Journal of Environmental Management 327 (2023) 116762 Contents lists available at ScienceDirect Journal of Environmental Management journal homepage: www.elsevier.com/locate/jenvman Research article Participatory multi-criteria evaluation of landscape values to inform wildfire management Gonzalo Gamboa a, *, Iago Otero b, Conchy Bueno a, Etel Arilla c, Helena Ballart d, Laura Camprubí e, Guillem Canaleta d, Gemma Tolosa c, e, Marc Castellnou c a Institute of Environmental Science and Technology, Universitat Autònoma de Barcelona, Building ICTA-ICP, C/ de Les Columnes s/n, Campus de La UAB, 08193, Cerdanyola del Vallès, Barcelona, Spain b Interdisciplinary Centre for Mountain Research, University of Lausanne, Site de Sion - Ch. de L’Institut 18, 1967, Bramois, Sion, Switzerland c Forest Actions Reinforcement Group (GRAF), Firefighters Body, Catalan Government, Avinguda de Serra Galliners, 112, 08290, Cerdanyola del Vallès, Barcelona, Spain d Pau Costa Foundation, Av. Mossèn Cinto Verdaguer, 42 A, Baixos 2A08552 Taradell, Barcelona, Spain e El Risell SCCL, C/ Rabassa, 46-48 A, 08024, Barcelona, Spain A R T I C L E I N F O A B S T R A C T Keywords: Climate change is expected to increase the number of days with meteorological conditions conducive to un­ Participatory multi-criteria evaluation controllable wildfires. Thus, it is necessary to strengthen the capacity of wildfire-prone regions to minimize the Landscape values adverse impacts of these wildfires by creating resilient landscapes. In this paper we develop a participatory multi- Wildfire management strategy criteria evaluation to identify and map landscape values and prioritize areas according to these values in the Adaptation to risk Montseny Biophere Reserve (Catalonia, NE Spain). Then, we draft a wildfire management strategy to protect the Resilience areas that have been prioritized through selected fuel reduction sectors that would reduce wildfire intensity. Finally, we emphasize the added value of a participatory multi-criteria evaluation in the adaptation to and management of expected megafires. We find that the integration of landscape values through participatory multi- criteria evaluation has the potential to alter wildfire management strategies by adding fuel reduction sectors and changing their implementation order. However, the implementation of the planned fuel reduction treatments faces socioeconomic and institutional barriers that call for a deeper engagement with transdisciplinary project design and transformative science. 1. Introduction matter of concern (San Miguel-Ayanz et al., 2018). Indeed, fire ser­ vices increasingly fear not being able to control devastating wildfires, Climate change is expected to substantially alter the Earth’s wildfire especially those affecting wildland-urban interfaces (Castellnou et al., activity (Moritz et al., 2012). Focusing down to Europe, its Mediterra­ 2019). nean part has witnessed an increase in the length of fire weather seasons Thus, there is an evident need to strengthen the capacity of wildfire- during the last decades (Jolly et al., 2015), and large wildfires have been prone regions to adapt to extreme wildfire risk. Adaptive capacity is here catastrophic for entire regions due to socioeconomic losses and human broadly understood as the ability to implement social processes that casualties (San Miguel-Ayanz et al., 2013). A sharp increase in the minimize the adverse impacts of expected wildfires via the creation of number of days with meteorological conditions conducive to extreme resilient landscapes, which would be able to withstand this perturbation wildfires is projected for the coming decades (Bowman et al., 2017), as without major losses in ecological and social functions (Prior and Erik­ well as in its summer burnt area (Turco et al., 2018; Moritz et al., 2012). sen, 2013; González-Hidalgo et al., 2014; Abrams et al., 2015; Fischer More northern latitudes have recently experienced unprecedented and Jasny, 2017). One potential way to strengthen adaptive capacity is wildfire events, turning the wildfire challenge into a continent-wide through participatory planning whereby public agencies, stakeholders, * Corresponding author. E-mail addresses: gonzalo.gamboa@uab.cat (G. Gamboa), iago.otero@unil.ch (I. Otero), conchybueno@gmail.com (C. Bueno), etel@gencat.cat (E. Arilla), helena@paucostafoundation.org (H. Ballart), laura@elrisell.cat (L. Camprubí), gcanaleta@paucostafoundation.org (G. Canaleta), gemma@elrisell.cat (G. Tolosa), incendi@yahoo.com (M. Castellnou). https://doi.org/10.1016/j.jenvman.2022.116762 Received 24 January 2022; Received in revised form 14 September 2022; Accepted 10 November 2022 Available online 1 December 2022 0301-4797/© 2022 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by- nc-nd/4.0/). G. Gamboa et al. Journal of Environmental Management 327 (2023) 116762 citizens and scientists design, implement and monitor fuel reduction 2. The study region treatments (Chapin et al., 2008; Butler and Goldstein, 2010; Everett and Fuller, 2011; Almstedt and Reed, 2013; Plana et al., 2015; Gazzard et al., The Montseny Biosphere Reserve (MBR) is located in the Catalan Pre- 2016; Otero et al., 2018). Fuel reduction treatments have the potential coastal mountain range (Fig. 1). It is made up of eighteen municipalities to lower wildfire spread rates and intensity and thus minimize their in the counties of La Selva, Osona and Vallès Oriental, between the adverse effects (Costa et al., 2011). provinces of Barcelona and Girona. The total area is 50,167 ha and it has A crucial challenge of participatory wildfire planning lies in ac­ a population of about 52 thousand inhabitants. The altitudinal gradient counting for the vast diversity of landscape values and functions and the presence of three biogeographical strata (Mediterranean, Euro- potentially affected by a wildfire (Rawluk et al., 2017). Diverse combi­ Siberian and Boreal-Alpine) give the MBR a great diversity of habitats nations of wildfire modelling and expert or participatory value assess­ and species in a relatively small territory. Far from being considered a ment have been developed with the aim of including landscape values purely ecological phenomenon, the diversity of the MBR is also the into adaptive management strategies (Morehouse et al., 2010; Alcasena result of anthropogenic use of its resources throughout history (Boada, et al., 2015; Otero et al., 2018). However, new efforts are needed to 2002; Roigé and Estrada, 2008). Activities such as agriculture, livestock build more inclusive evaluation methods and to ensure that the results of breeding and logging, present since the Neolithic in varying intensities, participatory planning are effective for the adaptation of wildfire-prone have acted as an additional source of landscape diversity. regions, beyond purely academic exercises (Otero et al., 2018). In the second half of the 20th century, socio-economic development In this paper we deploy a participatory multi-criteria evaluation decreased the importance of primary activities to the benefit of other process (Munda, 2004, 2008; Proctor and Drechsler, 2006) and sectors, such as industrial, tourist or residential activities (Sánchez, self-reflect on its potential to improve wildfire management. Our study 2010). The depopulation of farmhouses, the migration of their in­ shows that a participatory multi-criteria evaluation of landscape values habitants to towns and villages and the arrival of inhabitants whose may alter wildfire management strategies by adding fuel reduction economic activity is not related with the primary sector changed the sectors and/or changing their implementation order as compared to landscape: forests expanded, crops and pastures were reduced, and the purely technical considerations of wildfire behaviour and management. urban and industrial surface increased (Boada, 2002). At the same time, We also identify relevant factors influencing the outcomes of a partici­ the appreciation of nature by a growing urban population led, in 1978, patory multi-criteria evaluation of landscape values to inform wildfire to a land planning scheme that protects Montseny as a Natural Park, as management, and propose ways to deal with them in a transparent way. well as to the declaration of the MBR. Climate change comes to add to In the following, we introduce the study region and the project these transformations. In the second half of the 20th century, the (Section 2), we describe the methods employed (Section 3) and we average annual temperature increased by 1.2–1.4 ◦ C, with considerable present the main results (Section 4). Next, we reflect on our findings effects on ecosystems (Jump et al., 2007; Peñuelas et al., 2007; Peñuelas with particular attention on the added value of the participatory multi- and Boada, 2003). All these transformations and changes contribute to criteria evaluation process to derive lessons learnt (Section 5). Section 6 increase the vulnerability of the MBR to wildfires (Otero and Arilla, concludes. 