From famine foods to delicatessen: Interpreting trends in the use of wild edible plants through cultural ecosystem services

UNCORRECTED PROOF 1Surveys 2Q1 From famine foods to delicatessen: Interpreting trends in the use of wild 3edible plants through cultural ecosystem services 4Q2 Victoria Reyes-García a,b, ⁎, Gorka Menendez-Baceta c , Laura Aceituno-Mata c , Rufino Acosta-Naranjo d , 5Laura Calvet-Mir b,e , Pablo Domínguez b , Teresa Garnatje f , Erik Gomez-Bagetthun b,g , 6Manuel Molina-Bustamante c , Marta Molina d , Ramón Rodríguez-Franco d , Ginesta Serrasolses h , 7Joan Vallès h , Manuel Pardo-de-Santayana c 8 a Institució Catalana de Recerca i Estudis Avançats (ICREA), Barcelona, SpainQ3 9 b Institut de Ciència i Tecnologia Ambientals, Universitat Autònoma de Barcelona, 08193 Bellaterra, Barcelona, Spain 10 c Departamento de Biología (Botánica), Universidad Autónoma de Madrid, C/ Darwin 2, Campus de Cantoblanco, 28049 Madrid, Spain 11 d Departamento de Antropología Social, Universidad de Sevilla, Doña Marí de Padilla s/n 41004, Sevilla, Spain 12 e Internet Interdisciplinary Institute (IN3), Universitat Oberta de Catalunya, Spain 13 f Institut Botànic de Barcelona (IBB-CSIC-ICUB), Passeig del Migdia s.n., Parc de Montjuïc, 08038 Barcelona, Spain 14 g Norwegian Institute for Nature Research (NINA), Gaustadalléen 21, 0349 Oslo, NorwayQ4 15 h Laboratori de Botànica, Facultat de Farmàcia-Unitat associada CSIC, Universitat de Barcelona, Av. Joan XXIII, s.n., 08028 Barcelona, Spain 16 a b s t r a c t1 7 a r t i c l e i n f o 18 Article history: 19 Received 2 March 2015 20 Received in revised form 5 October 2015 21 Accepted 1 November 2015 22 Available online xxxx 23 Keywords: 24 Biocultural diversity 25 Ecosystem services 26 Ethnobotany 27 Local ecological knowledge 28 Natural resources 29 Spain 30The Millennium Ecosystem Assessment found a general decline in the consumption and gathering of wild edible 31plants, but some studies also observe a localized increase. Using information from interviews (n = 1133) in seven 32sites in the Iberian Peninsula and one in the Balearic Islands, we 1) identify current trendsin the consumption and 33gathering of wild edible plants (n = 56 plant-uses) and 2) analyze how cultural ecosystem services relate to such 34trends. Our data show a generalized decrease in the consumption and gathering of wild edible plants, although 35the trend changes significantly across plant-uses. Specifically, we found that –despite the overall decreasing 36trend–uses of wild edible plants that simultaneously relate to foods with high cultural appreciation and the 37recreational function of gathering remain popular. Our results signal that cultural services and values associated 38to the gathering and consumption of some wild edible plants are important factors explaining divergent trends 39across plant species. This finding reinforces the notion that cultural ecosystem services are deeply intertwined 40with other categories of services which can combine in complex, non-linear ways producing a variety of 41interdependent benefits. 42 © 2015 Published by Elsevier B.V. 4344 45 46 47 1. Introduction 48 Wild edible plants are defined as plant species collected in the wild 49 to be consumed as food or drink. Wild edible plants have been an inte- 50 gral part of human diet throughout history and around the world 51 (Behre, 2008; Hummer, 2013; Leonti et al., 2006; Schulp et al., 2014). 52 Although the Millennium Ecosystem Assessment found a general 53 decline in their consumption and gathering (MA, 2005), wild edible 54 plants continue to be consumed in many parts of the world, not only 55 in subsistence-oriented economies but often also in rural and even 56 urban areas in developed countries (Bharucha and Pretty, 2010; 57 Certomà and Tornaghi, 2015; Schulp et al., 2014). Because of their 58importance to income (Angelsen et al., 2014; Łukasz et al., 2013; 59Shumsky et al., 2014), nutrition (Mavengahama et al., 2013; Toledo 60et al., 2003), and food security (Bharucha and Pretty, 2010; Nolan and 61Pieroni, 2014; Redzic, 2010; Vinceti et al., 2013), wild edible plants are 62included in all major ecosystem service classifications as a type of 63provisioning service (see e.g., de Groot et al., 2002; Haines-Young and 64Potschin, 2013; MA, 2005; TEEB, 2010). 65Research suggests that, while wild edible plants were an important 66provisioning service in Europe until the 20th century (Kangas and 67Markkanen, 2001; Łukasz et al., 2013), in recent decades their gathering 68and consumption have decreased both in terms of quantity and diversi- 69ty (Bharucha and Pretty, 2010; MA, 2005; Tardío et al., 2005). The 70decrease in this provisioning service is concomitant with urbanization 71and associated rural exodus, modernization of lifestyles, industrializa- 72tion of food production, or loss of natural habitats, among others 73(Abbet et al., 2014; Bharucha and Pretty, 2010; Kalle and Soukand, 742013; Łukasz et al., 2013; Turner and Turner, 2008). Ecological Economics xxx (2015) xxx–xxx ⁎Corresponding author at: Environmental Sciences and Technology Institute (ICTAUAB), ICTA-ICP, Edifici Z Carrer de les Columnes, Universitat Autònoma de Barcelona, E- 08193, Bellaterra (Cerdanyola del Vallès-Barcelona), Spain Q5 . URL: http://icta.uab.cat/Etnoecologia/ (V. Reyes-García). ECOLEC-05166; No of Pages 9 http://dx.doi.org/10.1016/j.ecolecon.2015.11.003 0921-8009/© 2015 Published by Elsevier B.V. Contents lists available at ScienceDirect Ecological Economics jour na l ho me pa ge : www.elsevie r. co m/ lo ca te/ecole con Please cite this article as: Reyes-García, V., et al., From famine foods to delicatessen: Interpreting trends in the use of wild edible plants through cultural ecosystem services, Ecol. Econ. (2015), http://dx.doi.org/10.1016/j.ecolecon.2015.11.003 UNCORRECTED PROOF 75 Decreasing consumption and gathering trends, however, seem not 76 to be affecting all areas and all wild edible plants with the same intensi- 77 ty. For example, a recent research in Cantabria, north of the Iberian 78 Peninsula, found that local people assign a high value to wild fruits, 79 but not so much to wild vegetables, and that the consumption of some 80 wild edibles (i.e., the fruits of Quercus robux and Quercus ilex) is cultur- 81 ally stigmatized (Menendez-Baceta et al., 2012). The opposite trend is 82 reported for other wild species, like the sprouts of Asparagus acutifolius 83 which are increasingly harvested to be sold (Molina et al., 2012), or 84 other wild edible plants that have become local delicatessens and 85 markers of cultural identity (see e.g. Aceituno-Mata, 2010; Kalle and 86 Soukand, 2013). Some researchers have also highlighted the impor- 87 tance of the gathering of wild edible plants as a recreational activity 88 (Kangas and Markkanen, 2001; Schulp et al., 2014). In other words, 89 explanations on divergent trends in the use of wild edible plants in 90 Europe seem to revolve around the cultural services they provide, 91 where cultural services are defined as “non-material benefits people 92 obtain from ecosystems through spiritual enrichment, cognitive devel- 93 opment, reflection, recreation, and aesthetic experience”(MA, 2005: 94 894) or as “ecosystems' contributions to the non-material benefits 95 (e.g., capabilities and experiences) that arise from human–ecosystem 96 relationships”(Chan et al., 2012: 9). 