This is the accepted version of the journal article: Garteizgogeascoa, María; García-del-Amo, David; Reyes-García, Victoria. «Us- ing proverbs to study local perceptions of climate change : a case study in Sierra Nevada (Spain)». Regional environmental change, Vol. 20, Issue 2 (June 2020), art. 59. DOI 10.1007/s10113-020-01646-1 This version is available at https://ddd.uab.cat/record/264780 under the terms of the license Regional Environmental Change Using proverbs to study local perceptions of climate change: A case study in Sierra Nevada (Spain) --Manuscript Draft--Manuscript Click here to view linked References Using proverbs to study local perceptions of climate change: A case study in Sierra Nevada (Spain) María Garteizgogeascoa1, David García-del-Amo1, Victoria Reyes-García1,2 1 Institut de Ciència i Tecnologia Ambientals, Universitat Autònoma de Barcelona, 08193 Cerdanyola del Vallès, Spain 2 Institució Catalana de Recerca i Estudis Avançats, 08010 Barcelona, Spain María Garteizgogeascoa mariagarteiz@gmail.com David García del Amo David.Garcia.DelAmo@uab.cat Corresponding author Victoria Reyes-García ICREA Research Professor Institute of Environmental Science and Technology (ICTA-UAB) ICTA-ICP, Edifici Z Carrer de les Columnes Universitat Autònoma de Barcelona E-08193, Bellaterra (Cerdanyola del Vallès-Barcelona) Tel: +34 93 586 8976 E-mail: Victoria.reyes@uab.cat 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 1 63 64 65 ABSTRACT Local communities’ dependence on the environment for their livelihood has guided the development of indicators of local weather and climate variability. These indicators are encoded in different forms of oral knowledge. We explore whether people recognize and perceive as accurate one type of such forms of oral knowledge, climate- related proverbs. We conducted research in the Alta Alpujarra Occidental, Sierra Nevada, Spain. We collected climate-related proverbs and classified them according to whether they referred to the climatic, the physical, or the biological system. We then conducted questionnaires (n=97) to assess informant’s ability to recognize a selection of 30 locally relevant proverbs and their perception of the accuracy of the proverb. Climate-related proverbs are abundant and relatively well recognized even though informants consider that many proverbs are not accurate nowadays. Although proverbs’ perceived accuracy varied across informant’s age, level of schooling, and area of residence, overall proverb’s lack of reported accuracy goes in line with climate change trends documented by scientists working in the area. While our findings are limited to a handful of proverbs, they suggest that the identification of mismatches and discrepancies between people’s reports of proverbs (lack of) accuracy and scientific assessments could be used to guide future research on climate change impacts. Keywords: Andalucía (Spain); climate change impacts; ethnoclimatology; indigenous and local knowledge; proverbs. Length: Text: 5406; Figures: 2; Tables: 3. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 2 63 64 65 51 Introduction 1 2 52 In their quest to better understand local climate change impacts, both natural and 3 4 53 social scientists are challenged by the scarcity of grounded data. This has resulted in 5 6 54 climate scientists calling for the exploration of new data sources (Rosenzweig and 7 8 55 Neofotis 2013). Within this context, several authors have argued that local ecological 9 56 knowledge has an untapped potential to contribute to further our understanding of local 10 11 57 climate change impacts (Barnes et al. 2013; Savo et al. 2016). Researchers have 12 13 58 documented many instances in which Indigenous peoples or local communities with a 14 15 59 long history of interaction with the environment have developed complex knowledge 16 17 60 systems that allow them to detect changes in local weather and climatic variability, as 18 19 61 well as the impacts of such changes in the physical and the biological systems on which 20 62 they depend (e.g., Orlove et al. 2000; Fernández-Llamazares et al. 2015; Reyes-García 21 22 63 et al. 2016). Through observations of their immediate environment accumulated over 23 24 64 generations and continuously adapted to environmental and other changes local 25 26 65 populations have developed a large body of ethnoclimatological knowledge, defined as 27 28 66 “the comprehensive system of insights, experiences, and practices regarding climate 29 30 67 and local weather events, as well as their changes at different spatiotemporal scales” 31 68 (Reyes-García et al. 2018). 32 33 34 69 An important component of ethnoclimatological knowledge is weather 35 36 70 forecasting. Traditional weather forecasting, or the short- and long-term predictions of 37 38 71 weather variability, is typically based on the observation of local phenomena of 39 40 72 different types, which are used as weather indicators (King et al. 2008; Zuma- 41 Netshiukhwi et al. 2013; Kagunyu et al. 2016). For example, in several parts of the 73 42 43 74 Mediterranean region, short-term weather is forecasted through the observation of 44 45 75 physical phenomena (e.g., clouds, fog), astronomical conditions (e.g., the moon), or 46 47 76 plant and/or animal behavior (Lefale 2010; Enock 2013). In the same region, annual 48 49 77 weather is forecasted through observations of weather conditions during specific days of 50 51 78 the month of August, observations that are then extrapolated to the coming months, a 52 79 method in Spain known as cabañuelas (Mesa et al. 1997). Moreover, research shows 53 54 80 that some of these traditional weather forecasting methods may be accurate (e.g., Orlove 55 56 81 et al. 2000; Marin 2010; Lefale 2010). For example, Orlove and colleagues (2000) 57 58 82 showed that farmers in drought-prone regions of Andean South America make 59 60 83 observations of changes in the apparent brightness of stars in the Pleiades around the 61 62 3 63 64 65 84 time of the southern winter solstice. These observations are then used to forecast 1 2 85 interannual variations in summer rainfall and to decide when to plant the most important 3 4 86 crops. These researchers found that poor visibility of the Pleiades in June is caused by 5 87 an increase in high cirrus clouds which is indicative of an El Niño year, usually linked 6 7 88 to reduced rainfall during the growing season. In other words, the studied Andean 8 9 89 weather forecasting method seems to be based on a visible indicator of El Niño 10 11 90 variability. 12 14 15 92 methods have raised the interest of some atmospheric institutions, such as the Philippine 16 17 93 Atmospheric Geophysical and Astronomical Services Administration or the weather 18 19 94 agencies of New Zealand and Australia (Galacgac and Balisacan 2009). Out of this 20 21 95 interest, the Australian Bureau of Meteorology supports the collection of local weather 22 23 96 calendars (Parrotta and Agnoletti 2012) and Mozambique includes traditional weather 24 25 97 forecasting knowledge in its National Adaptation Plan for Action (Bambaige 2007). 