Shedding Light on Agritourism in Iberian Cross-Border Regions from a Lodgings Perspective

: Agritourism has been widely discussed in the literature. As the range of options for its interpretation increased, studies have come to focus their objectives on its systematization, evaluated with respect to the nature of the interactions between tourists and agriculture, the level of contact with agricultural activities, and the authenticity of the experience. Adding to this complexity is the association of agritourism with experiences related to the agri-food sector, as well as the typologies of accommodation. This research contributes to an improved understanding of agritourism based on evidence from a cross-border region between Portugal and Spain. We identify the links between agriculture and tourism by identifying different categories of agritourism activities via the application of geostatistical analysis. The main results highlight that landscapes, agricultural activities, and agri-food products have a positive inﬂuence in creating tourism products and services. Despite this result, passive agritourism activities prevail, such as tasting experiences, contact with endogenous products, enjoyment of the agricultural landscape, or on-farm entertainment. This reality highlights a gap related to authentic agritourism, pointing to the absence of or weak links to agricultural activities and weaknesses in the local “farm-to-table” channel. The grouping analysis results led to the creation of six homogeneous groups, three of which correspond to lodging establishments capable of creating agritourism. However, these groups represent only 39% of the supply. Given this reality, it is of utmost importance to implement policies that encourage tour operators to create agritourism products.


Introduction
The complexity associated with the concept of agritourism is due to various geopolitical and social factors that influence public policy [1]. Regulatory policies at the national/regional level, in the case of Portugal and Spain, have been guided by the European Union (EU), which, over the last few years, has evolved rural development strategies from the point of view of food production to the diversification of activities (income basis into non-agricultural activities) [2] that guarantee the quality of life of rural communities, responses to the needs of the host society, and, more recently, to the mitigation of climate change [3].
The present study presents results obtained in cross-border regions and intends to unravel the patterns and dynamics of agritourism. As the literature shows, tourism activities are pioneers in the process of transforming the frontiers and tourism landscape, which is an opportunity for the development of innovative tourism products [4]. Effectively since the 1990s, European institutions have been promoting cross-border cooperation, particularly with EU's regional policy and, in particular, with the help of the European Regional Development Fund (ERDF), Interreg programs [5], and LEADER initiative. This initiative undoubtedly encourages a more sustainable development with respect to rural areas where tems in some countries consider lodging as agritourism (e.g., Portugal, Italy, or Israel) [28] and, in some cases, may include all hospitality services (e.g., lodging, accommodation services, food services, and recreational activities). In this latter example, there are strong synergies with the agricultural sector as, in addition to accommodation in the agricultural exploitation, food services are provided, which enhances the opportunity to interact with local production and the opportunity to participate in farm activities.
The literature has also associated agriculture-based educational activities with agritourism [33], which are manifested in activities such as flora and fauna observations [34] and experiences involving social interactions between farmers and tourists [35]. According to other authors, the lodging service is the most important service offered in agricultural exploitations, followed by sensorial experiences with local products [36], which reinforces the role of local products in creating authentic touristic experiences. On the other hand, there are also authors who have argued that direct participation in agricultural activities (e.g., olive oil harvesting or animal breeding) are those that best translate into authentic experiences [37]. Facing this diversity, Srikatanyoo and Campiranon [38] proposed three distinct agritourism products according the motivations of the agritourist: relaxation within a rural setting; enjoying agricultural experiences; and enjoyment of the quality of life, relationships with local people and farmers and adventure. They suggest that agritourism is not a "standardized recreational activity", but is adaptable to different contexts, motivations, and farmers [39].
Faced with the diversity of experiences, the present study considers agritourism activities as those that "incentives for active and participatory interaction of the guest with the agricultural practices and farm life" [23:45], involving education/awareness regarding agriculture, local traditions, and the values of rurality [39], as well as contact with gastronomy, local products, and animals [40]. Therefore, in this study, a distinction is made between "active" and "passive" agritourism activities in order to better understand the patterns of agritourism supply.
From this point of view, it is important to note that agritourism plays an important role in promoting sustainable tourism and contributes to sustainable development. The linkages between agritourism and sustainable development have been widely discussed in the literature [41], based on the general idea that "small-scale and specialized forms of tourism have often been considered ideal development tools to sustain the fragile natural, economic and sociocultural environments of many rural areas" [42]. The promotion of sustainable agritourism is important to ensure the self-resilience of rural inhabitants, to preserve the natural environment, and to promote stable and profitable production activities [39]. The literature has demonstrated that agritourism contributes positively to the three dimensions of sustainability: From an economic perspective, agritourism can contribute to income diversification [34,43] and create new jobs [44]; within the social-cultural dimension, agritourism can contribute to recovering and preserving tangible and intangible agricultural heritage [45,46], as well as increasing the social status of rural inhabitants and farmers [44]; finally, in environmental terms, agritourism can help improve public infrastructure, protect natural habitats and ecosystems [44], and promote environmental awareness and sustainable consumption habits [30]. Another important contribution of agritourism is the recovery of agricultural activities and new territorial dynamics to support new organizational networks, where farmers, tourism operators, and public authorities jointly propose solutions for rural development [47]. Currently, the global food system is gaining weight in market dynamics [48], which has contributed to the loss of biodiversity, pollution, and climate change [49]. Thus, preserving small-scale food production provides a means to combat the abandonment of territories and preserve the landscape, which is, in fact, an important tourist resource that supports a great diversity of businesses, of which tourism has been taking advantage [31,50]. In general, agritourism is widely recognized as strategy to capitalize natural, cultural, agricultural, and social capital with a positive impact on rural livelihood in low-density territories [39].
Considering the above, this paper seeks to add to the literature relating to agritourism by detecting and characterizing relationships between agriculture and tourism. This subject has been devalued in the literature [51]; however, linking agriculture and tourism provides a way to make the tourism industry more economically inclusive [52,53], and can generate a multiplier effect that benefits the local economy [51] and nurture sustainable development [23], especially if one considers the authenticity of the "experience" [23].
Based on this premise, the aim of this study was to identify and characterize the linkages with agricultural activity from the perspective of accommodation establishments, which are often localized at or close to farms [54]. The main relations were observed with the goal of identifying the main agri-food supply chains and recreational activities promoted by accommodation establishments. The starting point for this research includes the studies of Fisher (2019) [48] and Anderson (2018) [53], which clearly state the need for identifying the characteristics of agricultural tourism supply chains in order to promote products and services that valorize authenticity. This authenticity has been clearly noted in the literature as a tool for sustaining regional identity and sustainability, enhancing the roles of the landscapes [55] or local products [56,57] provided through agriculture, which are oftentimes devalued by the demand as a tourist resource [58]. However, linking local production and tourism can allow tourism to become more economically and socially integrated [53]. For these reasons, we highlight the ties and the opportunities to develop a territory that has been particularly weakened due to low population density. With the present investigation, we expect to contribute useful information for tourism managers and local authorities, in addition to contributing information regarding the establishment of the agritourism concept and demonstrating its roles in sustainable development.
To enrich these aims, the following research questions were set: • RQ1. Which accommodation typologies exist, and what type of services do they promote in the study area?
Tourism in rural areas increased since the 1970s, both in terms of demand and supply [59]. Currently, the accommodation supply in rural areas has diversified in terms of size, typology, location, food services, and recreational activities offered [60]. Therefore, in the planning of economic and social development, it is important to understand the characteristics of the supply in each tourist destination.

