Social sensitivity analysis in conflictive environmental governance: A case of forest planning

Environmental Impact Assessment Review 65 (2017) 54–62

Contents lists available at ScienceDirect

Environmental Impact Assessment Review journal homepage: www.elsevier.com/locate/eiar

Social sensitivity analysis in conflictive environmental governance: A case of forest planning

MARK

Serafin Corrala,⁎, Montserrat Acostab a b

Department of Applied Economics and Qualitative Methods, University of La Laguna, Canary Islands 38200, Spain Department of Techniques and Projects in Engineering and Architecture, University of La Laguna, Canary Islands 38200, Spain

A R T I C L E I N F O

A B S T R A C T

Keywords: Environmental governance Conflict situations Social multi criteria assessment Participatory techniques Social sensitivity analysis

In environmental governance, there is a need to include tools that analyze the robustness of the assessment processes, as well as the validity of policies and measures. This requires a methodological framework that integrates formal techniques such as sensitivity analysis with what the authors have called social sensitivity analysis. The latter consists of participatory processes in which stakeholders analyze the robustness of the assessment process used as well as the validity of the results of such assessments. This methodology was applied to a procedure for assessing planning alternatives for forest tracks. Sensitivity analysis studies the technical robustness of the results of the assessment, as well as exploring the social validity of these results, thus facilitating processes of dialogue and consensus needed in decision-making in conflictive situations. These results are considered interesting, not only from the perspective of implanting polices, but also as a reference for other places with similar situations.

1. Introduction Forestry management and planning “often concern large areas, long time horizons and multiple stakeholders, which complicates the planning process and increases the uncertainty involved in it” (Kangas and Kangas, 2004, p. 169). Forest planning processes, as Funtowicz and Ravetz stated, can be characterised, “by uncertain facts, disputed values, high stakes and urgent decisions” (Funtowicz and Ravetz, 1991, p. 141). Thus, this planning is not free from situations where different stakeholders have conflicting interests, such as those described by Hiltunen et al. (2008), who proposed a management plan for natural resources, Pravat and Humphreys (2013) with their study of management and use of forests, or Acosta and Corral (2015) in their investigation into the use of forest tracks. These conflicts are also aggravated by the uncertainty that characterises environmental systems, themselves (Corral Quintana, 2004; Funtowicz and Ravetz, 1993; Funtowicz and De Marchi, 2000; Giampietro et al., 2006. These circumstances greatly hinder the application of traditional scientific methodologies to tackle environmental governance issues. In these cases, where uncertainty and ignorance are present and clashes between different interests occur, science must seek solutions that allow the participation of society (Ravetz, 2004). During such participation, there can be an exchange of views, discussion and sometimes consensus, thereby achieving an enriched

⁎

decision-making process. In this process, it is essential to identify the most relevant stakeholders (Kangas et al., 2014; Nordström et al., 2010). Failure to do so may result in a lack of information that will lead to inappropriate alternatives being chosen to solve problems (Nordström et al., 2010). In this sense, Buchy and Hoverman (2000) indicated that when different groups with different interests participate in consultations, meetings or negotiations there is an increase in participants' knowledge of alternatives thus, promoting better understanding between the parties. In addition, when the planning of forest resources is carried out with the participation of the communities concerned, these processes are more transparent and easily understood, as affirmed by Vainikainen et al. (2008). Thus, in recent decades, approaches that support decision-making and integrate tools through which stakeholders can be part of the planning process have proliferated. Guimarães and Corral (2002) reviewed how Decision Support Systems (DSS) have evolved from more technocratic approaches to assessment processes with varying levels of participation (Arnstein, 1969) in the search for mutual learning (Corral, 2011; Corral-Quintana et al., 2016; Funtowicz and De Marchi, 2000; Giampietro et al., 2006) or as Gibbons (1999, p. C82) coined “socially robust knowledge”. This need to involve stakeholders in the planning and decisionmaking has also been highlighted in the case of forest planning. There has been a steady evolution towards more inclusive processes such as

Corresponding author. E-mail addresses: [email protected] (S. Corral), [email protected] (M. Acosta).

http://dx.doi.org/10.1016/j.eiar.2017.04.003 Received 16 November 2016; Received in revised form 31 January 2017; Accepted 12 April 2017 0195-9255/ © 2017 Elsevier Inc. All rights reserved.

Environmental Impact Assessment Review 65 (2017) 54–62

S. Corral, M. Acosta

Fig. 1. Sensitivity Analysis Approach for socio-environmental issues.

2. Method

those conducted by Ananda (2007) related to policies of forest land use, Prell et al. (2009) linked to the use of national parks. Nordström et al. (2010) also investigated the role of participation in urban forest planning and finally Rosenberger et al. (2012) analyzed the issue of paying for access to forests. In recent decades, stakeholders have gone from being regarded as mere informants, to being involved in defining the issues of forest environments. In some cases, they have even defined the alternatives and criteria (e.g. Laukkanen et al. (2002) or Vainikainen et al. (2008)) that have served to address forest issues such as landscape planning, tourism and timber management. The integration of social actors undoubtedly enriches planning processes; however, there are still uncertainties related to socioenvironmental issues that need to be addressed regarding “technical (inexactness), methodological (unreliability), epistemological (ignorance), and societal (social robustness)” dimensions (Van Der Sluijs et al., 2005, p. 482). Often quantitative uncertainty and sensitivity assessment methods are used, although these only address the technical dimension. Thus, uncertainty mainstream methodologies such as Monte Carlo analysis or Bayesian updating alone are not suitable for environmental and societal issues because undeterminable uncertainties prevail over quantitative ones. According to Van Der Sluijs et al. (2005) although quantitative techniques are essential in any uncertainty analysis, they only provide a partial insight into what is a very complex usually mass of uncertainties (Pereira and Quintana, 2009; Van Der Sluijs et al., 2005; Van Der Sluijs et al., 2008). As mentioned, in situations where conflicting interests prevail, it is not enough to deal just with uncertainties of a technical nature (those related to the information available, the variables used and the model applied). In these cases, the legitimacy of the planning process is affected by uncertainties of epistemological and social dimensions, putting these processes into dispute and hampering decision-making. Given these characteristics, a methodology is implemented to explore the technical and social uncertainties that may arise in environmental governance issues. It can be seen that when developing forest planning processes or similar processes in which decisions are taken that may affect stakeholders, it is necessary to apply methodologies that allow society to participate as part of these processes. In particular, technical sensitivity analysis is necessary but not sufficient in these situations, where conflict may occur, as is the case with planning and management of natural resources. This paper proposes a methodology to explore the robustness of techniques and processes used in environment planning in conflictive situations by involving different stakeholders. Furthermore, the results and conclusions are presented from the application of this approach to a case of integrated assessment of forest track planning and management issues.