26 98 However, some researchers have started to call for caution in the use of traditional 27 28 99 weather forecasting methods, signalling that these methods might lose their predictive 29 30 100 ability in the current situation of rapid climatic changes (Weatherhead et al. 2010). 31 32 33 101 In this work, we explore whether local people recognize climate-related 34 35 102 proverbs, a traditional weather forecasting method, and whether they perceive such 36 103 proverbs as accurate. Proverbs are concise and short sentences known by many people 37 38 104 and containing locally accepted truths, teachings, and traditional points of view. 39 40 105 Proverbs typically appear as fixed, metaphorical, and easy to remember forms, for 41 42 106 which they are considered an effective form of storing and transmitting knowledge and 43 44 107 thus maintaining people-place relations (Sterling et al. 2017). Previous studies have 45 108 already explored the potential use of proverbs for biodiversity conservation and 46 47 109 management (e.g., Kurien 1998), including their potential as a source of information to 48 49 110 predict meteorological phenomena (Kanno et al. 2013). In this work, we advance this 50 51 111 line of research by analyzing trends in proverbs’ recognition and perceived accuracy, 52 53 112 and by exploring whether changes in the perceived accuracy of proverbs are in line with 54 55 113 climate change impacts documented by scientists in the study area. 56 57 114 58 59 60 115 Sierra Nevada and its people 61 62 4 63 64 65 116 We conducted research in Sierra Nevada, a rural and high mountainous region 1 2 117 (reaching 3482 m.a.s.l.) in Andalusia, southwestern Spain. Traditionally, Sierra Nevada 3 4 118 was occupied by farmlands, mainly devoted to herbaceous and mixed crops. Like in 5 119 other areas of Spain during the 20th century, the integration of local economies in global 6 7 120 markets and agricultural mechanization and intensification lead to the loss of 8 9 121 competitiveness of traditional agriculture, the decrease in job opportunities for young 10 11 122 generations, and urban migration in Sierra Nevada (Cozar Valero 2000). 12 13 14 123 Conservation policies also contributed to dramatically change land uses in Sierra 15 124 Nevada. The area was designated as Biosphere Reserve in 1986, then as Natural Park in 16 17 125 1989, and finally as National Park in 1999. These protection figures recognize the area 18 19 126 as one of the most important hotspots for biodiversity in the Mediterranean Region 20 21 127 (Lorite et al. 2007). The creation of the protected area occurred at the same time that 22 23 128 agro-silvopastoral uses were abandoned and the area experienced an active reforestation 24 129 with pine trees and increasing urbanization. In 2006, only 4.72% of the protected area 25 26 130 was dedicated to farmlands; traditional mixed crops had been replaced by a 27 28 131 monospecific arboriculture of olive and almond trees, and sheep transhumance was 29 30 132 much less frequent. Nowadays, although agriculture, livestock and beekeeping continue 31 32 133 to be present, the economic activity is rapidly shifting to the tertiary sector (e.g., 33 34 134 tourism, skiing) (Pérez-Luque et al. 2016). 35 36 135 Sierra Nevada is a relevant case to study how the accuracy of climate-related 37 38 136 proverbs might help in scientists’ quest to detect local climate change impacts for two 39 40 137 reasons. First, Sierra Nevada inhabitants have a history of dependency on natural 41 42 138 resources to maintain their livelihood, which also translates in an important pool of 43 44 139 local knowledge (Iniesta-Arandia et al. 2015). Livelihood activities in Sierra Nevada 45 140 were traditionally linked to local weather conditions and water management techniques 46 47 141 that –over centuries- shaped an engineered landscape dominated by water transport, i.e., 48 49 142 acequias, and water storage infrastructures, i.e., aljibes (Castillo Martín 1999). 50 51 143 Traditional agriculture, livestock and other subsistence activities were highly dependent 52 53 144 on the maintenance of this human modified landscape. 54 55 56 145 The second reason why Sierra Nevada provides a relevant case study is the 57 146 availability of scientific data that can be partly used as a base for comparison. In 2007, a 58 59 147 global change observatory (i.e., the Observatorio del cambio global de Sierra Nevada) 60 61 62 5 63 64 65 148 was stablished in the National Park. Scientists and technical staff from the Observatorio 1 2 149 collect information on climatology, phenological changes in plants, bird’s migration, 3 4 150 and changes in the cryosphere. Scientists at the Observatorio have also compiled 5 151 temporal series for some of those indicators. In line with predictions for the 6 7 152 Mediterranean region (Planton et al. 2012; IPCC 2014), data from this observatory 8 9 153 show that the area has experienced a decrease in rainfall and an increase in the average 10 11 154 annual maximum and minimum temperatures. Data also show that snow duration at 12 13 155 altitudes above 2500 m.a.s.l. has decreased of 3.8 days in 17 years and that the soil 14 15 156 thermic regime has changed, resulting in the disappearance of permafrost (Zamora 16 157 Rodríguez et al. 2015). Furthermore, a delay in the flowering seasons has been detected, 17 18 158 specially affecting early flowering plants at high altitudes. Finally, while net values of 19 20 159 bird species diversity have not changed significantly, bird species that predominated in 21 22 160 the 1980s (such as Sylvia communis) are less abundant nowadays and new species, like 23 24 161 Saxicola rubicola, Prunella modularis or Sylvia conspicillata, have appeared (Zamora 25 162 Rodríguez et al. 2015). The Observatorio disseminates the most significant findings of 26 27 163 its research through scientific publications and local mass media (e.g., regional journals 28 29 164 such as Granada hoy or Ideal de Granada), but it does not have a dissemination 30 31 165 strategy directly targeting local populations. 32 33 34 166 Methods 35 36 167 Data from this work come from a bibliographic compilation of proverbs and a 37 38 168 face-to-face survey (n=97) conducted during June and July 2018. Data collection 39 40 169 received the approval of the Universitat Autònoma de Barcelona (CEEAH-3367). We 41 42 170 collected the free, prior, and informed consent of all participants before starting the 43 44 171 survey (Supplementary Material 1). 