•
RQ2. Are agritourism activities an option provided by accommodation establishments in the study area?
The increased tourism supply in rural areas coincides with the global crisis, which is affecting rural areas and has triggered strategies for the economic diversification of agricultural explorations, providing an opportunity to enhance the value of rural heritage [59]. An example of niche rural tourism is agritourism. This typology ranges from lodging in agricultural exploration to visits to agricultural explorations [59]. Given the diversity of existing recreational activities, as well as the typology of the lodging supply, we attempt to identify the existing recreational activities in the study's territory in order to perceive how agricultural activity is integrated into recreational activities.
• RQ3. Which are the main resources that serve as the basis for dynamizing the recreational activities that promote the differentiation of the lodging offer?
The competitive advantages of tourism destinations depend on several resources, from physical to human, historical, and cultural resources [61]. This leads to respect for territorial capital. The articulation of the singularities of this capital and the differentiation of the tourist offer are fundamental elements in the generation of quality tourism. This approach emphasizes the importance of endogenous resources and intersectoral relationships as determining factors in driving innovation and sustainability in the tourism sector.

•
RQ4. Are there visible links between agriculture and tourism in the study area?
The literature highlighted that the visibility of the relationship between the tourism sector and local agriculture is very tenuous [52]. On the other hand, it has been argued that the competitiveness of tourism depends on its ability to strengthen relationships with local suppliers-in particular, food supply companies [62]-thus contributing to local economic development [53] and reinforcing the image of the destination [51]. The linkages between tourism and agriculture are not limited to economic flows between sectors, but they are also visible in relationships established between human and nature. In this regard, the literature demonstrates that, particularly in remote areas, the linkages between tourism and agriculture have been central to sustainable development trajectories [2].
These networks have particular relevance in rural areas, where farmers may take advantage of opportunities to transfer their agri-food products to new markets (e.g., theme fairs, traditional trade stores, restaurants, accommodation, and directly to tourists). Thus, the present study aims to identify the local patterns and dynamics that translate the linkages between tourism and agriculture. This analysis intends to clarify how lodgings relate to the sector either by also relying on food production or animal husbandry or as they maintain linkages with local production in the supply chains or promote recreational activities with agriculture and the agri-food sector as centers of tourist dynamics.

The Case Study
In cross-border regions, as a result of political constructions, the respective territorial identities tend to become devoid of one purpose or another. For this reason, these regions have a special interest in destination marketing [63]. The boundary between Portugal and Spain is a geographical area with particular interest in this context, especially considering the following factors: It is supposed to be one of the most fixed and stable borders in the world [64], the trans-boundary landscapes are the product of a set of processes and interactions that occur in this space and influence the local culture [65], and there is a long tradition of co-operation across the borders of Iberian municipalities, which work alongside many initiatives in terms of regional issues [66].
Border areas offer countless opportunities for tourism, inviting the tourist to come into contact with different cultures, gastronomies and local products, and ways of life, as well as providing access to valuable historical and cultural heritage. However, these territories are often devalued by national public policies and regional tourist strategies due to the difficulties of operationalizing common actions with the capacity to treat it as a "single destination" that covers the territories of two countries [67].
However, the Iberian cross-border local preserves natural areas with high environmental value, remarkable (and still under-valued) monumental heritage, and assets of cultural interest that must be preserved and revalued in favor of tourist demand for low-intensity destinations [68]. This area is ideal for guests seeking calm locations, remoteness, silence, and to learn more about rural traditions in a destination that crosses different cultures.
In recent decades, several cross-border cooperation initiatives have been developed to promote actions to enhance the quality of life of local inhabitants. These include investments into infrastructure, trans-boundary research, and dissemination through cultural and scientific events for the enhancement of tourist products based on both natural and cultural heritage. The networks between trans-boundary municipalities are visible in several projects supported by EU Structural Funds co-financing the development of cross-border regions in the form of Interreg. There are several examples from Tagus International Natural Park and the Historical Villages Network to local initiatives such as the "Periferias Festival".
In the study area, two relatively consolidated tourist axes can be identified (Alcántara (Spain)-Idanha-a-Nova (Portugal) and Valéncia de Alcántara (Spain)-Marvão (Portugal)-Castelo de Vide (Portugal). However, in the absence of a touristic strategy to structure the offering based on local resources, this study intends to contribute to the potential of agritourism evaluation via the valorization of linkages between agriculture and tourism. As this is a territory where agriculture still preserves production models in an extensive model and dominates small-scale agriculture, we believe that agritourism could be the key to promoting sustainable local development and create innovative touristic products.
Administratively, the study area corresponds to the border municipalities of NUT III Cáceres (12 municipalities) and all contiguous municipalities in Portugal (which includes four municipalities of NUT III Beira Baixa and four municipalities of NUT III of Alto Alentejo), known as "raia" in Portugal or "raya" in Spain ( Figure 1). The human geography of this territory is marked by a low rate of occupation, population aging, and progressive de-population as a consequence of migration in recent decades [69]. Isolation and weak accessibility contribute to low economic dynamism and low investment in public policies. In total, 115,806 inhabitants are registered in the area, of which 64% are concentrated in the urban municipalities of Castelo Branco and Portalegre in Portugal border ( Figure 1) over a total area about 8000 km 2 , corresponding to a population density of 14 inhabitants/km 2 ( Figure 1).

