Breaking Brands: New Boundaries in Rural Destinations

: Tourism destinations are generally delimited and branded based on administrative boundaries, which act as artiﬁcial barriers that may reduce the competitiveness of the tourism sector. Increasingly, literature is taking a demand perspective (i.e., tourist spatial behaviour) when identifying and promoting destinations. This perspective can help to promote destinations more efﬁciently, particularly in rural areas, where most tourism ﬂows depend on private vehicle and which do not take into account administrative boundaries. These ﬂows are therefore highly conditioned by the geography of the area, hosting capacity and the cumulative effect of attractions. This research centres on brand creation from a tourist perspective, particularly how tourists consume a destination. Els Ports (Spain), a rural mountain area divided into multiple administrative divisions, each marketing its own brand, is taken as a case study. Recently, destination managers have seen the opportunity for regional cooperation and taken steps to cross traditional boundaries to market the area better. This study uses GIS techniques to compare tourist travel patterns with brand boundaries and new cooperative initiatives. The ﬁndings provide material for discussion on the branding strategy of Els Ports and the need to rebrand rural tourism destinations into functional tourism areas.


Introduction
Tourism destinations, particularly destination branding, are a key field of interest for both academia and practitioners. Despite the importance of the term "tourism destination", academics continue to debate how to define it. The management-oriented research perspective considers tourism destinations as agglomerations of facilities and services designed to meet the needs of tourists [1]. This, however, contrasts with the reality of established destinations.
Currently, tourism destinations are defined by the administrative boundaries of the corresponding territorial administrations. Branding based on administrative areas leads to static brands distinguished only by their borderlines. This is an observable trend in Spain, where institutional divisions by local, regional and national authorities have led to administrative areas becoming homogeneous, closed, tourist-branded spaces.
However, destination areas may transcend political boundaries, which can lead to certain inconsistencies that affect neighbouring tourism actors and tourists alike. Previous studies on tourist mobility provide evidence that tourist visits are not necessarily restricted to the official boundaries of a destination. Therefore, defining destination brands based on administrative boundaries may not be the most appropriate spatial configuration for optimal promotion and destination development [2].
As tourists play a critical role in the process of defining a tourism destination, the literature on destination management is increasingly pointing to the importance of understanding how destinations are consumed in order to adapt them to consumer needs [3][4][5][6].
Tourist flows activate or deactivate places, thus contributing to the shape, dimension, and structure of the destination [7][8][9]. However, a search of the literature reveals a paucity of previous studies examining tourist movements from a marketing perspective [10]. Existing studies focus on identifying viable segments of the market rather than aiming at rethinking the boundaries of the branded area.
Only a few studies have put into question the use of administrative boundaries as the branding determinant of a destination area or proposed alternative branding solutions. These alternative strategies focus on either geographical marketing criteria [2] or the stakeholders' perspectives [11]. This shows a lack of branding studies advocating the functionality of the branded area by focusing on the tourist perspective.
Furthermore, branding literature pays little attention to regional or smaller areas, and only a few branding studies refer to rural destination regions [11][12][13]. Product attributes of rural tourism are inherently attached to the concepts of nature, culture and gastronomy and have the potential to appeal to more independent consumers involved in the destination selection process [10]. The majority of rural destinations are also immersed in a highly competitive tourism marketplace, facing both financial and human resource marketing challenges. This highlights the importance of using an appropriate branding strategy, which can guarantee the future social and economic development of tourism in these areas [10].
To fill this gap, this paper proposes abandoning the discourse of the so-called formal destination based on administrative areas and replacing it with a new branding strategy focused on making the area more functional for the tourists' use, which must facilitate more appropriate tourism development in rural areas. Thus, the main aim of this study is to identify the functional destination based on tourists' intra-destination travel patterns in a transboundary rural area as a previous step of the branding process.
Current destination brands in the transboundary rural area of Els Ports, located between the Spanish regions of Catalonia, Aragon and Valencia, are compared to determine how tourists actually consume this area geographically. The analysis centres on revealing whether the destination boundaries and the first transboundary co-development initiatives in the area coincide with tourist visitation patterns.
Results of the analysis show that neither the boundaries of the formal destination nor initial steps to rebrand coincide with destination consumption. Tourists consume a destination without stopping at administrative boundaries, and functional consumption also tends to be highly conditioned by the location of accommodation hubs. From here, tourists make side trips to areas close by, aided by good road networks. This shows the scope for latent opportunities in the area of Els Ports. Branding and promotion should focus on the functionality of the area to be more efficient, taking advantage of the cumulative effect of attractions and the centrality of accommodation hubs.