2015). Fig. 1. Montseny Biosphere Reserve. Source: Own elaboration. 2 G. Gamboa et al. Journal of Environmental Management 327 (2023) 116762 3. Methodology corresponding fire propagation model. First, the type was defined based on the analysis of the following aspects: the causes of wildfires, the most Our methodology is characterized by the joint work between scien­ important episodes, the types of fires, the danger (i.e. the possibility of a tists and actors to find solutions to a particularly problematic situation wildfire, assessed from past ignitions, orography and vegetation), the (large wildfires), where uncertainty, conflicting interests and urgency vulnerability (which indicates the damage that a fire can cause and is shape public decision-making and call for an extended-peer community assessed on the basis of the presence of elements such as settlements, to face complex problems (Funtowicz and Ravetz, 1993). infrastructures or protected areas within forest lands or less than 500 m To do so, we followed the methodological steps presented in Fig. 2, from them) (Generalitat de Catalunya, 2020) the synoptic conditions which combine scientific interdisciplinary work and public participa­ (Duane and Brotons, 2018) boosting wildfires and the homogeneous tion. Interdisciplinarity integrates theories and methods from different regime zones (i.e. areas of the territory with similar fire frequencies and scientific disciplines to deal with technical incommensurability, i.e., homogeneous propagation patterns considering previous wildfires) multiple (disciplinary) descriptions of the same system. Public partici­ (Generalitat de Catalunya, 2018; Castellnou et al., 2009; Costa et al., pation deals with social incommensurability, i.e., incorporating multiple 2011; Piqué et al., 2011). and legitimate values that exist in society (Munda, 2004). (See Next, we defined containment polygons according to the worst-case Tables SM1 and SM2 that show the different sources of information scenario defined previously. Containment polygons are the territorial used, from technical studies on wildfire risks to deliberations with units that firefighters use to assess the pattern of spread of fires, predict regional and local actors). their extent and manage wildfires. Containment polygons are defined To integrate expert and lay knowledge, a participatory process was based on basins of different sizes, their limits are defined following designed at three different levels following Otero et al. (2018): regional, either ridges or the valley bottoms, depending on the type of wildfire: local and individual. The first two refer to the administrative capacity to bigger wildfires entail bigger containment polygons. Within a polygon, implement policies, and/or to carry out socio-economic and environ­ wildfire is expected to burn with a similar behaviour (Castellnou et al., mental management. Regional actors are those that perform their 2019) and, when fire enters to a containment polygon, it is expected to functions at the whole MBR and beyond, and local actors operate within burn down at once and completely. Third, we built a qualitative fire the MBR (e.g. at municipal scale). Therefore, social actors were expected propagation model for the selected fire type and the containment to provide knowledge according to the territorial scale they operate. polygons. It was defined according to the Campbell prediction system Regional actors include those involved in wildfire prevention and (Campbell, 1995) and indicates the intensity with which the fire would territorial governance (e.g. technicians from the Territorial services of pass from one polygon to another. The intensity depends on the orien­ the Catalan Department of Agriculture, the Wildfire Prevention Tech­ tation of the slope (sunny or shady), the fire movement (uphill or nical Municipal Office of the Barcelona Provincial Council and of the downhill) and direction (upwind or downwind). Thus, a distinction can Montseny Natural Park), actors who could provide information on be made between maximum, medium, or low vectorial alignment of fire landscape values (e.g. regional museums), as well as regional associa­ propagation forces (slope, wind and orientation), which generate situ­ tions (e.g. Federations of Forest Defence Associations, Association of ations characterized by being outside, at the limit, or within the extin­ Forest Owners). Actors at local level included local institutions (e.g., guishing capacity, respectively. For example, a fire that spreads on a municipalities) and local associations (e.g., Forest Defence Associations, sunny slope, uphill and downwind is rated 3 out of 3 (maximum align­ Tourist Business Association). We organized several meetings with ment). With adverse weather and fuel conditions, it would be out of regional and local actors, either to validate the methodology, to identify extinguishing capacity of fire services. landscape values and to present and discuss results. At the individual level, we invited citizens to participate in three exhibitions set up in 3.2. Identification of landscape values municipal markets to get their opinion about places they value most (See Table SM2, which synthesizes the different meetings and activities Landscapes are characterized by its biophysical features as well as by carried out, their objectives and the participants). the socio-cultural perceptions of it (Termorshuizen and Opdam, 2009). Following this idea, we understand landscape values as an attachment or 3.1. Definition of the wildfire propagation scenario feelings that people develop with places, which are based on aspects of the landscape that are important to people (held values) or are the This stage was aimed at defining the type of fire that can cause the expression of the relative importance of the place in relation to other largest damage to the MBR, the containment polygons and the places (assigned values) ((Brown and Reed, 2000; Brown and Weber, Fig. 2. Methodological process, levels of participation and aim of meetings and workshops. Note: The numbers in brackets indicate the sections of this article in which the different steps are explained. 3 G. Gamboa et al. Journal of Environmental Management 327 (2023) 116762 2012; Penning-Rowsell, 1981)). According to (Brown and Weber, 2012), 3.3. Mapping landscape values landscape values merge what is important for the individual (held value) to what is important for the individual in the physical landscape The information on landscape values collected from expert knowl­ (assigned value). Landscape values can be instrumental or edge and public opinion (Section 3.2) was introduced into a Geographic non-instrumental. The former refers to what provides tangible benefits Information System. Participatory GIS has been considered an adequate to people such as economic value through socio-economic activities (e.g. approach to map and analyse landscape values for land use management agriculture, tourism, forest management), and the latter refers to (Fagerholm and Käyhkö, 2009). Studies identifying landscape values something that is worth in itself. In fact, some landscape values can be usually analyse their geographical patterns – spatial distribution, fre­ considered as landscape features that have not necessarily a quantifiable quency, intensity/density – the relative importance and the spatial worth, importance, esteem or utility, but that strongly attract and con­ relation between values (Reed and Brown, 2003; McIntyre et al., 2004; nect individuals to particular areas (Morehouse et al., 2010). The Brown, 2004; Fagerholm and Käyhkö, 2009; Garcia-Martin et al., 2017) concept of landscape value can capture the holistic character of and analyse the benefits provided by the landscape to human wellbeing landscape-related benefits and values (Bieling et al., 2014) and would (Fagerholm and Käyhkö, 2009; McIntyre et al., 2004; Brown, 2004; match people’s way of framing and expressing landscape services, Garcia-Martin et al., 2017). In this case we apply a participatory GIS benefits and values (Bieling et al., 2014). framework to identify and map