97 The argumentative line of this paper is that cultural ecosystem ser- 98 vices and values associated to the consumption and gathering of wild 99 edible plants might help interpreting divergent trends in the use of 100 these plants. Using information from seven sites in the Iberian Peninsula 101 and one in the Balearic Islands, we first identify current trends in the 102 consumption and gathering of wild edible plants and then analyze 103 how different cultural ecosystem services relate to such trends. Our 104 expectation is that the consumption and gathering of species associated 105 to cultural services and values would be more prevalent than the 106 consumption and gathering of species lacking such association. 107 2. Methods 108 Data were sampled in seven sites of the Iberian Peninsula and one of 109 the Balearic Islands, a region with a long tradition in the consumption of 110 wild edible species (Leonti et al., 2006; Tardío et al., 2006). Sampling 111 was conducted in two phases. In the first phase, we compiled an inven- 112 tory of wild edible plants consumed in each area. In the second phase, 113 we conducted a systematic survey on past and present consumption 114 and gathering of selected species. For the purpose of in this work, we de- 115 fine wild edible plants as plant species that are collected in the wild to 116 be consumed as food or drink. Our definition includes native species 117 growing in their natural habitat as well as naturalized species 118 (i.e., species planted in the past, no longer managed but still harvested). 119 2.1. Site Selection 120 The site selection was based on several criteria. First, we focused on 121 areas where local people traditionally gathered wild edible plants. Sec- 122 ond, we aimed to cover some of the ecological and cultural diversity of 123 Spain, although we are aware of the impossibility of being exhaustive 124 in such criterion. Third, we selected sites where wild plants could be col- 125 lected near people's homes, e.g. from crop fields, wild areas or hedge- 126 rows (González et al., 2011; Stryamets et al., 2012). Fourth, in none of 127 the sites legal restrictions affected the gathering of the selected species. 128 About 50% of one of the sites, Doñana, is protected (Gómez-Baggethun 129 et al., 2010), but survey data were collected in villages with non- 130 protected surroundings. Last, we selected sites where members of the 131 team had either conducted previous ethnobotanical work or had 132 contacts that facilitated the realization of such work. 133 We worked in a total of eight sites; six in mountain regions and two 134 other. The six sites in mountain regions include: Alta Vall del Ter, a valley 135 on the southern flanks of the eastern Pyrenees mountain range; Alt 136 Empordà, the easternmost region of the north of Catalonia, where the 137Pyrenees descend through a plain to meet the Mediterranean Sea; 138Gorbeialdea, a mountainous region of southern Biscay in the Basque 139Country; Sierra Morena Extremeña, an area in the low and middle height 140mountain regions of southern Extremadura; Sierra Norte de Madrid, in 141the Central range that crosses the north of Madrid province, 70 km 142north of Madrid city; and east-central Asturias, an Atlantic valley on 143the northern slopes of the Cantabrian range. One site was conducted 144in a plain territory: Doñana, marshlands, dunes, and pine forest area in 145south-western Andalusia touching the Atlantic Ocean. Finally, one site 146was settled in eastern Mallorca, the largest island in the Mediterranean 147Balearic archipelago, east of the Iberian Peninsula. With the exception 148of east-central Asturias and Gorbeialdea, which belong to the 149Euro-Siberian region, all sites are placed in the Mediterranean biogeo- 150graphical basin (Fig. 1). 1512.2. Phase 1: Inventory 152In each study area, we started by compiling an inventory of wild ed- 153ible plants. For Alta Vall del Ter (Rigat et al., 2009), Alt Empordà (Parada 154et al., 2011), Gorbeialdea (Menendez-Baceta et al., 2012), Sierra Norte de 155Madrid (Aceituno-Mata, 2010), and east-central Asturias (San Miguel 156López, 2004) we used data from previous fieldwork. For Sierra Morena 157Extremeña,Doñana, and eastern Mallorca, we conducted fieldwork to 158elaborate the inventory and interviewed people locally recognized as 159knowledgeable about wild edible plants (Davis and Wagner, 2003). 160We asked them to list all the wild edible plants in the area and, for 161each plant listed, to provide all relevant information regarding its gath- 162ering and consumption: past and present use, mode of consumption, 163processing techniques, symbolic attachment, and the like. 164Based on Tardío et al. (2006), information regarding edible uses of 165wild plants was categorized as 1) fruit (when the fresh or dry fruit is 166eaten, raw or cooked), 2) vegetable (when any of the vegetative parts 167is consumed, raw or cooked), 3) beverage (when any part of the species 168is used to prepare liquor or infusions), and 4) seasoning (when any part 169of the plant is used for food seasoning). 1702.3. Phase 2: Survey 171Between 2012 and 2013, we conducted a survey. As many wild edi- 172ble plants have more than one edible use (for example, the fruits of 173Rubus ulmifolius are consumed raw, but they are also used to elaborate 174liqueurs), we selected only the most popular use. Thus, in our survey 175we only asked for the most popular use of each wild edible plant 176(plant-use). 1772.3.1. Plant-use Selection 178Since we worked in eight areas with marked cultural and ecological 179differences, we could not use the same survey in all the areas, but rather 180performed site-specific selections. To ensure comparability, we used the 181same criteria to select plant-uses in each site. To narrow the selection, 182we first identified species with a prominent edible use (versus other 183uses, such as medicinal or ornamental) and not locally gathered for 184large-scale commercialization, but rather mostly for self-consumption 185or exchange. In each site-specific survey, we included the four catego- 186ries of use (fruit, vegetable, beverage, and seasoning). To keep the 187length of the survey at around 40 min/informant, we limited the survey 188to seven plant-uses, so –in total–we asked about 56 plant-uses (7 plant- 189uses ∗8 areas = 56; considering the same plant-use in different areas as 190different observations). The final list of plant-uses is given in Table 1, 191where we also report the scientific name of the species with taxa au- 192thorities, growth form, and voucher number. 1932.3.2. Sample Selection 194We collected survey data from 1133 informants (between 100 and 195180 per site) mostly recruited in villages or small towns. After ap- 196proaching a person, we first explained our goals and requested consent 2V. Reyes-García et al. / Ecological Economics xxx (2015) xxx–xxx Please cite this article as: Reyes-García, V., et al., From famine foods to delicatessen: Interpreting trends in the use of wild edible plants through cultural ecosystem services, Ecol. Econ. (2015), http://dx.doi.org/10.1016/j.ecolecon.2015.11.003 UNCORRECTED PROOF 197 to ask some questions. A total of 310 people (21% of the people 198 approached) refused to participate: 50% because of lack of time, 25% be- 199 cause of lack of interest, and 14% because of lack of knowledge. The re- 200 maining 11% gave other reasons or simply did not give any clear answer. 201 In each site, the sample was stratified according to criteria that might af- 202 fect use and consumption of wild edible plants. Specifically, we aimed at 203 having 1) 50% men and 50% women (Grasser et al., 2012; Kangas and 204 Markkanen, 2001), 2) 33% of informants in the each of the three age cat- 205 egories selected (b40; 41–60; and N61) (Cornara et al., 2009), and 3) be- 206 tween 15 and 30% of the population in the agricultural sector, depending 207 on the site (Hadjichambis et al., 2008; Idolo et al., 2010) (Table 2). 