This research contributes to the discourse on functional destination branding and promotion by giving an example of the potential held by rural destinations to rebrand themselves in accordance with how tourists actually consume them geographically. This is an excellent opportunity for rural destinations to excel, with more efficient marketing that better meets tourists' needs.

Destinations from an Administrative-Based Criterion
An existing body of studies on cross-boundary areas reveals the barrier international and regional borders pose for tourism development [14][15][16][17]. This barrier effect is particularly relevant in the case of international, non-permeable borders, where neighbouring areas cannot add value by offering visitors easy access to attractions and services located on the other side of the border. In the case of permeable boundaries, the boundary effect is still pronounced at both regional and local levels [18].
Following the St. Gallen Consensus on Destination Management [19], Destination Management Organisations (DMO) are in charge of marketing and promoting destinations areas, alongside other multiple activity domains. Unfortunately, most DMOs depend on Sustainability 2021, 13, 9921 3 of 19 public administrations; thus, DMOs apply administrative criteria when branding and promoting the destination. Although branding does not necessarily need to follow administrative or physical boundaries, in practice, DMOs manage brand identity, meshing everything the administrative territory contains into one single brand. This represents a colourless mass, distinguishable only by its borders [3,5,20].
An increasing number of studies support the obsolescence of administrative-based destinations [3][4][5]14]. Boundaries can trigger inter-council rivalry and even distrust, which can become an issue for the Regional Tourism Organization [21]. Furthermore, neighbouring tourism actors found where the borders touch may suffer from a lack of co-development initiatives and inconsistencies in regulations, policies and promotion because they belong to different administrative systems [15,17,[22][23][24].
From a demand perspective, tourists do not restrict their visits based on administrative boundaries unless there is a physical impediment [3,5,6,14]. Nowadays, new communication technologies offer tourists a wide range of available information sources in addition to the traditional channels [25,26]. Input from several channels contributes to creating a tourist's own image of a destination and initiates the decision-making process, which makes them feel more empowered to organize their own trip rather than use traditional promotional channels [10]. Thus, the marketed area may not coincide with the space represented or spatial flows of what tourists do at the destination [6,27]. This underlies the idea that tourism destinations should move towards a more dynamic model based on how tourists actually consume a space [3][4][5][6]14].
This inconsistency emerges from a previously unsolved debate about how to define a tourism destination. The complexity of this debate has led to a general acceptance in both academia and in practice that regional and local tourism destinations are an appropriate unit of analysis, and other alternatives are not taken into consideration [28]. Thus, destinations are very often interpreted from a geographical-administrative point of view, including destinations' self-recognition as places determined by the tourist offerings [20,29].
In view of the criticism of administrative-based destinations, a new geographical layer needs to be established, adding more complexity to an already complex branding process [30,31]. The constructionist perspective supports an alternative viewpoint of the destination, asserting that destinations should not be defined by maps or political decisions. Unlike the modernist view, this perspective recognizes that it is hard to define the boundaries of a destination as they are being constantly produced and reproduced through complex practices and discourses. Thus, destinations are not stable, closed systems; they are the result of social practices beyond the traditional production-consumption dichotomy. They are under constant negotiation and renegotiation and are very much affected by the role of tourists as co-producers [1]. Many questions arise at this point that stem from the need to decide the criteria and size of the destination in order to identify its boundaries and design a successful branding strategy.

Destination Branding Limits Strategy
Destinations need to create and develop brands to help their positioning and differentiation. This is done by building an emotional relationship between the destination and potential visitors with the final aim of attracting tourists [32,33], and involves developing a brand for a determined geographical area, taking into consideration both supply and demand. In fact, a brand is extrinsically linked to its geographical knowledge. This knowledge generates place meaning and represents the components of an area in a unitary way, differentiating a destination from other competing areas [11].
The early 2000s saw a tendency for micro-branding and multiple-layer strategies; however, the excess of brands may saturate potential tourists, confusing them and lessening marketing effectiveness [11]. In rural destinations, in particular, the strategy of micro-branding is inappropriate, as DMOs have to face the marketing cost in a highly competitive tourism marketplace [10]. In contrast, cooperative branding across multiple rural communities has proved successful and produces a consistent attributes-based image, as perceived by tourists [34]. In fact, the synergistic relationships among several tourist nodes and the surrounding area eventually generate a critical mass that can effectively position a destination in the market [22].
Branding studies encompass the strategy of building a consensual vision of a destination whilst taking into consideration the residents and the area's economic activities [11]. This represents a unique combination of product characteristics with both functional and non-functional added value [27,35], which aims for a consistent, attributes-based image using cooperative branding [34]. This is important in order to promote an attitude of cooperation between competing businesses in the destination, and without which the strategy would fail [36]. The complexity of this lies in the DMO's lack of direct control over the actual delivery of the brand promised by the local tourism community and tourism stakeholders [31,37].
