Since 2013, the olive landscapes have gradually degenerated due to the spread and establishment of Xylella fastidiosa subsp. pauca (hereafter Xf) in Apulia, southern Italy. From 2013 to 2019, a total of approximately 54,000 hectares of olive orchards in the south of this region have been seriously damaged, and their restoration will progressively regenerate the economic, social, cultural and environmental nonmarket benefits. Since there is a willingness to restore the affected landscape in the best interest of the local citizens, this research aims to predict their preference heterogeneity and willingness to pay (WTP) to improve this landscape and continue research and experimentation in relation to Olive Quick Decline Syndrome Disease by the bacterium. For this purpose, a choice experiment method is used. The social field survey includes a representative sample of 683 respondents in three major cities (Foggia, Bari and Lecce) of Apulia region. The results reveal that for the local citizens interviewed, the most appreciated olive landscape services are cultural heritage and aesthetic values. In addition, the findings revealed citizens’ positive appreciation of improving the damaged olive landscape, while respondents are not willing to pay a premium for research. The results show that the average value that Apulians are willing to pay for landscape restoration is about 5.7 million of € per year. Further, this research has implications for land use planners in the study area, which faces issues of harmful pathogen management and land revival.

alien species, biological invasion impact, choice experiment, economic costs, ecosystem services, environmental changes, social perception, willingness-to-pay

Olive-growing is recognized as a multifunctional ecosystem in Italy. It shapes the landscape of the countryside with a particularly visual spatial representation (Lanfranchi and Giannetto 2012). Furthermore, this aesthetic value is not only the absolute non-market feature of this ecosystem, but provides considerable socio-cultural services (Severini 2006), economic effects (Viganò 2006), ecological benefits (Bernetti et al. 2006; Torquati et al. 2006), and conservation of the agro-biodiversity (Corrado et al. 2011; Fernández-Habas et al. 2018). However, since 2013, a part of this rural landscape has increasingly lost these attributes in the Salento Peninsula of Apulia in southern Italy. Consequently, a total of approximately 22 million plants have been affected and approximately 6.5 million olive trees died (Beck et al. 2019), causing real damage to the landscape and identity of the study area.

The loss is due to Olive Quick Decline Syndrome Disease (OQDS, previously known as “CoDiRO”, recently named “De Donno”) (Saponari et al. 2013, 2018) caused by Xylella fastidiosa subsp. pauca (hereafter Xf), a quarantine plant pathogen for the EU area, where stringent and specific regulations apply. So, Apulian olive trees of Lecce province were affected by Xf and characterized by leaf scorching, scattered desiccation of twigs and branches conferring thus a burned aspect and subsequent tree mortality (Saponari et al. 2013) as shown in Pictures 1, 2. The most severely and impressively affected olives are the centuries-old trees of the local highly susceptible cultivars Cellina di Nardò and Ogliarola salentina which marked by (OQDS) from incipient signs of infection to plant death. Monumental trees with a skeletal appearance, severely pruned to promote new vegetation, have shown dramatic decline and severe symptoms of desiccation where the few shoots produced are already dead. In the field, remarkable differential response to natural infection by Xf (Giampetruzzi et al. 2017) was revealed in adjacent rows of olive trees of cultivars Leccino and Ogliarola salentina and comparative analysis of the transcriptome of these varieties was carried out to investigate the reasons for this differential behavior (Giampetruzzi et al. 2016). Moreover, in experimental fields and controlled conditions for the evaluation of resistance/susceptibility of the varieties/species, through mechanical inoculation with bacterial cultures, strain “De Donno”, does not infect grapevines and citrus whereas it multiplies readily in oleander, olive seedlings and rooted cuttings of cv. Cellina di Nardò and to a much lesser extent in other olive cultivars like Coratina, Frantoio, Leccino (European Food Safety Authority 2015; Saponari 2016). However, one of the best solutions was the search for resistant varieties which are a highly desirable approach with the aim of the possible reconstitution of the Salentinian olive industry based on a set of cultivars that can replace for the largely predominant and highly susceptible ones. Leccino and FS-17 have already been identified and a patch-graft with these resistant varieties on old infected olive trees was piloted, considering this cultural practice as an encouraging starting point in the research for resistant material.

Picture 1.

Olive leaf scorching due to Xylella fastidiosa invasion in Apulia region. Source: Infoxylella.

Picture 2.