The proposed methodology aims to explore the robustness of forest planning processes. The social uncertainty arising from these planning processes is often characterised by systemic uncertainty and disagreements among the stakeholders involved. To do this, an approach that combines formal and informal methods of analysis (see Fig. 1) is proposed. On the one hand, a sensitivity analysis is used to obtain a technical validation of the parameters in the forest planning process. On the other, complementary approaches are employed in which stakeholders assess the robustness of the process, the methods applied and the results obtained from forest planning assessment to achieve social validation of the process. This methodology will allow: - the robustness of the procedures and processes used to be analyzed. In this sense, although not all results are accepted by all stakeholders, “their generation process is an open and transparent process in which the views of all parties are included” (Corral Quintana, 2004, p. 193). - dialogue and debate among stakeholders to improve decisionmaking procedures and enables the needs and concerns of all involved to be met. 2.1. Sensitivity analysis techniques Quantitative aspects of uncertainty have been debated and numerous techniques proposed to measure it (see Mowrer et al., 1996). Sensitivity analysis (SA) has been defined as “a process that aims to assess the response of a model to changes in input parameters” (Ligmann-Zielinska and Jankowski, 2008). Technically, it partitions the results of model's components and parameters to identify the key determining variables (Smith and El-shaarawi, 2002) through the assessment of small changes to input parameters on assessment outcomes (Crosetto and Tarantola, 2001; Munda, 1994; Saltelli et al., 2008; Tarantola, 2008) and, therefore, validating the robustness of the results. There are a large number of approaches to perform a sensitivity analysis. Thus, for instance, the one-factor-at-a-time approach (OAT) explores the effect that changes of a factor produces on the outcome (Bailis et al., 2005; Campbell et al., 2008; Murphy et al., 2004). However, this approach is not able to find out relations between input variables (Czitrom, 1999). Generally speaking, there are two main techniques to approach SA: a) global SA and b) local SA. Other classifications, such as Saltelli et al. (2000) focus on the capability rather than the methodology of a specific technique. Global SA refers to the techniques that pursue the quantification of output uncertainties resulting from simultaneous parameter changes 55

Environmental Impact Assessment Review 65 (2017) 54–62

S. Corral, M. Acosta

environmental planning processes, thereby helping to respond to conflicts among users. For example, Garmendia and Gamboa (2012) detected when analyzing the results of an assessment exercise on the alternatives that one of the alternatives that was given priority in technical analysis was the least valued by stakeholders. Therefore, it would be unlikely to be effectively implemented given the opposition and controversy. SSA is performed using different inclusive techniques (i.e. focus groups, citizen-juries …) in which stakeholders through dialogue could debate not only the issue framing, the information used, the evaluation procedure but also the results. In addition, with this type of process stakeholders have the opportunity to reach consensus situations that would not have been otherwise possible, as in the case of traffic planning (Hernández-González and Corral Quintana, 2016) and water management (Paneque et al., 2009). In this sense, Pereira et al. (2005) suggest the tools used in decision making TIDDD (Tool to Inform Debates, Dialogues & Deliberations) should be understood as conceptual tools to develop debates, dialogues and discussions among social partners. Through these tools not only is knowledge provided to generate and organize discussion processes, but also to complement the knowledge with the information arising from the process. In fact, according to Hernández-González and Corral Quintana (2016) the benefits of actors' inclusion in environmental governance are various: “(a) ownership of policies, (b) better decisions in terms of sustainability and the inclusion of community values, (c) credibility of public agencies, and (d) faster planning implementation” (HernándezGonzález and Corral Quintana, 2016, p. 205) This is what Susskind and Elliott (1983) described more than thirty years ago as “coproduction” of knowledge. In addition, SSA decreases the technocratic nature of assessment processes in decision-making (Hernández-González and Corral, 2017). Since it not only allows stakeholders to trace the information obtained at the beginning of the assessments, but also all the way through to the results of the assessment. This situation gives the planning processes a feeling of proximity and transparency, thereby contributing to people's confidence in the results. Traceability is an aspect that is present in various areas (e.g. in the worldwide web through the tracks Internet users leave through the use of digital technologies (Wasmer and Woll, 2011) or in monitoring product manufacturing (Sima et al., 2016). This traceability is also considered a quality attribute in software development (Winkler and Pilgrim, 2010) and in food safety, thereby preventing commercial fraud, as it verifies products' origins (Galimberti et al., 2013; Verbeke et al., 2007).

(Turányi, 1990). This technique is able to test single variable changes and combinations with others. A first example of global SA is the Fourier Amplitude Sensitivity Test (FAST) developed by Cukier et al. (1973) to examine the sensitivity of solutions when all rate coefficients are varied simultaneously. A Monte Carlo algorithm can also be used to analyze the sensitivity of a function with respect to random set of variables (Sobol', 1990). On the other hand, local SA refers to the assessment of the effects of small changes of parameters on many responses (Turányi, 1990). One of the first attempts to develop local SA have been the adjoint techniques intended to assess the sensitivity of the response to variations in a model's parameters (Cacuci, 1981a, 1981b). Among the DSS used in natural resources planning, there are examples of sensitivity analyses applied to validate models' robustness and the results obtained (Ananda and Herath, 2003; Locatelli et al., 2008). In some of these investigations, the methodology used to conduct these assessments is unclear (Rohaendi et al., 2012). In other cases, they go further providing a broader discussion of the analysis carried out (Hernández-González and Corral, 2017), such as in several cases dealing with transport planning issues (Hernández-González and Corral Quintana, 2016) and desertification (Corral-Quintana et al., 2016; Corral et al., 2015). Thus, in those cases where an analysis of the technical and methodological uncertainties in natural resources planning and specifically in forest planning should be highlighted. These are grouped according to the objectives pursued during the uncertainty assessment performed for better comprehension. Among those studies involving local SA, in which parametric changes were made, Ananda and Herath (2003) analyzed the results obtained in forest planning using an analytical hierarchical multicriteria method. They showed that changing the weighting of decision criteria affected priorities, thus concluding that using different weightings resulted in different positions of the planning options. Similarly Henao et al. (2012) evaluated the strength of an energy solution through a sensitivity analysis applied to initially established weightings. They used the multi-criteria method of compromise programming. The results obtained were compared with the initial ones thus identifying whether the initial classification could be modified. By contrast, Locatelli et al. (2008) showed that changes made to criteria weightings hardly affected the results of the assessment of planning alternatives for reforestation impacts under a scheme of payment for the use of the environment in the north of Costa Rica. Similarly, Rohaendi et al. (2012) gave different weightings to the criteria in the case of Sawahlunto, West Sumatera, Indonesia, where they assessed five possible land uses including agriculture, recreation, residential, landfill and forestry. They also found that the initial results were not altered by parametric changes.