Despite this, it is the tourist who perceives the image of functional and psychological attributes of the destination [33], and destination branding is considered to be successful when it meets tourist expectations [32]. Thus, a destination brand should consider both perspectives: the consensus for projecting the destination identity and the tourists' image of the destination; otherwise, the brand may not be well defined [11].
Place branding should not entail creating a new identity for a destination but of understanding how tourists perceive the destination to be able to influence the location decisions made by potential visitors [27]. This includes all aspects that make up the destination at all functional and emotional levels [27,31]. Thus, to build a successful branding strategy, the tourist should be the designer as well as the final evaluator of the process [36]. To do so, marketing and branding should entail understanding the dynamics of tourist demand by analysing the perception of space in order to understand the correspondence between the name of the area and the meaning that this signifies outside its borders [20].
Seen from this perspective, destination branding should not only be about choosing the right techniques, which require visual and verbal associations with the destination, such as a name, term, sign, symbol or design. Rather, it should entail understanding the market, paying attention to practise and performance, and creating spatiality through production and consumption practices [1,27,30,37].
Although many studies have referred to the intangible, cognitive and emotional elements of destination branding, scant attention has been paid to the functional elements of the destination. An increasing number of researchers have recommended the need to understand the tourists' perspective of destination limits in order to adapt the destination to consumers' needs [4][5][6]19]. Since destinations should meet the tourists' perspective, the question now is where to draw the boundaries of the destination and which stakeholders should be involved in creating a consensual vision of the destination.
In the field of destination management, some studies stress the essential role tourists play in the process of defining a tourism destination. They suggest focusing on the representational space and the spatial practice created by tourists' travel patterns as a tool for identifying tourism destination boundaries [5,6,19,20,27]. The tourists' "recreational destination" can be represented as a network consisting of different nodes connected to each other through tourist trips ( [38], p. 2). Through their flows, tourists activate and deactivate places, which contributes to the shape, dimension, and structure of the destination [7][8][9]. Adding up these connections would enable a macro-spatial analysis of intra-destination movements [38].
Considering these previous studies and acknowledging the need to involve the perspective of tourists in the branding process [2], identifying the destination area to be branded from the perspective of tourist functionality is precedent. In fact, the destination is where the entire tourist experience takes place, in that it has the availability of attractions, facilities and services, and enables successful marketing and promotion, which may expand or shrink, and may gain deeper or more superficial meanings [11,20]. Therefore, identifying a destination from the perspective of the consumer should be the first step in the branding process [2]. After the destination is identified, it should encompass the strategy of building a consensual vision of the destination taking society and economic activities into consideration [36] Finally, the destination will gain meaning by seeing tourist images through specific production and consumption processes that connect people to the world by contextualizing their experiences with the tangible and intangible attributes of the destination [1,27].
In a rural destination, tourists mostly travel using their own car due to a lack of public transport. Using a private car leads to a touring behaviour where tourists build their own itineraries along the routes of the destination, linking several nodes and taking advantage of the cumulative effect of clustered attractions [22,46,47]. Therefore, the predominant travel pattern is base-camp or hub-and-spoke, where tourists are settled in a central accommodation or base-camp and make side trips to surrounding areas [6,39,46,48]. Moreover, the scarcity of support facilities in rural destinations [22] makes tourism more dependent on a symbiotic relationship with the support services offered by base camps. This pattern meets with the definition "local destination" ( [43], p. 405), which is considered as "the area containing products and activities that could normally be consumed on a day trip from the heart of the destination".
Tourist accommodation provides the basis for trips, and visits to attractions in a given area are not random [49]. Thus, to detect the destination from the perspective of tourists, it is essential to gather knowledge about which attractions are connected to accommodation hubs through their trips, as well as factors that may affect these patterns. The relationship between tourist accommodation and visitation patterns is expressed by the territoriality of day trips. Tourists' exploration can be categorized according to how far they venture from the accommodation point [43]: (1) no movement, where tourists remain at the accommodation; (2) convenience-based movement, characterized by exploring the immediate proximity of the accommodation; (3) concentric exploration, consisting of side trips around the accommodation influence area, and (4) unrestricted wide movement, where tourists venture further afield.
The distance travelled to an attraction is one of the most important factors influencing travel patterns. As noted by the distance decay concept, demand for attractions declines with the distance travelled [40,50], but only a few studies have documented distances that tourists venture from their accommodation points [41,44,48]. Regarding time distance in nature-based areas, Paulino et al. [6] found that most visits within a destination take place in the immediate vicinity of the accommodation and its influence area, which is within 30 min driving distance. Regarding geographical distance, a study by Smallwood et al. [48] found that most tourists travelled less than 20 km from their accommodation. Despite this general trend, other factors, such as choice of attractions, also influence how far tourists venture from their accommodation.