208 2.3.3. Survey Design 209 Our survey included three sections. In the first section, we asked 210 socio-demographic data (age, sex, occupation). In the second section, 211 we asked about past and present consumption and gathering of the 212 seven selected plant-uses. We started by showing the informant a visual 213 stimulus (a picture, a voucher, or the fresh plant) where the edible part 214 could be easily recognized. We then asked for the local name of the 215 plant. If the person did not recognize the plant, we provided him/her 216 with the local name and asked again if the person knew it. If the infor- 217 mant did not know the species, we moved to the following visual stim- 218 ulus in the survey. If the person recognized the species, we asked about 219 its uses; again, when informants did not report the selected use, we 220 moved on to the next plant. When informants listed the wild edible 221 use, we asked about present (last 12 months) and past consumption 222 and about the main way of obtaining the species (i.e., gathering, gift, 223 or the market). 224In the third section of the survey, we asked informants to tell us their 225level of agreement on a set of statements related to a selection of pre- 226determined cultural services and values associated with such plant-use, 227including heritage, place and identity values (e.g., considered a local tradi- 228tion), health values, perceptional benefits (tasteful), and recreational ele- 229ments associated with gathering and preparation (e.g., perceived time 230invested in gathering and preparing it, link to leisure). All statements 231were evaluated in a scale from 1 (do not agree) to 5 (completely agree). 2322.4. Data Analysis 233To assess trends in the consumption of wild edible plants, we aggre- 234gated information by site. We first calculated the proportion of infor- 235mants who recognized each species, irrespectively of whether they 236knew about their uses, and the proportion of informants who men- 237tioned their edible use. Then, we assessed changes between past and 238present consumption, calculating the difference between people who 239reportedly consume the plant now minus the people who reportedly 240consumed it in the past, divided by the total number of people who con- 241sumed the plant in the past. We call this measure consumption index. 242Put it formally Consumptions¼ X iN i¼i1 SpsEat−X iN i¼i1 EatPast X iN i¼i1 EatPast 244244 Fig. 1. Map of the study areas. 3V. Reyes-García et al. / Ecological Economics xxx (2015) xxx–xxx Please cite this article as: Reyes-García, V., et al., From famine foods to delicatessen: Interpreting trends in the use of wild edible plants through cultural ecosystem services, Ecol. Econ. (2015), http://dx.doi.org/10.1016/j.ecolecon.2015.11.003 UNCORRECTED PROOF t1:1Table 1 t1:2Specific uses of wild edible species included in the survey, by study area. t1:3Scientific name (family; growth form) Folk name a Herbarium voucher Part used Plant-use included in survey (brief explanation of elaboration) t1:4Alt Empordà t1:5Arbutus unedo L. (Ericaceae; tree) Cirera d'arboç BCN29836 Fruit Fruits (eaten raw) t1:6Cynara cardunculus L. (Asteraceae; perennial herb) Preó BCN29860 Inflorescence Seasoning (to curdle milk) t1:7Foeniculum vulgare Mill. (Apiaceae; perennial herb) Fonoll BCN29867 Young shoot Vegetable (snack) t1:8Juglans regia L. (Juglandaceae; tree) Nous BCN29877 Unripe fruit Beverage (to make alcoholic spirits) t1:9Origanum vulgare L. (Lamiaceae; subshrub) Orenga BCN29742 Flowering aerial part Seasoning t1:10 Reichardia picroides (L.) Roth (Asteraceae; perennial herb) Cosconilla BCN29933 Young leaf Vegetable t1:11 Rubus ulmifolius Schott (Rosaceae; shrub) Móra BCN29938 Fruit Fruits (raw or cooked in marmalade) t1:12 t1:13 Alta Vall del Ter t1:14 Carlina acanthifolia All. subsp. cynara (Pourr. ex Duby) Arcang. t1:15 (Asteraceae; perennial herb) Carlina BCN24738 Inner part of the inflorescence receptacle Vegetable (snack) t1:16 Cynara cardunculus L. (Asteraceae; perennial herb) Flor d'empresorar BCN24759 Inflorescence Seasoning (to curdle milk) t1:17 Fragaria vesca L. (Rosaceae; perennial herb) Maduixa BCN24889 Fruit Fruits t1:18 Juglans regia L. (Juglandaceae; tree) Nous BCN24908 Fruit Beverage (to make alcoholic spirits) t1:19 Origanum vulgare L. (Lamiaceae; subshrub) Orenga BCN24939 Flowering aerial part Seasoning t1:20 Rubus ulmifolius Schott (Rosaceae; shrub) Móra BCN24978 Fruit Fruits (raw or cooked in marmalade) t1:21 Taraxacum dissectum (Ledeb.) Ledeb. (Asteraceae; perennial herb) Xicoina BCN25016 Young leaf Vegetable t1:22 t1:23 Doñana t1:24 Asparagus acutifolius L. (Asparagaceae; shrub) Espárrago, espárrago triguero BCN29976 Young shoot Vegetable t1:25 Chamaerops humilis L. (Arecaceae; palm shrub/tree) Palmito BCN23832 Young shoot Vegetable t1:26 Glycycrhiza glabra L. (Fabaceae; perennial herb) Palodú, palo arazú BCN47726 Rhizome Vegetable (chewed as snack) t1:27 Mentha pulegium L. (Lamiaceae; shrub) Poleo BCN28895 Flower Beverage t1:28 Rubus ulmifolius Schott (Rosaceae; shrub) Zarzamora, mora MA729323 Fruit Fruit t1:29 Scolymus hispanicus L. (Asteraceae; perennial herb) Tagarnina MA852821 Aerial part Vegetable t1:30 Thymbra capitata (L.) Cav. (Lamiaceae; subshrub) Tomillo BCN20616 Flowering shoot Seasoning t1:31 t1:32 Eastern Mallorca t1:33 Chamaerops humilis L. (Arecaceae; palm shrub/tree) Garballó BCN 23832 Apical shoot Vegetable t1:34 Cichorium intybus L. (Asteraceae; perennial herb) Cama-roja BCN 29660 Young leaf Vegetable t1:35 Crithmum maritimum L. (Apiaceae; perennial herb) Fonoll marí BCN104272 Leaf Vegetable t1:36 Cynara cardunculus L. (Asteraceae; perennial herb) Card de formatjar BCN 29860 Inflorescence Seasoning (to curdle milk) t1:37 Foeniculum vulgare Mill. (Apiaceae; perennial herb) Fonoll BCN 95541 Shoot Beverage t1:38 Quercus ilex L. (Fagaceae; tree) Aglà BCN103497 Fruit Fruit t1:39 Rubus ulmifolius Schott (Rosaceae; shrub) Móra d'abatzer BCN 29938 Fruit Fruit (raw or cooked in marmalade) t1:40 t1:41 East-central Asturias t1:42 Crataegus monogyna Jacq. (Rosaceae; shrub) Maluca, espinera ESM141 Fruit Fruit (as snack) t1:43 Fragaria vesca L. (Rosaceae; perennial herb) Meruétanu, abeyuétanos, freses silvestres ESM171 Fruit Fruit t1:44 Mespilus germanica L. (Rosaceae; tree) Carápanu MP920 Fruit Fruit (as snack) t1:45 Prunus spinosa L. (Rosaceae; shrub) Andrín ESM111 Fruit Beverage t1:46 Rubus ulmifolius Schott (Rosaceae; shrub) Mora ESM304 Fruit Fruit (raw or cooked in marmalade) t1:47 Rumex acetosa L. (Polygonaceae; perennial herb) Agrieta, chupes ESM126 Young shoot and basal leaf Vegetable (snack) t1:48 Vaccinium myrtillus L. (Ericaceae; shrub) Arándanu ESM93 Fruit Fruit t1:49 t1:50 Gorbeialdea t1:51 Fagus sylvatica L. (Fagaceae; tree) Pago GM776 Young leaf Vegetable (chewed as a snack) t1:52 Laurus nobilis L. (Lauraceae; tree) Ereinotza GM737 Leaf Seasoning t1:53 Prunus spinosa L. (Rosaceae; shrub) Arranokan GM723 Fruit Beverage (to elaborate an alcoholic spirit ‘pacharan’) t1:54 Pyrus cordata Desv. (Rosaceae; tree) Basomakatz GM718 Fruit Fruit (eaten raw as snack) t1:55 Rubus ulmifolius Schott (Rosaceae; shrub) Masusta GM766 Fruit Fruit (raw or cooked in marmalade) t1:56 Rumex acetosa L. (Polygonaceae; perennial herb) Bedar garratza GM668 Young leaf Vegetable (chewed as a snack) t1:57 Urtica dioica L. (Urticaceae; perennial herb) Asun GM719 Aerial part Vegetable (cooked) t1:58 t1:59 Sierra Morena Extremeña t1:60 Asparagus acutifolius L. (Asparagaceae; shrub) Espárrago, espárrago triguero BCN29976 Young shoot Vegetable t1:61 Foeniculum vulgare Mill. (Apiaceae; perennial herb) Hinojo BCN29867 Young shoot Vegetable (snack) t1:62 Helichrysum stoechas (L.) Moench (Asteraceae; perennial herb/subshrub) Manzanilla real o grande BCN29872 Flowering aerial part Beverage t1:63 Rubus ulmifolius Schott (Rosaceae; shrub) Zarzamora, mora MA729323 Fruit Fruit (raw or cooked in marmalade) t1:64 Rumex pulcher L. (Polygonaceae; perennial herb) Romaza, cocina verde BCN26671 Basal leaf Vegetable (snack) t1:65 Scolymus hispanicus L. (Asteraceae; perennial herb) Tagarnina MA852821 Basal leaf pealed Vegetable t1:66 Thymus mastichina (L.) L. (Lamiaceae; subshrub) Tomillo salsero BCN34644 Flowering aerial part Seasoning t1:67 t1:68 Sierra Norte de Madrid t1:69 Armeria arenaria subsp. segoviensis (Gand. ex Bernis) Nieto Fel. t1:70 (Plumbaginaceae; perennial herb) Patas de cigüeña, majuletas, patas de milano MA450678 Peduncle of inflorescence Vegetable (as a snack) t1:71 Crataegus monogyna Jacq. (Rosaceae; shrub) Majoleto, majuelo, espino majulero MA729324 Fruit Fruit (snack) t1:72 Prunus spinosa L. (Rosaceae; shrub) Endrino MA729279 Fruit Beverage t1:73 Rubus ulmifolius Schott (Rosaceae; shrub) Zarza, zarzamora MA729323 Fruit Fruit (raw or cooked in marmalade) 4V. Reyes-García et al. / Ecological Economics xxx (2015) xxx–xxx Please cite this article as: Reyes-García, V., et al., From famine foods to delicatessen: Interpreting trends in the use of wild edible plants through cultural ecosystem services, Ecol. Econ. (2015), http://dx.doi.org/10.1016/j.ecolecon.2015.11.003 UNCORRECTED PROOF where Sp s Eat refers to the plant-use consumption now and 245 Sp s EatPast to the plant-use consumption in the past. A positive number 246 would indicate an increase in the number of consumers over time, a 247 negative number would indicate a decrease, and a number close to 248 zero no changes. The gathering index was constructed in a similar way. 249 We also calculated a market origin index as the difference between 250 a) the number of informants who obtain the plant-use from the market 251 now divided by the total number of informants who consume it now, 252 minus b) the number of informants who obtained the plant-use from 253 the market in the past divided by the total number of people who con- 254 sumed it in the past. High values indicate an increase in the proportion 255 of people depending on the market to obtain the plant-use. MarketOrigins¼ X iN i¼i1 SpsBuy X iN i¼i SpsEat − X iN i¼i1 SpsBuyPast X iN i¼i1 SpsEatPast 257257 To analyze trends while simultaneously considering consumption 258 and gathering, we performed a hierarchical cluster analysis with Ward 259 agglomerative technique (Kaufman and Rousseeuw, 1990). The proce- 260 dure clusters items (plant-uses in our case) according to the calculated 261 distance between pairs of observations regarding some selected criteria 262 (here consumption and gathering indices). Distances between objects 263 are represented in a dendrogram where objects are joined together in 264 a hierarchical fashion from the most similar to the most different 265 regarding the consumption and gathering indices. We interpret the dif- 266 ferent clusters as representing the different trends in consumption 267 and gathering of wild edible plants. 268 In our last step, we explored relations between the clusters and peo- 269 ple's evaluations of the cultural services provided by plant-uses in each 270 cluster. For each plant-use, we first calculated the percentage of people 271 who partially (=4) or totally (=5) agree with each statement in our 272 questionnaire. We then used a Kruskal–Wallis test to examine whether 273 such percentages varied across the different clusters. To detect differ- 274 ences between clusters, we ran multiple comparisons using a post hoc 275 Dunn test (Dunn, 1964). For all the calculations we used the full sample. 276 For the statistical analysis we used STATA 11.1 for Windows (Stata 277 Corporation, Texas, USA). 2783. Results 2793.1. Knowledge, Consumption and Gathering of Wild Edible Plants 280Overall, 50 out of the 56 species in our surveys were recognized by at 281least half of the people interviewed (see Supplementary material). A re- 282markable exception is Rumex acetosa in Gorbeialdea, recognized only by 28328% of the informants. While the recognition of the selected species was 284rather generalized, we found variation between sites, with higher levels 285of recognition in Doñana, Alta Vall del Ter, and Sierra Morena 286Extremeña. Less people identified the selected plants as edible; thus, 287only 40 out of the 56 species in the survey were recognized as edible 288by at least half of the informants. Remarkable cases are Fagus sylvatica, 289Crataegus monogyna and Urtica dioica in Gorbeialdea, which were recog- 290nized by 93%, 87% and 99% of the informants, but only 7%, 13% and 31% 291identified them as edible. 292The analysis of the consumption index (potentially ranging between 2931 and −1) suggests an overwhelming general decrease in the consump- 294tion of wild edible plants. From all the plant-uses in thesurvey, only one, 295the vegetable use of A. acutifolius in Sierra Morena Extremeña, has 296experienced an increase in consumption. The consumption of all the 297other plant-uses in all the other sites has decreased, including the con- 298sumption of the same species in Doñana. Overall, 14 plant-uses had a 299decrease in consumption index higher than 0.75, and 32 had a decrease 300in consumption index higher than 0.50. 301Our analysis further suggests that, from a given plant-use, trends in 302consumption vary from one area to another. Thus, the consumption of 303the fruits of R. ulmifolius, a plant-use included in all the surveys, varies 304from −0.15 in Eastern Mallorca to −0.70 in Doñana. It is also worth no- 305ticing that overall trends are dissimilar between sites. For example, 306while four or five of the seven plant-uses in the surveys in Alt Empordà, 307Alta Vall de Ter, and Doñana had a consumption index N−0.5, the seven 308plant-uses included in the survey in Gorbeialdea and five of the 309plant-uses included in the survey in Sierra Norte de Madrid had a 310consumption index b−0.75. 311The decrease in gathering appears even more pronounced than the 312decrease in consumption. None of the plants in our surveys experienced 313an increase in gathering related to the particular use selected and only 314four had a gathering index N−0.25 (indicating a very low decrease). 315The three plant-uses with values in the gathering index close to zero 316(Origanum vulgare,R. ulmifolius, and A. acutifolius) also have very low 317decrease in consumption. Furthermore, of the 56 plant-uses in our 318survey, 38 (68% of the total) had a gathering index ≤−0.50. 319Despite the general decreasing trend in gathering, we found differ- 320ences between sites. In Gorbeialdea and eastern Mallorca all the plant- 321uses but one have gathering indices ≤− 0.50. Similarly, in east-central 322Asturias, four out of the seven plant-uses included in the survey had a 323gathering index ≤−0.75. 