landscape values, but we haven’t In this study, we identify components of the landscape that are considered the frequency and intensity with which the participants valued by participants. Components can be classified in forms, re­ expressed the values. lationships and practices, which are dynamic and coevolve in time. When mapped, different elements were assigned different scores Forms refer to physical and measurable aspects of the landscape, according to the importance within each of the 5 categories: biodiver­ including natural (e.g. landform, vegetation, water bodies) or anthro­ sity, socioeconomic activities, heritage, leisure, and public values. For pogenic (e.g. croplands, bridges, roads) elements. Relationships instance, within the biodiversity category, elements of natural heritage encompass people–people interactions, people–landscape interactions, with high and very high conservation interest were given a score of 4 and ecological relationships within the landscape. Practices include both and habitats of technical priority were given a score of 3. These values human practices and ecosystem processes (Stephenson, 2008). were assigned by the research group and validated with experts We combined data from both expert and lay knowledge. The data participating in the project. These scores are relative within each cate­ collected was iterated between groups of actors in the different meetings gory, i.e. there is no relation between the scores assigned to elements of held along the project, and the database with the landscape values grew one category and the scores assigned to elements of another category. both qualitatively and quantitatively. Expert knowledge on landscape Five vector maps (one for each category) were generated. Then, a ras­ values was mainly obtained from the databases of the Public use plan of terization process of each vector map took place, in which the pixels of the Montseny natural park (Montseny Natural Park,), stored in the the raster maps have a score according to the elements contained in that Montseny Information system. The Public use plan contains information pixel. For instance, a pixel in the biodiversity raster map containing a on habitats and biodiversity, socio-economic activities within the park, monumental tree (score 1 in Table 1) and located within a technical built heritage and public use. The database of built heritage was vali­ priority habitat (score 3 in Table 1) had an aggregated score of 4. Then, dated and complemented by a technician of the Montseny Ethnologic the score of each containment polygon under each category was calcu­ Museum. Information on habitats and biodiversity was validated and lated by aggregating the scores of all the pixels contained in it. complemented by technicians from the Biodiversity and Environment The mapping of values was carried out only in the “core” and section of the Department of Territory and Sustainability, Catalan gov­ “buffer” areas of the MBR, which represents 63.7% of the total (Fig. 1). ernment, and technicians from the Natural Science Museum of Gran­ Outside this area (“transition zone”), the availability of data was partial, ollers (sources of information are presented in Table SM1). Then, the and a systematic comparison between containment polygons was not identified landscape values were classified in the following four cate­ possible. gories: biodiversity, socioeconomic activities, built heritage and leisure, which were validated and used in subsequent meetings with the social 3.4. Prioritization of the territory actors. Meetings and public exhibitions were carried out to identify land­ The containment polygons identified in Step 3.1 were prioritized scape values from the local actors and individual citizens. Meetings with according to the number and diversity of values contained. First, local actors included a presentation of the preliminary results of the containment polygons are ranked within each category according to the project (e.g., wildfire regime, fire propagation model, landscape values aggregated score calculated through the rasterization process. That is, already mapped based on expert knowledge). Then, the participants of within each category, containment polygons were ranked according to the meetings were split in three thematic groups and were asked to the amount of landscape values contained in them. Second, to obtain the discuss and locate in the map their most valued places. Inspired by final ranking of polygons, we applied a multi-criteria evaluation Bieling et al. (2014) we asked three open questions: What are the areas, method. Multi-criteria evaluation is an analytical framework to support places or activities you value most? What area would you not want to decision-making (Munda, 2008). It allows the analyst to compare a set of lose in case of a big wildfire? What containment polygon would you alternatives under a set of multidimensional criteria. In this case, the protect first in case of wildfire? Answering these questions also facili­ containment polygons are the alternatives to evaluate and compare, and tated the identification of landscape values not included in the data­ the different categories of values are the evaluation criteria. Thus, a bases. Local actors were also asked to prioritize the categories defined by multi-criteria impact matrix is constructed with n rows and m columns, the expert group and reflect on missing categories. where n is the number of containment polygons to prioritize and m, the With this information, we introduced a new category called “Public number of categories of values. Therefore, the multi-criteria impact values”, which includes the places and (localized) practices valued by matrix has been populated with the ranking of containment polygons for local actors and individual citizens. Since the participants of the meet­ the biodiversity, socioeconomic activities, heritage, leisure and public ings and exhibitions are not a representative sample of the population, values categories. we did not calculate the frequency of landscape values, and places and This information can be qualitatively analysed or a mathematical activities were counted only once, even though several places such as algorithm can be applied to aggregate criteria and obtain a ranking of emblematic locations of the MBR were mentioned by many people. polygons. There is a great diversity of multi-criteria methods, each one with its advantages and disadvantages. In this case, a simplification of the PROMETHEE method (Brans et al., 1986) was used (More 4 G. Gamboa et al. Journal of Environmental Management 327 (2023) 116762 Table 1 Scores assigned to each element of the landscape identified by experts and general public. ScoreA Biodiversity Heritage Socio-economic activities Leisure Public values 5 − Buildings of heritage interest protected by − Highly frequented law areas − Hiking trails 4 − Elements with high and very high conservation interest 3 − Technical priority habitatsB − Churches and hermitages − Private hunting areas, − Potential habitat of Calotriton high captures arnoldiC 2 − Urban cores and − Private hunting areas, − Highly households medium captures valued sites 1 − Monumental trees − Residential buildings − Camping sites, recreational − Private hunting areas, − Valued sites − Open spaces: Harvest meadows, − Agrarian infrastructure of heritage interest areas and restaurants low captures Cropland, Cleared forest, Pastures − Buildings and facilities of the of the MBR − Cropland and pastures − Other elements with heritage interest (e.g. − Managed forest plots springs, graveyards, archaeological sites) − Resource extraction sites − Historic resource extractions sites (e.g. − Industrial areas. mines, lime kilns) Notes: A) The scores are relative within each category, i.e. there is no relation between the scores assigned to elements of one category and the scores assigned to elements of another category. The scores used in each category were defined together with experts participating in the project. B) Rivers, creeks and riparian areas are included in the Technical priority habitats. C) The potential habitats of the Calotriton arnoldi are explicitly included here because it is endemic of the Montseny park. information on the method can be found in the Supplementary mate­ model showed in Fig. 3. rial). It is a non-compensatory method, which uses weights as impor­ tance coefficients and is easy to understand and communicate to 4.2. Landscape values of the MBR non-experts; which are desirable characteristics when a multi-criteria model is used in decision-making