As attractions are essential to the tourist experience, they significantly influence tourists' willingness to travel widely or narrowly within a destination [6,42,51]. Thus, iconic attractions are more likely to draw tourists from further away than secondary attractions [6,41]. Furthermore, other aspects of attractions, such as their spatial distribution, their characteristics or the inter-attraction distances, are significant factors influencing both tourists' travel patterns and the distances travelled from their accommodation [43].
Areas that provide a range of tourist facilities and infrastructure tend to attract a greater number of tourists, conditioned by the accommodation, which determines if an area can be developed as a destination [4,6,41,51].
Topography and road networks, particularly in rural destinations, strongly influence tourist flows within a destination [5,6,48]. Despite the importance of attractions, accommodation, and spatial characteristics of a destination, the literature points to a wide range of other factors that influence the distances travelled by tourists from the selected accommodation. These include distance travelled from home, budget, travel group composition, length of stay, trip purpose, familiarity with the destination, markers, personal choices, tourists' sociocultural background and cultural distance, availability of public transport, and level of tourism intermediation [45,52,53]. Theorizing the degree of influence of each factor is highly complex. Therefore, examining tourists' territoriality patterns directly is a reliable source of data that can be used to identify the destination from the perspective of tourists in the first step to designing the best branding strategy.

Case Study
This study explores the case of the rural mountain area around Els Ports (Spain). Els Ports massif covers around 100,000 hectares, with a maximum altitude of 1450 m.a.s.l. Figure 1 shows the administrative divisions of three different Spanish provinces which straddle the range (Castellón, Tarragona and Teruel), each one belonging to different autonomous communities (Valencia, Catalonia and Aragon, respectively), six counties, and twenty-one municipalities. Urban settlements are mostly found in the valleys around the mountain range. The most populated towns nearby are Tortosa (33, Traditionally, the branding and management of Els Ports have been segmented and conditioned by the multiple administrative boundaries dividing this relatively small rural area. Having to deal with multiple brands and sub-brands has made it difficult to reach the final customer. However, culturally, the area surrounding Els Ports massif has historical ties, and currently, the area's main cultural interests can be found in its architectural heritage, traditions, agriculture, and gastronomy. These ties between the different administrative territories have been perceived as an opportunity for regional cooperation. As area. Having to deal with multiple brands and sub-brands has made it difficult to reach the final customer. However, culturally, the area surrounding Els Ports massif has historical ties, and currently, the area's main cultural interests can be found in its architectural heritage, traditions, agriculture, and gastronomy. These ties between the different administrative territories have been perceived as an opportunity for regional cooperation. As such, in 2013, the neighbouring areas initiated steps to break the traditional administrative boundaries so they can market tourism in the area in a more cooperative way. First, the towns of Alcañiz (Aragon), Morella (Valencia) and Tortosa (Catalonia) have collaborated to create the "Route of the Three Kings", which aims to promote their valuable heritage by taking advantage of historical ties between these three towns. Right after, another cooperative project between three counties surrounding the Els Ports Massif was initiated: Baix Ebre, and Terra Alta (Catalonia) and Matarranya (Aragon). The main objective of this cooperation was to jointly promote the "Val de Zafan cycling greenway", which runs along an old railway line crossing mountains and rivers. Both projects have been quite successful in branding terms and are relatively well-known among tourists visiting the area today, particularly the cycling greenway.
These two projects led to greater collaboration, which involved the counties of Terra Alta (Catalonia) and Matarranya (Aragon) and the municipality of Morella (Valencia) (See Figure 2). The aims of the projects were not only to promote a particular product but to create a tourism brand that projects the territory as one, going beyond the classic administrative boundaries. The project involved the public sector in each of the three areas, as well as tourism associations and businesses and wine and olive oil Denominations of Origin. Nowadays, marketing mostly consists of a brochure, press trips, a website, participation in local fairs, and a biannual newspaper publication with activities and tourism resources available in the area.  Figure 2). The aims of the projects were not only to promote a particular product but to create a tourism brand that projects the territory as one, going beyond the classic administrative boundaries. The project involved the public sector in each of the three areas, as well as tourism associations and businesses and wine and olive oil Denominations of Origin. Nowadays, marketing mostly consists of a brochure, press trips, a website, participation in local fairs, and a biannual newspaper publication with activities and tourism resources available in the area.  and part of the Catalan SCI in Els Ports Natural Park. Furthermore, since 2013, the Catalan Natural Park makes up the core and buffer areas of the Terres de l'Ebre Biosphere Reserve, a maritime-terrestrial Reserve that extends through the entire final section of the Ebro River basin to the sea.