45 46 172 Proverbs compilation 47 48 49 173 Drawing on Hong et al. (2016), to select proverbs we followed a two-step 50 51 174 procedure. In our first step, we consulted different national, regional, and local 52 53 175 compilation books and paremiology webs (see Supplementary Material 2). We 54 55 176 transcribed proverbs from our sources into a file, grouping proverbs with similar 56 177 formulations. We then selected those referring to weather, seasons, wild fauna and flora, 57 58 178 and/or agriculture. Our preliminary list included 665 proverbs. We refined this list by 59 60 179 excluding strictly agricultural proverbs (i.e., proverbs that explain how or when to plant 61 62 6 63 64 65 180 or harvest), as agricultural production is affected by many drivers. This procedure 1 2 181 reduced our list to 366 proverbs. In the second step we reduced the list by eliminating 3 4 182 proverbs without local relevance. To do so, we interviewed eight local key informants, 5 183 three women and five men, selected because they had previously participated in the 6 7 184 collection of proverbs/songs/stories, because they maintained traditional weather 8 9 185 forecast practices (cabañuelas), or because they were locally recognized as experts. We 10 11 186 asked key informants whether they knew the proverbs in our list and we only retained 12 13 187 proverbs recognized by at least two of the key informants. Our final list includes 283 14 15 188 proverbs. 16 17 189 Following Reyes-García et al. (2019), we classified proverbs as climatic if they 18 19 190 referred to temperature, precipitation, air masses, or seasonal events; physical if they 20 21 191 referred to soil moisture, snow cover, or ice sheet; and biological if they referred to 22 23 192 fauna and flora phenology or abundance. Each proverb could be included in more than 24 25 193 one category. For example, we classified the proverb Wet December, frozen January as 26 194 referring both to the climatic and the physical systems. Most of the proverbs in our list 27 28 195 (n=166 out of 283) referred exclusively to the climatic system, 55 referred to the 29 30 196 biological, and 26 to the physical system. The remaining proverbs simultaneously 31 32 197 referred to more than one system. Thus, 21 proverbs simultaneously referred to the 33 34 198 climatic and the biological system, 11 to the climatic and the physical systems, three to 35 199 the biological and physical systems, and one of the proverbs included information 36 37 200 referring to the three systems. 38 39 40 201 Sample 41 42 43 202 Our sampling strategy for the face-to-face survey includes village and 44 45 203 informants selection. To reduce variability in the weather conditions experienced by 46 204 informants answering the survey, we selected neighboring municipalities in the Alta 47 48 205 Alpujarra Occidental, in the southern face of Sierra Nevada (Figure 1). To reduce 49 50 206 potential differences in perception generated by dependency in different livelihood 51 52 207 activities, we selected villages with more than 300 inhabitants, located inside the Sierra 53 54 208 Nevada Natural Park, and at an altitude higher than 500 m.a.s.l. Specifically, we applied 55 56 209 the survey in five study areas corresponding with five different ravines or watersheds, 57 210 for a total of 12 villages. 58 59 60 211 INSERT FIGURE 1 61 62 7 63 64 65 212 We directed our survey to people who had lived in the area for 25 years or more, 1 2 213 assuming that they would be knowledgeable about local proverbs and able to comment 3 214 on their current accuracy. We used a convenience sampling strategy to target a stratified 4 5 215 sample of respondents, aiming to interview a similar number of people working on the 6 7 216 primary (e.g., agriculture, livestock, apiculture) and other sectors (e.g., transport, 8 9 217 tourism). To select potential interviewees, in each village we first had a meeting with 10 11 218 the mayor to introduce ourselves and to present our intended work. During that meeting, 12 13 219 we asked the mayor to introduce us to people who could be interested in participating in 14 220 the study. We then used snowball sampling to contact other potential survey 15 16 221 respondents (Huntington 2000). We also approached people in the street and asked them 17 18 222 whether they wanted to participate in a survey. If a person said s/he was willing to 19 20 223 participate, we first asked for the time living in the area and main profession, and only 21 22 224 interviewed people who fitted our sampling criteria. 23 24 25 225 Survey 26 27 226 Our survey had two distinct sections. In the first section, we asked about 28 29 227 informants’ sociodemographic characteristics. Particularly, we asked about informants’ 30 31 228 age (in years), sex (i.e., women and men), schooling (i.e., people who had never 32 33 229 attended school versus people who had), primary sector and ravine (i.e., Bérchules, 34 35 230 Trevélez, La Taha, Poqueira, and Lanjarón). As in this area people who work in the 36 231 primary sector typically engage in different tasks simultaneously, we could only 37 38 232 differentiate between informants whose economic activity was in the primary sector 39 40 233 versus people with occupations in other sectors. 41 42 43 234 In the second section of the survey, we asked informants about a list of 30 44 45 235 randomly selected proverbs from the full list of 283 identified proverbs. From the 30 46 236 proverbs selected, 10 referred to the climatic, 10 to the physical, and 10 to the biological 47 48 237 system. We then asked informants whether they knew each of the proverbs and whether 49 50 238 they thought they were accurate. We created a variable called recognition that was 51 52 239 coded as (1) when the person knew the proverb and (0) otherwise. To limit variation in 53 54 240 responses associated to people having different interpretations of proverbs, after the 55 56 241 question on recognition, we commented with informants the meaning of the proverb, 57 242 and then asked the informant whether s/he thought the proverb was accurate nowadays. 58 59 243 We coded responses to this second question in a variable called accuracy which took 60 61 62 8 63 64 65 244 the value of (0) when the person said the proverb was not accurate, (1) when the person 1 2 245 thought it was accurate, (2) when the person did not know if it was accurate, and (3) 3 4 246 when the person thought it was accurate only sometimes. 5 6 247 Data analysis 7 8 9 248 We start the analysis calculating the percentage of proverbs refereeing to the 10 11 249 climatic, the physical, and the biological system recognized in each ravine. Then, to 12 13 250 identify the correlates of proverbs’ recognition and accuracy, for each informant we 14 251 calculated 1) the number of proverbs recognized and 2) the number of proverbs 15 16 252 considered as i) accurate, ii) not accurate, iii) sometimes accurate, and iv) for which the 17 18 253 person ignored accuracy. We then aggregated these percentages across the three 19 20 254 categories of proverbs (i.e., climatic, physical, and biological). 