Border Landscapes and Territory Identity
The regional identity of this territory is strongly marked by human characteristics but, above all, by the physical geography of the territory. The weak accessibility and distance from urban regions helped preserve the unique natural values of this territory, which today justify the existence of nature protection areas.
The study area integrates natural heritage areas with natural protection status. Likewise, the study area includes border municipalities that are part of the Tagus/Tajo International Transboundary Biosphere Reserve, classified by UNESCO, with some overlapping areas of the Natura 2000 reserve and protected landscape areas, such as the Malcata, São Mamede, Gardunha, and São Pedro. These areas are dominated by typical Mediterranean forest species, including Quercus suber L., Quercus rotundifolia Lam., Arbutus unedo L. 1753, Quercus coccifera L., Cistus ladanifer L., Genisteae (Bronn) Dumort 1827, Lavandula, Pistacia terebinthus L. and so on.
Agroforestry predominated in a large part of the territory, with emphasis on the agro-silvo-pastoral system designated by "Dehesa" in Spanish territories and "Montado" in Portugal. It is a unique landscape occupying around 3.5-4 million ha in Spain and Portugal collectively [70] and, in total, occupies an area of about 339,899 ha in the study area. In this generally extensive production system, there are utilizations related to cork stopper, acorns, the raising of grazing animals (e.g., cattle, sheep, and Iberian pigs), and an enormous quantity of endogenous products, such as wild mushrooms, wild esparto, aromatic, and medicinal plants. These landscapes have abundant diversity as well as ecological benefits at the level of landscape patterns that offer diverse values to society, appreciated for both production and non-production functions, thus making it positively evaluated for its touristic potential [71].
In the study area, olive groves occupy around 9% (50,977 ha). These olive groves are characterized by a low density of trees, rainfed regimes, and scattered patterns of occupation. The traditional olive groves occupy the slopes of the main rivers (Tagus, Erges/Eljas, Ponsul, and Ocreza), in terracing marked by drystone walls that preserve native cultivars (e.g., Galega and Manzanilla-Cacereña), and at borders of settlements. Due to these characteristics, the olive groves create a paradigmatic cultural landscape [72]. Olives and olive oil represent a product category with characteristics that project a particular image in the minds of potential tourists and allow for a diversity of activities related to gastronomic or educational activities [73].
Unique landscapes and quality products can be found in this territory. The diversity of products is highlighted by a quality seal with local branding, particularly Products with Denomination of Origin (PDO) and Protected Geographical Indication (PGI), such as the "PDO Products of Dehesa Extremadura", "PDO Lamb of Extremadura" in Spain, and "PDO Beira Baixa Cheese" (Castelo Branco, Vila Velha de Ródão, Penamacor, and Idanha-a-Nova) in Portugal. The traditional olive tree, although with less expression, provides numerous high-quality products, including olive oil with "PDO Olive Oil Beira Baixa" (Portugal) and "PDO Olive Oil Gata-Hurdes" (Spain). In the municipalities of Nisa, Marvão, Castelo de Vide, and Portalegre-the southern area of the study territory in portuguese border-quality products include chestnuts ("PDO Marvão-Portalegre"), cherries ("PDO Cereza de São Julião"), beef ("PDO Carnalentejana"), and wine ("PDO Alentejo-Portalegre"). Effectively, food products and gastronomic specialties become the vehicle for a closer understanding of local cultures, and in this area, it is possible find a huge diversity of quality agri-food products. This context can contribute positively to develop a "culture of a place". According the literature, a culture of a place is more attractive, emotional, and seeks the sustainable development of this destiny via the maintenance of productive functions, contributing to the multi-functionality of rural areas [74].

Tourism Resources
The cross-border territory possesses unique landscape qualities [75], where the combination of natural aspects and heritage and architectural configurations supports the development of various rural, natural, and cultural tourism activities [68]. The following tourism resources are highlighted:

•
The craftsmanship and the popular culture are manifested in expressions of music ("Adufe" of Idanha-a-Nova), pottery ("Olaria pedrada of Nisa"), weaving ("Portalegre tapestry", "traditional embroidery of Castelo Branco", "traditional embroidery of Nisa"), and cork extraction, with associated traditional techniques and knowledge. There are several museums and art interpretation centers that preserve these local techniques and knowledge. • Protected natural areas (Tagus International Natural Park, Zona de Interés Regional Sierra de San Pedro, Natural network of Malcata, Gardunha and São Mamede Natural Park), natural monuments (Porta de Ródão; Figure 2c), and special bird protection area with characteristic habitats that promote various ecotourism activities, such as bird watching, hiking and nature walks, or photographic safaris. • Rivers, reservoirs, and thermal waters that invite moments of enjoyment, water sports (Albufeira da Apartadura in Marvão, Albufeira da Meimoa in Penamacor), boat trips on the Tagus River (cais fluviais in Vila Velha de Ródão, Malpica do Tejo and Lentiscais, Santiago de Alcántara, Alcántara), and thermal spas (Monfortinho, Idanha-a-Nova). • Rural architecture characterized by the use of stone (e.g., "Choça of Marvão"). To enjoy natural and cultural resources, there are several tour operators on each side of the border that offer birdwatching, astrotourism, horse rides, cycling and walking tours, and/or other active sports tours.

Accommodation Supply and Touristic Demand
Accommodation infrastructure is an important element of tourism products. Figure 1 shows the accommodation supply, according to their categories (rural hotels, local accommodation, country houses, and agritourism in Portugal and rural accommodation in Spain) in each municipality of the study area. The available statistics point to the existence of 168 rural tourism accommodation facilities with 1113 beds [76][77][78], mostly distributed across the municipalities of Idanha-a-Nova, Portalegre, and Nisa ( Figure 1). In relation to the total, only 11% of the available beds correspond to the category of agritourism.
In order to identify the main patterns of tourism activity in this cross-border territory, we gathered national statistics regarding the demand. Reference data from 2019 indicated that there were 296,661 overnight stays/year [78,79]. In 2020, due to the pandemic crisis, there was a drop in touristic demands of about 78%. However, it should be noted that Idanha-a-Nova managed to increase the number of stays during the pandemic year against the general trend [80], probably positively influenced by the municipality strategy promoting this destination through the campaign "Idanha in family," offering an agri-food box 1 . This is a great example of promoting tourism while, at the same time, supporting local production.

Agritourism Resources
The agritourism potential was observed according to the main landscape characteristics, where the traditional and extensive agri-food production modes are preserved. Another important agritourism resource is local food and local gastronomy, which valorizes fresh products. Field observations revealed some important activities that valorize agriculture activities. Briefly, the agritourism potential in the study area was identified with regard to the presence of the following agritourism resources: • Local products are celebrated in numerous gastronomic festivals (e.g., the Cheese Festival in Valencia de Alcántara and Castelo Branco; the Chestnut Festival in Marvão, "Mormentera" Festival in Alcántara; and the Olive oil Festival in Malpica do Tejo) and are recognized in specific pedestrian routes, such as PR12 "Olive Tree" in Castelo Branco or the "Ham Route" in Extremadura.

•
The cultural landscape is related with agricultural uses, particularly the agro-silvopastoral system (dehesa/montado) and traditional olive groves. • Agritourism programs encourage tourists to participate, for example, in olive picking, grape harvesting, or learning about endogenous species and products. • Agritourism lodging and other lodging typologies located at active farms enhance agritourism experiences.
Due the characteristics of cultural and historic aspects, landscapes, human geography, and the quality and richness of local food, this area has great potential to develop a unique touristic strategy, particularly based on agricultural landscapes and local food as the main basis for tourism.

Research Design
A methodology with four stages was designed ( Figure 3). In the first step, a literature review was performed to identify the main gaps and support the questionnaire design according to the objectives of this study. We carried out observation field trips and the collection of photographs to characterize the study area, as well as the compilation of statistical and cartographic data that allowed for the characterization of the territory. In the second step, the questionnaire was sent to accommodation managers in order to collect data and information that allow for the identification of the main characteristics and to detect linkages with agricultural activity. Therefore, a database was created using the Excel and SPSS version 27 software in order to support the statistical analysis. Both databases enabled the authors to design and develop a Geographical Information System application using ArcGis v.10.8., which brings together geographical and touristic patterns with dynamic information as presented in Section 2.6.
The third stage consisted of data analysis and treatment. Descriptive analysis techniques, as well as parametric and non-parametric statistical tests, were applied. Furthermore, we conducted grouping analysis to determine the linkages between agriculture and tourism with respect to the characteristics of the territory. All tests were conducted as previously performed in the literature, as detailed in Section 2.6.

Questionnaire Design
A questionnaire addressed to the accommodation households was designed and implemented. Based on the aims of the study, the survey was developed to collect information that would allow for characterization of the accommodation in study area, the identification of their relationships with agriculture sector, and the determination of the existing agritourism activities. This was organized into the following sections (Appendix A): (a) general profile of accommodations, (b) agricultural activity, (c) food products suppliers, (d) sale of local products, (e) restaurant, (f) agritourism, (g) partnerships, (h) general profile, and (i) general opinion (Table A1). The survey was tested, and its completion took an average of 30-60 min (personal interviews) or about 10 min (online questionnaire). To test the reliability of the questionnaire, the Cronbach's alpha value was calculated. The Cronbach's alpha for the treated questions was 0.938, indicating very satisfactory levels of internal consistency and reliability for the questionnaire and its dimensions [81].