Although the main aim of these protection figures is not related to branding, they can assist in identifying areas. Furthermore, some of them, such as the Biosphere Reserve and the Natural Parks, work actively to diversify and distinguish tourism products in the area and project the sector internationally. For example, the Biosphere Reserve promotes activities related to ecotourism, bird watching, cycling, wine tourism and gastronomic tourism and has an associated brand that adds value to the agri-food products and tourist services in the area.
Furthermore, Els Ports Natural Park (Catalonia) is preparing an application to obtain the European Charter for Sustainable Tourism (ECST) and is considering adhering companies that operate in the vicinity of the Natural Park. This will extend the positive effects of ECTS in the area that is effectively part of the massif and its closest area of influence, without strictly limiting it to the Catalan administrative boundaries. However, adhesion to ECST only affects private businesses and not cooperative management with the neighbouring Natural Park of la Tinença de Benifassà.
Moreover, a previous case study in this region detected the potential for a destination in the north-western area of the massif from the perspectives of the geographical location of services and attractions [54], as well as tourist's geographical consumption [5,6]. Based on this analysis and the existing cooperation between the administrative areas, the data analysis for on the north-west of the massif and its surrounding areas compares the official administrative-based brands and the new branding initiatives with the tourists' actual travel patterns, which show the destination consumed.

Data
Innovative data collection methods, such as passive mobile positioning, were discarded due to the rural characteristics of the case study and connection dead spots [8]. Instead, traditional "in situ" surveys were used to collect primary data for their proven reliability and avoid excessive geographical micro-scaling of tracking techniques [55].
Optimum survey locations were identified to avoid possible deviations stemming from the selection of survey places, which included both accommodation and attractions. First, attraction survey locations were selected after being identified as important attractions in a content analysis of guide books [54]. Second, relevant accommodation points were selected after considering official capacity registers (beds by municipality). The number of survey days in each location reflected the number of accommodation beds and the number and popularity of attractions. Additionally, to obtain more representative samples, other significant locations were included halfway through the survey period due to the high number of responses.
The sample used in this study included a total of 835 completed questionnaires answered between July and October 2018. The selection of survey participants was carried out randomly, but in order to meet with accepted definitions of tourist, focused exclusively on leisure tourists, excluding those visiting for business purposes and tourists who had just arrived at the destination. Furthermore, day-trippers and long-stay tourists (over 60 nights) were excluded; the former for not representing the hub-and-spoke travel patterns, which need to stay-overnight [39] and the latter for tending to experience the destination in a similar way to residents [56]. Participants were asked to identify their accommodation location and the attractions they had visited in that location from a list of preselected locations identified through content analysis of guidebooks explained above. However, other accommodation points and attraction places mentioned by the tourists were added to the answers in order to build a data base from the tourists' perspective. Furthermore, the surveys included several sociodemographic questions (See Table 1) with the aim of checking the sample profile, which confirms that tourists' profile of the area fits with the typical characteristics of rural destinations: local tourists travelling with their own cars and showing a low level of intermediation of the trip [57].

Methods
Individual flows data extracted from the survey were aggregated into an asymmetric matrix to count the number of nominations of each connection between different places, representing accommodation (rows) and attractions (columns). Each cell represented the frequency of flows from a single accommodation point to an attraction. These connections (edges that contain "start id", "attraction id" and "number of nominations") were stored in a text file. Moreover, the locations (nodes that contain "id", "location name", and "coordinates") were stored in a second text file. Nodes are a point layer providing geographic information, whilst edges are a non-spatial layer containing information on the strength or weight of a flow between two specific nodes.
The research compared tourist travel patterns within the destination with the current boundaries, as well as the new cooperative initiatives through network analysis and map representation.
The crucial first step in comparing tourist travel patterns at the destination with the current administrative boundaries was to create flow line features from the nodes and edges layers. This was carried out using the free QGIS 3.18 software by creating a virtual layer following the SQL query (Table 2). This query joins, via node IDs, the geographic information from the nodes layer with the flow weights in the edges table. In the last line, start and end node IDs were checked to confirm they were different to avoid zero-length lines.