21 22 23 255 We explored the sociodemographic correlates of individual proverb recognition 24 256 by running a multivariate regression with the number of proverbs recognized as 25 26 257 dependent variable against informant’s sociodemographic information (i.e., age, sex, 27 28 258 schooling, economic activity sector, and ravine) as potential explanatory variables, 29 30 259 while controlling for ravine fixed-effects. Since our dependent variables are the count of 31 32 260 proverbs (recognized or accurate), we used a Poisson regression, typically used for 33 34 261 modeling events where the outcomes are discrete data with non-negative integer values 35 262 36 (Cameron and Trivedi, 1998). We control for ravine’s fixed-effects, or invariant 37 263 characteristics that might affect the estimated association, by including a set of dummies 38 39 264 for the five ravines. To test whether our results hold for proverbs referring to climate 40 41 265 change impacts on different systems, we ran a set of similar models, but using as 42 43 266 dependent variables the disaggregated variables that capture the number of climatic, 44 45 267 physical and biological proverbs recognized by the informant. All regressions include 46 268 clusters by ravines, to indicate that the observations may be correlated within ravines, 47 48 269 but independent between them. 49 50 51 270 We use the same approach to explore the correlates of accuracy. In this analysis, 52 53 271 we use as dependent variable the number of proverbs the informant considered not 54 55 272 accurate. As for the previous analysis, we ran regressions for the variable capturing 56 273 overall proverbs’ lack of accuracy as well as regressions considering the number of 57 58 274 climatic, physical, and biological proverbs that are not accurate. 59 60 61 62 9 63 64 65 275 In all the analyses we report p-values <0.10 as indicator of statistical 1 2 276 significance. However, as our sample size is small and some of our groups of 3 4 277 heterogeneous size (e.r., 20 women vs. 77 men), the coefficients and statistical 5 278 significance of our estimates should be read with caution. For the statistical analysis, we 6 7 279 used STATA for Windows, version 13. 8 9 10 280 Results 11 12 13 281 Descriptive analysis 14 282 On average, survey respondents were 65.1 years old. We interviewed 20 women 15 16 283 and 77 men. As much as 40 of the informants had never attended school and 37 had 17 18 284 only completed primary school. From the people interviewed, 44 worked in the primary 19 20 285 sector and 53 in other economic sectors. The number of informants was similarly 21 22 286 distributed across ravines, except for Lanjarón where we could only conduct 11 23 24 287 interviews. 25 288 26 27 289 Sociodemographic correlates of proverb’s recognition 28 29 290 On average, 83.3% of informants recognized the 10 climatic proverbs in our list, 30 31 291 a percentage considerably higher than for proverbs referring to the physical (44.54%) 32 33 292 and the biological systems (36.29%) (Table 1). Indeed, some proverbs referring to the 34 35 293 climatic system were known by most people in the sample (i.e., proverbs Cli1, Cli7, 36 294 Cli9, and Cli10 were known by more than 95% of informants) and only one (Cli6) was 37 38 295 only known by 54% of the informants. Comparatively, the recognition of proverbs 39 40 296 referring to the physical system was lower. Thus, only one of the proverbs referring to 41 42 297 the physical system (Phy7) was known by all informants whereas the rest were known 43 44 298 by less than 60% of informants. One of the proverbs (Phy6) was only recognized by 45 46 299 14.4% of the informants, being the proverb least recognized. Only two of the proverbs 47 300 referring to the biological system (i.e., Bio6 and Bio9) were recognized by more than 48 49 301 half of the informants. Proverbs refereeing to the biological system generally displaying 50 51 302 the lowest level of recognition (Table 1). 52 53 54 303 INSERT TABLE 1 55 56 57 304 Proverb recognition varied across informants, some people recognizing more 58 305 proverbs than others. On average, informants recognized 16.4 out of 30 proverbs in our 59 60 306 survey (SD 5.36), although one person only recognized seven proverbs and another 61 62 10 63 64 65 307 person knew 29. The number of proverbs recognized differed between informants from 1 2 308 the five ravines, people in Bérchules recognizing more proverbs than people in the other 3 4 309 four ravines and people in Lanjarón displaying the lowest level of recognition, 5 310 particularly for proverbs referring to the physical and biological systems (Figure 2). 6 7 8 311 INSERT FIGURE 2 9 10 11 312 Results from multivariate analysis partly confirm these findings. Thus, when 12 13 313 using the number of proverbs recognized by the informant as a dependent variable, we 14 15 314 found that age was not associated in a statistically significant way with proverb 16 315 recognition. We found that people who had received school education recognized fewer 17 18 316 proverbs than people who had never gone to school and that informants in all other 19 20 317 ravines knew fewer proverbs than informants in the ravine of Bérchules (Table 2, 21 22 318 column 1). 23 24 25 319 INSERT TABLE 2 26 27 28 320 Different sociodemographic characteristics are associated to the recognition of 29 321 climatic, physical, and biological proverbs. Thus, older informants recognized more 30 31 322 climatic proverbs than younger informants (column 2), but age was not associated to the 32 33 323 recognition of proverbs referring to the physical or the biological systems. Women 34 35 324 recognized fewer proverbs referring to the climatic system than men (column 2), but 36 37 325 there were no statistically significant differences between women and men in the 38 326 recognition of proverbs referring to the physical (column 3) and the biological systems 39 40 327 (column 4). School attendance was not related to informant’s level of recognition of 41 42 328 climatic proverbs, but it was negatively associated to level of recognition of proverbs 43 44 329 related to the physical and biological systems. Informants working in the primary sector 45 46 330 recognized more proverbs related to the physical system than informants working in 47 48 331 other sectors. Finally, although people in Bérchules generally recognized more proverbs 49 332 than people in all the other ravines, regression results show that people in Poqueira 50 51 333 recognized more proverbs related to the climatic system than people in Bérchules (Table 52 53 334 2, column 2). 54 55 335 56 57 336 Sociodemographic correlates of proverbs’ accuracy 58 59 337 Overall, levels of proverb’s perceived accuracy are lower than levels of 60 338 proverb’s recognition. On average, 51.1% of the proverbs in the list were considered 61 62 11 63 64 65 339 accurate nowadays by people interviewed, while 36.6% of the proverbs were considered 1 2 340 not accurate, and 3.2% were considered accurate only sometimes. For 9.0% of the 3 4 341 proverbs an answer could not be provided due to uncertainty. In general, proverbs 5 342 referring to the climatic system were the ones considered less accurate (42.4% not 6 7 343 accurate and 6.3% sometimes accurate), followed by proverbs referring to the biological 8 9 344 system (33.2% not accurate and 1.4% sometimes accurate). On average, proverbs 10 11 345 referring to the physical system were considered the most accurate (55.4% accurate). 