3243.2. Trends in the Use of Wild Edible Species 325Based on the visual inspection of the dendrogram resulting from 326cluster analysis, we classified plant-uses into three clusters. The first 327cluster (Table 3, Group A) is the smallest (n = 11, ≈20% of the total) 328and includes species for which the selected uses have experienced a Table 1 (continued) Scientific name (family; growth form) Folk name a Herbarium voucher Part used Plant-use included in survey (brief explanation of elaboration) t1:74 Sierra Norte de Madrid t1:75 Rumex papillaris Boiss. & Reut. (Polygonaceae; perennial herb) Acedera, azadera MA852820 Basal leaf Vegetable t1:76 Scolymus hispanicus L. (Asteraceae; perennial herb) Cardillo MA852821 Basal leaf Vegetable t1:77 Thymus zygis Loefl. ex L. (Lamiaceae; subshrub) Tomillo salsero, tomillo MA784735 Flowering shoot Seasoning t1:78 a Folk names are in the following languages: Catalan in Alt Empordà, Alta Vall del Ter and Eastern Mallorca, Spanish in Doñana, Sierra Morena Extremeña and Sierra Norte de Madrid, t1:79 Asturian in East-central Asturias and Basque in Gorbeialdea. t2:1Table 2 t2:2Sample description, by study area. t2:3Study area N % women % per age group % agriculture t2:4b40 41–60 N61 t2:5Alt Empordà 101 48 38 27 36 15 t2:6Alta Vall del Ter 100 51 18 36 46 22 t2:7Doñana 150 53 28 35 37 44 t2:8Eastern Mallorca 152 45 38 30 32 6 t2:9East-central Asturias 150 42 7 31 63 33 t2:10 Gorbeialdea 150 49 35 35 30 22 t2:11 Sierra Morena Extremeña 150 48 26 33 41 31 t2:12 Sierra Norte de Madrid 180 52 30 42 28 11 t2:13 Total 1133 48 28 34 38 21 5V. Reyes-García et al. / Ecological Economics xxx (2015) xxx–xxx Please cite this article as: Reyes-García, V., et al., From famine foods to delicatessen: Interpreting trends in the use of wild edible plants through cultural ecosystem services, Ecol. Econ. (2015), http://dx.doi.org/10.1016/j.ecolecon.2015.11.003 UNCORRECTED PROOF 329 small decrease in consumption (average consumption index = −0.12) 330 and a relatively low decrease in gathering (average gathering 331 index = −0.35), at least in relation to the other groups (Fig. 2). Plant- 332 uses in this group include the fruits of Fragaria vesca and R. ulmifolius 333 (one occurrence), the use for seasoning of O. vulgare,Mentha pulegium, 334 Thymbra capitata,Thymus mastichina and Thymus zygis, the use for bev- 335 erages of Juglans regia and the vegetable use of A. acutifolius. Because 336 overall they continue to be widely used plants, we name this group 337 ‘popular’plant-uses. 338 The second cluster (Table 3, Group B) includes 29 plant-uses (≈52% 339 of the total) with intermediate values. In contrast with ‘popular’plant- 340 uses, we found a steeper decrease in the consumption and gathering 341 of plant-uses in this group (−0.52 and −0.58). Plant-uses in this 342 group include the fruits of R. ulmifolius (seven occurrences), Prunus 343 spinosa (two occurrences), F. vesca,Arbutus unedo,Q. ilex, and Mespilus 344 germanica; the vegetable use of Taraxacum dissectum,Chamaerops 345 humilis,Cichorium intybus,Crithmum maritimum,U. dioica,F. sylvatica, 346 and Scolymus hispanicus; the use for seasoning of Laurus nobilis and 347 Foeniculum vulgare; and the use for beverage of J. regia. We call this 348 group ‘gradually abandoned’uses. 349 Finally, the third cluster (Table 3, group C), composed by 16 plant- 350 uses (≈28%), experience the strongest decrease in consumption 351 (−0.79) and gathering (−0.86). Plant-uses in this group are varied 352 and include the fruits of C. monogyna and Pyrus cordata and the use as 353 vegetable of Reichardia picroides. However, many of the plant-uses in 354 this group refer to vegetable uses, mainly consumed as snacks while 355in the field (F. vulgare,Carlina acanthifolia,Vaccinium myrtillus, 356R. acetosa,Armeria arenaria). We call this group ‘mostly abandoned’ 357plant-uses. 3583.3. Cultural Services and Wild Edible Species Use 359We next explore cultural services and values associated with the 360three clusters identified. The percentage of informants who recognized 361species was significantly different among the three clusters, using 362Kruskal–Wallis, χ 2 = 20.8, p b.0001 (Table 3). A post hoc Dunn test 363showed that the percentage of people who recognized species in the 364mostly abandoned cluster (65%) differed significantly (p b.001) from 365those who recognized species in the gradually abandoned (87%) and 366popular (95%) clusters (Fig. 3). Results are similar for the variable that 367capture the percentage of informants recognizing the species in each 368cluster as edible (χ 2 = 18.78, p b.0001 for the Kruskal–Wallis test), 369with statistically significant differences between the cluster of mostly 370abandoned plant-uses (in which 50% identified species as edible) and 371the clusters of gradually abandoned (77.2%) and popular (91.5%) 372plant-uses (p b.001 for both comparisons). We also found differences 373in the three clusters regarding the number of informants who report 374to buy such species now versus the past (χ 2 = 9.50, p b.009 for the 375Kruskal–Wallis test). Statistically significant differences were found be- 376tween the cluster of popular plant-uses (which had an average market 377index of 0.14) and the clusters of gradually (0.07; p = .07) and mostly 378abandoned (0.4; p = .002) plant-uses (Table 3). t3:1Table 3 t3:2Characterization of groups resulting from the hierarchical cluster analysis. t3:3Variables Total Group A Group B Group C Kruskal–Wallis t3:4Popular (n = 11) Gradually abandoned (n = 29) Mostly abandoned (n = 16) χ 2 p-Value t3:5Mean of variables used to create cluster t3:6Consumption index a −0.53 −0.12 −0.52 −0.84 45.51 .0001 t3:7Gathering index a −0.62 −0.35 −0.58 −0.87 36.98 .0001 t3:8 t3:9Values of independent variables across clusters t3:10 Current status t3:11 Recognize a,b 82.30 95.42 86.91 64.90 20.80 .0001 t3:12 Edible a,b 72.27 91.53 77.23 50.04 18.78 .0001 t3:13 Market index a 0.08 0.15 0.07 0.04 9.50 .009 t3:14 Cultural heritage b t3:15 The use is traditional in this area 58.15 71.93 61.11 43.31 8.19 0.02 t3:16 It is good for health 45.44 63.59 49.88 24.94 15.46 0.0004 t3:17 It tastes good 44.90 59.39 50.71 24.41 15.71 0.0004 t3:18 It is only eaten in times of famine 15.06 11.54 16.17 15.45 1.13 0.57 t3:19 Recreation b t3:20 I gather it for leisure 20.18 31.09 21.92 9.54 8.17 0.02 t3:21 Gathering is time consuming 12.25 9.45 16.98 5.62 4.92 0.08 t3:22 Preparing is time consuming 8.67 4.08 13.02 4.08 3.14 0.21 t3:23 a See definitions in the Supplementary material. t3:24 b Cells represent the percentage of informants who partially (=4) or totally (=5) agree with each of the statements. Fig. 2. Consumption and gathering indexes, by group. Fig. 3. Proportion of informants who know and identify as edible plants in the three groups. 6V. Reyes-García et al. / Ecological Economics xxx (2015) xxx–xxx Please cite this article as: Reyes-García, V., et al., From famine foods to delicatessen: Interpreting trends in the use of wild edible plants through cultural ecosystem services, Ecol. Econ. (2015), http://dx.doi.org/10.1016/j.ecolecon.2015.11.003 UNCORRECTED PROOF 379 Results from Kruskal–Wallis test show that the percentage of infor- 380 mants who agree with statements indicating cultural appreciation was 381 significantly different among the three clusters (p b.05) for all the var- 382 iables, except for agreement with the statement that such plant-uses 383 were only consumed in times of famine, variable for which we did not 384 find statistically significant differences among clusters (p = .57). A se- 385 ries of multiple comparisons using post hoc Dunn tests showed that 386 the differences regarding the perceptions of plant-uses as traditional, 387 healthy and tasty were statistically significant when comparing mostly 388 abandoned plant-uses with both gradually abandoned and popular 389 plant-uses (p b.05 or lower for all comparisons). 390 Regarding the recreation function, the percentage of informants who 391 gather wild plants as a hobby was significantly different among the 392 three clusters (χ 2 = 8.17, p b.02, Table 3), with statistically significant 393 differences between the cluster of mostly abandoned plant-uses 394 (9.54) and the clusters of gradually abandoned (21.91%) and popular 395 (31.09%) plant-uses. We also found statistically significant differences 396 between clusters regarding the percentage of informants who agree 397 with the statement that gathering the selected species is time consum- 398 ing (χ 2 = 4.92, p b.08, Table 3), but not in the percentage of informants 399 who agree with the statement that preparing the selected species is 400 time consuming. Regarding gathering time, the Dunn test suggests 401 that differences are statistically significant only when comparing 402 plant-uses in the gradually abandoned (16.98) and mostly abandoned 403 clusters (5.62), with popular plant-uses somewhere in between 404 (9.45%). 