Once the flow lines were established, we used data-driven line width and added feature blending. In this regard, a geometry generator was used to convert the straight lines into curved arrows, adding a third point to the line feature. Furthermore, we extended the geometry generator to crop the lines at the beginning and end, as indicated in Table 3. Table 3. Geometry generator code used to convert the straight lines into curved arrows and to crop lines at the beginning and end. Once the tourist flow maps were generated (Figures 3 and 4), we used the QNEAT3 plugin, a Python-written tool useful for advanced network analysis. To be specific, the distance between the 10 accommodation hubs and the 80 tourist attractions was calculated in order to obtain distance-decay graphs ( Figure 5). To do so, we used the road network of the National Geographic Institute of Spain. As tourists are "outcome" oriented, transit time is seen more as a friction factor in rural areas than kilometre distance [6,43,58]. We used the Isochrone Area algorithm to calculate accessibility from the accommodation nodes to the rest of the area under analysis. The output of this algorithm was a polygonal layer indicating the distance from the accommodation nodes in intervals of 10 min ( Figure 6). distance between the 10 accommodation hubs and the 80 tourist attractions was calculated in order to obtain distance-decay graphs ( Figure 5). To do so, we used the road network of the National Geographic Institute of Spain. As tourists are "outcome" oriented, transit time is seen more as a friction factor in rural areas than kilometre distance [6,43,58]. We used the Isochrone Area algorithm to calculate accessibility from the accommodation nodes to the rest of the area under analysis. The output of this algorithm was a polygonal layer indicating the distance from the accommodation nodes in intervals of 10 min ( Figure  6).

Results and Discussion
In this section, the output from the data analysis in three thematic areas is presented and discussed. First, results of the transboundary travel patterns are provided. Second, the effect of distance decay in travel patterns within a destination is presented. The consumed destinations are then contrasted with the official branded areas. This includes ad-

Results and Discussion
In this section, the output from the data analysis in three thematic areas is presented and discussed. First, results of the transboundary travel patterns are provided. Second, the effect of distance decay in travel patterns within a destination is presented. The consumed destinations are then contrasted with the official branded areas. This includes administrative bases and branded areas related to protected areas, as well as with the recent transboundary branding initiatives. Finally, the branding strategy the area should implement is discussed. The most significant figures and tables have been selected to explain the results.

Travel Patterns within a Destination Highly Attached to Accommodation Nodes
The high level of interconnectivity seen in Figure 3 between accommodation nodes and attractions on the Aragon and Catalan sides of the mountain range demonstrated that visitation patterns mostly extend along the southern part of Terra Alta and the eastern part of the Matarranya counties, regardless of administrative boundaries [5,6,54].
A closer look at the area with a high frequency of flows ( Figure 4) shows the bounding effect accommodation points exert on the way destinations are consumed geographically. This indicates that accommodation nodes should play a central role in the branding process of a rural area. In fact, the typical touring behaviour of rural areas showed the need to establish a base camp provided with support services. Thus, this proves that locations providing a range of tourist facilities and infrastructure tend to attract a greater number of flows due to the significant role played by accommodation, and which also determines whether the surrounding area can be developed as a destination [4,6,39,[46][47][48]59]. This results in geographical consumption highly dependent on a symbiotic relationship with the support services offered [22], as seen in Horta de Sant Joan, Arnes, Vall-de-roures and Beseit. Furthermore, the inexistence of accommodation nodes in the surrounding areas explained the lack of tourist flows, despite existing attractions with development potential.

Distance Decay as a Constrictor Factor in Travel Patterns within a Destination
Since destination nodes play a crucial role in the branding process, it is essential to discover how far tourists travel from them so the boundaries of the branded area can be established. As tourists are outcome-oriented, the time-distance travelled from accommodation points to attractions visited was an important friction factor that helped determine the limits of the destination from the consumption perspective, as seen in Figure 5 [6,42,43,58]. Here, the temporal isochrones from the main accommodation nodes showed the effect of distance decay when visiting neighbouring attractions.
The first main result showed that during their visit, tourists mostly stayed within the convenience of the area surrounding their accommodation [6,41,42]. Most flows from accommodation points do not exceed 30 min (Figures 5 and 6). However, some flows to neighbouring areas coincide with renowned attractions, such as Alcañíz, the Ebro Delta and Tortosa, which demonstrates tourists' willingness to travel longer distances to see places that are unique or highly attractive [6,41,43].
This tendency for closeness can also be applied to the accommodation points themselves, which are clustered within the convenient exploration category within a maximum distance of 30 min. Although none of the accommodation nodes have over 600 beds, the symbiotic relationship generated by the sum of them and their closeness ends up generating a critical mass capable of effectively positioning a destination in the market and generating a larger number of flows [22].
The second noticeable result concerns the effect of topography and the road network in visitation patterns, as observed in previous studies [5,6,48,57]. Due to the predominance of car use in rural destinations [46], results demonstrated that flows are generally negatively affected by the existence of mountain ranges (Both Els Ports and other minor ridges), acting as a barrier to flows, and positively affected by the road network. In fact, the road network and physical constraints of the area, together with the lack of public transport, induce tourists to engage in touring behaviour [46], following similar itineraries and visiting the same places (Figures 4 and 6).