12 13 346 The climatic proverbs considered less accurate are Cli5, referring to extreme 14 15 347 variability in September rains (considered accurate by only 15.5% of informants), Cli1, 16 348 referring to how clouds shape can predict rain (21.7%), and Cli10, referring to how 17 18 349 weather in March and April influences flowering in May (23.7%). The proverbs 19 20 350 referring to the physical system considered less accurate were Phy3, which refers to the 21 22 351 time when snow can be found at different altitude levels (8.3%), Phy6, which relates to 23 24 352 the presence of fog during March with the risk of frost in May (9.3%), and Phy4, 25 353 referring to the frequency of frost during the month of May (41.2%). Finally, the 26 27 354 proverbs referring to the biological system perceived as less accurate were Bio4, 28 29 355 referring to the fructification of the chestnut tree (35.1%), Bio10, referring to the 30 31 356 relation between the flowering period of the olive tree and the female partridge’s 32 33 357 reproduction season (51.1%), and Bio8 referring to the flowering period of the genus 34 35 358 Carduus (52.2%). 36 359 As for the variable recognition, results from multivariate analysis examining the 37 38 360 sociodemographic correlates of accuracy also suggest that there is variation in the 39 40 361 sociodemographic characteristics of informants associated to the perceived accuracy of 41 42 362 proverbs (Table 3). Thus, when using as a dependent variable the total number of 43 44 363 proverbs that the person considers as inaccurate, we found that older informants 45 46 364 perceived as inaccurate more proverbs than younger informants. We also found that 47 365 men perceived as inaccurate more proverbs than women, and that people working in the 48 49 366 primary sector perceive more proverbs as inaccurate as people working in other sectors. 50 51 367 Schooling was not related to the total number of proverbs considered as inaccurate by 52 53 368 an informant, but ravine was: informants in Poqueira and Lanjarón considered as 54 55 369 inaccurate more proverbs than informants in the ravine of Bérchules, while informants 56 57 370 in La Taha considered as inaccurate less proverbs than informants in the ravine of 58 371 Bérchules (Table 3, column 1). 59 60 61 62 12 63 64 65 372 INSERT TABLE 3 1 2 3 373 The analysis by proverbs referring to different systems also suggests that 4 374 different sociodemographic characteristics are associated to the perception of accuracy 5 6 375 of climatic, physical, and biological proverbs. Age was related with a higher report of 7 8 376 inaccurate proverbs related to the climatic and the physical, but not to the biological 9 10 377 system (column 4). Women considered climatic (column 2) and biological (column 4), 11 12 378 but not physical (column 3) proverbs less inaccurate than men (Table 3). People who 13 14 379 had attended school reported less inaccuracy in proverbs related to the physical system, 15 380 and people working in the primary activity reported more inaccuracy on proverbs 16 17 381 related to the climatic and the biological systems. Finally, there was considerable 18 19 382 variation across ravines and types of proverbs (Table 3). 20 21 22 383 Discussion 23 24 384 In this work, we assess the level of recognition and accuracy of proverbs 25 26 385 referring to the climatic, physical, and biological systems in Sierra Nevada, an area 27 28 386 where people continue to depend on the environment for their livelihood and where 29 30 387 local climate change impacts have been documented. Before commenting on the main 31 32 388 results of the analysis, we signal two potential caveats of our work. First, our results 33 34 389 might suffer from omitted variable bias. For example, the mode of transmission, e.g., 35 whether the proverb was learned from elders or neighbors could be a relevant predictor 390 36 37 391 for our two dependent variables, recognition and perceived accuracy. However, we did 38 39 392 not collect information on how respondents came to learn about proverbs. Lack of 40 41 393 inclusion of this and other potentially relevant variables might bias our results. Second, 42 43 394 our results might also suffer from measurement error. Particularly, our measure of 44 45 395 accuracy refers only to the present-day, without questioning whether proverbs were 46 396 accurate at any point in the past. The difference in perceived past accuracy and present 47 48 397 accuracy of the selected proverbs could be a more powerful indicator if the results are to 49 50 398 be interpreted as perceptions of climate change. Further work in this line should aim to 51 52 399 address these caveats. 53 54 55 400 The first result from this work is that proverbs are generally well recognized 56 401 among people in our sample. Thus, while there is variation in the level of proverbs’ 57 58 402 recognition, most proverbs are recognized by more than half of the informants. 59 60 403 Moreover, according to both, the number of proverbs in our original list and the number 61 62 13 63 64 65 404 of proverbs recognized by survey respondents, proverbs related to the climatic system 1 2 405 seem to be predominant in the area. Thus, contrary to what studies in other regions of 3 4 406 Granada have shown (Benítez Cruz 2009), in the Alta Alpujarra Occidental 5 407 ethnoclimatological knowledge seems tightly linked to climatic elements such as 6 7 408 clouds, fog, or rainbows. Indeed, during our interviews, several informants mentioned 8 9 409 indicators of short-term weather forecast related to the presence of clouds or fog in 10 11 410 specific mountains. For example, one informant in La Taha said that “when the fog 12 13 411 stops at Coto Corona, it will rain” and another informant from Trevélez said that “when 14 15 412 there is fog in Piedraventana, there will be a cold rain the following way”. Furthermore, 16 413 during the surveys, informants mentioned local climatic proverbs not included in our 17 18 414 original list (see Supplementary Material 3). Our results then suggest that climate- 19 20 415 related proverbs, continue to be recognized in the study area. 21 22 23 416 The second finding from this work is that many informants consider that 24 25 417 proverbs, and particularly climate-related proverbs, are not accurate nowadays. It should 26 418 be noted that, as we only asked about present accuracy, our data do not allow us to 27 28 419 discern between proverbs which have become inaccurate and proverbs which might 29 30 420 have never been accurate in the area. However, while conducting interviews, some 31 32 421 informants made statements such as: “I no longer pay attention to water signals because 33 34 422 they are no longer credible”, or “In the past, cattle used to announce the rain; but now 35 423 they only know when it rains after they get wet, as rain now is unpredictable”, which 36 37 424 suggest indeed a general loss of accuracy of traditional weather forecasting methods. 