Olive trees mortality caused by Xylella fastidiosa in Apulia region. Source: Infoxylella.

This bacterium (Sequence Type 53) originating on coffee plants in Costa Rica is driven by an insect-vector, Philaenus spumarius, which is widely distributed in Apulia region (Bosco 2014; European Food Safety Authority 2015; Martelli et al. 2015). Further, Frem et al. (2020) predicted that this bacterium may invade new European and Mediterranean countries. However, the bacterium was also detected in Neophilaenus campestris and Euscelis lineolatus indicating the potential vectoring roles of these insects for the spread of the bacterium in Apulian region (Elbeaino et al. 2014) while the spittlebug P. spumarius was the most abundant species found in orchards on both weeds and olive trees in Italy (Cornara et al. 2017). To better understand the role of the host plant on the ecology of the infections, numerous plant species have been shown to be colonized by Xf in southern Apulia, including almond, oleander, cherry and other species (Saponari et al. 2013, 2014). This may confirm the impacts of host plant species on the efficiency of transmission of Xf by P. spumarius and therefore the ability of the bacterium to spread rapidly and trigger an outbreak. Xf colonizes the xylem network of the trees (Wells et al. 1987).

As a consequence, by destroying the rural landscape in the study area, Xf is gradually inducing changes in olive trees landscape which provides a set of economically valuable goods and provisioning services such as food diet (olive and olive oil), regulating services that affect climate, biodiversity, sustainable agriculture and cultural services that provide recreational, aesthetic, and educational and research features. Thereby, these benefits contribute towards daily life and human well-being in terms of health, security, good life and social relations. Ultimately, linking these attributes to human wellbeing constitutes an integral part of the economic analysis that is applied when decisions are necessary for the concerned stakeholders to manage a biological invasion process (Emerton and Howard 2008). In this framework, the Figure 1 highlights the links between Xf invasion, olive landscape services and public’s well-being. Changes in ecosystem services and human well-being, give rise to assess the public’s preferences about landscape recovery.

Figure 1.

Overview of the impact of Xylella fastidiosa on olive landscape services based on the general illustration of Emerton and Howard 2008.

Furthermore, the provision of ecosystem services is not perceived or observed similarly by all social groups (Sardaro et al. 2016a). Therefore, assessing their preferences about ecosystem recovery such as olive landscape restoration constitutes an integral part of the economic analysis of invasive species that is applied when policy decisions should be taken to manage biotic invasions (Emerton and Howard 2008). Moreover, Apulian residents as taxpayers would pay additional costs for landscape recovery and rehabilitation. Meanwhile, residents’ preferences or opinions must be taken into consideration in the decision-making process (Haltia 2015) in order to comprehensively assess landscape (Tagliafierro et al. 2013). Without this consideration, the prevention of wasteful and imprudent resource allocation cannot be guaranteed. Given this, this research aims to provide an assessment of local citizens’ preferences for different olive landscapes changes and for pursuit of research associated to Xf. Also, the study provides their WTP for the improvement of the damaged olive landscape in Apulia within an economic analysis of biotic invasive framework. Burgess et al. (2012) points out that citizens’ preferences in relation to a non-excludable and non-rivalrous environmental non-market public good, are commonly examined by economists through non-market evaluation techniques (Figure 2) which are classified in two groups: revealed preference methods (i.e. market price, cost-based, hedonic pricing and travel cost) and stated preference methods (i.e. contingent valuation method, CVM and choice experiment, CE). The first group is limited to market goods and services, but based on observed user behavior. The second group can be applied to all goods and services to capture all use and non-use values, but based on hypothetical situations of users and non-users (Plan Bleu 2016). Among these techniques, CE is preferred for rural landscape evaluation (Scarpa and Cicia 2000; Scarpa et al. 2007), and is used for the purpose of this study for the following reasons: (i) it is suitable for evaluating multi-attribute ecological goods (Arriaza-Balmón et al. 2006), (ii) it minimizes strategic bias across a set of choices (Bennett and Blamey 2001; Hanley et al. 2001), (iii) it captures total economic value (use and non-use values) of the ecosystem, and (iv) it allows consideration of public preferences in the context of environmental goods such as olive landscapes (Sardaro et al. 2016b). Unlike CE, where the choice set is composed of at least three options (status quo and two hypothetical alternatives), CVM is based on only two possibilities (status quo and one hypothetical alternative).

Figure 2.