Despite these major patterns revealing the centrality of accommodation nodes, and the power of attraction of natural areas (i.e., Parrissal, Pesquera, Toll de Vidre and Estrets), we detected two specific nodes that bring more dynamism to the destination: the Green Trail in the north-eastern side of the massif; and a complex dedicated to the motor industry called Motorland, in Alcañiz. Although this study did not analyse data on the tourists' profiles, the map of flows showcases the synergies between these two attractions and the area of Els Ports massif, demonstrating the existence of the two different market segments it attracts. One segment is tourists interested in doing the greenway by bicycle and visiting the villages connected through this cycling route (Vall-de-roures, Cretes, Horta de Sant Joan, Bot, Pinell de Brai, Xerta and Tortosa); the other is tourists travelling by motorbike who are interested in visiting Motorland in Alcañiz and touring around the twisting roads of the massif. Therefore, for branding purposes, these segments should also be considered.

Comparing Branded Areas with Actual Travel Patterns
The initiative undertaken by the Three Territories is similar to the solution we propose. They decided to break the administrative boundaries of the traditional destinations. It adopted cooperative branding across multiple rural communities to build an image based on consistent rural attributes, which is optimal for addressing marketing costs [10,37]. Furthermore, the initiative has managed to involve the private tourism sector, both in conceptualization, by building a consensual vision of the transboundary destination and through daily contact with tourists, thus transforming tourism attractions and accommodation into information points [11,36].
However, during the process of brand conceptualization, the dynamics of tourist demand were not fully considered. Although it was acknowledged that tourist flows do not stop at borderlines, tourists' perceptions of space represented by their visitation patterns in the area were not taken fully into account [3,5,6,14,20]. Therefore, the Three Territories brand fails to accurately represent the image tourists have of the destination. The results show that the Morella municipality and the north-western part of Terra Alta County are not visited by tourists staying in the area. In contrast, decentralized locations, such Alcañiz, seem to be better connected to the destination core.
The intention of the ECST initiative Els Ports Natural Park (Catalonia) to incorporate tourism-oriented businesses located in the influence areas appears a good proposal to integrate the management of this natural area regardless of its political boundaries. Unfortunately, their brand is already defined by the declaration of the protected area and leaves aside the neighbouring Natural Park of la Tinença de Benifassà. Involving the private sector in the destination may be a positive tool for building a common vision from the stakeholders' perspective. However, in terms of branding, promotion, on the whole, may not be totally fruitful as it fails to reflect the tourists' perspective. Els Ports massif prevents flows from connecting between the two sides of the mountain range; thus, the resulting brand (which includes both sides of the mountain range) would not represent tourists' perception of space [3,5,6,14,20]. However, incorporating businesses in the Aragon area, together with the north-western part of the Catalan Natural Park, is a viable initiative, as tourist travel patterns show that they consider the area a destination.

Branding Strategy Proposal
According to the literature, place branding should consist of understanding how tourists perceive the destination [27]. Therefore, to bring the tourists perspective into the branding process, the branding proposal must entail understanding the dynamics of the tourist demand [2,20] whilst paying attention to the representational space and the spatial practice created by tourists' travel patterns [5,6,19,20,27,30,37].
Following this perspective and considering that results showed visitation patterns extending far beyond the present destination brands defined by administrative boundaries, the area should seriously consider developing a transboundary destination, at least for branding purposes. The branded transboundary destination should include Matarranya county and the southern part of Terra Alta county [22,46,47].
Since micro branding in rural destinations is not desirable due to the cost, we advocate for branding that takes in the area outlined and that relies on cooperation between the various DMOs and rural communities [10,34]. Although some public stakeholders may be reluctant to take up transboundary cooperation in order not to break their status quo, successful branding needs to encompass the strategy of building a consensual vision of the destination [11,36]. This is not an easy task; however, other rural destinations are already successful examples of cooperative branding [34]. In fact, in this area, the location of the accommodation nodes has already demonstrated the successful synergistic relationships between them by generating a critical mass capable of effectively positioning a destination in the market.
Results prove that accommodation nodes play an essential role in the representational space and spatial practice created by tourists' travel patterns [6,19,20,27,41]. Thus, the branding process will have to be initiated considering the centrality of accommodation nodes, not only in functional terms but also in their role in destination personality and information deliverers. The proximity of the four accommodation nodes in the area, together with the lack of support services in surrounding areas, and the cumulative effect of the clustered attractions of the area, lead to this latent transboundary destination. Furthermore, the cooperation between accommodation nodes and the surrounding attractions will be of vital importance to generate a feeling of destination and to promote an attitude of cooperation between competing businesses within the destination [36].