405 4. Discussion 406 We start the discussion by acknowledging some limitations of this 407 study. A first important limitation relates to sample selection biases. 408 To select informants, we used a convenience sample by soliciting partic- 409 ipation from people in public places, e.g., parks, bars, and grocery stores. 410 Convenient sampling precludes us from drawing conclusions about the 411 larger population (Babbie, 2009). Furthermore, about 21% of the people 412 approached declined to participate. Given that some of these people ar- 413 gued that they lacked knowledge on wild edible plants, our findings 414 might indeed underrepresent the real magnitude of the decreasing 415 trend in the use of wild edible plants. We argue, however, that this 416 was the only ethical way to conduct the survey, and that –given that 417 much research on wild edible plants–is largely conducted with local 418 experts, this first approach to capture a larger part of the population 419 provides valuable insights for the purposes of our research. 420 Two additional caveats relate to our survey. First, our questions only 421 gather people's perceptions. Whether wild edibles were actually con- 422 sumed in the past with the frequency reported by informants is an 423 open question. However, given the lack of other empirical data, it is 424 the best estimation we can have. Second, many of the variables mea- 425 sured are intertwined, even if we attempted to measure them indepen- 426 dently. For example, we found that a large proportion of informants 427 were not able to identify or recognize as edible species with ‘mostly 428 abandoned’uses. The finding is not surprising, as gathering is clearly re- 429 lated to the abilities to identify and recognize wild plants as edible 430 (Pilgrim et al., 2008). While such abilities might be less clearly related 431 to consumption (wild edible plants can also be obtained by means not 432 requiring identification abilities such as gifts or the market), the possi- 433 bility that those variables are closely interrelated remains high. 434 Keeping those caveats in mind, we now discuss the main findings of 435 this work. First, we found an overall, generalized decrease in the con- 436 sumption and gathering of wild edible plants. In fact, we only find an in- 437 crease in the consumption of one of the plant-uses analyzed: the 438 consumption of Asparagus in Sierra Morena Extremeña, a plant-use 439 that is a strong marker of cultural identity and place attachment 440 (Acosta-Naranjo and Díaz-Diego, 2008). We found no instance of 441 increase in gathering of any wild edible plant. 442Several authors have argued that such general trend is concomitant 443with urbanization and modernization of lifestyles (González et al., 2011; 444Hadjichambis et al., 2008; Seeland and Staniszewski, 2007; Tardío et al., 4452005). Even in rural areas, as the sites studied here, most people nowa- 446days rely on foods obtained through the market (Abbet et al., 2014; 447Kalle and Soukand, 2013; Łukasz et al., 2013), which imply that the gen- 448eral decrease in the consumption and gathering of wild edible species 449relates to the overall drop in the provisioning services they use to pro- 450vide. In such context, the question that remains, however, is ‘why the 451consumption and gathering of some wild edible plants (about 20%) 452remains relatively popular?’ 453The analysis of the uneven trends in the consumption and gathering 454of wild edible plants helps us answer such question. Data presented in 455Fig. 4 suggest that plant-uses in the ‘popular’and ‘gradually abandoned’ 456clusters are relatively similar in most criteria except two: the market 457index and the gathering time. These two cluster together contrast 458sharply with the cluster of ‘mostly abandoned’plant-uses. We first 459discuss the differences between the first two clusters and then the 460differences between those two and the last one. 461We found two main differences between the first two clusters. First, 462‘popular’plant-uses have the highest average market index, suggesting 463an increased dependency on the market for obtaining the species. Plant- 464uses in this cluster include the use for seasoning of O. vulgare and 465T. zygis, now easily available in the market. Moreover, some plants in 466this cluster, like A. acutifolius, are sold by gatherers in informal local 467markets. Second, more informants reported gathering of species with 468uses falling in the ‘gradually abandoned’cluster as time consuming. 469‘Popular’plant-uses included several fruits and plants for seasoning, 470whereas ‘gradually abandoned’plant-uses included many species used 471as vegetables, which require long preparations. For example, although 472our ethnographic information suggests that S. hispanicus, a wild vegeta- 473ble present in three of the study areas, is highly valued, it systematically 474fell within the category of ‘gradually abandoned’. The preparation of 475such vegetable requires peeling the thorny leaves, a time consuming 476process that might discourage some gatherers. Thus, the two factors 477that seem to explain why some plant-uses remain relatively ‘popular’ 478while others are being ‘gradually abandoned’relate to the increasing 479availability of some plants in formal and informal markets and to 480required time investment for gathering. 481Those factors alone, however, do not explain the difference between 482species in those two clusters and ‘mostly abandoned’plant-uses. For ex- 483ample, many of the ‘mostly abandoned’plant-uses are snack foods, and 484therefore did not require long gathering and preparation times. Then, 485what explains that some uses remain relatively popular, while some 486others are being ‘mostly abandoned’? Some researchers have argued 487that the decrease in the consumption of wild foods relates to the fact 488that they are perceived as food of the poor, a safety net, or a reserve 489food in case of famine (e.g. Hedge et al., 1996; Łukasz et al., 2013; 490Pouta et al., 2006; Senaratne et al., 2003). This, however, does not 491seem to be the case in our sites, as –on average–only 15% of informants 492agreed that wild edible plants are only eaten in times of famine, the 493percentage being similar across the three clusters. 494Our analysis unravels that, indeed, the cultural ecosystem services 495and values associated with different wild edible species can be a critical 496factorin explaining different trends in their consumption and gathering. 497For example, in contrast with plant-uses in the ‘popular’and ‘gradually 498abandoned’groups, less informants agreed with statements regarding 499cultural appreciation (e.g., being traditional in the area, healthy, or 500tasty) when such statements referred to ‘mostly abandoned’plant- 501uses. Similarly, the gathering and consumption of ‘popular’and 502‘gradually abandoned’plant-uses are more frequently identified as 503leisure activities than the gathering and consumption of ‘mostly 504abandoned’plant uses. Moreover, when all explanations provided are 505taken together, non-use values, such as those associated with cultural 506identity and heritage values seem to be –at least–as important as 507cultural services more frequently accounted for in the literature on 7V. Reyes-García et al. / Ecological Economics xxx (2015) xxx–xxx Please cite this article as: Reyes-García, V., et al., From famine foods to delicatessen: Interpreting trends in the use of wild edible plants through cultural ecosystem services, Ecol. Econ. (2015), http://dx.doi.org/10.1016/j.ecolecon.2015.11.003 UNCORRECTED PROOF 508 cultural ecosystem services and wild edible plants, such as recreation 509 (Schulp et al., 2014). 