processes (Janssen and Munda, The process of identifying and mapping values showed a remarkable 1999; Munda, 2004, 2008). Compensation refers to the possibility that amount and diversity of landscape values, highlighting both the social very good evaluations under some criteria can offset very bad evalua­ and ecological importance of the MBR. It also showed the values at stake tions under other criteria. Non-compensatory methods do not allow this when designing a strategy for wildfire management in a context of offsetting and, as a result, the best ranked polygons are those presenting climate change. The maps presented in Fig. 4a–f are the result of this a balanced diversity of landscape values across categories. process for the different categories of landscape values. Biodiversity (Fig. 4a) acquires the highest values in the central areas of the massif, such as the Pla de la Calma (where livestock activities have 3.5. Development of a wildfire management strategy given rise to meadows and moors of very high ecological value), and around Turó de l’Home or Matagalls, the highest iconic peaks of MBR. In The next step comprises a study of the areas that should be managed these areas we find the juxtaposed presence of different elements of high in order to help the fire extinction services to change the behaviour of a conservation interest, such as habitats of the Montseny brook newt convective fire with west wind (i.e. the wildfire propagation scenario), (Calotriton arnoldi), meadows, mature forests, alders, unique flora, considering the values and priorities identified in the previous stage. mosses, fungi, lichens and some birds and very high numbers of nesting First, the fire propagation model was used to identify the fire multiplier points. In contrast, the peripheral area has lower values. polygons, that is, those containment polygons with a significant number Built heritage (Fig. 4b) is spread across the MBR, which testifies the of connections in maximum alignment of propagation factors to other intense use that humans have made of Montseny, both historically and polygons. Second, Management Promotion Areas (MPAs) were proposed today. Built heritage is distributed in a more or less ubiquitous way, with to close fire entrances to and exits from multiplier polygons. A MPA is an some empty areas and others of high concentration, such as the Tordera area where active forest management to reduce fuel load is encouraged. valley or the Arbúcies stream. This is done to reduce the alignment of the propagation factors from Socio-economic activities (Fig. 4c) are present in much of the terri­ maximum to medium or low. Proper management of these areas would tory and acquire maximum values where we find inhabited areas or reduce the intensity of the fire and give options to extinguishing services farmhouses with primary sector activities. It highlights a wide range of to stop or manage the wildfire. Third, a fire management strategy was primary activities (farms with forestry activity, crops and pastures). developed to address any possible fire access route in maximum align­ Regarding leisure (Fig. 4d), higher values are mainly concentrated in ment to the prioritized polygons, to reduce fire intensity to extinguishing the southern periphery of the valued area. There, high-frequented places capacity. The strategy also established an order for the implementation and especially productive hunting areas are juxtaposed, generating the of the MPAs: temporal priority was given to actions that close the en­ maximum values. trances of the fire to the prioritized polygons, which would reduce the Finally, the category of Public values, developed mainly from the impact on the most important landscape values. Finally, the strategy was contributions of local actors, highlights several areas that must be discussed and validated with the regional actors. considered when planning fire prevention in the MBR (Fig. 4e). Among them we find Santa Fe, Turó de l’Home, the plain of La Calma, the hill of 4. Results Tagamanent, the castle of Montsoriu and the stream of Arbúcies. The participants mentioned several reasons for the importance of these 4.1. Wildfire propagation scenario and containment polygons areas, such as these are places of identity, popular historical places, or places where social and traditional gatherings take place. Based on the analysis of the region’s wildfire regime, a west wind convection wildfire was defined as the worst-case scenario for MBR, which lead us to define the containment polygons and the propagation 5 G. Gamboa et al. Journal of Environmental Management 327 (2023) 116762 Fig. 3. Containment polygons and propagation model for a west wind convection wildfire. Note: The red arrows represent high-intensity propagation (outside the extinction capacity); orange arrows , of medium intensity (at the limit of the extinction capacity), and green arrows , of low intensity (within the extinction capacity). Containment poly­ gons are identified with numbers. Dark grey line delimit the “buffer” and “core” zones of the MBR (target of the valuation). Source: Own elaboration based on the expert knowledge of the GRAF (Gen­ eralitat de Catalunya, 2018). (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.) Fig. 4. Prioritization of containment polygons by categories. a) Biodiversity; b) Heritage; c) Socio-economic activities; d) Leisure; e) Public values; f) Climate change. Darker colors indicate higher priority. Source: own elaboration based on scores indicated in Table 2. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.) 6 G. Gamboa et al. Journal of Environmental Management 327 (2023) 116762 4.3. Multi-criteria analysis of landscape values 4.4. Wildfire management strategy Table 2 presents the multi-criteria impact matrix, which indicates the The wildfire management strategy through Management Promotion ranking of containment polygons under each category. To obtain the Areas (MPAs) is a risk reduction approach to deal with large forest fires final ranking (extreme right column of Table 2), we applied the multi- and is aimed at reducing the impact of fire on the landscape values set criteria method explained in Section 3.4 and in the supplementary ma­ out in the previous sections. Fig. 5 shows the MPAs, which would be terial. A first comparison assigning equal weights (i.e. 20%) to all cat­ used to limit the extent of a convective fire with west winds, to facilitate egories was performed. Then, to check the robustness of the results, a the stabilization of fire tails and flanks, and to ensure access to extin­ sensitivity analysis was carried out by changing the weights of the cat­ guishing means. This is done by reducing the alignment of propagation egories: five additional comparisons were run in which we assigned, in factors from maximum to medium or low, so fire propagation is under each case, a weight of 30% to one category and 17,5% to the rest. When the extinguishing capacity of fire services. applying different weights, only minor changes occurred, and the 10 It should be noted that the MPAs are located mostly in the western highly ranked containment polygons remained almost the same. and central part of the MBR, as they are designed to prevent a fire The results of polygon prioritization for different value categories propagating eastward from growing. show a general pattern: priority polygons tend to be in the centre and Three types of MPAs can be seen in Fig. 5, which have been defined southwest of the massif (Fig. 4a–f). to manage a large convection wildfire driven by west winds starting in polygons 1 or 8 (located further west of the MBR). “MPAs technical Table 2 Ranking of containment polygons under each category and final multi-criteria ranking. Note: The first column indicates the different containment polygons, which are ranked according to the different categories of values in the subsequent columns. The last column indicates the ranking of polygons when applying the multi-criteria method and all categories are weighted equally. The cells highlighted in grey indicate the first ten polygons in the rankings under each category. 