Data Collection
The research was conducted on cross-border regions (Centro region of Portugal and Extremadura, autonomous community of Spain) comprising a total of 20 municipalities. The information on demand was obtained from surveys and personal interviews. Data collection began in February 2020. However, limitations imposed by the COVID-19 pandemic situation conditioned the face-to-face data collection. To overcome this difficulty, online questionnaires were designed and distributed to all accommodations in the database. However, the response rate was very low (5%). Personal interviews resumed from January 2021 to June 2022 and were subject to the regulations in force. The target group for this research was the holders of accommodations. Accommodation households were randomly chosen regardless of their category and according to their denomination:

•
In Portugal, we covered rural accommodation types, specifically "countryside houses", "agritourism", "rural hotels" (Decree-law n. º 80/2017 of 30 June, that constituting the common diploma to all tourist enterprises), and "local accommodation" (Decree-law n. º 62/2018 of 22 August, which indicates the rules for tourist accommodation in residential buildings). Around 40% of the sample comprised "local accommodation", while "country houses" represented 42%.
• In Spain, we covered rural accommodations, specifically "rural hotels" and "rural accommodations" (Decree-law n. º 65/2015 of 14 April), that establish and classify rural tourism accommodation in the Autonomous Community of Extremadura. The sample comprised about 18% of accommodations.
The data used were obtained from the Statistical Office Tourism of Portugal (last updated on 31 December 2021) [79] and Extremadura Tourism official website, which is run by the regional government (last updated on 31 December 2021) [78]. The initial database included 251 accommodation establishments. The basic information was retrieved from these sites, and later georeferenced each existing business that was officially open. The selected accommodations were first contacted by phone and email in order to introduce the type of research and to obtain consent for their anonymous participation in it. At the same time, this step allowed us to validate whether the accommodation was active. We verified that only 202 accommodations were active, representing 80% of the supply announced in the official data. This situation probably resulted from the negative impact of the COVID-19 pandemic. About 17% of the total did not wish to participate in the data collection process. Due to the COVID-19 pandemic situation, it is important to highlight some aspects observed that possibly influenced the availability of accommodation managers to collaborate in data collection, namely the following:

•
Unstable phase for the sector; • Business model changed or adapted; • New services provided; • New investments; • Suspension or cessation of activity decisions.
It is important to note that despite not being the objective of this work, the literature argues that agrotourism is an opportunity for rural territories to captivate tourist demand in the post-COVID-19 era [82]. Therefore, the present study will be an important contribution to reveal what exists and what resources can be relied upon to enhance this destination.
In general, the sample comprised 168 accommodations. After this validation, the online questionnaire was sent to the validated accommodations. Due to difficulties in collecting the questionnaires, a large part of the questionnaires required a personal visit to the accommodations to fill them. In this case, much of the data were collected via personal interviews.

Sampling
For research purposes, 168 survey questionnaires were distributed, of which 89 surveys were collected and processed from the field, covering 53% of the total cases. The number of surveys proved to be sufficient to determine the goodness of the proposed methodology, considering that the margin of error with 95% confidence in the most unfavorable case was 7.1%, while that in the most favorable case was 4.3%.
Profile of Respondents Table 1 shows the sociodemographic characteristics of the sample. The sample comprised accommodation managers (N = 89), and the variables gender, age, level of education, study level, and time spent in business were recorded. About 41% of the managers were female and 59% were male. The average age of the participants was 52 years old (Sta. Dev., 10.5). Effectively, the ages were mostly between 50 and 59 years old (37%), with a considerable number of managers over 60 years old (22%). Most participants studied to a high-school education level (74%); however, only 7% had qualifications related to tourism and 21% had qualifications related to agriculture. Agricultural activity 50 56 It is important to note that most managers had another occupation, and they were only working part-time in their accommodation business. About 56% of managers participated in agricultural activities, of which only 30% had businesses in this area, while 70% conducted agriculture activities for self-consumption purposes. Only 13% stated agricultural investments to diversify the funding sources, while 20% had agricultural activity to recover and valorize the family heritage or to create their own job.

Data Analysis
The most appropriate method to fulfil the research objectives was considered as quantitative and geographical analysis. We followed the steps below to answer the research questions guiding this investigation:

•
For descriptive statistics, we present arithmetic means and standard deviations, and frequencies and percentages are used to describe qualitative variables. This procedure was used to answer the first and second research questions-which accommodation typologies exists and what type of services they promote-thus describing the promoted agritourism activities (Sections 3.1 and 3.2).

•
To address the third research question, which allowed us to identify the main resources used by the accommodations to promote recreational activities, the quantitative and qualitative results were analyzed via cross-tabulations by Pearson Chi-square testing (significance level α = 0.05). The value of the Chi-Square statistic indicates whether or not there exists a statistical relationship between variables in the cross-classification table. This technique has been widely applied in the literature [21,83].

•
To answer the fourth research question (Section 3.4), related to linkages between tourism, agriculture, and their spatial patterns, we applied a geostatistical technique (i.e., Grouping analysis). This method allowed us to identify homogenous groups with the same characteristics of variables, conveying relevant information contributing to regional development [84]. With this method, the R 2 value is calculated, which represents the efficiency of the grouping analysis. Both the R 2 value and number of groups vary as functions of spatial constraints [85].
This approach allowed us to delimit the spatial patterns and dynamics observed in the territory. To apply the clustering analysis technique, the territorial analysis unit was the locations of the lodgings. Then, the Delaunay Triangulation method was used, with the Euclidean distance as a reference. This method generates Thiessen polygons, in which the vertices are the existing centers; that is, the common axes between points. According to the literature, the clustering analysis methodology constructs a connectivity graph representing the neighborhood relationships between entities, generating a minimal expansion tree. This reflects the spatial structure of the lodgings and the common varieties linked in the analysis. Subsequently, the number of clusters containing the best possible solution, as recommended by the F statistic index, can be determined. Initially, the first minimal expansion tree was created with two groups. However, these groups do not provide the opportunity to distinguish the similarities and differences between them. Analyzing the obtained results, an F Statistic of 9.8 was the best value, indicating that the creation of six groups can best document the specific relationships between tourism and agriculture in the territory. For the application of this technique, 15 variables were used ( Table 2). The selection of variables was made after numerous analyses, discarding the variables with the least explanatory relevance regarding the relationship between agriculture and tourism. For the final selection, variables with an R 2 value greater than 0.12 were considered when the defined groups were obtained. The variables patents in the Table 2, included in the model had, as a unit of measurement, 0 (no) or 1 (yes) for the following cases: If the accommodation also has an agricultural activity (AgriProd), if the accommodation is located on a farm (AccFarm), if the accommodation is located on an active farm (AccFarmA), if they have animal husbandry (Animals), if they have agri-food production only for self-consumption (PSelfCons), if they have agri-food production for sale (PSell), and if the accommodation has local networks with farmers, restaurants, or other agri-food tourism/sector agents. The variables expressed as quantities were as follows: Recreational activities that value agriculture (AgriAgri), contact with animals (AgriAnim), contact with local products and gastronomy (AGroGast), agritourism activities that enhance education/awareness for rurality and sustainability (AgriEdu), and finally, active agritourism activities (that is, those that enhance the involvement and direct the participation of the guest in the daily management activities of the farm (e.g., feed animals, olive oil harvesting, and so on)). Variables expressing the diversity of agricultural cultures and fresh/processed products by the lodging owners were also considered, as well as variables that allowed for determining whether guests have access to self-produced products or local products, either by direct consumption during their stay or via the possibility of acquiring them directly at the lodging for consumption after their stay. The relationships between agriculture and tourism are detailed in Figure 4.