510 The interpretation that the association with cultural ecosystem ser- 511 vices relates to different trends in the consumption and gathering of 512 wild edible plants matches well with previous research findings and 513 with our own ethnographic information. Previous research has 514 highlighted that the gathering and consumption of wild edible 515 plants play a significant role in maintaining local culture, identity 516 (Pardo-de-Santayana and Gómez-Pellón, 2002; Schunko and Vogl, 517 2010; Seeland and Staniszewski, 2007), and spirituality (Hummer, 518 2013). Similarly, in some of our sites, we observed that some uses of 519 wild edibles seem to be maintained due to a revival of traditions linked 520 to their cultural construction as “typical”foods. This is the case of 521 species used to elaborate liqueurs, as the use of walnut in a traditional 522 Catalan beverage (ratafia) or the fruits of P. spinosa macerated in alcohol 523 in Basque Country (patxaran). The finding also meshes with previous 524 research highlighting that wild edible plants have remained more im- 525 portant in countries in which wild food is important in the traditional 526 cuisine, versus countries where traditional cuisine is dominantly based 527 on agricultural products (Schulp et al., 2014). Thus, identitarian- 528 gastronomic traditions seem to help maintaining alive the gathering 529 and consumption of some wild edible plants (see also Leonti et al., 530 2006; Pieroni and Price, 2006). 531 5. Conclusion 532 Our data show a generalized decrease in the consumption and 533 gathering of wild edible plants in all study sites. However, we also 534 found that the assessed trend is uneven and changes significantly across 535plant-uses. Specifically, we found that –despite the overall decreasing 536trend–uses of wild edible plants that simultaneously relate to foods 537with high cultural appreciation and the recreational function of gather- 538ing remains popular. While the overall decrease in the consumption of 539wild edibles might be concomitant with forces related to urban, indus- 540trial, and post-industrial lifestyles in which wild edible plants have 541lost their historically important role as provisioning services (Abbet 542et al., 2014; Bharucha and Pretty, 2010; Kalle and Soukand, 2013; 543Łukasz et al., 2013; Turner and Turner, 2008), cultural services and 544values associated to the gathering and consumption of some wild edible 545plants seem to explain divergent trends across species. In sum, even if 546wild edible plants are a provisioning ecosystem service, in our study 547sites (and, we may dare to say, in other sites with modern food produc- 548tion and supply systems) their role as a provisioning service is nowa- 549days marginal or negligible, and in most cases no longer accounts for 550continuity in their use. It is primarily through their bundling with 551cultural ecosystem services and non-use values that the persistence in 552the consumption and gathering of some wild edible plants can be ex- 553plained. Our finding reinforces the notion that ecosystem services 554tend to combine in complex and non-linear ways and, more specifically, 555that cultural ecosystem services are deeply bundled with the other 556categories of ecosystem services (Gould et al., 2014; Gould et al., 5572015; Milcu et al., 2013), often producing a wide range of interdepen- 558dent benefits. 559Acknowledgments 560Research was funded by the Programa de Ciencias Sociales y 561Humanidades del Ministerio de Economia y Competitividad (España) Fig. 4. Percentage of informants who mostly or totally agree with statements regarding A) cultural heritage and B) recreation values of wild edible plants. 8V. Reyes-García et al. / Ecological Economics xxx (2015) xxx–xxx Please cite this article as: Reyes-García, V., et al., From famine foods to delicatessen: Interpreting trends in the use of wild edible plants through cultural ecosystem services, Ecol. Econ. (2015), http://dx.doi.org/10.1016/j.ecolecon.2015.11.003 UNCORRECTED PROOF 562 (CSO2011-27565), with supporting funding from Diputación de Bada- 563 joz, Kent University and Catalan government (514SGR2014). Gómez- 564 Baggethun received partial funding from the FP7 EU-funded project 565 OpenNESS (EC-308428). We thank all the informants who participated 566 in the research. We thank L. García-Cervigón, X. Martínez, M. Mayans, A. 567 Noguera, E. Carrió, and M. Parada for data collection and J. Tardío and 568 two anonymous reviewers for comments to previous versions of this 569 article. Reyes-García thanks the Dryland Cereals Research Group at 570 ICRISAT-Patancheru for providing office facilities. 571 References 572 Abbet, C., Mayor, R., Roguet, D., Spichiger, R., Hamburger, M., Potterat, O., 2014. 573 Ethnobotanical survey on wild alpine food plants in lower and central Valais 574 (Switzerland). J. Ethnopharmacol. 151, 624–634. 575 Aceituno-Mata, L., 2010. Estudio etnobot nico y agroecológico de la Sierra Norte de 576 Madrid. Universidad Autónoma de Madrid. 577 Acosta-Naranjo, R., Díaz-Diego, J., 2008. Y en sus manos la vida. Los cultivadores de las 578 variedades locales de tentudía. Centro de Desarrollo Comarcal de Tentudía, Tentudía, 579 Extremadura. 580 Angelsen, A., Jagger, P., Babigumira, R., Belcher, B., Hogarth, N.J., Bauch, S., Boerner, J., 581 Smith-Hall, C., Wunder, S., 2014. Environmental income and rural livelihoods: a 582 global-comparative analysis. World Dev. 64, S12–S28. 583 Babbie, E.R., 2009. The Practice of Social Research. Wadsworth, Independence, Kentucky, 584 USA. 585 Behre, K.E., 2008. Collected seeds and fruits from herbs as prehistoric food. Vegetation 586 History and Archaeobotany 17, 65–73. 587 Bharucha, Z., Pretty, J., 2010. The roles and values of wild foods in agricultural systems. 588 Philos. Trans. R. Soc., B 365, 2913–2926. 589 Certomà, C., Tornaghi, C., 2015. Political gardening. Transforming cities and political 590 agency. Local Environ. 20, 1123–1131. 591 Chan, K.M.A., Guerry, A.D., Balvanera, P., Klain, S., Satterfield, T., Basurto, X., Bostrom, A., 592 Chuenpagdee, R., Gould, R., Halpern, B.S., Hannahs, N., Levine, J., Norton, B., 593 Ruckelshaus, M., Russell, R., Tam, J., Woodside, U., 2012. Where are cultural and social 594 in ecosystem services? A framework for constructive engagement. Bioscience 62, 595 744–756. 596 Cornara, L., La Rocca, A., Marsili, S., Mariotti, M.G., 2009. Traditional uses of plants in the 597 Eastern Riviera (Liguria, Italy). J. Ethnopharmacol. 125, 16–30 2009. 598 Davis, A., Wagner, J.R., 2003. Who knows? On the importance of identifying “experts” 599 when researching local ecological knowledge. Hum. Ecol. 31, 463–489. 600 de Groot, R.S., Wilson, M., Boumans, R., 2002. A typology for the description, classification 601 and valuation of ecosystem functions, goods and services. Ecol. Econ. 41, 393–408. 602 Dunn, O.J., 1964. Multiple comparisons using rank sums. Technometrics 6, 241–252. 603 Gómez-Baggethun, E., Mingorria, S., Reyes-García, V., Calvet, L., Montes, C., 2010. 604 Traditional ecological knowledge trends in the transition to a market economy: 605 empirical study in the Donana natural areas. Conserv. Biol. 24, 721–729. 606 González, J.A., García-Barriuso, M., Amich, F., 2011. The consumption of wild and 607 semi-domesticated edible plants in the Arribes del Duero (Salamanca–Zamora, 608 Spain): an analysis of traditional knowledge. Genet. Resour. Crop. Evol. 58, 991–1006. 