7 G. Gamboa et al. Journal of Environmental Management 327 (2023) 116762 Fig. 5. Management promotion areas (MPAs) iden­ tified taking into account the behaviour of a convec­ tive fire with west wind and the prioritization of values in the Montseny Biosphere Reserve. The map differentiates between those MPAs identified only with technical criteria related to wildfire propagation (brown) and those MPAs that were added after analyzing the data on landscape values from the participatory process (orange). A third layer (yellow) shows an additional set of MPAs that could be added to protect the urban centers of the periphery (which were not included in the participatory evaluation, see methods). Source: Own elaboration based on the knowledge of the GRAF, (Generalitat de Catalunya, 2018). (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.) Fig. 6. Diagram of the propagation of two convective fires with west wind through the containment polygons (ignition points in polygons 1 and 8). Management promotion areas are schematically represented. It also indicates the possible perimeters of the fire depending on the success of the management strategy. criteria” are those MPAs that were defined according to the criteria of would be focused on preventing the spread towards polygons 7, 14, 15 the fire services, without considering the most valued polygons. The aim and 16. If this is not achieved, polygons 17 and 18 can be protected by of these MPAs would be to avoid fire entering to and leaving from cutting access from 16 to 17 and from 14 to 15 to 18 and 19. This would polygons 4, 14 and 18, which are multipliers polygons (See Figs. 3 and also protect polygon 24, another highly prioritized one. For a fire that 6). “MPAs landscape values” are those MPAs that were defined to treat starts in polygon 8, the initial goal would be to contain it within the any possible path of fire in full alignment towards the most valued polygon itself. If the fire goes to polygon 13 (a multiplier), efforts would polygons (18, 19, 24, 17, 21 and 4) and/or reduce the intensity of fire be put to prevent it from passing to polygons 14 and 19. It was proposed down to extinguishing capacity. Finally, the “MPAs urban” are those to start working in the innermost MPAs of the massif (close to the pri­ MPAs defined to protect urban areas in the southern slope of the Mon­ ority central polygons) and end with the most peripheral MPAs. tseny massif (these were not included in the participatory multi-criteria The definition of the fuel reduction treatment, target vegetation evaluation, see section 3 Methodology). structure, and operational usefulness of each MPA were left for a suc­ For a fire that starts in polygon 1, the first goal would be to avoid fire cessive planning phase. Regarding treatments, forest management with exceeds polygons 1, 2, 3 and 5. In case the fire reaches polygon 4, efforts machinery, prescribed burns, agriculture, extensive livestock, and a 8 G. Gamboa et al. Journal of Environmental Management 327 (2023) 116762 combination of these (e.g., clearing with machinery to leave the target people), overcoming potential drawbacks of purely participatory pro­ structure and subsequent maintenance with prescribed burn or exten­ cesses. For instance, we now think that highly frequented areas (Table 1) sive grazing) were proposed. These actions were considered as an op­ should not be included as a landscape value. Most of these places are portunity to reactivate the primary sector in the MBR. However, during very well connected with roads and public transport, which determines the last project meeting, when this strategy was presented to social ac­ which places are visited and which ones are not. In fact, most of highly tors, several participants expressed their concern about the lack of visited areas coincide with the parking lots where information points are funding to implement the MPAs. Several alternatives were proposed: to located (and visitors are registered). In any case, it is important to un­ explicitly consider MPAs as an infrastructure against wildfires at the derline that our participatory process can be used to identify valued Catalan level (and budget them accordingly), to prioritize MPAs in calls places, but not the degree to which a place is valued. The reason is that for forest management subsidies, a scheme of payment for ecosystem for the latter to be possible, we would need to reach a representative services to forest owners that implement MPAs, or to include the MPAs sample of the population and to assume that respondents have complete in the Fire prevention plan of the MBR and its municipalities. information about the benefits they obtain from landscape. Even in this case, the actors not willing or not able to participate, as well as future 5. Discussion generations, would be underrepresented. In addition, the extremely broad range of values potentially affected by the expected wildfires We developed a method to include landscape values in the design of makes it difficult to establish what is the “population” that should be a wildfire management strategy aimed at adapting a fire-prone region to sampled to estimate the value of landscape. As indicated by Otero et al. new climatic conditions, which can be associated with more intense (2018) this should not constitute a problem as long as participatory fires. processes are able to build legitimacy among actors and citizens in a The work presents two important novelties regarding the method­ long-term transformative engagement (See also Otero, 2022). ology developed by Otero et al. (2018). First, the creation of a map (Geographic Information System) in which each pixel is associated to a 5.2. Multi-criteria evaluation landscape value, as it was done by Morehouse et al. (2010). In theory, this allows wildfire managers to overlap the landscape values with Regarding the multi-criteria evaluation, it should be noted that the different maps of fire containment polygons (designed to deal with optimal methods for this type of project are the so-called non-compen­ different types of fires and climate change scenarios) and other layers satory. Non-compensatory methods do not allow very good results under with relevant information for spatial planning. Potentially interesting one criterion to compensate for very bad results under other criteria, as layers that could be added to refine our exercise are energy and water the criteria may represent the values, goals, and priorities of different infrastructures (Alcasena et al., 2015). social actors (Munda, 2004, 2008). The chosen multi-criteria method Second, the incorporation of multi-criteria analysis, which allows the avoids trade-offs between categories of landscape values. However, the automated prioritization of polygons using a series of criteria that can be chosen method to calculate the value of each polygon under each adjusted with successive phases of social participation. Multi-criteria category (Section 3.3) does allows intra-category compensation, since it evaluation has been used to allocate management areas based on fire works as a linear aggregation procedure. This means, for instance, that potential and logistical factors important to fire fighters (Gonzalez-O­ elements of biodiversity with high and very high conservation interest labarria et al., 2019). With our method, the allocation of such areas can (that were given a score of 4 Table 1) equates to four monumental trees be done by also considering landscape values from the affected (which were given a score of 1), or that a legally protected church (with communities. a score of 5) burning down in a wildfire is the same than five fountain (with a score of 1) being burn down. 5.1. Participation and valuation Even though the scores given to different elements with different value were validated by experts in our workshops, the trade-offs Several approaches to identify landscape values have been devel­ mentioned above are difficult to assess even for an expert in the corre­ oped, from purely expert approaches based on specific academic disci­ sponding field. Moreover, it should be noticed that the scoring scheme plines to more grounded approaches based on public participation. (i.e. spread of scores, number of elements scored differently, and the Purely expert-based analysis may fail to reflect the values expressed by value of the scores themselves) could influence the outcomes of the those who feel belonging to the landscape, since the choice of the multi-criteria evaluation process. To test this, we performed a sensitivity discipline and of the assessment method would determine what is analysis by changing the spread of scores, and no changes occurred in perceived and considered as valuable (Stephenson, 2008). In fact, the the prioritization of containment polygons. pre-analytical decisions (i.e. those made before data collection and However, for future experiences we recommend carrying out a two- analysis) about the relevant attributes used to describe and represent a