38 39 425 Moreover, the lack of reported accuracy of several of the proverbs selected often 40 41 426 goes in line with trends documented by the Observatorio de Cambio Global de Sierra 42 43 427 Nevada (Zamora Rodríguez et al. 2015; Pérez-Luque et al. 2016) and other trends 44 45 428 reported in local, regional, and national scientific literature. Thus, for 19 out of the 30 46 429 selected proverbs, we find agreement between proverbs’ perceived accuracy and 47 48 430 indicators of climate change reported in the scientific literature. For example, lack of 49 50 431 accuracy of proverbs indicating changes in precipitation regimes (Cli1, Cli2, Cli3), the 51 52 432 start of the snow period (Phy3), frequency of frost events (Phy4, Phy 5, Phy 6), or 53 54 433 blackbird (Turdus merula) abundance (Bio1), match well with scientific reports for the 55 56 434 area (see Supplementary Material 4). 57 435 For seven out of the 30 proverbs studied we found disagreement between the 58 59 436 two sources of information. For example, the scientific literature available reports an 60 61 62 14 63 64 65 437 increase of seasonal temperatures while less than 50% of informants consider proverbs 1 2 438 related to these trends as inaccurate (e.g., Cli4, Cli8 and Cli9). These discrepancies, 3 4 439 however, deserve further analysis. For three out of the seven proverbs in which there is 5 440 no agreement (Phy2, Phy5 and Bio3), the comparison had to be done with scientific 6 7 441 literature from outside Spain due to lack of more local reports, which weakens the term 8 9 442 of the comparison. Similarly, at least for one of the proverbs (i.e., Phy9 referring to soil 10 11 443 humidity), the mismatch could be explained by the use of different scales of 12 13 444 observation, as the reference used (i.e., Oliva et al. 2014) studied soil temperature at 14 15 445 higher altitudes (e.g., above 3000 m.a.s.l) than the villages surveyed. 16 446 Finally, for five of the proverbs in our list, we did not find any scientific source 17 18 447 reporting the same change. This is the case, for example, for proverbs referring to 19 20 448 changes in cultivated species fruiting times (Bio2 and Bio4). Although some studies 21 22 449 have shown the relation between fruiting times and climate change (Lebourgeois et al. 23 24 450 2018), such type of work has not been conducted in our study area. All in all, 25 451 mismatches and lack of correspondence between information from the two knowledge 26 27 452 systems could be then considered as potential entry points for additional research (e.g., 28 29 453 Marin 2010; Tengö et al. 2014). 30 31 454 The third important finding of this work relates to variation in level of proverb’s 32 33 455 recognition and perceived accuracy across informants. Consistent with previous work, 34 35 456 our study generally shows that individual characteristics are associated to proverb 36 457 recognition and perceived accuracy (Uekermann et al. 2008; Je and Wright-Harp 2011). 37 38 458 From the several association found, some deserve further discussion. First, we found 39 40 459 that respondents with formal schooling have a lower level of proverb’s recognition, a 41 42 460 finding that dovetails with work suggesting that formal education is not complementary 43 44 461 to forms of traditional knowledge, particularly when traditional knowledge is not taught 45 46 462 at schools (Reyes-García et al. 2010; McCarter et al. 2014). Second, our results suggest 47 463 that older people perceive more proverbs as inaccurate than younger people, probably 48 49 464 because age give a longer basis for comparison (Fernández-Llamazares et al. 2015) and 50 51 465 a more concrete comprehension of proverbs (Uekermann et al. 2008). Finally, our 52 53 466 results also suggest that people who work in the primary sector perceive more proverbs 54 55 467 as inaccurate, probably because their dependence from and daily interaction with the 56 57 468 environment makes them more aware of changes. 58 59 60 61 62 15 63 64 65 469 In the same line, a consistent finding in our analysis is that proverb’s recognition 1 2 470 varies between ravines, with Lanjarón displaying the lowest level of proverb 3 471 recognition. While the difference could be attributed to sampling issues (i.e., we had a 4 5 472 smaller sample size for Lanjarón), it could also be explained by different levels of 6 7 473 rurality. Lanjarón is the ravine with the highest percentage of cultivated area (22.7%), 8 9 474 largely dominated by olive and almond trees, as well as the ravine with the highest 10 11 475 conversion to the tertiary sector, with a 65.4% higher occupation hotel capacity than 12 13 476 Poqueira, the second most touristic ravine (INE 2017). Lanjarón has diversified also its 14 477 economy to the secondary sector, locating the fourth biggest company of mineral water 15 16 478 of Spain. It is also the first village of the mountain range, closer to Granada (province´s 17 18 479 capital) and the seaport. In contrast, Bérchules, the ravine with the highest level of 19 20 480 proverb recognition, is the ravine with the highest percentage of herbaceous crops 21 22 481 cultivated area (2.6%). 23 24 25 482 Conclusion 26 483 In this work, we explore whether people in Sierra Nevada recognize climate- 27 28 484 related proverbs and consider them accurate. We found that climate-related proverbs are 29 30 485 abundant and relatively well maintained in the study area, although informants generally 31 32 486 consider that many of them are not accurate. We found both alignments and mismatches 33 34 487 between people’s perceptions of proverbs accuracy and information provided by 35 488 scientific assessments of climate change impacts in the area. While our findings are 36 37 489 limited to a handful of proverbs, the evidence is enough to suggest that the study of 38 39 490 local perceptions of proverbs accuracy could enrich our understanding of climate 40 41 491 change impacts. Particularly, the identification of mismatches and discrepancies 42 43 492 between people’s perceptions and scientific assessments should be used, not to dismiss 44 45 493 one of the two bodies of knowledge, but rather to guide future research. 