Regarding the distance decay, travel patterns showed that the branded area can include attractions up to 40 min driving distance, in line with previous rural studies [6,47]. Furthermore, renowned satellite attractions for which tourists are willing to travel longer distances to visit [6,41,43], such as Alcañiz, should also be considered for inclusion in the brand; maybe not as part of the destination core, but as a complementary attraction which contributes to generating a critical mass apt to effectively position the destination in the market [22].
Regarding brand personality, building a colourless brand distinguished only by its borderlines is not optimal, as this fails to reflect the image of the destination perceived by tourists. The brand should strive to understand the multifaceted dynamics of the tourist demand [3,5,20]. The brand's meaning takes into account tourists' images by contextualizing their experiences with the tangible and intangible attributes of the destination [1,27]. Therefore, brand personality should consider both the characteristics and specific products offered in the area and the various tourist experiences interacting with them in order to capture the attention of different market segments. Accordingly, the branded area should act as an umbrella brand for particular products depending on tourism niches, for instance, the road cycling product, which is one of the fruitful transboundary branding initiatives, and the Motorland segment, which is as yet undiscovered in branding terms. These two segments would help better define the brand and reduce the gap between the projected identity of the destination and the perceived tourists' image of the destination [11,32]. Furthermore, they are elements that differentiate them from other rural areas with different landscapes, but with some cultural and natural activities and a similar gastronomy.

Conclusions
The literature review showed that most tourism destination brands are defined in terms of administrative limits and fail to consider the tourist perspective of the destination. Furthermore, most branding studies tend to pay more attention to intangible, cognitive and emotional elements of destination branding than aspects of functionality. This study contributes to defining the area to be branded in a rural mountain area from the tourists' perspective. Acknowledging the extensive use of private vehicles and tourist behaviour involving touring in rural areas, this study centres on the real tourist flows to discover the representational space and the spatial practice created by tourists' travel patterns. To do so, we chose Els Ports region, which is presently multi-branded with each municipality, county, province, and Natural Parks following their own administrative criteria.
The analysis of tourists' flows in this rural area highlighted the differences in contrast with previous studies in urban areas. The importance of accommodation in rural areas is crucial, whilst in urban areas is much less relevant. The selection method of the accommodation hub by tourists is a need in future research. This element is also related to transport mode. Whilst urban areas are more attached to public transport, rural areas depend almost totally on private transport systems. Rural destinations can also have different market segments behaving differently, even if they share the accommodation hub. In this regard, deeper analysis on different segments can be conducted even to plan different brands if needed.
Considering the road network and the distance decay, the results of this study showed that the limits of a destination can be drawn 30-40 min travel distance from accommodation nodes. Other rural destinations can also benefit from these results for building destination brands as an alternative to time-consuming data gathering. They can identify the main accommodation nodes of an area on a map, drawing the time impedance to define an approximate representation of the functional tourism destination. In the case of the present study, the results defined a functional area comprising mostly of Matarranya county (Aragon) and the southernmost part of Terra Alta County (Catalonia).
Regarding the consumer perspective, in particular, destination brands should also be permeable, including or excluding accommodation nodes or attractions according to tourist preferences. For example, renowned satellite attractions should be considered as travel patterns revealed that tourists are willing to travel longer distances to visit unique, prominent attractions.
The resulting destination brands would represent a new functional area that contrasts with the administrative one and is based on the perspective of the consumer. This new area to be branded would be more effective and demand-oriented, as well as able to reach more public by virtue of joint forces and the scale economies produced. This implies a greater effort in terms of continuous collaboration between neighbouring areas and their stakeholders and continuous analysis of the areas' demand, but at the same time brings a better knowledge of the market and a more efficient way of branding the destination For the methodology of this branding process, we considered a generic perspective of travel patterns but specific and differentiated segments that were explored. Therefore, this should be a further step to be considered in future research. In this vein, the concept of multilayer brands should be explored, taking different targets into consideration. Adapting to different markets and concentrating efforts of different stakeholders could generate an even higher degree of competitiveness and effectiveness, in contrast to the multiple brands following administrative boundaries.
Thus, this paper is a starting point of a line of research aimed at filling a gap in the literature by examining tourists' movements from a branding perspective. The authors are aware that this research is limited to rural areas. The geographical characteristics, the low tourist density and the connection dead spots typical from rural destination conditions required data collection through in-situ surveys, whilst other techniques and larger samplings may be possible in massive destinations. This prevents the data inference of single attractions. However, at the same time, this study shines a light on rural destinations, which are more often forgotten by researchers for having lesser impact and limitations related to data gathering. Furthermore, this problem can be solved by applying a systematic collection of data on an annual basis that would help to better fine-tune the limits. Data Availability Statement: Publicly available datasets were analyzed in this study. This data can be found here: https://dugi-doc.udg.edu (accessed on: 29 August 2021).