Summary of various techniques used to assess the economic, social and ecological economic impacts of invasive species. The diagram assembles two categories of methods (revealed preferences and stated preferences) where Choice Experiment belonging to the second category was considered in our study.

There have been several studies of valuation of environmental services and damages in the past decades. These include the valuation of a damaged ecological public good through CVM as a useful approach in public policy formulation (Portney 1994), environmental valuation through CE (Hanley et al. 2001; Scarpa et al. 2007; Campbell and Hynes 2011; Hasund et al. 2011), assessment of environmental damage in monetary terms through the WTP (Johansson 1990), valuation approaches for natural ecological functions (de Groot et al. 2002), the elicitation of factors affecting citizens’ behavior towards the prevention of environmental damage (Torgler and Garcia-Valiňas 2007), the estimation of WTP for rural landscape changes (Campbell 2007), individuals’ preferences for agri-environmental services (Garrod et al. 2014), and economic quantitative valuation of damages due to climate change (Auffhammer 2018). At the Italian level, one specific study has quantified the economic and landscape impact of Xf in the Salento area (Sardaro et al. 2015). However, this previous research is based on a direct income loss approach, rather than on stated preference methods. The added-value of the present research is twofold. First, the monetary values of non-market olive landscapes features, based on individuals’ preferences, have never been yet assessed in the affected area. Second, the present paper enriches the scientific literature that uses CE to capture society’s perception of research and experimentation activities, considered as an attribute in this study, to improve the representation of olive landscapes. The results have consequences for policy with respect to the recovery Plan (see below) for the damaged olive landscape and for pursuing ongoing research activities related to Xf.

The findings explored in the Results provide a clear picture of respondents’ opinions on the olive landscape, their preferences for the improvement of affected orchards, and their socio-economic profiles. Here we connect the observed results to the existing literature and derive some policy reflections from our findings. Firstly, our analysis highlights an extremely high level of perception of the cultural heritage and aesthetic benefits of the olive landscape. Previous research papers also assumed the importance of the assessment of cultural heritage values (Tengberg et al. 2012), as a subcategory of cultural ecosystem services (Hølleland et al. 2017) and aesthetic perception (Tribot et al. 2018) in landscape evaluation for sustainable land use planning and ecological restoration management. Rodríguez-Entrena et al. (2017) proposed the use of olive orchards to improve the aesthetic and visual quality of rural landscape’s green cover. Secondly, our analysis shows that Apulia’s inhabitants are willing on average to pay for alternative features in order to restore and revive the region’s olive orchards. Obviously, the majority of respondents were found more likely to choose “landscape changes” over the current situation of degenerated land with diseased olive trees. Using plants and trees is hence preferred by citizens to mitigate the deterioration of rural landscape (Frontuto et al. 2020). Four types of alternative landscape features were proposed to the local public in the study area.

The highest WTP for landscape change was found for the landscape with Xf-resistant olive cultivars in a dispersed or sparse production system, followed by mixed productive crops, and then by mixed productive crops with Mediterranean bushes. Given this, the present study underlines the relative public preferences to crop diversity in landscape configuration heterogeneity as outlined by Hass et al. (2018) and sustains existent evidence to biodiversity conservation and ecosystem features. However, the substitution of olive orchards with native plants or Mediterranean bushes alone certainly requires less maintenance throughout the year, provides a wildlife habitat (plant reproduction, bee abundance) and allows landscape conservation (Slattery et al. 2003).

Meanwhile, this type of landscape gives the lowest direct incomes for local farmers. These types of landscapes were selected over a monoculture cropping system planted with Xf-resistant olive cultivars in a traditional production system. On the other hand, the landscape with resistant-Xf olive cultivars in an intensive production system was relatively the least attractive option for land restoration. The lack of preferences for this kind of landscape seems to decrease local public utility. This perception is in agreement with a recent study (Arata et al. 2020), in which the mean WTP of Italian inhabitants in Lombardy region (northern Italy), is only €6.7 to increase the green area by 7%. Local citizens attach a specific reconversion to other crops which were already in the past but appear to be gradually replaced by oil olive monoculture. These preferences diversify Apulia’s agricultural landscape and enhance biodiversity in this region.