46 494 47 48 495 Acknowledgments 49 50 51 496 Research leading to this work has received funding from the European Research 52 53 497 Council (ERC) under grant agreement No 771056-LICCI-ERC-2017-COG and from the 54 55 498 Spanish government through a grant of the Ministry of Economy and Competitiveness 56 (CSO2014-59704-P). García-del-Amo and Reyes-García acknowledge financial support 499 57 58 500 from the Spanish Ministry of Economy and Competitiveness, through the “María de 59 60 501 Maeztu” Programme for Units of Excellence in R&D (MdM-2015-0552). We thank 61 62 16 63 64 65 502 researchers from the Observatorio del cambio global de Sierra Nevada and all the people 1 2 503 who kindly collaborated with us during fieldwork. 3 4 504 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 17 63 64 65 505 References 1 2 506 Bambaige, A. (2007). National Adaptation Strategies to Climate Change Impacts. A 3 507 Case Study of Mozambique. Human development report, 2008. 4 5 508 Barnes J, Dove M, Lahsen M, Mathews A, McElwee P, McIntosh R, Moore F, O’Reilly 6 7 509 J, Orlove B, Puri R, Weiss H, Yager K (2013) Contribution of anthropology to the 8 510 study of climate change. Nat Clim Chang 3:541–544. doi: 10.1038/nclimate1775 9 10 511 Benítez Cruz G (2009) Etnobotánica y etnobiología del poniente granadino. 11 12 512 Cameron AC, Trivedi PK (1998). Regression analysis of count data. Cambridge 13 513 University Press. 14 15 514 Castillo Martín A (1999) Aguas y acequias en La Alpujarra (Sierra Nevada). In: II 16 17 515 Conferencia de La Alpujarra. Ed Rosua y Catedra UNESCO, Granada, pp 151–164 18 19 516 Cozar Valero ME (2000) Los recursos humanos en un espacio natural protegido: Sierra 20 517 Nevada. Cuad geográficos la Univ Granada 30:325–346 21 22 518 Enock CM (2013) Indigenous Knowledge Systems and Modern Weather Forecasting: 23 519 Exploring the Linkages. J Agric Sustain 2(2) 24 25 520 Fernández-Llamazares Á, Díaz-Reviriego I, Luz AC, Cabeza M, Pyhälä A, Reyes- 26 27 521 García V (2015) Rapid ecosystem change challenges the adaptive capacity of local 28 522 environmental knowledge. Glob Environ Chang 31:. doi: 29 523 10.1016/j.gloenvcha.2015.02.001 30 31 524 Galacgac, E. S., & Balisacan, C. M. (2009) Traditional weather forecasting for 32 525 sustainable agroforestry practices in Ilocos Norte Province, Philippines. Forest 33 34 526 Ecology and Management, 257(10), 2044-2053. 35 36 527 Hong J, Park C, Shin Y (2016) Protocol to Survey Proverbs, Old Sayings and 37 528 Community Rules Related to Ecosystem Services in ASEAN Countries. In: 38 529 Extended Extracts of Workshop on Traditional Knowledge for Ecosystem Services 39 530 in ASEAN Countries – Folk Culture: Proverbs, Old Sayings and Community 40 41 531 Rules. APAFRI, Kuala Lumpur, Malaysia 42 43 532 Huntington HP (2000) Using traditional ecological knowledge in science: methods and 44 533 applications. Ecol Appl 10:1270–1274. doi: 10.1890/1051- 45 534 0761(2000)010[1270:UTEKIS]2.0.CO;2 46 47 535 Institutio Nacional de Estadística (INE) (2017) Andalucía pueblo a pueblo. Fichas 48 536 municipales. 49 50 537 http://www.juntadeandalucia.es/institutodeestadisticaycartografia/sima/ficha.htm? 51 538 mun=18901. Accessed 14 May 2019 52 53 539 Iniesta-Arandia I, del Amo DG, García-Nieto AP, Piñeiro C, Montes C, Martín-López B 54 540 (2015) Factors influencing local ecological knowledge maintenance in 55 541 Mediterranean watersheds: Insights for environmental policies. Ambio 44:285– 56 57 542 296. doi: 10.1007/s13280-014-0556-1 58 59 543 IPCC (2014) Climate Change 2014: Synthesis Report. Contribution of Working Groups 60 544 I, II, and III to the Fifth Assessment Report of the Intergovernmental Panel on 61 62 18 63 64 65 545 Climate Change. Geneva, Switzerland 1 2 546 Je B, Wright-Harp W (2011) Cultural and Generational Factors Influencing Proverb 3 547 Recognition. Contemp Issues Commun Sci Disord 38:26–35 4 5 548 Kagunyu A, Wandibba S, Wanjohi JG (2016) The use of indigenous climate forecasting 6 549 methods by the pastoralists of Northern Kenya. Pastoralism 6:7. doi: 7 8 550 10.1186/s13570-016-0054-0 9 10 551 Kanno H, Sakura I T, Shinjo H,Miyazaki H, Ishimoto Y, Saeki T, Umetsu C, Sokotela 11 552 S, Chiboola M (2013) Indigenous Climate Information and Modern Meteorological 12 553 Records in Sinazongwe District, Southern Province, Zambia. Japan Agric Res Q 13 JARQ 47:191–201. doi: 10.6090/jarq.47.191 14 554 15 16 555 King DNT, Skipper A, Tawhai WB (2008) Māori environmental knowledge of local 17 556 weather and climate change in Aotearoa – New Zealand. Clim Change 90:385– 18 557 409. doi: 10.1007/s10584-007-9372-y 19 20 558 Kurien J (1998) Traditional ecological knowledge and ecosystem sustainability: new 21 559 meaning to asian coastal proverbs. Ecol Appl 8:S2–S5. doi: 10.1890/1051- 22 23 560 0761(1998)8[S2:TEKAES]2.0.CO;2 24 25 561 Lebourgeois F, Delpierre N, Dufrêne E, Cecchini S, Macé S, Croisé L, Nicolas M 26 562 (2018) Assessing the roles of temperature, carbon inputs and airborne pollen as 27 563 drivers of fructification in European temperate deciduous forests. European 28 564 Journal of Forest Research, 137(3), 349-365. doi: 10.1007/s10342-018-1108-1 29 30 565 31 32 566 Lefale PF (2010) Ua ‘afa le Aso Stormy weather today: traditional ecological 33 567 knowledge of weather and climate. The Samoa experience. Clim Change 100:317– 34 568 335. doi: 10.1007/s10584-009-9722-z 35 36 569 Lorite J, Navarro FB, Valle F (2007) Estimation of threatened orophytic flora and 37 570 priority of its conservation in the Baetic range (S. Spain). Plant Biosyst - An Int J 38 39 571 Deal with all Asp Plant Biol 141:1–14. doi: 10.1080/11263500601153560 40 41 572 Marin A (2010) Riders under storms: Contributions of nomadic herders’ observations to 42 573 analysing climate change in Mongolia. Glob Environ Chang 20:162–176. doi: 43 574 10.1016/J.GLOENVCHA.2009.10.004 44 45 575 McCarter J, Gavin MC, Baereleo S, Love M (2014) The challenges of maintaining 46 576 indigenous ecological knowledge. Ecol Soc 19:art39. doi: 10.5751/ES-06741- 47 48 577 190339 49 50 578 Mesa S, Delgado AB, Blanco E (1997) Ritos de lluvia y predicción del tiempo en la 51 579 España mediterránea. In: Goloubinoff M, Katz E, Lammel A (eds) Antropología 52 580 del clima en el mundo hispanoamericano. Abya Yala, Quito, Ecuador, p 93 53 54 581 Oliva M, Ortiz AG, Salvador F, Salvà M, Pereira P, Geraldes M (2014) Long-term soil 55 56 582 temperature dynamics in the Sierra Nevada, Spain. Geoderma, 235, 170-181. doi: 57 583 10.1016/j.geoderma.2014.07.012 58 59 584 60 61 62 19 63 64 65 585 Orlove BS, Chiang JCH, Cane MA (2000) Forecasting Andean rainfall and crop yield 1 586 from the influence of El Niño on Pleiades visibility. Nature 403:68–71. doi: 2 587 10.1038/47456 3 4 588 Parrotta, JA, Agnoletti M (2012). Traditional forest-related knowledge and climate 5 6 589 change. In Traditional Forest-Related Knowledge, Springer, Dordrecht. pp. 491- 7 590 533. 