Main Characteristics of Lodging Supply
About of 41.5% of accommodations were founded during the period 2015-2019 and about 23.5% were created during the pandemic crisis that were previously planned. This reveals the opportunity to increase the accommodation supply in low-density territories, in response to demand trends [6]. Effectively, mostly accommodations were integrated into natural reserves (74%) and 45% were close to the border, with areas characterized by low populational and business density (Table 3). Notably, 50 cases were localized on a farm, of which 80% were active.  18 20 The offer of lodging in rural areas, designated by "Country Houses", was the most representative in the lodging offer, with 34% of the available beds ( Table 4). The rural tourism typology is more oriented towards the valorization of the local way of life, promoting recreational outdoor activities that allow contact with nature, culture, and proximity to local populations and rural traditions. Previous studies confirmed that there is increasing interest in rural areas as leisure spaces, promoting a return to the origins and the "authentic", which is often associated with nostalgia for the "good old days" [20]. This was followed by "Local accommodation", a typology that proliferated in recent years and diversified tourism offerings in the Portuguese territory. On the other hand, the "Rural Hotel" typology took second place in terms of the number of beds available in the territory (27%). To understand which accommodation typology predominates, according to the activities they promote, we followed the classification used in previous studies [86]. The results showed that most of the lodgings fell into the category "Rural tourism" (57%); that is, in general, they promote contact with local traditions, history, and culture while the Bed and Breakfast (B&B) is the center of the experience. This is followed by the "Leisure tourism" category (30%), which promotes contact with the way of life of the farmer and the rural community. Only 5% of the lodgings promoted "ecological experience". However, this typology offers access of fresh products. Finally, only 9% of the lodgings fell into the "agricultural experience" category ( Figure 5). In this case, only seven lodgings were included in the Agritourism typology in which B&B and pool are the main facilities and it only had 11% of the beds in the study territory (Table 4). In fact, a considerable part of the existing lodgings was centered on "Bed and Breakfast" services (68%), while 44% offered only lodging services with access to all house divisions, which means that this typology is not linked to any program of activities or access to local products promoted by accommodation managers (Table 5). Only 23% of the lodgings in the territory had tourist packages (i.e., providing other recreational activities or access to tourist infrastructure). Birdwatching, walking or cycling tours, photographic tours, or boat trips on the Tagus River were highlighted. In addition to the dynamization of personalized activities that qualify the offers, only 23% offered tour guide service and 12% provided a meal service. The present study indicates that the typology of accommodation can influence tourism activities and services provided in the region. For example, observing the typology of lodging activities and services, it is important to highlight that 65% of the lodgings had a breakfast service. However, the "Local accommodation" typology was the one that registered a tendency centered only on the "bed" and was, in general, devoid of a large part of tourist services. Among the accommodations offering a breakfast service, 45% had their own production to serve during the stays, while 73% had local networks to provide local products. Among their own production, fresh seasonal products (fruits and vegetables), jam (use of seasonal fruits and vegetables), and olive oil were highlighted. The local products most consumed in the lodgings were bread, cheese, and traditional cakes. According to previous studies, this pattern contributes to the development of a territory, making it possible to strengthen local agri-food chains and to reinforce the role of high-quality products [87]. At the same time, it allows for the creation of a destination identity based on the culture and history of the place where the agri-food is produced [51]. As gastronomy is an integral part of the tourist experience, food and access to local products should be a central concern in the structure of the touristic supply. This will make it possible to add value to tourism products and promote the sustainability of tourism [88].
According to the data in Table 5, tourism services provided by the lodgings were, in general, reduced. It can be seen that most of them were committed to an offer that allows the guest to enjoy the facilities of the premises by providing access to the swimming pool (61%), bicycles (31%), and access to and enjoyment of landscaped spaces and green areas (27%).
It is important to note that some patterns identified during the data collection period may have been strongly influenced by the pandemic crisis, which resulted in new services and/or changes in accommodation management routines, namely the following:

•
Offers based on the accommodation product and limited offer of recreational activities involving personal contact; • New ways of supplying breakfast: breakfast baskets and breakfast served in the room; • Enhancement of local gastronomy: the sale of local products at the accommodation, and access to the vegetable garden; • Local networks between accommodation and catering: meal delivery service at the accommodation, and meals prepared with guests.

Recreational Activities: Is Agritourism an Option?
The collected data revealed that 63% of the sample (56 lodging establishments) provided activities related to agritourism. However, it is important to note that only 9% of cases carried out the dissemination of agritourism products and services with an active and regular communication strategy-that is, communicating activities, products, and encouraging contact with agricultural activity and/or their own production. This means that most activities referred to by the participants are carried out informally and, in general, are requested by guests. This reality means that the value associated with the service provided and the opportunity to create a positive impact on the territory is lost. In fact, when asked about the value of the activities promoted, most stated that they are included in the value of the accommodation or depend on the allocated resources.
It is also important to note that most lodging managers who accumulate functions with agricultural activity declare that they are committed to extensive agri-food production (35%), traditional methods (17%), and rainfed agriculture (26%). These data are extremely interesting insofar as they can form the basis for the offer of agri-ecotourism products that value not only local products but also traditional methods that favor quality products and environmental protection. This information is relevant and can contribute to the promotion of sustainable products and the enhancement of the local potential.
Despite the informal nature of most promoted activities, there were patterns and dynamics in the linkages between agriculture and tourism that are interesting to explore. In particular, if one considers that agritourism takes, as its main impact, the valorization of the fresh agri-food products, it can also be thought of as a vehicle for education and awareness regarding the emerging challenges faced by rural areas and traditional products, with respect to climate change.
A number of agritourism activities have been referenced in the literature [86]. To facilitate the collection of information, several activities were listed in the questionnaire and subsequently grouped into categories according to a previous study [89]: (a) enjoyment of the agricultural landscape and entertainment on the farm (24%); (b) tasting experiences/contact with endogenous products (21%); (c) contact with animals (19%); (d) valuing local agriculture and rural livelihoods (14%); (e) learning about the farm lifestyle and activities or rural traditions (12%); and (f) doing something, especially with the traditional process (11%). These results confirm that there exists a great diversity of patterns in agritourism activity offerings, with emphasis on activities related to enjoying the agricultural landscape and contact with local products.
As shown in Table 6, there was a predominance of activities that value agricultural activities (37%), including experiences such as olive picking, harvesting, agriculture, or visits to wineries and vineyards. Agritourism activities that promote education/awareness were the second most popular (24%). These include experiences involving learning about the processes of cork extraction and valorization of the forest, beekeeping, contact with farmers and the local population, and debates or lectures related to sustainability. These experiences offer opportunities to enhance the value of local and sustainable products [19,30] while, at the same time, triggering more sustainable lifestyles [90]. In third position were experiences valuing gastronomy and local products (22%). Finally, activities that promote contact with animals (17%) were the least offered. Even though the existing offers contemplates the approach to agricultural activities and local products, it was found that the majority fell into the category of passive agritourism (68%), as most of the activities place the participant in the role of a mere observer. This result reveals that active tourism-which promotes direct contact with agricultural activity-is still little explored and under-valued as a recreational offer.
It would, therefore, be advisable to reinforce that these activities are conducted informally and, in general, without any return for the farmer/host manager. This is a reality that must urgently be improved.
It is important to note that these activities are mostly (54%) organized by the accommodation manager, while 46% of the activities are organized through local partnerships. The main business partners include tour operators, public authorities, and local farmers (representing 13% of the cases). However, as mentioned above, most lodgings do not offer these structured products, and the activities are conducted informally or as requested by the guests. Only 35% of the lodgings charged for agritourism activities, with an average value of around EUR 27.60/person.