609 Gould, R.K., Ardoin, N.M., Woodside, U., Satterfield, T., Hannahs, N., Daily, G.C., 2014. The 610 forest has a story: cultural ecosystem services in Kona Hawai'i. Ecol. Soc. 19. 611 Gould, R.K., Klain, S.C., Ardoin, N.M., Satterfield, T., Woodside, U., Hannahs, N., Daily, G.C., 612 Chan, K.M., 2015. A protocol for eliciting nonmaterial values through a cultural 613 ecosystem services frame. Conserv. Biol. 29, 575–586. 614 Grasser, S., Schunko, C., Vogl, C.R., 2012. Gathering “tea” — from necessity to connectedness 615 with nature. Local knowledge about wild plant gathering in the Biosphere Reserve 616 Grosses Walsertal (Austria). J. Ethnobiol. Ethnomed. 8. 617 Hadjichambis, A.C., Paraskeva-Hadjichambi, D., Della, A., Giusti, M.E., De Pasquale, C., 618 Lenzarini, C., Censorii, E., Gonzales-Tejero, M.R., Sanchez-Rojas, C.P., Ramiro-Gutierrez, 619 J.M., Skoula, M., Johnson, C., Sarpaki, A., Hmamouchi, M., Jorhi, S., El-Demerdash, M., 620 El-Zayat, M., Pieroni, A., 2008. Wild and semi-domesticated food plant consumption in 621 seven circum-Mediterranean areas. Int. J. Food Sci. Nutr. 59. 622 Haines-Young, R., Potschin, M., 2013. Common International Classification of Ecosystem 623 Services (CICES): Consultation on Version 4, August–December 2012 EEA Framework 624 Contract No EEA/IEA/09/003. 625 Hedge, R., Suryaprakash, S., Achoth, L., Bawa, K.S., 1996. Estraction of non-timber forest 626 products in the forest of BR Hills. 1. Contribution to rural income. Econ. Bot. 50, 627 243–250. 628 Hummer, K.E., 2013. Manna in winter: indigenous Americans, huckleberries, and 629 blueberries. Hortscience 48, 413–417. 630 Idolo, M., Motti, R., Mazzoleni, S., 2010. Ethnobotanical and phytomedicinal knowledge in 631 a long-history protected area, the Abruzzo, Lazio and Molise National Park (Italian 632 Apennines). J. Ethnopharmacol. 127, 379–395. 633 Kalle, R., Soukand, R., 2013. Wild plants eaten in childhood: a retrospective of Estonia in 634 the 1970s–1990s. Bot. J. Linn. Soc. 172, 239–253. 635Kangas, K., Markkanen, P., 2001. Factors affecting participation in wild berry picking by 636rural and urban dwellers. Silva Fennica 35, 487–495. 637Kaufman, L., Rousseeuw, J.P., 1990. Finding Groups in Data: An Introduction to Cluster 638Analysis. Wiley, New York. 639Leonti, M., Nebel, S., Rivera, D., Heinrich, M., 2006. Wild gathered food plants in the 640European Mediterranean: a comparative analysis. Econ. Bot. 60, 130–142. 641Łukasz, Ł., Zovko Končić, M., Miličević, T., Dolina, K., Pandža, M., 2013. Wild vegetable 642mixes sold in the markets of Dalmatia (southern Croatia). J. Ethnobiol. Ethnomed. 9. 643MA, 2005. Ecosystems and Human Well-being: A Framework for Assessment. Island 644Press, Washington, D.C., USA. 645Mavengahama, S., McLachlan, M., de Clercq, W., 2013. The role of wild vegetable species 646in household food security in maize based subsistence cropping systems. Food Sec. 5, 647227–233. 648Menendez-Baceta, G., Aceituno-Mata, L., Tardío, J., Reyes-García, V., Pardo-de-Santayana, 649M., 2012. Wild edible plants traditionally gathered in Gorbeialdea (Biscay, Basque 650Country). Genet. Resour. Crop. Evol. 59, 1329–1347. 651Milcu, A.I., Hanspach, J., Abson, D., Fischer, J., 2013. Cultural ecosystem services: a 652literature review and prospects for future research. Ecol. Soc. 18. 653Molina, M., Pardo-de-Santayana, M., García, E., Aceituno-Mata, L., Morales, R., Tardío, J., 6542012. Exploring the potential of wild food resources in the Mediterranean region: 655natural yield and gathering pressure of the wild asparagus (Asparagus acutifolius 656L.). Span. J. Agric. Res. 10, 1090–1100. 657Nolan, J.M., Pieroni, A., 2014. Introduction to Special Issue on Food Security in a Changing 658World Journal of Ethnobiology SI, 4–6. 659Parada, M., Carrio, E., Valles, J., 2011. Ethnobotany of food plants in the Alt Emporda 660region (Catalonia, Iberian Peninsula). J. Appl. Bot. Food Qual.-Angewandte Botanik 66184, 11–25. 662Pardo-de-Santayana, M., Gómez-Pellón, E., 2002. Ethnobotany: traditional management 663of plants and cultural heritage. An. Jard. Bot. Madr. 60, 171–182. 664Pieroni, A., Price, L., 2006. Eating and Healing. Traditional Foods as Medicine. The Haworth 665Press, New York. 666Pilgrim, S.E., Cullen, L.C., Smith, D.J., Pretty, J., 2008. Ecological knowledge is lost in 667wealthier communities and countries. Environ. Sci. Technol. 42, 1004–1009. 668Pouta, E., Sievanen, T., Neuvonen, M., 2006. Recreational wild berry picking in Finland — 669reflection of a rural lifestyle. Soc. Nat. Resour. 19, 285–304. 670Redzic, S., 2010. Use of wild and semi-wild edible plants in nutrition and survival of 671people in 1430 days of siege of Sarajevo during the war in Bosnia and Herzegovina 672(1992–1995). Coll. Anthropol. 34, 551–570. 673Rigat, M., Bonet, M.A., García, S., Garnatje, T., Vallès, J., 2009. Ethnobotanical studies in the 674high river Ter valley (Pyrenees, Catalonia, Iberian Peninsula). Non-crop food vascular 675plants and crop food plants with medicinal properties. Ecol. Food Nutr. 48, 303–326. 676San Miguel López, E., 2004. Etnobotánica de Piloña (Asturias): cultura y saber popular 677sobre las plantas en un concejo del Centro-Norte Asturiano. Departamento de 678Biología. Universidad Autónoma de Madrid, Madrid. 679Schulp, C.J.E., Thuiller, W., Verburg, P.H., 2014. Wild food in Europe: a synthesis of 680knowledge and data of terrestrial wild food as an ecosystem service. Ecol. Econ. 681105, 292–305. 682Schunko, C., Vogl, C.R., 2010. Organic farmers use of wild food plants and fungi in a hilly 683area in Styria (Austria). J. Ethnobiol. Ethnomed. 6. 684Seeland, K., Staniszewski, P., 2007. Indicators for a European cross-country state-of-the- 685art assessment of non-timber forest products and services. Small-scale For. 6, 686411–422. 687Senaratne, A., Abeygunawardena, P., Jayatilake, W., 2003. Changing role of non-timber 688forest products (NTFP) in rural household economy: the case of Sinharaja World 689Heritage site in Sri Lanka. Environ. Manag. 32, 559–571. 690Shumsky, S., Hickey, G.M., Johns, T., Pelletier, B., Galaty, J., 2014. Institutional factors 691affecting wild edible plant (WEP) harvest and consumption in semi-arid Kenya. 692Land Use Policy 38, 48–69. 693Stryamets, N., Elbakidze, M., Angelstam, P., 2012. Role of non-wood forest products for 694local livelihoods in countries with transition and market economies: case studies in 695Ukraine and Sweden. Scand. J. For. Res. 27, 74–87. 696Tardío, J., Pardo-de-Santayana, M., Morales, R., 2006. Ethnobotanical review of wild edible 697plants in Spain. Bot. J. Linn. Soc. 152, 27–71. 698Tardío, J., Pascual, H., Morales, R., 2005. Wild food plants traditionally used in the province 699of Madrid, central Spain. Econ. Bot. 59, 122–136. 700TEEB, 2010. The Economics of Ecosystems and Biodiversity: Ecological and Economic 701Foundations. Earthscan, London. 702Toledo, V., Ortiz-Espejel, B., Cortes, L., Moguel, P., Ordoñez, M.D., 2003. The multiple use of 703tropical forests by indigenous peoples in Mexico: a case of adaptive management. 704Conserv. Ecol. 7, 9. 705Turner, N.J., Turner, K., 2008. “Where our women used to get the food”: cumulative effects 706and loss of ethnobotanical knowledge and practice; case study from coastal British 707Columbia. Botany 86, 103–115. 708Vinceti, B., Termote, C., Ickowitz, A., Powell, B., Kehlenbeck, K., Hunter, D., 2013. The 709contribution of forests and trees to sustainable diets. Sustainability 5, 4797–4824. 710 9V. Reyes-García et al. / Ecological Economics xxx (2015) xxx–xxx Please cite this article as: Reyes-García, V., et al., From famine foods to delicatessen: Interpreting trends in the use of wild edible plants through cultural ecosystem services, Ecol. Econ. (2015), http://dx.doi.org/10.1016/j.ecolecon.2015.11.003