3. Case study 2.2. Social sensitivity analysis framework The proposed methodology was applied to an assessment process of policy alternatives aimed at addressing the problems of regulating alternative uses of forest tracks on insular territories and resolving conflicts that arise among forest users. Tenerife Island has 2000 km of forest tracks and a network of 200 km that have been established for vehicle circulation for recreational purposes, with restrictions depending on the characteristics of the vehicles. There are several issues that make it difficult to plan the transit of users by this type of route. One of them is its location, given that parts of them traverse the Parque Natural de Corona Forestal which is considered a protected natural area. In addition, this Park is declared Ecological Sensitivity Area, which leads us to consider a planning that must conform to the regulations that govern this type of space. The consideration as Ecological Sensitivity Area implies the existence of natural, cultural or landscape values that are sensitive to deterioration or susceptible to ruptures in their natural balance. Another issue is the effect that it has had on the population, the implementation of the rules that regulate the motorized transit on these routes. The population has been divided among those who consider that the measure violates their rights as

Since social values are involved in so many planning processes (Munda, 2008), social sensitivity analysis (SSA) is needed. The main idea behind SSA is to return the planning assessment results to stakeholders to explore the social compliance of any policies or measures to be carried out, promoting a reflexive dialogue among the social actors involved. Following Corral Quintana, SSA “is, essentially, a social validation of assessment results” (Corral Quintana, 2004) that assess the robustness of both the assessment and the results obtained based on social actors' to stakeholders' understanding. “The term SSA refers to the most frequently applied sensitivity analysis aimed at testing the robustness of the results of a model or system in the presence of uncertainty, as well as at understanding the relationships between input and output variables in a system or model. This validation process is carried out through the involvement of the stakeholders” (Hernández-González and Corral Quintana, 2016, p. 206). Thus, SSA could also be defined as a form of Social Traceability allowing stakeholders to be linked from the beginning to the end, in 56

Environmental Impact Assessment Review 65 (2017) 54–62

S. Corral, M. Acosta

Source: Adapted from Acosta and Corral (2015) Fig. 2. Integrated assessment methodology applied in the case study. Source: Adapted from Acosta and Corral (2015).

between them. For instance, while Quad bikes are generally looking for speed and irregular track positions, Recreational area users prefer to enjoy quiet walks safely and away from the noise of the engines, because in the end, they go to these environments to escape from the noise of cities, an issue that is hampered by the presence of Quad bikes. Equally, Horse-riders also come into conflict with the motor vehicles, because the high speed driving and the noise that they produce, causes the horses to lose control, putting in danger the safety not only of the animal but also that of the rider. As for management issues, they focused on aspects that had to do with the services offered in the forests of the island, for example, greater vigilance to be able to have better quality visits within a framework of security. Improved infrastructure was also requested, such as the firmer, more solid forest tracks to counteract the erosion caused by continuous vehicular traffic and the weather. Thus two assessments, one for planning alternatives and one for management using the NAIDE multi-criteria method (Munda, 1995; Paneque et al., 2009) were performed. The NAIADE assessment of alternatives was carried out by “means of pairwise comparison with respect to each assessment criterion generating a ranking of alternatives” (JRC, 1996, p. 8). NAIADE produces a structure of alternatives according to different criteria, thus, “the final ranking is based on the Φ+ and Φ − values for each of the alternatives. Φ + is based on the better and much better preference relations with a value going from 0 to 1 indicating how alternative a is “better” than all other alternatives. Φ − is based on the worse and much

users, since traditionally they have always been able to move freely on them, and those who fully support this regulation. More details on this issue and further assessment are available from Acosta and Corral (2015). The assessment in this case was based on the integration of three tools (see Fig. 2): Institutional Analysis (IA), Participatory Techniques (PT) and Multi-criteria Assessment methods (MA). Initially, IA was conducted with the dual purpose of first identifying the most relevant stakeholders, and providing an initial view about perceptions and positions. Second, relations between the various stakeholders can be identified, and existing knowledge about the problem can be determined (Corral Quintana, 2004). Stakeholders portrayed through the IA, a set of alternatives and the criteria for assessment, through conducting several rounds of questionnaires and interviews. Through these participatory actions, it was noted that stakeholders raised various issues related to forest tracks that could be grouped into two groups, that is, issues concerning planning and, on the other hand, issues related to management (see Table 1). This meant that two assessments were actually required, when initially it was thought that there was just one problem. Planning alternatives were related to actions that govern the movement of motor vehicles in the forest environment, because conflicts often arose between vehicle users and non-vehicle users concerning issues such as speed or noise. Table 2 presents a list of collectives that make use of the forest tracks. The diversity of these users leads to a heterogeneity of interests, which often leads to conflicts 57

Environmental Impact Assessment Review 65 (2017) 54–62

S. Corral, M. Acosta

Table 1 Definition alternatives in forest planning and management. Problem

Alternative

Description

Management Alternatives

Limit use by areas.

Carry out a classification of the forest tracks, so that each zone can have an exclusive use, that is, there will be tracks that can only be used by hikers and walkers, cyclists and horsemen on horseback; tracks for exclusive use of recreation with motor vehicles. Control access to the forest environment and provide the necessary measures to improve the surveillance so that the current regulations are enforced. Carry out cleaning and maintenance of forest tracks and firebreakers, as well as silvicultural treatments and works to improve water drainage. Have support points (identified with number, name, … etc.) in strategic areas to help with possible emergencies. Distribute garbage collection points. Exclusive transit for emergency vehicles.

Improve surveillance and control of access. Improve infrastructures. Meeting and assistance points.

Planning Alternatives

Rubbish collection points. Traffic circulation restricted to emergencies Unrestricted traffic circulation. One-way systems. Pre-paid traffic circulation charge. Don't do anything; hereinafter known as “BAU”.