Resources Based on the Promotion of Agritourism Activities
To determine the resources used to promote agritourism activities, a survey including the item's characteristics of the lodging, infrastructures, location, and agricultural activity promoted by the lodging managers was carried out. This information made it possible to identify seven categories of resources based on the characteristics of the landscape, agriculture, agri-food, local culture, rural and intangible heritage, rural settlements, and non-specific rural activities. To verify whether there exists an association between agritourism activities and the basic resources for the organization/dynamization of recreational activities, Pearson's Chi-Square was calculated via cross-tabulations ( Table 7).
The results indicated a significant association with the resources associated to the landscape (Pearson Chi-Square p = 0.010), agriculture (Pearson Chi-Square p = 0.003), and agri-food (Pearson Chi-Square p = 0.003). In more detail, there was a positive association with the valorization of the following resources (Table S1) • As local human capital, "local knowledge" is more relevant in the design of agritourism activities (Pearson Chi-Square p = 0.010); • As expected, localization in "agricultural exploitation" is important to promote agritourism activities (Pearson Chi-Square p < 0.001); • For non-rural resource typologies, "online sales of agri-food products" (Pearson Chi-Square p = 0.031), "digital content related with agritourism" (Pearson Chi-Square p < 0.001), and facilities, such as "swimming pools" (Pearson Chi-Square p < 0.001), were highlighted.

Linkages between Agriculture and Tourism: Which Territorial Patterns and Dynamics Exist?
The agricultural landscape has interesting characteristics, providing potential for the development of agritourism activities. These landscapes are a source of wildlife, food production, and biodiversity preservation-resources that can give rise to a wide diversity of recreational activities while, at the same time, generating income for farmers. In the study area, about 50% of the lodgings were located on farms, thus being able to benefit from a diversity of resources. This result indicates that agritourism activities are not exclusive of lodging designated as "agritourism", which suggests that other accommodation management models can take advantage of their location, the landscape resources, and agricultural activities. This scenario provokes interest in perceiving which agritourism patterns and dynamics exist in the territory. To determine the links between agriculture and tourism, spatial clustering obtained via Delaunay Triangulation was applied, considering the Euclidean distances between the georeferenced lodgings. Initially, the model generated two groups (Figure 6a), making it unstructured. Based on the Pseudo F-Statistic value of 9.10, we proceeded to apply the same technique, instead generating six groups that reflect the different patterns and dynamics of the relationship between tourism and agriculture (Figure 6b). This analysis provided the spatial relationship of agritourism patterns and dynamics for each group, and these are distinguished by the characteristics of the lodgings, services, and agrotourism activities. The application of the clustering analysis to the 15 variables analyzed resulted in a satisfactory model as, according to the coefficient of determination (R 2 ), only one variable was below 20% (agritourism activities based on local gastronomy). The variables that contributed most to the definition of the delimited groups made it possible to determine links between tourism and agricultural practices. These links were particularly visible for accommodations with the following characteristics (Table 8): (a) Accumulated functions associated with agriculture, including animal husbandry, food production, and the existence of agricultural crops; (b) Offering recreational activities that promote contact with agriculture and local traditions, favored by the location in the agricultural exploration. In greater detail, the variable "agri-food products" contributed most to the characterization of groups 1, 2, and 5. In this case, the products of the olive tree and of Dehesa/Montado stand out, particularly olive oil, honey, jam, meat, and aromatic plants. In the case of the variable "Agritourism activities (agriculture)", it is more representative in groups 2 and 3, while the variable "Active agritourism" stands out in groups 1 and 3. However, when the data were analyzed in an integrated manner, it was observed that the agritourism is still little explored in the accommodations represented in groups 4 and 6-which represent 62% of the cases-for which weak links with agriculture were detected.
This technique allowed us to generate box plots of grouped variables (Appendix B- Figure A1), documenting the characteristics of each group. In addition to the characteristics of each lodging, activity, and service, it was possible to describe the different patterns and dynamics of agritourism. According to the main characteristics of the accommodations, it was possible to observe the heterogeneity of the dimension of the groups, distinguished by the following characteristics: • Group 1 "Lodging with production/processing agri-food activity": This group represents 7% of the lodgings, which presented more robust linkages with agriculture, particularly in variables "products" and "crops", which had higher representativeness (5.50 and 2.50, respectively). Among the main crops, the olive tree stood out, and olive oil and honey were among the main products. The main resources supporting the agritourism activities detailed in this group were autochthonous varieties and breeds. The most outstanding agritourism activities in this group were related to gastronomy and contact with animals. This group has enormous potential for boosting agritourism activities, thus enhancing the value of agriculture, especially by promoting sustainable practices and the preservation of local varieties that are being abandoned or replaced. • Group 2 "educational agritourism that values olive groves": This group includes 9% of the lodgings, in which a greater contribution of the variable "products" was verified, and in which the olive grove had an important weight. This group includes lodgings with a significant role in the dynamization of agritourism activities that promote education and environmental awareness. Activities such as guided visits to olive groves, thematic lodging dedicated to olive oil, and visits to olive mills have been developed. This group presents enormous potential for the development of the olive tourism, once both the landscape elements (olive groves) and the derived products (olive oil) are integrated into the accommodation supply. Among the main basic resources for the dynamization of agritourism activities are the cultural resources of the landscape and the autochthonous varieties. • Group 3 "active agritourism that values enotourism": This group represents 6% of the lodgings, with greater weight of variables "AgroAgri", "AgroA", and "Crops", particularly regarding vineyards and other Mediterranean crops. Active agritourismthat is, promoting direct contact with agricultural activity-is particularly important in activities related to enotourism. It is also interesting to note that the variable animal husbandry was highly valued in this group, contributing to the dynamization of agritourism, especially with respect to promoting contact with animals. This activity is closely related to the valorization of the agro-silvo-pastoral landscape (Dehesa/Montado) characterizing the territory. • Group 4 "rural tourism": When analyzing the contribution of each variable in each group, it was observed that group 4 presented insignificant weights in all variables. This group represents 25% of the analyzed lodgings, among which Local Accommodation typology lodgings stand out. This group is characterized by a weak relationship between tourism and agriculture and, therefore, by a weak contribution to the development of qualified agritourism offers. • Group 5 "agritourism that values Dehesa/Montado": This group represents 17% of the lodgings, which are characterized by the valorization of Dehesa/Montado products, including meat, honey, and other wild products. Extensive animal husbandry is also a characteristic of some of the cases represented in this group. These characteristics are interesting to the extent that they enhance outdoor activities in a unique, biodiverse ecosystem with multiple traditions.
• Group 6 "passive or non-existent agritourism": When analyzing the contribution of each variable in each group, it was observed that group 6 had insignificant weights in all variables. This group represented most of the accommodations (36%), and was characterized by weak or non-existent relationships between tourism and agriculture and, therefore, scarce contributions to the development of qualified agritourism offers. One of the detailed potentialities that can contribute to the development of new agritourism products is the establishment of collaborative networks with local farmers, strengthening the supply of structured and qualified products. This typology seems particularly suitable for the demand that is only looking for accommodation and not looking for any activities.
Despite the diversity of characteristics between groups, the territorial analysis of the groups presented an interesting configuration, associating the groups to the following territorial references: Malcata and Gata mountains (Group 1); the valorization of the proximity to the border and to the Tagus river, where the olive tree plays an important role in the definition of cultural landscapes (Group 2); and São Mamede mountain (Groups 4 and 5).
However, in the case of group 6, about 34% of the accommodations were included in the UNESCO Reserve. This fact may be an indicator of the fragile relationship between tourism and agriculture, as rural and nature tourism modalities are more established. Therefore, taking into account that this is an area with agro-silvo-pastoral Dehesa/Montado land use over 48% [89], it would be interesting to explore the segments of agri-ecotourism, thus promoting sustainable production and the valorization of local and seasonal products [19,91].
In general, the results obtained in the Grouping Analysis showed that it is possible to explore the relationship between agriculture and tourism to derive a new configuration, which promotes the agritourism supply. For the structuring of qualified products, it is important to value local knowledge, which also stands out as a basic resource in the definition of agritourism products, as well as to encourage networking between accommodation establishments and other farmers.