However, this kind of reconversion would be possible in conditions of profitability and economic sustainability over the new productions. Nevertheless, our findings support the results of Howley (2011), who found that intensive farming landscapes were less attractive for respondents. However, the dispersed production system provides fewer economic and social benefits than intensive production, as noted by Sardaro et al. (2016a). Furthermore, despite the fewer studies addressed to olive landscape valuation, the results of this research are relatively representative. A previous study by Marangon et al. (2008) assessed WTP for the introduction of olive trees into the landscape. The contingent valuation method applied on a hill region between Italy and Slovenia showed an average WTP of €25.59 household/year. Rodrígues-Entrena et al. (2017) stated that visitors’ WTP per year is around €6.52 in terms of landscape restoration with parking fees to improve the aesthetic quality of the land. Torres-Miralles et al. (2017) estimated the WTP to sustain olive-growing with a natural reserve in Andalusia (Spain). Using CVM, the mean WTP was €37 household/year. Sardaro et al. (2016a) evaluated the benefits provided by olive landraces in Apulia, revealing that WTP for landscape preservation was €207 per year/family.

Furthermore, a number of studies elicited individual preferences in relation to environmental issues (such as landscape preferences) and multi-functionality of agriculture (Abler 2004; Dachary-Bernard and Rivaud 2013). Tempesta and Vecchiato (2017) have already reviewed previous Italian studies for evaluating landscape benefits. They found that WTP ranged from €2.8 to €74.3 per household/year. Ciaian and Paloma (2011) found that the mean WTP was €149 per ha, by using a meta-analysis technique of agriculture landscape valuation in EU. Our results highlight the relation between the mean WTP values and the distance from the degenerated landscape. Lecce residents live closer to the studied area and are relatively willing to pay more than Foggia residents (Figure 1). This proximity issue was analyzed by del Saz-Salazar and Menéndez (2007), who found a positive correlation between WTP and residence proximity. This issue was also stressed by Arata et al. (2020) as needing to be explored in future studies. The third contribution from the present study is the use of research as a conceptual contribution in designing choice experiment. Studies using this type of attribute are absent from the literature review for evaluation of ecosystem services.

Environmental issues such as restoration of damaged landscape are of crucial importance for land use development plans at regional, national and European levels (European Landscape Convention 2000). This convention urges each EU Member State “to assess the landscapes thus identified, taking into account the particular values assigned to them by the interested parties and the population concerned”. In consequence, by assessing WTP and preferences of the local public towards alternative features or changes in landscape through the stated preferences approach, our study constitutes an efficient prerequisite tool to support local policy-makers on the allocation of financial resources in the best interest of the local inhabitants in the study area. This evidence was recently noted by Rewitzer et al. 2017; Arata et al. 2020 and Frontuto et al. 2020. Assessment of the social value of the damaged olive landscape in Puglia is based on landscape changes and pursuit of the ongoing research and experimentation on Xf. The latter attribute is a specific challenge for the Italian landscape, given the economic impact of Xf subsp. pauca on olives. Reinforcement of the ongoing research on the production of transgenic olive plants and on vector control (i.e. tillage, weed management, use of insecticides, trapping, monitoring and surveillance) is necessary to reduce the economic impact of Xf. Schneider et al. (2020) have highlighted the importance of the strengthening research to reduce the future spread rate of Xf from 5.18 to 1.1 km/year on Italian, Spanish and Greek olives. In fact, the production of resistant cultivars of olives is an important axis of research to cope with Xf invasion. In addition, through the replantation of potential damaged landscape by using resistant olive cultivars, the hypothetical economic impact of Xf on olives in these countries can also be reduced from a range of €3.58 to 8.69 billion (without replanting resistant olive cultivars) to a range of €2 to 4.13 billion (by replanting resistant cultivars), over a period of 50 years. Thus, the present study has important policy implications for the current regional recovery action plan (Mipaaft 2020) in the study area, particularly for the restoration and revival of damaged olive orchards, control of Xf vectors and enhancement of research and experimentation aimed at finding a cure for this disease. Lastly, the present choice experiment model could be enhanced by involving other landscape attributes related to the level or types of crops for reconversion, the safeguarding of ancient/monumental olive trees, and communication activities. This research could also be extended to cover the preferences of tourists and of local entrepreneurs for whom improvement of the olive groves would have positive impacts on income.

This research was supported by CURE-XF, an EU-funded project, coordinated by CIHEAM Bari (H2020 – MSCA – RISE. Reference number: 634353). Thanks also to Sarah Jane Christopher of UNIBA for her careful proof-reading of the manuscript and to the anonymous Referees for their valuable suggestions.