Unrestricted transit through forest tracks. Establish a one-way transit system in each of the forest tracks. Payment of an annual license fee (for vehicles that frequently travel on the forest tracks) or an access fee (for vehicles that circulate sporadically on the forest tracks). Business as usual.

Table 2 Stakeholder communities according to their area of action and type of activity. Source: Acosta and Corral (2015). Stakeholder communities according to their area and type of activity Area of action

Type of activity Decision-makers

Local

- Tenerife Island Council - Town halls

Regional

- Canarian government

Motorized sports

- 4 wheel drive vehicles - Motorbikes - Quad bikes

Non-motorized sports

-

Surveillance safety/emergencies and rescue

Horse-riders Mountaineers Cyclists Triathletes Hikers Hunters

National

- Environmental agents - BRIFOR (Forest Fire Brigade prevention and extinguishing

- Safety and Emergency Committee of the Canarian Government - Emergency and Security Coordination Centre 112 - LAND ARMY - SEPRONA (Nature Protection Service)

worse preference relations, its value ranging from 0 to 1 indicates how a is “worse” than all other alternatives” (JRC, 1996, p. 9). The assessment of alternatives is performed in NAIADE using one of the following three operators: “Minimum”, “Zimmerman-Zysno” and “Simple product” operators (see Fig. 3). The choice of the operators determines the degree of compensability among criteria. The first one allows the greatest degree of compensability, the second one implies no compensability at all and the third one permits a certain compensability, which is determined by the parameter γ (γ ranges from 0 to 1) (from 1 minimum compensation to 0 maximum compensation). Zimmermann and Zysno (1983) showed that subject's judgments are best represented as a “combination of the two membership functions that do not fit any of the rules proposed for conjunction and disjunction” (Munda, 1995, p. 106). Additionally, NAIADE used four parameters in the assessment: the number of semantic iterations, number of iterations in integral calculation and the minimum requirement for fuzzy relation. Through parametric changes in the number of iterations (the first two parameters) a possibility or a similarity degree of equality between two fuzzy sets were changed. The larger the semantic distance, the smaller the possibility degree of equality, specifically more iterations mean more precision but a slower calculation.

Firms

Others

- Private firms (Forestry workers) - Private firms (tourism) - Pine-needle gatherers - Bee-keepers - Water management

-

- Public sector firms (forestry work

- PROFOR (Association of Forestry Professionals in Spain

Experts Recreational area users Farm owners Forest residents

Fig. 3. NAIADE assessment parameters and operators.

From the assessment of the alternatives for both planning and management issues, two alternative rankings were obtained (see Fig. 4). 58

Environmental Impact Assessment Review 65 (2017) 54–62

S. Corral, M. Acosta

Forest planning

Forest management

Alternatives: A: BAU B: Only emergencies C: Unrestricted traffic D: One-way tracks E: pre-paid charge circulation

Table 3 Variation of parameters of operators.

Alternatives: A: Limit Use by Areas B: Access control C: Improve infrastructures D: Meeting Points E: Sanitation Points

Parameters

Initial values

Variations of parameters in: minimum, Zimmerman-Zysno, simple product

Number of interaction semantic distance Number of interaction integral calculation Minimum requirement for fuzzy relation

100

100–1000

100

100–1000

α = 0.4

0.2–0.4–0.6

Fig. 1). Thus, a first exercise consisted of changing the number of semantic iterations and of the interaction integral calculation, as well as the minimum requirement for fuzzy relation, for the values of the three available NAIADE operators: “Minimum”, “Zimmerman-Zysno” and “Simple product” were different to the ones used in Acosta and Corral (2015) (see Table 1). This technical analysis was followed by a SSA exercise based on stakeholders' inputs elicited during a focus group exercise. The SA is performed by comparing the ranking of alternatives resulting from parametric variations and operators with the initial values of each parameter and operator used in the case study (see Table 3). Thus, in the iterations both the semantic distance as well as the integral calculation of 100 to 1000 was changed. In addition, while in Acosta and Corral (2015) α was equal to 0.4 by default, in this analysis the minimum requirement will be relaxed to α = 0.2, and then made stricter to α = 0.6. Each of these variations was repeated for different operators, either for the Minimum operator (which gives no compensation), the single product or the Zimmermann-Zysno operator (ɣ) which allows for varying degrees of compensation, from 0 (as a minimum compensation) to 1 (as maximum). In this case, ɣ was modified from the initial ɣ = 0.4 to ɣ = 0.2 (lower compensation) and ɣ = 0.6 (higher compensation). With each variation, new rankings of alternatives were generated that were then compared with those initially obtained; thereby establishing whether the initial ranking really was robust (see Fig. 4). There were 25 results for each assessment carried out (planning and management) (see Table 4).

Source: Adapted from Acosta and Corral (2015) Fig. 4. Ranking of planning and management alternatives. Source: Adapted from Acosta and Corral (2015).

Regarding the planning evaluation, the best alternative was E (around 90% of these criteria were characterised as “good.”) followed by A and B. ones in the ranking, both at the same level. With regard to the management assessment, the best alternative was C); followed by A and B. Regarding planning alternatives, the BAU alternative was positioned at the top of the classification (less than 65% of the socioeconomic criteria were evaluated as “good” during the assessment) Despite the presence of conflicts in the use of forest tracks, social actors consider that it is possible to continue with the current situation provided that normative revisions are adapted to the needs of the environment and the demands of society are taken into consideration. The technical robustness of these results and their acceptance and social stability are discussed in the next section. 4. Results and discussion. The robustness analysis carried out consists of both a SA and SSA (see

Table 4 Results of the sensitivity analysis for forest planning and management assessment alternatives. Operator

Number of iterations in semantic distance

Number of iterations in integral calculation

Minimum requirement for fuzzy relation

Minimum

100 1000 100 100 100 100 100 100 1000 100 100 100 1000 100 100 100 1000 100 100 100 100 1000 100 100 100

100 100 1000 100 100 100 100 100 100 1000 100 100 100 1000 100 100 100 1000 100 100 100 100 1000 100 100

0.4 0.4 0.4 0.2 0.6 0.4 0.4 0.4 0.4 0.4 0.2 0.6 0.4 0.4 0.2 0.6 0.4 0.4 0.2 0.6 0.4 0.4 0.4 0.2 0.6

