jellyfish presence to be recorded and transmitted in real-time to both users and coastal managers (Canepa et al. 2013).
The provision of information to the coastal user is a fundamental aspect that needs to be included in mitigation management plans. The perception of jellyfish by society is generally negative (Bosch-Belmar et al. 2017;Palmieri et al. 2014;Vandendriessche et al. 2016), even for non-threatening species such as P. noctiluca, where the public's reaction greatly outweighs the actual risk associated with this species. Public perceptions depends on their previous knowledge, their views on the potential dangers, their cultural background and whether they feel the enjoyment of their activities will be compromised (Lucas, Gelcich, and Uye 2014). The provision of information on the identification of jellyfish species, the magnitude of the severity of potential stings, the advised behavior in events of jellyfish encounters and the treatment to follow in sting cases can contribute toward the change in jellyfish perception by the coastal user toward a less negative view (Baumann and Schernewski 2012). Information can be displayed both on-site or through mobile phone applications. The coupling of mobile phone applications with real time information on the abundance of jellyfish around the coastline could prove an extremely useful tool for coastal users to avoid the unpleasant effects of jellyfish presence (Marambio et al. 2013). Recent advances in technology and artificial intelligence have allowed for the development of inexpensive underwater camera systems that automatically identify and quantify jellyfish species and their associated numbers (Gauci, Deidun, and Abela 2020;Martin-Abadal et al. 2020). This technology allows for the development of early-warning systems that could simultaneously inform coastal users and managers. Allowing the latter to make informed decisions on when it might be preferrable to take actions, such as temporal beach closures in events of high numbers of jellyfish to avoid associated detrimental impacts and negative image association. The development and implementation of continuous and real-time information systems are elements that have been previously advocated (Bordehore et al. 2016;Brodeur et al. 2016) and would offer a sense of security to coastal users.

Conclusions
Results from the study presented here indicate that an increase in jellyfish numbers could affect the experience and choice of holiday destination by tourists, potentially leading to a significant decrease in the volume of tourism in an area frequently affected by jellyfish outbreaks. This situation can be counteracted through the adoption of pro-active and preventive jellyfish management plans that include the provision of information, the presence of lifeguard and health services, the potential deployment of jellyfish exclusion nets. In addition, the implementation of early warning systems based on real-time continuous recording systems are aspects that should be considered by coastal destination administrations to promote a sense of tourist safety. Our study highlights that such measures have a significant potential in reducing or avoiding the loss of tourists in jellyfish-prone destinations, preventing their migration to other less affected destinations.

Declaration of interest
The author of this paper declares no conflict of interest.