stage multi-criteria evaluation process to rank containment polygons: an system, determine the methods of observation, the disciplines and, intra-category and an inter-category multi-criteria evaluation using the consequently, the results of the analysis (Kovacic and Giampietro, same non-compensatory multi-criteria method. Let’s consider, for 2015). In our study, we have included an extra category of landscape instance, the intra-category comparison for the biodiversity category. In values that come directly from the participatory process with local ac­ this case, 3 criteria would be used to rank the containment polygons: i) tors and citizens. In this sense, the use of the concept of landscape values number of elements with high and very high conservation interest; ii) was straight forward; there was little need to explain its meaning to the technical priority and potential habitats of high interest, and iii) participants, who could easily express what they value most from the open spaces. In this way, it wouldn’t be necessary to assign a score to landscape. In this way, we have expanded the pre-analytical categories different landscape elements, and the ranking within each category defined according to expert knowledge or recorded in expert-based would be obtained without allowing compensation between high and databases, by adding elements such as local or village festivities low-valued elements of that category. Then, the final ranking of and/or fountain locations. Also, the inclusion of non-expert knowledge containment polygons would be based on the rankings obtained for each by means of participatory workshops was useful to perform a quality category, as it has been done in this study. check and validate the methodology, the landscape values categories We have learnt that there are several factors that would influence the and the quality of the information used in the study. outcomes of a participatory multi-criteria evaluation: the quality of the At the same time, the inclusion of expert knowledge enabled us to information populating the impact matrix (which is related to the identify landscape values beyond those areas that are well connected landscape values included in the evaluation), the spread of the scores, with roads and transport services (and thus are easily accessed by the weight of the criteria, and the multi-criteria aggregation method. In 9 G. Gamboa et al. Journal of Environmental Management 327 (2023) 116762 general terms, the content of the valuation was contingent upon the landscape values, the prioritization of wildfire containment polygons by actors included and their availability to provide input. As such, it was applying a multi-criteria evaluation method and the development of a necessarily non-exhaustive and relevant values potentially influencing wildfire strategy based on MPAs. The implementation of MPAs is, in public priorities could have been left out, as it was also found in Otero turn, a great opportunity to boost the primary sector and build a mosaic et al. (2018). In this sense, we think that transparency (to make as­ landscape resilient to large wildfires, but a greater engagement with sumptions visible) and an inclusive participatory process are adequate transformative governance and transdisciplinary project design is ways of improving knowledge of the problem at hand and of controlling necessary. the quality of the information and assumptions used (Funtowicz and Ravetz, 1993; Gamboa, 2006; Munda, 2008; Garmendia and Stagl, 2010; Credit author statement Garmendia et al., 2010). We have carried out a multi-level and multi-actor participatory process, with several iterations allowing par­ Gonzalo Gamboa Conceptualization, Methodology, Software, ticipants to raise their concerns on the methodology, the problem Visualization, Formal analysis, Investigation, Data curation, Writing – structuring and the analysis. original draft, Writing – review & editing, Project administration, Iago Where is the added value of participatory multi-criteria evaluation? Otero Conceptualization, Methodology, Writing – original draft, To answer this question, it is worth recalling that the MPAs target Writing – review & editing, Supervision, Project administration, Fund­ strategic areas located in the borders of the containment polygons to ing acquisition, Conchy Bueno Visualization, Investigation, Data prevent fire entering into them. Within a containment polygon, wildfire curation, Etel Arilla Investigation, Helena Ballart Investigation, Laura is expected to burn with a similar behaviour and assumed to consume it Camprubí Investigation, Guillem Canaleta Investigation, Writing – completely. Therefore, the output of the multi-criteria evaluation (i.e., review & editing, Gemma Tolosa Investigarion, Marc Castellnou ranking of most valued polygons) determines both which MPAs should Conceptualization be implemented and the order of implementing them (from the inside out of the MBR), to protect the most valued polygons. Our study shows Funding that the participatory multi-criteria evaluation of landscape values has the potential to alter wildfire management strategies by adding fuel This research has been funded by the Biodiversity Foundation of the reduction sectors and changing their implementation order as compared Ministry for the Ecological Transition, through the call of climate change to purely technical considerations of wildfire behaviour and manage­ adaptation projects (PRCV00643). ment. Therefore, it can be considered as a spatially cost-effective adaptation strategy, which minimizes the loss of landscape values in the face of unavoidable catastrophic fires. Declaration of competing interest However, the MPAs that were planned during the project have not yet been implemented. One the one hand, this would have required The authors declare that they have no known competing financial more and better networking with local authorities so that they could interests or personal relationships that could have appeared to influence integrate the MPAs in their wildfire prevention schemes (Otero et al., the work reported in this paper. 2020). On the other hand, this is linked to the current residential and tourist function of the MBR, where primary activities have lost much of Data availability the economic viability they had in the past (Boada, 2002). In fact, most of the participants in the study considered it necessary to promote the Data will be made available on request. primary sector and a “rural” life-style. In our meetings, it was argued that living in and from the landscape (e.g. farming, grazing and forestry Acknowledgements activities) would both increase surveillance capacity and generate a landscape mosaic resilient to wildfires, as has been argued elsewhere We would like to thank all institutions and people that have partic­ (Mancini et al., 2018; Oliveira and Zêzere, 2020). ipated in the project. Special thanks to Gemma Font (Ethnological Moving from deliberative exercises to concrete action is one of the Montseny Museum, la Gabella), Anna Sanitjas (Diputació de Girona), biggest challenges to reduce the risk of fires in a context of climate María Barrachina (Montseny Natural Parc), Roser Maneja, Martí Boada, change. In fact, the viability of the creation of a resilient landscape Sònia Sánchez, Montserrat Puigdomènech and Isabel Lopera (Institute of through the promotion of primary activities (Aquilué et al., 2020; Otero, Environmental Science and Technology, Universitat Autònoma de Bar­ 2011; Regos et al., 2016); ; depends on larger changes in socioeconomic celona) and Anna Badia (Department of Geography, Universitat models and food systems (Otero and Nielsen, 2017). Autònoma de Barcelona). Thus, to adapt European and other landscapes to climate change and wildfires of increasing intensity, the use of participatory multi-criteria Appendix A. Supplementary data evaluation should be complemented by a stronger effort in trans­ disciplinary research design (Lang et al., 2012; Moser, 2016) and Supplementary data to this article can be found online at https://doi. transformative governance (Visseren-Hamakers et al., 2021), to coor­ org/10.1016/j.jenvman.2022.116762. dinate policies across sectors (wildfire management, biodiversity con­ servation, agriculture, etc.) for the creation of functional and resilient References landscapes. Abrams, J.B., Knapp, M., Paveglio, T.B., Ellison, A., Moseley, C., Nielsen-Pincus, M., Carroll, M.S., 2015. Re-envisioning community-wildfire relations in the U.S. West as 6. Conclusions adaptive governance. Ecol. Soc. 20 https://doi.org/10.5751/ES-07848-200334. Alcasena, F.J., Salis, M., Ager, A.A., Arca, B., Molina, D., Spano, D., 2015. Assessing This work developed a method to develop a wildfire management landscape scale wildfire exposure for highly valued resources in a mediterranean area. Environ. Manag. 