Zimmerman-Zysno

Simple product

Compensation

0.6 0.4 0.2 0.6 0.6 0.6 0.6 0.2 0.2 0.2 0.2 0.4 0.4 0.4 0.4

59

Ranking of forest planning assessment alternatives

Ranking of forest management assessment alternatives

E, E, E, E, E, E, E, E, E, E, E, E, E, E, E, E, E, E, E, E, E, E, E, E, E,

C, C, C, C, C, C, C, C, C, C, C, C, C, C, C, C, C, C, C, C, C, C, C, C, C,

A, B, C, D A, B, C, D A, B, C, D A, B, C, D B, A, C, D A, B, C, D A, B, C, D A, B, C, D A, B, C, D A, B, C, D A, B, C, D A, B, C, D A, B, C, D A, B, C, D A, B, C, D B,A, C, D A, B, C, D A, B, C, D A, B, C, D B, A, C, D A, B, C, D A, B, C, D A, B, C, D A, B, C, D B, A, C, D

A, A, A, A, A, A, A, A, A, A, D, A, A, A, A, A, A, A, A, A, A, A, A, A, A,

B, D, B, D, B, D, B, D, D, B, D, B, D, B, B, D, D, B, D, B, E, A, B, D, B, D, B, D, B, D, D, E, D, B, D, B, B, D, D, E, B, D, B, D, B, D, B, D, D, E,

E E E E E E E E E E B E E E E B E E E B E E E E B

Environmental Impact Assessment Review 65 (2017) 54–62

S. Corral, M. Acosta

Table 5 List of groups that participated in the focus-group. Collective groups

Number of participants

Public managers Users (making use motor vehicle) Users (without using motor vehicle) Surveillance, security, emergency and rescue Private companies Total number of participants

2 1 2 3 3 11

Zysno with compensation 0.2 (4 matches out of 5) and Simple Product (4 matches out of 5). In view of these results, the technical robustness of the outcomes obtained in the initial assessment is confirmed. After analyzing the technical stability of the results, the next step was validation by the stakeholders. It was decided to use participatory focus groups. This technique allows qualitative data to be obtained through group discussions (Morgan, 1997). Owing to the large number of groups involved in the initial study (28 in total), and considering that, according to some authors, focus groups require a small number of participants (for example, (a) between 6 and 12 participants (Aigneren, 2002); (b) 6 to 10 (Escobar and Bonilla-Jimenez, 2009); (c) between 7 and 10, with optimal number being no less than 4 (Calvente and Rodríguez, 2000); (d) between 4 and 10 people, with an optimal number of 6–8 (Huerta, 1977); (e) between 6 and 12 participants (Mahlau et al., 2002); (f) between 4 and 10 (Rodríguez and Cerdá, 2002)), we decided to select a low number of participants. To do this, two criteria were taken into account to form the groups: (1) inviting those who could nourish the process the most, (2) representing all the stakeholder collectives. Finally, there were 11 participants representing each of the collectives (see Table 5). During the focus group discussions, the ranking from the assessment of the closest alternative to the actual planning one was dealt with first. Subsequently, there was further discussion on the issues concerning management alternatives. Since both the alternatives and the criteria had emerged from surveys conducted in the first phase of the assessment (Acosta and Corral, 2015), participants in the focus group only knew about the information that each had contributed in the aforementioned surveys, therefore they were unaware of the proposals of other participants. Despite this, in the of validation ranking closest to the planning alternative, all alternatives were well received by all stakeholders, agreeing that the best alternative was “Traffic circulation by a pre-paid charge”(E) in those specific areas where users receive a service, such as enjoying specific activities or unique landscape areas. This result agrees with that obtained both in the initial analysis and with the SA technique. In addition, the collective “Monitoring, Security, Emergency and Rescue” proposed the establishment of “mixed” alternatives to consider different forest activities and/or different forest areas. This might help adapt alternatives to new situations and demands by society. In this sense, the “Managers” group proposed a new joint alternative to consider the “BAU” alternative (which was ranked second in the initial ranking) whenever there were reviews of existing regulations to adapt to changes and new needs the forest system. This new proposal, “BAU” with regulatory review, was welcomed and supported by all the groups present. In addition, all stakeholders agreed on the importance of disclosure and information on forest regulations due to unawareness, the population has in this area, as well as on promoting awareness campaigns on the importance of forests in the education system. Regarding the views on the ranking of alternatives closest to the management one, the discussion focused on the option to “improve infrastructure” because it was felt that improving infrastructure for a cyclist is very different to the improvement needed from the view a 4 × 4vehicle user. The consensus regarding “improving surveillance and control of access” also stood out; this was seen as meeting the need

Fig. 5. Alteration in the ranking of planning alternatives derived from parametric change.

For planning, using the three NAIADE operators, 84% of rankings coincided with the initial one obtained (see Fig. 5). Although a priori, these data may indicate a slight lack of consistency in the results, in the remaining 16% of alternative rankings the difference is the change of order between the second alternative and third in the ranking. This corresponds to the BAU and “Traffic restricted to emergencies”, which in the initial analysis were both at the same level in the ranking. For “Minimum” and “Simple product” operators four out of the five combinations produced matching rankings; while in the case of Zymerman-Zysno operator with its three variants, derived from the types of compensation (0.6, 0.4 and 0.2), two changes occurred in the ranking of the 15 cases analyzed. In view of these results, the robustness of the results obtained in the analysis for the rankings of planning alternatives is considered justified. As for the Analysis of Technical Sensitivity conducted for the management results, in the three NAIADE operators, it is concluded that 56% of the resulting rankings (see Fig. 6) were consistent with those obtained in the initial research. By contrast, in 40% of cases a change of order among the alternatives located in third and fourth position, corresponding to the actions, Traffic restricted to emergencies (B) and One-way system on forest tracks (D) occurred. In 4% of cases a change occurred in four alternatives, although the best positioned in the rankings from the alternative analysis was not altered. It is emphasized that the greatest number of coincidence has been derived from Minimum operators (4 matches out of 5), Zimmerman-

Fig. 6. Alteration in the ranking of management alternatives derived parametric changes.