Discussion
The tourism system of the cross-border territory is complex, in which areas where the tourism is apparently consolidated prevail (e.g., the Castelo de Vide-Marvão-Valéncia de Alcántara and Alcántara-Idanha-a-Nova axes), based on cultural tourist dynamics which coexist with other realities, often disconnected from recreational activities and local attractions. This reality makes it difficult to define specific tourist typologies, especially in a territory integrated into a protected natural area, where the natural and cultural heritage require reinvention as a touristic resource. In this regard, the literature demonstrated that agritourism might play an important role in heritage preservation [46], and for this reason, it should be considered as a strategy for sustainable local development.
The obtained results indicated that the agritourism supply in the study area values agriculture (37%), educational/awareness activities related to agriculture (27%), local food and gastronomy (22%), and contact with animals (17%). However, these activities are mostly (68%) based on passive contact with agriculture [21], which can be justified in terms of several factors: Limitations imposed by the pandemic crisis, small business scale, devaluation of the touristic product by supply or demand, or limited access to base resources. Regarding this issue, it is important to make some considerations about the concept of agritourism. Despite the discussion around the concept of agritourism [22,24,25], "authentic agritourism" has been indicated as an authentic experience "that promotes understanding of agricultural practices of small-scale farms and increases the understanding of its social relevance" [23]; that is, there should be a clear understanding that agricultural activities are part of the tourist experience, which allow the tourist to revive memories and satisfy their nostalgia [92]. This approach is closely related to the aesthetics and sensorial experiences of the guests as attractive touristic resources [23]. In response to this trend, this investigation highlighted the need to improve the agritourism supply, facilitating the development of a strong and articulated strategy to redefine agricultural tourism based on the following ideas: Linking farming activity (own production/transformation) to recreational activity at the accommodation; adapting the production model to tourism activities, taking advantage of crop cycles and seasons; protecting and valuing the local knowledge and endogenous breeds of plants and animals; and promoting sustainable and traditional agriculture practices. From this point of view, the above can serve as unique ingredients promoting "authentic agritourism" anchored on the agritourist characteristic of "atmosphere", which does not call into question whether an agritourism activity comprises an active or passive experience. In detail, the literature suggested that the concept of experiencing atmosphere is influenced by a farm's structure, farm life, the involvement of the guest in the activities of the farm, and the access to and enjoyment of farm products [23]. Thus, the present research indicates that, even more important than the existence of an offer of recreational activities based on "active" agriculture, a positive impact on the education and provoke changes to sustainable habits in the daily life of guests should be promoted. Likewise, the creation of new agritourism experiences should be encouraged for accommodations that have no link to agriculture. Moreover, it may be more beneficial to create an interconnected network between farmers and tourism service providers. This would greatly increase the impact of agritourism on the territory.
Another important contribution of this study was identified and characterized in terms of the relationship between agriculture and tourism in order to perceive the potential of agritourism in the cross-border regions between Portugal and Spain which have been affected by population decline for many years. Effectively, small-medium-sized businesses predominate, which face difficulties caused by the pandemic crisis, as well as those related to overcoming the challenging characteristics of non-massive destinations. Against expectations, this study provided evidence of the resilience and adaptability of the sector facing the challenges in terms of creating or adapting new products and services to overcome the pandemic crisis and to develop new opportunities. Once located in rural areas, the valorization of agriculture resources is evident. Within the context in which sustainability has become important, agritourism activities can serve as a key strategy promoting this area's sustainability, particularly for the following reasons: Tourism activities must be developed in a manner complementary to agriculture; furthermore, all endogenous resources should be combined with natural and cultural resources, in order to strengthen the cross-border identity, with multiple effects on various economic sectors. The literature has demonstrated that these ingredients have direct impacts on job creation and economic performance [2].
The results obtained highlighted the role of the landscape in promoting agritourism activities. The literature has highlighted the interest in traditional agricultural landscapes for their historical and cultural value, while emphasizing the importance of more sustainable practices for their contribution to biodiversity conservation [93]. Another crucial resource highlighted is related to agricultural activities. A positive association between the "olive grove" (agricultural crops resources-based) and agritourism activities was identified, as well as with "olive oil" (agri-food resource-based). This result can be confirmed as the olive grove is identified as a scenery associated with recreational activities, especially pedestrian routes in the olive grove and observation/participation in olive picking. Furthermore, the valorization of the olive oil was particularly related to activities ranging from visits to olive mills to tasting the olive oil (product with PDO classification). This type of tourism segment, recognized in the literature as "olive-oil tourism [94] or "oleotourism" [95], has registered an increase in demand [96]. Thus, the opportunity for the agents in the territory to define qualified products articulated with other experiences in this destination may be reinforced. As the literature demonstrated, the combination of "oleotourism" with other rural activities related to nature, local heritage, and culture [87] can have a synergistic effect on demand and increased tourist satisfaction [50]. In particular, the case of enotourism also stands out. Our study revealed that this tourism segment provides an interesting contribution to the promotion of agritourism activities, enhancing the supply of tourism products that value local gastronomy and traditions, as evidenced by the literature [97].
Another contribution of the present study was the production of territorial information that makes it possible to visualize unified tourist spaces that are similar to one another [85]. This information is important for helping tourism planners, especially in creating homogeneous and differentiated products based on their endogenous potential [40]. The generated model was based on six distinct groups. Despite their small size, groups 1, 2, 3, and 5 (39%) provided offers that match the potential of agriculture, agricultural products, and local relations. These groups present interesting characteristics and dynamic activities that make the landscape enjoyable for outdoor activities from spring through to winter (e.g., harvesting, olive picking, sheep shearing, mushroom picking, sowing, and so on). On the other hand, groups 4 and 6, which comprised 61% of the cases, revealed that the potential of agritourism remains underutilized in a large part of the territory. This reveals the need for a concerted territorial strategy between Portugal and Spain to enhance border tourism based on agriculture and its products. This strategy could include the following actions: (a) the enhancement of the multi-functionality of agricultural holdings; (b) the creation of annual programs with tourist offers that take advantage of the natural cycle of the main local crops; (c) the enhancement of gastronomic and autochthonous products; and (d) the creation of educational programs based on the natural and cultural values of the agricultural landscape.
On the other hand, among the main gaps of this investigation, we highlight the following: the fact that the data collection period was during the pandemic crisis and the fact that the consequent possible influence on the activities or services provided may have influenced the results; the reduced number of participants; and the reduced number of variables to explain the patterns and dynamics of agritourism applied in the grouping analysis, as well as the territorial arrangements that excluded similar cases due to their territorial distance. This situation was verified, for example, in some cases of agritourism or countryside house accommodation typologies with links between agriculture, which were thus excluded from groups with active agritourism or with a greater relationship with agricultural activity. According to the literature [85], groups were formed according to the Euclidean distance, and it was not possible to evaluate the real distance between lodgings according to communication routes. According to this limitation, one proposal for future research is to apply a network analysis technique as a criterion to establish the concept of real distance. Despite these limitations, the analysis obtained indicated that the relationships between agriculture and tourism detected were, in general, tenuous due to invisibility in search engine results that demonstrated a non-structured agritourism supply. In addition to this result, the dynamics observed suggest that there is enormous potential to structure the accommodation supply based on the potential of the landscape, agriculture, and agri-food products.