55, 1200–1216. https://doi.org/10.1007/s00267-015-0448- strategy based on landscape values, using a participatory multi-criteria 6. evaluation methodology. It integrated expert and non-expert knowl­ Almstedt, Å., Reed, M.G., 2013. Introducing a framework for good and adaptive edge by means of participatory workshops and meetings, the review of governance: an application to fire management planning in Canada’s boreal forest. technical documents and the inclusion of expert knowledge-based da­ For. Chron. 89, 664–674. https://doi.org/10.5558/tfc2013-119. Aquilué, N., Fortin, M.J., Messier, C., et al., 2020. The potential of agricultural tabases. Our method included the definition of the worst-case wildfire conversion to shape wildfire regimes in mediterranean landscapes. Ecosystems 23, (wildfire propagation scenario), the identification and mapping of 34–51. https://doi.org/10.1007/s10021-019-00385-7. 10 G. Gamboa et al. Journal of Environmental Management 327 (2023) 116762 Bieling, C., Plieninger, T., Pirker, H., Vogl, C.R., 2014. Linkages between landscapes and against large forest fires in the Catalonia region of Spain. For. Ecol. Manag. 432, human well-being: an empirical exploration with short interviews. Ecol. Econ. 105, 612–622. https://doi.org/10.1016/j.foreco.2018.09.039. 19–30. https://doi.org/10.1016/j.ecolecon.2014.05.013. Janssen, R., Munda, G., 1999. Multi-criteria evaluation for quantitative, qualitative and Boada, M., 2002. In: El Montseny: Cinquanta Anys D’evolució Dels Paisatges. fuzzy evaluation problems. In: van der Bergh, J.C.J.B. (Ed.), Handbook of Publicacions de l’Abadia de Montserrat, Barcelona. Environmental and Resource Economics. Cheltenham/Northhampton, MA, Edward Bowman, D.M.J.S., Williamson, G.J., Abatzoglou, J.T., Kolden, C.A., Cochrane, M.A., Elgar, pp. 837–852. Smith, A.M.S., 2017. Human exposure and sensitivity to globally extreme wildfire Jolly, W.M., Cochrane, M.A., Freeborn, P.H., Holden, Z.A., Brown, T.J., Williamson, G.J., events. Nat Ecol Evol 1, 1–6. https://doi.org/10.1038/s41559-016-0058. Bowman, D.M.J.S., 2015. Climate-induced variations in global wildfire danger from Brans, J.P., Vinke, Ph, Mareschal, B., 1986. How to select and how to rank projects: the 1979 to 2013. Nat. Commun. 6, 7537. https://doi.org/10.1038/ncomms8537. Promethee method. Eur. J. Oper. Res. 24 (2), 228–238. https://doi.org/10.1016/ Jump, A.S., Hunt, J.M., Peñuelas, J., 2007. Climate relationships of growth and 0377-2217(86)90044-5. establishment across the altitudinal range of Fagus sylvatica in the Montseny Brown, G., 2004. Mapping spatial attributes in survey research for natural resource Mountains, northeast Spain. Ecoscience 14 (4), 507–518. https://doi.org/10.2980/ management: methods and applications. Soc. Nat. Resour. 18 (1), 17–39. https:// 1195-6860(2007)14[507:CROGAE]2.0.CO;2. doi.org/10.1080/08941920590881853. Kovacic, Giampietro, 2015. Beyond “beyond GDP indicators:” the need for reflexivity in Brown, G., Reed, P., 2000. Validation of a forest values typology for use in national forest science for governance. Ecol. Complex. 21, 53–61. https://doi.org/10.1016/j. planning. Forest Science 46 (2), 240–247. https://doi.org/10.1093/forestscience/ ecocom.2014.11.007. 46.2.240. Lang, D.J., Wiek, A., Bergmann, M., et al., 2012. Transdisciplinary research in Brown, G., Weber, D., 2012. Measuring change in place values using public participation sustainability science: practice, principles, and challenges. Sustain. Sci. 7, 25–43. GIS (PPGIS). Applied Geography 34, 316–324. https://doi.org/10.1016/j. https://doi.org/10.1007/s11625-011-0149-x. apgeog.2011.12.007. Mancini, L.D., Corona, P., Salvati, L., 2018. Ranking the importance of Wildfires’ human Butler, W.H., Goldstein, B.E., 2010. The US Fire Learning Network: springing a rigidity drivers through a multi-model regression approach. Environ. Impact Assess. Rev. 72, trap through multiscalar collaborative networks. Ecol. Soc. 15, 21. 177–186. https://doi.org/10.1016/j.eiar.2018.06.003. Campbell, D., 1995. The Campbell Prediction System: A Wild Land Fire Prediction and McIntyre, N., Yuan, M., Payne, R.J., Moore, J., 2004. Development of a values-based Communication System. Available online: http://cps.emxsys.com. approach to managing recreation on Canadian crown lands. In: 2nd International Castellnou, M., Pagé, J., Miralles, M., Piqué, M., 2009. Tipificación de los incendios Conference on Monitoring and Management of Visitor Flows in Recreational and forestales de Cataluña. Elaboración del mapa de incendios de diseño como Protected Areas. June 16–20, 2004, Rovaniemi, Finland. Available online: htt herramienta para la gestión forestal. 5o Congreso Forestal Español: Montes y p://www.metla.fi/julkaisut/workingpapers/2004/mwp002-14.pdf. sociedad: saber qué hacer. àvila. Sociedad Española de Ciencias Forestales: Junta de Montseny Natural Park. Public use plan. https://parcs.diba. Castilla y León. Available online: https://www.recercat.cat/bitstream/handle/2072 cat/es/web/montseny/pla-d-us-public#a. /281988/2009_Castellnou_CongrAvila.pdf?sequence=1. Morehouse, B.J., O’Brien, S., Christopherson, G., Johnson, P., 2010. Integrating values Castellnou, M., Prat-Guitart, N., Arilla, E., Larrañaga, A., Nebot, E., Castellarnau, X., and risk perceptions into a decision support system. Int. J. Wildland Fire 19, Vendrell, J., Pallàs, J., Herrera, J., Monturiol, M., Cespedes, J., Pagès, J., 123–136. https://doi.org/10.1071/WF08064. Gallardo, C., Miralles, M., 2019. Empowering strategic decision-making for wildfire Moritz, M.A., Parisien, M.-A., Batllori, E., Krawchuk, M.A., Van Dorn, J., Ganz, D.J., management: avoiding the fear trap and creating a resilient landscape. Fire Ecol 15, Hayhoe, K., 2012. Climate change and disruptions to global fire activity. Ecosphere 31. https://doi.org/10.1186/s42408-019-0048-6. 3, 1–22. https://doi.org/10.1890/ES11-00345.1. Chapin, F.S., Trainor, S.F., Huntington, O., Lovecraft, A.L., Zavaleta, E., Natcher, D.C., Moser, S.C., 2016. Can science on transformation transform science? Lessons from co- McGuire, A.D., Nelson, J.L., Ray, L., Calef, M., Fresco, N., Huntington, H., Rupp, T.S., design. Curr. Opin. Environ. Sustain. 20, 106–115. https://doi.org/10.1016/j. DeWilde, L., Naylor, R.L., 2008. Increasing wildfire in Alaska’s boreal forest: cosust.2016.10.007. pathways to potential solutions of a wicked problem. Bioscience 58, 531–540. Munda, G., 2004. Social Multi-Criteria Evaluation (SMCE): methodological foundations https://doi.org/10.1641/B580609. and operational consequences. Eur. J. Oper. Res. 158 (3), 662–677. https://doi.org/ Costa, P., Castellnou, M., Larrañaga, A., Miralles, M., Kraus, D., 2011. Prevention of 10.1016/S0377-2217(03)00369-2. Large Wildfires Using the Fire Types Concept. Departament de Interior, Generalitat Munda, G., 2008. Social Multi-Criteria Evaluation for Sustainable Economy. Springer, de Catalunya. GRAF Technical Unit. Available online: https://lessonsonfire.eu/en/ New York. document/guide-prevention-large-wildfires-using-fire-types-concept. Oliveira, S., Zêzere, J.L., 2020. Assessing the biophysical and social drivers of burned Duane, A., Brotons, L., 2018. Synoptic weather conditions and changing fire regimes in a area distribution at the local scale. J. Environ. Manag. 264, 110449 https://doi.org/ Mediterranean environment. Agric. For. Meteorol. 253, 190–202. https://doi.org/ 10.1016/j.jenvman.2020.110449. 10.1016/j.agrformet.2018.02.014. Otero, I., 2022. Social-ecological transformation to coexist with wildfire: reflecting on 18 Everett, Y., Fuller, M., 2011. Fire safe councils in the interface. Soc. Nat. Resour. 24, years of participatory wildfire governance. In: Ruiz-Mallén, I., March, H., 319–333. https://doi.org/10.1080/08941920903313835. Satorras, M. (Eds.), Urban Resilience to the Climate Emergency: Unravelling the Fagerholm, N., Käyhkö, N., 2009. Participatory mapping and geographical patterns of Transformative Potential of Institutional and Grassroots Initiatives. Springer. the social landscape values of rural communities in Zanzibar, Tanzania. Fennia-Int. Otero, I., 2011. Matadepera, Barcelona Metropolitan Region, Spain: Peasant Memory for J. Geogr. 