60

Environmental Impact Assessment Review 65 (2017) 54–62

S. Corral, M. Acosta

Additionally, through validation, understanding among stakeholders is facilitated, approaching positions and achieving compromise solutions. Thus, during the discussion on planning alternatives, those attending the focus group considered the same alternative as the initial analysis to be relevant. This situation of consensus was unthinkable at the beginning of this investigation, because during the interviews and surveys conflicts between different collectives was perceived. Development of social sensitivity analysis has allowed this consensus to emerge, reducing the marked divergence between views perceived earlier in the process. As an example of this disparity, one can mention the widespread perception held in initial interviews among the forest sports collective that does not make use of motor vehicle. These conflicts have been reduced through dialogue generated during the development of the focus group session. Finally, although the purpose of the focus group was not to make decisions on the implementation of alternatives, managers have witnessed the widespread demand and interest of participants in forest track planning. This will help give them the notion of how to continue addressing forestry issues so as to avoid, as far as possible, rejections by the population (such as in this case study, the resolution governing the motorized traffic by forest tracks Tenerife, which was perceived negatively by respondents). It has also been found that control of attendance by members of the different collectives in the focus group is a matter that can easily escape the hands of the researcher, because although in our case, this was controlled, an unexpected illness provoked one of the collectives (Others) to remain unrepresented. This setback at the last minute did not allow the attendance of another member of this group, so it was left without representation in the focus group. Where a mixture of (partial) knowledge, assumptions, and ignorance are involved, science should look for solutions to overcome these limitations by means of public participation (Ravetz, 2004). Therefore, sensitivity analysis should be expanded to include approaches where the decision-making processes become relevant (Munda, 2005). For instance, who decides in a MCA assessment the criteria that should assess a range of alternative options? Who decides the direction of these criteria? Clearly, these are decisions that are beyond scientists and, therefore, should be collectively assessed through a new social contract between the scientific community and society (Gibbons, 1999). The authors conclude by recommending, in cases of natural resource planning, the use of social sensitivity analysis to give greater strength to assessments of policy-making. Such analysis not only makes the process more transparent, but also promotes more stable decisions.

to achieve environmental promotion and as the best alternative for forest conservation. The implementation of this alternative would require a substantial increase in resources. Currently, there are 36 agents to control 100,000 ha, according to the collective of managers. This was relevant information for users and companies, as forest managers affirmed the need for increased surveillance in the forests. The focus group debate led to the formation of two groups among participants. One group, consisting of “Sport (using motor vehicle)”, “Business” and "Sport (without the use of motor vehicles)" giving priority to the alternative “Improving surveillance and control of access” while the other group, consisting of “Managers” and “Monitoring, Security, Emergency and Rescue” considered it necessary to prioritize the alternative “Delineate use by zones”. Thus, in relation to planning alternatives, all agreed to implement an alternative “BAU” when contemplating regulatory reviews so that any review will adapt to the changes and new needs of the forest system. However, as to management alternatives, there was a clear division between stakeholders, as groups “Sport (using motor vehicle)”, “Business” and “Sport” (without using vehicles motor) gave priority to the alternative “Improving surveillance and access control”, while the other groups “Managers” and “Monitoring, Security, Emergency and Rescue” favoured prioritizing the alternative “Delineate the areas of use”. Despite this division in the focus group, there was a willingness of participants to improve the development of all activities through dialogue with the other groups and the opportunity to share their needs in the forest environment. 5. Conclusions This research highlights the need to involve stakeholders in validating results derived from assessment processes of public policy alternatives. Assessment processes used in the governance of natural resources usually involve systemic uncertainty and strong opposing interests. In this sense, these issues cannot be analyzed in isolation from the social context in which they occur, because they are strongly influenced by interests, perspectives, opinions, knowledge and different perceptions (Corral Quintana, 2009; Corral et al., 2015; CorralQuintana et al., 2016). In conclusion, the development of social sensitivity analysis is essential for validating the results of an assessment process. Without this type of analysis, it can be said that a technocratic character marks assessments, an aspect this paper has sought to avoid since the beginning. Moreover, when issues affecting society are being dealt with, the participation of stakeholders should be allowed. A key benefit of this approach is the greater consensus of views among the different collectives that has been made possible through dialogue. In addition, the presence of the Social Traceability is considered critical because it not only contributes transparency to the process, but also conveys confidence to the people involved. In view of the results, performing a social validation of planning alternatives is become relevant. While sensitivity analysis ensures the technical robustness of the assessment of planning alternatives, it is through social sensitivity analysis that the discussion of the results with stakeholders leads to the discovery of new aspects of the problem and possibly unexpected actions. SSA processes foster knowledge sharing, thereby overcoming misinformation and helping to reach consensus, all generated through dialogue between social actors. For example, in the present case, during the debate on forest management, stakeholders proposed completely new alternatives. New alternatives might be considered both a relevant finding and a challenge. A relevant finding because it improves knowledge about the issue at hand in ways not considered before, and a challenge because this implies that planning and management processes might be seen as iterative ones, in which new knowledge is shared and discussed. Particularly, in this case study the new alternative was incorporated to the assessment, discussed and evaluated in a later stage.

References Acosta, M., Corral, S., 2015. Participatory multi-criteria assessment of forest planning policies in conflicting situations: the case of tenerife. Forests 6 (11), 3946–3969. Aigneren, M., 2002. La técnica de recolección de información mediante los grupos focales. Revista Electrónica La Sociología En Sus Escenarios 6. Ananda, J., 2007. Implementing participatory decision making in forest planning. Environ. Manag. 39 (4), 534–544. Ananda, J., Herath, G., 2003. The use of analytic hierarchy process to incorporate stakeholder preferences into regional forest planning. Forest Policy Econ. 5 (1), 13–26. http://dx.doi.org.accedys2.bbtk.ull.es/10.1016/S1389-9341(02)00043-6. Arnstein, S.R., 1969. A ladder of citizen participation. J. Am. Inst. Plann. 35 (4), 216–224. Bailis, R., Ezzati, M., Kammen, D.M., 2005. Mortality and greenhouse gas impacts of biomass and petroleum energy futures in Africa. Science (New York, N.Y.) 308 (5718), 98–103. Buchy, M., Hoverman, S., 2000. Understanding public participation in forest planning: a review. Forest Policy Econ. 1 (1), 15–25. http://dx.doi.org.accedys2.bbtk.ull.es/10. 1016/S1389-9341(00)00006-X. Cacuci, D.G., 1981a. Sensitivity theory for nonlinear systems. I. Nonlinear functional analysis approach. J. Math. Phys. 22 (12), 2794–2802. Cacuci, D.G., 1981b. Sensitivity theory for nonlinear systems. II. Extensions to additional classes of responses. J. Math. Phys. 22 (12), 2803–2812. Calvente, M.G., Rodríguez, I.M., 2000. El grupo focal como técnica de investigación cualitativa en salud: Diseño y puesta en práctica. Aten. Primaria 25 (3), 181–186. Campbell, J.E., Carmichael, G.R., Chai, T., Mena-Carrasco, M., Tang, Y., Blake, D.R., ... Stanier, C.O., 2008. Photosynthetic control of atmospheric carbonyl sulfide during the growing season. Science 322 (5904), 1085–1088. http://dx.doi.org/10.1126/