Conclusions
In this paper, the agritourism phenomenon was investigated with the aim of providing an original contribution to the debate on relationships between agriculture and tourism, demonstrating that agritourism is not exclusive to any specific lodging typology and is not a standardized recreational activity; in contrast, it is adaptable to different contexts or specific atmospheres, as well as the motivations of both the visitors and farmers.
The main results revealed that the agritourism in the study area remains invisible in the strategies of tourism dynamization (cross-border regions between Portugal and Spain), and most agritourism activities are not publicized via an integrated communication strategy focused on agritourism products or services. This reality leads to a loss of the opportunity to create authentic and differentiated products and, consequently, to the loss of the value of the services provided. Despite this evidence, the obtained results were surprising.
First, agritourism activities are not exclusive to the agritourism accommodation typology; namely, there were other accommodation typologies that also had links to agricultural activities, using them as a resource to create different experiences that involve guests in agriculture and rural daily life. These were mostly activities where the guests are mere spectators, thus being able to enjoy spaces of silence and harmony with nature.
Second, the supply of agritourism activities rely on the agricultural landscape, traditional crops, and traditional products to create distinctive programs that combine a wide diversity of activities. Therefore, relevant new products can act as a driving force to valorize this territory, which transcends cultures, crosses borders, crosses natural areas, and where local products have a relevant territorial vocation.
Third, visible linkages between agriculture and tourism revealed themselves with different scales of guest involvement in farm activities. We observed supply profiles based only on accommodation, bed and breakfast services, and participation in recreational activities. These patterns reveal opportunities to choose a remote destination in order to enjoy the quiet environment or to have experiences to learn more about agricultural and rural environments.
Fourth, the methodology adopted revealed that there are patterns of lodgings that take advantage of their locative advantages, particularly agri-food resources. It should be noted that the case of olive groves and dehesa/montado have agritourism potential that is still underused in some areas.
Finally, a notable part of the tourism managers (50%) developed activities related to agriculture. This is an important contribution to local development, as a consequence of new investments to recover familiar heritage, endogenous resource protection, and promoting agriculture activity as complementary activity in the face of inconstant tourist flows. In particular, the production of olive oil and the valorization of extensive animal husbandry and the products of the agro-silvo-pastural system of Dehesa/Montado (cork, honey, meat) were highlighted. This pattern provides an opportunity to valorize the working farm as a central and characteristic feature of a distinct form of tourism in particularly in these cross-border region.
In conclusion, this investigation suggested some directions for the promotion of sustainable development in the considered cross-border region, which will ultimately depend on the joint actions of trans-border public entities and local businesses in terms of drawing up and implementing a tourism strategy that focuses on agritourism products as a complement to rural, natural, and cultural tourism products. This strategy should take, as its starting point, the characteristics of the territory and its natural vocation. In this case, the opportunity for each tourism operator to develop robust strategies that complement lodging and recreational activities, in which agriculture should be a strategic resource, can be made more democratic. This reality makes agritourism as a recreational activity possible regardless of the type of accommodation available. This will make it possible to create tourism products that include different services, from lodging and catering to recreational activities. This will require a clear and robust commitment to a local development policy that supports small-scale production, the multi-functionality of rural areas, the enhancement of endogenous species, and sustainable models at the expense of intensive agriculture. Funding: This study is part of the research carried out during the execution of the project "Agritourism in the dehesas of Extremadura: an opportunity to increase agricultural income and population fixation in rural areas", and its code is IB20012. This research was funded by the Ministry of Economy, Science and Digital Agenda of the Regional Government of Extremadura and the ERDF (European Regional Development Fund). Dora I. R. Ferreira was the beneficiary of a nationally funded individual PhD scholarship through the Foundation for Science and Technology (FCT, Portugal) with reference number 2020.05487.BD.

Data Availability Statement: Not applicable.
Acknowledgments: The authors would like to thank the interviewees for the information provided and express their sincere appreciation to the reviewers.

Conflicts of Interest:
The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analysis, or interpretation of the data; in the writing of the manuscript; or in the decision to publish the results. Table A1. Variables collected in the survey to describe the supply of rural lodging and to identify the relationships with the agricultural sector.

Type
Factor Levels

Main characteristics
Types of accommodations Only one option: Rural hotel; Local Accommodation; Agritourism; Rural accommodation; Country Houses.
Year Numeric

Location
Single Choice: Rural/urban area; small towns; natural areas; agricultural operation Elements of the landscape Multiple answers: Olive grove; Orchard; Vineyard; Pasture/Montado (agro-silvo-pastoral system).

Main infrastructures
Multiple answers: Stone wall; Local varieties; Pastures; Rural roads and single trails; Beehives; Traditional oven; Mills; Gardens; Vernacular architecture; Others.

No. of beds Numeric
Main services/activities Multiple responses: Swimming pool; Bicycle; Garden; Kitchen access; Meals on request; Breakfast included in price; Tour guide; Advantages of access to local/regional cultural infrastructures; Experience and tour packages    Benefits of proximity to another country/culture Descriptive Appendix B Figure A1. Parallel box plot obtained in Grouping Analysis using the ArcGIS software.