187 (1), 43–60. Available online: https://fennia.journal.fi/article/view/ Wildfire Prevention: Recovery of Social-Ecological Memory for Building Resilience 3703. to Wildfires. Report for the United Cities and Local Governments Committee on Fischer, A., Jasny, L., 2017. Capacity to adapt to environmental change: evidence from a Social Inclusion, Participatory Democracy and Human Rights. Inclusive Cities network of organizations concerned with increasing wildfire risk. Ecol. Soc. 22 Observatory. Available online: https://serval.unil.ch/fr/notice/serval:BIB_B75 https://doi.org/10.5751/ES-08867-220123. 0744BDD86. Funtowicz, S.O., Ravetz, J.R., 1993. Science for the post-normal age. Futures 25 (7), Otero, I., Arilla, E., 2015. Prevenció d’incendis forestals i nou model territorial. Sitja del 739–755. https://doi.org/10.1016/0016-3287(93)90022-L. Llop 38, 22–23. Gamboa, G., 2006. Social multi-criteria evaluation of different development scenarios of Otero, I., Nielsen, J.Ø., 2017. Coexisting with wildfire? Achievements and challenges for the Aysén region, Chile. Ecol. Econ. 59 (1), 157–170. https://doi.org/10.1016/j. a radical social-ecological transformation in Catalonia (Spain). Geoforum 85, ecolecon.2005.10.014. 234–246. https://doi.org/10.1016/j.geoforum.2017.07.020. Garcia-Martin, M., Fagerholm, N., Bieling, C., et al., 2017. Participatory mapping of Otero, I., Castellnou, M., González, I., Arilla, E., Castell, L., Castellví, J., Sánchez, F., landscape values in a Pan-European perspective. Landsc. Ecol. 32, 2133–2150. Nielsen, J.Ø., 2018. Democratizing wildfire strategies. Do you realize what it means? https://doi.org/10.1007/s10980-017-0531-x. Insights from a participatory process in the Montseny region (Catalonia, Spain). PLoS Garmendia, E., Stagl, S., 2010. Public participation for sustainability and social learning: One 13, e0204806. https://doi.org/10.1371/journal.pone.0204806. concepts and lessons from three case studies in Europe. Ecol. Econ. 69 (8), Otero, I., Gamboa, G., Bueno, C., Canaleta, G., Tolosa, G., Ballart, H., Camprubí, L., 1712–1722. https://doi.org/10.1016/j.ecolecon.2010.03.027. Vilalta, O., Arilla, E., Castellnou, M., 2020. Democratitzar la presa de decisions sobre Garmendia, E., Gamboa, G., Franco, J., Garmendia, J.M., Liria, P., Olazabal, M., 2010. els incendis forestals i adaptar-se al canvi climàtic. Un mètode aplicat a la reserva de Social multi-criteria evaluation as a decision support tool for integrated coastal zone la biosfera del Montseny. Quaderns Agraris 7–42. https://doi.org/10.2436/ management. Ocean Coast Manag. 53 (7), 385–403. https://doi.org/10.1016/j. 20.1503.01.109. ocecoaman.2010.05.001. Penning-Rowsell, E.C., 1981. Fluctuating fortunes in gauging landscape value. Progress Gazzard, R., McMorrow, J., Aylen, J., 2016. Wildfire policy and management in England: in Human Geography 5 (1), 25–41. https://doi.org/10.1177/030913258100500102. an evolving response from Fire and Rescue Services, forestry and cross-sector groups. Peñuelas, J., Boada, M., 2003. A global change-induced biome shift in the Montseny Phil. Trans. R. Soc. B 371, 20150341. https://doi.org/10.1098/rstb.2015.0341. mountains (NE Spain). Global Change Biol. 9 (2), 131–140. https://doi.org/ Generalitat de Catalunya, 2018. Zones homogènies de règim d’incendis forestals. 10.1046/j.1365-2486.2003.00566.x. Departament d’Interior, Generalitat de Catalunya. Available online: https://interior. Peñuelas, J., Ogaya, R., Boada, M., Jump, A.S., 2007. Migration, invasion and decline: gencat.cat/ca/arees_dactuacio/bombers/foc-forestal/incendis_forestals/zones-h changes in recruitment and forest structure in a warming-linked shift of European omogenies-de-regim-dincendis-forestals/. beech forest in Catalonia (NE Spain). Ecography 30, 829–837. https://doi.org/ Generalitat de Catalunya, 2020. Mapa de Protecció Civil de Catalunya (Civil protection 10.1111/j.2007.0906-7590.05247.x. map of Catalonia). Generalitat de Catalunya. Available online: https://pcivil.icgc. Piqué, M., Valor, T., Castellnou, M., Pagés, J., Larrañaga, A., Miralles, M., Cervera, T., cat/pcivil/v2/index.html. 2011. In: Integració del risc de grans incendis forestals (GIF) en la gestió forestal: González-Hidalgo, M., Otero, I., Kallis, G., 2014. Seeing beyond the smoke: the political Incendis tipus i vulnerabilitat de les estructures forestals al foc de capçades. Center of ecology of fire in Horta de Sant Joan (Catalonia). Environ. Plann. 46, 1014–1031. Forest Ownership. Department of Agriculture, Livestock, Fishing and Food. https://doi.org/10.1068/a45600. Generalitat de Catalunya.. Available online: http://cpf.gencat.cat/web/.content/ Gonzalez-Olabarria, J.R., Reynolds, K.M., Larrañaga, A., Garcia-Gonzalo, J., Busquets, E., or_organismes/or04_centre_propietat_forestal/06-Publicacions/publicacions_tec Pique, M., 2019. Strategic and tactical planning to improve suppression efforts 11 G. Gamboa et al. Journal of Environmental Management 327 (2023) 116762 niques/colleccions/orgest/parametres_i_indicadors_de_suport_a_la_gfs/orgest_inte San Miguel-Ayanz, J., Durrant, T., Boca, R., Libertà, G., Branco, A., de Rigo, D., graci__del_risc_de_gif_en_la_gf/docs/orgest_incendis_cpf_pdf_reduida.pdf Ferrari, D., Maianti, P., Artés Vivancos, T., Costa, H., Lana, F., Löffler, P., Nuijten, D., Plana, E., Martín, D., Font, M., Serra, M., Molina, D., 2015. Social Factor and Territorial Ahlgren, A.C., Leray, T., 2018. Wildfires in Europe, Middle East and North Africa Dimension of Wildfire Risk Management: Managing Societal Involvement and Cross- 2017. (No. EUR 29318 EN), JRC Technical Reports. Publications Office of the Sectoral Planning. FIREfficient Project. Available online: http://firefficient.ctfc.cat European Union. /wp-content/uploads/2014/03/deliverable15_Wildfire-social-assesment-an Sánchez, S., 2010. Análisi socioecolÒgica a la vall de Santa Fe (massís del Montseny): La d-governance.pdf. transformació del paisatge a través de la història ambiental. Doctoral Thesis. Prior, T., Eriksen, C., 2013. Wildfire preparedness, community cohesion and Universitat Autònoma de Barcelona. Available online: http://www.tdx.cat/handle/1 social–ecological systems. Global Environ. Change 23, 1575–1586. https://doi.org/ 0803/3717. 10.1016/j.gloenvcha.2013.09.016. Stephenson, J., 2008. The Cultural Values Model: an integrated approach to values in Proctor, W., Drechsler, M., 2006. Deliberative multi-criteria evaluation. Environ. Plann. landscapes. Landsc. Urban Plann. 84 (2), 127–139. https://doi.org/10.1016/j. C Govern. Pol. 24, 169–190. https://doi.org/10.1068/c22s. landurbplan.2007.07.003. Rawluk, A., Ford, R.M., Neolaka, F.L., Williams, K.J., 2017. Public values for integration Termorshuizen, J.W., Opdam, P., 2009. Landscape services as a bridge between in natural disaster management and planning: a case study from Victoria, Australia. landscape ecology and sustainable development. Landsc. Ecol. 24, 1037–1052. J. Environ. Manag. 185, 11–20. https://doi.org/10.1016/j.jenvman.2016.10.052. https://doi.org/10.1007/s10980-008-9314-8. Reed, P., Brown, G., 2003. Values suitability analysis: a methodology for identifying and Turco, M., Rosa-Cánovas, J.J., Bedia, J., Jerez, S., Montávez, J.P., Llasat, M.C., integrating public perceptions of ecosystem values in forest planning. J. Environ. Provenzale, A., 2018. Exacerbated fires in Mediterranean Europe due to Plann. Manag. 46 (5), 643–658. https://doi.org/10.1080/0964056032000138418. anthropogenic warming projected with non-stationary climate-fire models. Nat. Regos, A., Aquilué, N., López, I., et al., 2016. Synergies between forest biomass Commun. 9, 3821. https://doi.org/10.1038/s41467-018-06358-z. extraction for bioenergy and fire suppression in mediterranean ecosystems: insights Visseren-Hamakers, I.J., Razzaque, J., McElwee, P., Turnhout, E., Kelemen, E., Rusch, G. from a storyline-and-simulation approach. Ecosystems 19, 786–802. https://doi.org/ M., Fernández-Llamazares, Á., Chan, I., Lim, M., Islar, M., Gautam, A.P., 10.1007/s10021-016-9968-z. Williams, M., Mungatana, E., Karim, M.S., Muradian, R., Gerber, L.R., Lui, G., Liu, J., Roigé, X., Estrada, F., 2008. El mas al Montseny: La memòria oral. Temes d’Etnologia de Spangenberg, J.H., Zaleski, D., 2021. Transformative governance of biodiversity: Catalunya 16. Departament de Cultura i Mitjans de Comunicació, Barcelona. insights for sustainable development. Curr. Opin. Environ. Sustain. 53, 20–28. Generalitat de Catalunya. https://doi.org/10.1016/j.cosust.2021.06.002. San Miguel-Ayanz, J., Moreno, J.M., Camia, A., 2013. Analysis of large fires in European Mediterranean landscapes: lessons learned and perspectives. For. Ecol. Manag. 294, 11–22. https://doi.org/10.1016/j.foreco.2012.10.050. 12