61

Environmental Impact Assessment Review 65 (2017) 54–62

S. Corral, M. Acosta

group). Revista Española De Estudios Agro-Sociales Y Pesqueros 194. Morgan, D.L., 1997. The Focus Group Guidebook. Sage publications, London. Mowrer, H.T., Czaplewski, R.L., Hamre, R., 1996. Spatial Accuracy Assessment in Natural Resources and Environmental Sciences: Second International Symposium. USDA Forest Service, Rocky Mountain Research Station, Fort Collins, CO, USA May 21-23, 1996. (General Technical Report RM-GTR-277, p. 728 ()). Munda, G., 1995. Multicriteria Evaluation in a Fuzzy Environment: Theory and Applications in Ecological Economics. Physica-Verlag. Munda, G., 2005. “Measuring sustainability”: a multi-criterion framework. Environ. Dev. Sustain. 7 (1), 117–134. Munda, G., 2008. Social Multi-criteria Evaluation for a Sustainable Economy. Springer. Munda, G., 1994. Fuzzy Information in Multicriteria Environmental Evaluation Models. Joint Research Centre, Europ. Commission, Inst. for Systems Engineering and Informatics. Murphy, J.M., Sexton, D.M., Barnett, D.N., Jones, G.S., Webb, M.J., Collins, M., Stainforth, D.A., 2004. Quantification of modelling uncertainties in a large ensemble of climate change simulations. Nature 430 (7001), 768–772. Nordström, E., Eriksson, L.O., Öhman, K., 2010. Integrating multiple criteria decision analysis in participatory forest planning: experience from a case study in northern Sweden. Forest Policy Econ. 12 (8), 562–574. http://dx.doi.org.accedys2.bbtk.ull.es/ 10.1016/j.forpol.2010.07.006. Paneque, Salgado P., Corral Quintana, S., Guimaraes Pereira, A., del Moral Ituarte, L., Pedregal Mateos, B., 2009. Participative multi-criteria analysis for the evaluation of water governance alternatives. A case in the costa del sol (málaga). Ecol. Econ. 68 (4), 990–1005. Pereira, Â.G., Quintana, S.C., 2009. 3 pillars and 1 beam: quality of river basin governance processes. Ecol. Econ. 68 (4), 940–954. Pravat, P.S., Humphreys, D., 2013. Using a multilevel approach to analyse the case of forest conflicts in the Terai, Nepal. Forest Policy Econ. 33, 47–55. Prell, C., Hubacek, K., Reed, M., 2009. Stakeholder analysis and social network analysis in natural resource management. Soc. Nat. Resour. 22 (6), 501–518. Ravetz, J., 2004. The post-normal science of precaution. Futures 36 (3), 347–357. Rodríguez, M.P., Cerdá, J.M., 2002. Paso a paso en el diseño de un estudio mediante grupos focales. Aten. Primaria 29 (6), 366–373. Rohaendi, N., Flacke, J., Sagala, S.A.H., 2012. Developing Land Use Alternatives for Mined Land: A Case Study of Sawahlunto, West Sumatera, Indonesia . Rosenberger, R.S., Needham, M.D., Morzillo, A.T., Moehrke, C., 2012. Attitudes, willingness to pay, and stated values for recreation use fees at an urban proximate forest. J. For. Econ. 18 (4), 271–281. Saltelli, A., Chan, K., Scott, E.M., 2000. Sensitivity Analysis. Wiley, New York. Saltelli, A., Ratto, M., Andres, T., Campolongo, F., Cariboni, J., Gatelli, D., 2008. Global Sensitivity Analysis: The Primer. John Wiley & Sons. Sima, G., Lile, R., Dan, G., Muncut, E., 2016. Management, traceability and control of industrial processes. In: Soft Computing Applications. Springer, pp. 971–980. Smith, E.P., El-shaarawi, A.H., 2002. Uncertainty analysis. In: Piegorsch, W.W. (Ed.), Enclcyopedia of Environmentrics. John Wiley, Chichester, pp. 2283–2297. Sobol', I.Y.M., 1990. On sensitivity estimation for nonlinear mathematical models. Matematicheskoe Modelirovanie 2 (1), 112–118. Susskind, L., Elliott, M., 1983. Paternalism, conflict, and coproduction: Learning from citizen action and citizen participation in western Europe. In: Susskind, L., Elliott, M., Associates, A (Eds.), Paternalism, Conflict, and Coproduction: Learning From Citizen Action and Citizen Participation in Wester Europe. Springer Science + Business Media, LLC, New York, pp. 3–34. Tarantola, S., 2008. Global Sensitivity Analysis. The Primer, Wiley, England. Turányi, T., 1990. Sensitivity analysis of complex kinetic systems. tools and applications. J. Math. Chem. 5 (3), 203–248. Vainikainen, N., Kangas, A., Kangas, J., 2008. Empirical study on voting power in participatory forest planning. J. Environ. Manag. 88 (1), 173–180. http://dx.doi.org/ 10.1016/j.jenvman.2007.02.004. Van Der Sluijs, Jeroen, P., Craye, M., Funtowicz, S., Kloprogge, P., Ravetz, J., Risbey, J., 2005. Combining quantitative and qualitative measures of uncertainty in modelbased environmental assessment: the NUSAP system. Risk Anal. 25 (2), 481–492. Van Der Sluijs, J., Douguet, J.M., O'Connor, M., Pereira, Â.G., Quintana, S.C., Maxim, L., Ravetz, J.R., 2008. Qualité de la connaissance dans un processus délibératif. Natures Sciences Sociétés 16 (3), 265–273. Verbeke, W., Frewer, L.J., Scholderer, J., De Brabander, H.F., 2007. Why consumers behave as they do with respect to food safety and risk information. Anal. Chim. Acta 586 (1), 2–7. Wasmer, E., Woll, C., 2011. Scientific project, sciences po| LIEPP laboratoire interdisciplinaire d'evaluation des politiques publiques or in english, interdisciplinary research center for the evaluation of public policies. In: MPRA Paper, pp. 31434. Winkler, S., Pilgrim, J., 2010. A survey of traceability in requirements engineering and model-driven development. Softw. Syst. Model. (SoSyM) 9 (4), 529–565. Zimmermann, H., Zysno, P., 1983. Decisions and evaluations by hierarchical aggregation of information. Fuzzy Sets Syst. 10 (1), 243–260.

science.1164015. Corral Quintana, S., 2004. Una metodología integrada de exploración y comprensión de los procesos de elaboración de políticas públicas. Universidad de La Laguna. Corral Quintana, S., 2009. A quality assurance framework for policy-making: proposing a quality assurance assistant tool (QAAT). In: Science for Policy. New Challenges, New Opportunities, . Corral, S., 2011. Gobernabilidad hídrica y la directiva marco del agua. (XIII reunión de economía mundial). Corral, S., Legna-de la Nuez, D., de Lara, D.R.M., 2015. Integrated assessment of biofuel production in arid lands: Jatropha cultivation on the island of Fuerteventura. Renew. Sust. Energ. Rev. 52, 41–53. Corral-Quintana, S., Legna-de la Nuez, D., Verna, C.L., Hernández, J.H., de Lara, D.R.M., 2016. How to improve strategic decision-making in complex systems when only qualitative information is available. Land Use Policy 50, 83–101. Crosetto, M., Tarantola, S., 2001. Uncertainty and sensitivity analysis: tools for GIS-based model implementation. Int. J. Geogr. Inf. Sci. 15 (5), 415–437. Cukier, R., Fortuin, C., Shuler, K.E., Petschek, A., Schaibly, J., 1973. Study of the sensitivity of coupled reaction systems to uncertainties in rate coefficients. I theory. J. Chem. Phys. 59 (8), 3873–3878. Czitrom, V., 1999. One-factor-at-a-time versus designed experiments. Am. Stat. 53 (2), 126–131. Escobar, J., Bonilla-Jimenez, F.I., 2009. Grupos focales: Una guía conceptual y metodológica. Cuadernos Hispanoamericanos De Psicología 9 (1), 51–67. Funtowicz, S., De Marchi, B., 2000. Ciencia posnormal, complejidad reflexiva y sustentabilidad. la complejidad ambiental. (México, Siglo). Funtowicz, S., Ravetz, J.R., 1991. A new scientific methodology for global environmental issues. Ecol. Econ. 10, 137. Funtowicz, S., Ravetz, J.R., 1993. La ciencia posnormal: Ciencia con la gente Icaria editorial . Galimberti, A., De Mattia, F., Losa, A., Bruni, I., Federici, S., Casiraghi, M., ... Labra, M., 2013. DNA barcoding as a new tool for food traceability. Food Res. Int. 50 (1), 55–63. Garmendia, E., Gamboa, G., 2012. Weighting social preferences in participatory multicriteria evaluations: a case study on sustainable natural resource management. Ecol. Econ. 84, 110–120. Giampietro, M., Mayumi, K., Munda, G., 2006. Integrated assessment and energy analysis: quality assurance in multi-criteria analysis of sustainability. Energy 31 (1), 59–86. Gibbons, M., 1999. Science's new social contract with society. Nature 402, C81–C84. Pereira, Guimarães A., Corral Quintana, S., Funtowicz, S., 2005. GOUVERNe: new trends in decision support for groundwater governance issues. Environ. Model. Softw. 20 (2), 111–118. Guimarães, A., Corral, S., 2002. From technocratic to participatory decision support systems: responding to the new governance initiatives. J. Geogr. Inf. Decis. Anal. 6 (2), 95–107. Henao, F., Cherni, J.A., Jaramillo, P., Dyner, I., 2012. A multicriteria approach to sustainable energy supply for the rural poor. Eur. J. Oper. Res. 218 (3), 801–809. Hernández-González, Y., Corral Quintana, S., 2016. An integrated assessment of alternative land-based passenger transport policies: a case study in tenerife. Transp. Res. A Policy Pract. 89, 201–214. Hernández-González, Y., Corral, S., 2017. An extended peer communities' knowledge sharing approach for environmental governance. Land Use Policy 63, 140–148. Hiltunen, V., Kangas, J., Pykäläinen, J., 2008. Voting methods in strategic forest planning—experiences from metsähallitus. Forest Policy Econ. 10 (3), 117–127. Huerta, J.M., 1977. Los grupos focales. Universidad de Puerto Rico, Recinto de Mayagüez, Colegio de Ciencias Agrícolas, Servicio de Extensión Agrícola. JRC, 1996. NAIADE Manual and Tutorial-Versión 1.0. ENG. ispra site: Joint Research Centre of the European Commission . Kangas, A.S., Kangas, J., 2004. Probability, possibility and evidence: approaches to consider risk and uncertainty in forestry decision analysis. Forest Policy Econ. 6 (2), 169–188. http://dx.doi.org.accedys2.bbtk.ull.es/10.1016/S1389-9341(02)00083-7. Kangas, A., Heikkilä, J., Malmivaara-Lämsä, M., Löfström, I., 2014. Case Puijo—evaluation of a participatory urban forest planning process. Forest Policy Econ. 45 (0), 13–23. http://dx.doi.org.accedys2.bbtk.ull.es/10.1016/j.forpol.2014. 03.007. Laukkanen, S., Kangas, A., Kangas, J., 2002. Applying voting theory in natural resource management: a case of multiple-criteria group decision support. J. Environ. Manag. 64 (2), 127–137. Ligmann-Zielinska, A., Jankowski, P., 2008. A framework for sensitivity analysis in spatial multiple criteria evaluation. In: Cova, T.J. (Ed.), Geographic Information Science. Proceedings 5th International Conference GIScience 2008. vol. 5266. Berlin/Heidelberg, Springer, pp. 217–233 Park city, UT, September 23–26. Lecture notes in computer science. Locatelli, B., Rojas, V., Salinas, Z., 2008. Impacts of payments for environmental services on local development in northern costa rica: a fuzzy multi-criteria analysis. Forest Policy Econ. 10 (5), 275–285. Mahlau, M., Briz, J., de Felipe, I., 2002. Actitud y comportamiento de consumidores alemanes respecto al aceite de oliva: Un análisis de sesiones de grupo objetivo (focus

62