8 9 591 Pérez-Luque AJ, Bonet FJ, Zamora R, Barea-Azcón JM, Aspizua R, Sánchez-Gutiérrez 10 592 FJ (2016) Señales del cambio global en el sitio LTER-Sierra Nevada. Rev 11 593 Ecosistemas 25:65–71. doi: 10.7818/RE.2014.25-1.00 12 13 594 Planton S, Lionello P,Artale V, Aznar R, Carrillo A, Colin J, Congedi L, Dubois C, 14 15 595 Elizalde A, Gualdi S, Hertig E, Jacobeit J, Jordà G, Li L, Mariotti A, Piani C, Ruti 16 596 P, Sánchez-Gómez E, Sannino G, Sevault F, Somot S, Tsimplis M (2012) The 17 597 Climate of the Mediterranean Region in Future Climate Projections. Clim Mediterr 18 598 Reg 449–502. doi: 10.1016/B978-0-12-416042-2.00008-2 19 20 599 Reyes-García V, Fernández-Llamazares Á, Guèze M, Garcés A, Mallo M, Vila-Gómez 21 22 600 M, Vilaseca M (2016) Local indicators of climate change: the potential 23 601 contribution of local knowledge to climate research. Wiley Interdiscip Rev Clim 24 602 Chang 7:109–124. doi: 10.1002/wcc.374 25 26 603 Reyes-García V, Fernández-Llamazares À, Gueze M, Gallois S (2018) Does weather 27 forecasting relate to foraging productivity? An empirical test among three hunter- 604 28 29 605 gatherer societies. Weather Clim Soc 10:. doi: 10.1175/WCAS-D-17-0064.1 30 31 606 Reyes-García V, García-del-Amo D, Benyei P, D. García-del-Amo, P. Benyei, Á. 32 607 Fernández-Llamazares, K. Gravani, AB. Junqueira, V Labeyrie, X Li, DM. Matias, 33 608 A McAlvay, PG Mortyn, A Porcuna-Ferrer, A Schlingmann, R Soleymani-Fard 34 609 (2019) A collaborative approach to bring insights from local observations of 35 36 610 climate change impacts into global climate change research. Curr Opin Environ 37 611 Sustain. 39: 1-8. 2019. 38 39 612 Reyes-García V, Kightley E, Ruiz-Mallén I, Fuentes-Peláez N, Demps K, Huanca T, 40 613 Martínez-Rodríguez MR (2010) Schooling and local environmental knowledge: 41 614 Do they complement or substitute each other? Int J Educ Dev 30:. doi: 42 43 615 10.1016/j.ijedudev.2009.11.007 44 45 616 Rosenzweig C, Neofotis P (2013) Detection and attribution of anthropogenic climate 46 617 change impacts. Wiley Interdiscip Rev Clim Chang 4:121–150. doi: 47 618 10.1002/wcc.209 48 49 619 Savo V, Lepofsky D, Benner JP, Kohfeld KE, Bailey J, Lertzman K (2016) 50 620 Observations of climate change among subsistence-oriented communities around 51 52 621 the world. Nat Clim Chang 6:462–473. doi: 10.1038/nclimate2958 53 54 622 Sterling EJ, Filardi C, Toomey A, Sigouin A, Betley E, Gazit N, Newell J, Albert S, 55 623 Alvira D, Bergamini N, Blair M, Boseto D, Burrows K, Bynum N, Caillon S, 56 624 Caselle JE, Claudet J, Cullman G, Dacks R, Eyzaguirre PB, Gray S, Herrera J, 57 625 Kenilorea P, Kinney K, Kurashima N, Macey S, Malone C, Mauli S, McCarter J, 58 59 626 McMillen H, Pascua P, Pikacha P, Porzecanski AL, de Robert P, Salpeteur M, 60 627 Sirikolo M, Stege MH, Stege K, Ticktin T, Vavev R, Wali A, West P, Winter KB, 61 62 20 63 64 65 628 Jupiter SD (2017) Biocultural approaches to well-being and sustainability 1 629 indicators across scales. Nat Ecol Evol 1:1798–1806. doi: 10.1038/s41559-017- 2 630 0349-6 3 4 631 Tengö M, Brondizio ES, Elmqvist T, Malmer P, & Spierenburg M (2014) Connecting 5 6 632 diverse knowledge systems for enhanced ecosystem governance: the multiple 7 633 evidence base approach. Ambio 43(5), 579-591. 8 9 634 Uekermann J, Thoma P, Daum I (2008) Proverb interpretation changes in aging. Brain 10 635 Cogn 67:51–57. doi: 10.1016/j.bandc.2007.11.003 11 12 636 Weatherhead E, Gearheard S, Barry RG (2010) Changes in weather persistence: Insight 13 from Inuit knowledge. Glob Environ Chang 20:523–528. doi: 14 637 15 638 10.1016/J.GLOENVCHA.2010.02.002 16 17 639 Zamora Rodríguez RJ, Pérez-Luque AJ, Bonet FJ, Barea-Azcón JM, Aspizua R (2015) 18 640 La huella del cambio global en Sierra Nevada: Retos para la conservación. 19 641 Consejería de Medio Ambiente y Ordenación del Territorio. Junta de Andalucía 20 21 642 Zuma-Netshiukhwi G, Stigter K, Walker S (2013) Use of Traditional Weather/Climate 22 23 643 Knowledge by Farmers in the South-Western Free State of South Africa: 24 644 Agrometeorological Learning by Scientists. Atmosphere (Basel) 4:383–410. doi: 25 645 10.3390/atmos4040383 26 27 646 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 21 63 64 65 647 Figure captions 1 2 648 Fig. 1 Map of the study area 3 4 Fig. 2 Percentage of proverbs recognized in the five ravines, per category 649 5 6 7 650 Tables 8 9 651 Table 1 Recognition and accuracy of proverbs included in our survey instrument 10 11 652 Table 2 Results of Poisson regression showing sociodemographic correlates of proverb 12 13 653 recognition 14 15 654 Table 3 Results of Poisson regression showing sociodemographic correlates of proverb 16 655 accuracy 17 18 19 656 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 22 63 64 65 657 Fig. 1 Map of the study area 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 658 55 56 57 659 58 59 60 61 62 23 63 64 65 660 Fig. 2 Percentage of proverbs recognized in the five ravines, per category 1 2 661 3 4 5 100 90 6 90 84.35 82.86 7 80.91 78.26 8 80 9 10 70 61.88 11 58.95 60 54.35 55.71 % TOTAL 12 51.05 46.96 47.54 42.86 % Clim 13 50 45.45 14 41.43 36.52 35.79 % Phy 15 40 32.73 16 27.83 % Bio 30 22.73 17 18 20 19 20 10 21 0 22 Bérchules Trevélez La Taha Poqueira Lanjarón 23 662 24 25 663 26 27 28 664 29 30 665 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 24 63 64 65 666 Table 1 Recognition and accuracy of proverbs included in our survey instrument 1 2 Code Proverb Recognition Accuracy (%) 3 (%) Yes No Sometimes Don’t 4 know 5 [Cli1] 96.91 21.65 68.04 6.19 4.12 Cielo aborregado, a los tres días 6 remojado 7 Fleecy sky, in three days soaked 8 [Cli2] Cuando marzo mayea, mayo marcea 91.75 42.27 35.05 12.37 24.74 9 When March feels like May, May 10 feels like March 11 12 [Cli3] Mañanitas nubladas, tardes 62.88 49.48 23.71 17.53 12.37 13 paseadas Cloudy mornings, sunny evenings 14 15 [Cli4] En febrero busca la sombra el perro 93.81 52.58 42.27 4.12 1.03 16 y en marzo el perro y el amo 17 In February the dog looks for shade, 18 and in March the dog and the master 19 [Cli5] Septiembre o lleva los puentes o 60.82 15.46 79.38 3.09 2.06 20 seca las fuentes 21 September either carries bridges or 22 dries up fountains 23 [Cli6] Cuando los pájaros van en banda, 53.61 68.04 11.34 0 20.62 24 los hombres de parranda 25 When birds flock together, men go 26 on spree (a flock of birds flying 27 together is an indicator of heavy 28 rains) 29 [Cli7] Febrerillo el loco, sacó a su padre al 98.97 70.1 26.8 2.06 1.03 30 sol y lo apedreó 31 Crazy February, took his father to 32 sunbathe and stoned him (meaning 33 that weather in February is very 34 variable, in which sunny days can 35 alternate with hail) 36 [Cli8] Agosto, frío en rostro 78.35 62.89 30.93 5.15 1.03 37 August, cold in the face 38 [Cli9] Septiembre, el que no tenga ropa 95.88 52.58 38.14 9.28 0 39 que tiemble 40 If you don’t have clothes in 41 September, be ready to tremble 42 [Cli10] Marzo ventoso y abril lluvioso sacan 100 23.71 70.1 6.19 0 43 a mayo florido y hermoso 44 March winds and April showers 45 bring May flowers 46 83.3 45.88 42.37 6.29 5.46 AVERAGE CLIMATIC 47 48 [Phy1] Enero claro y heladero 56.7 76.29 21.65 2.06 0 49 January, clear sky and frosty 50 [Phy2] Es preciso prevenir la helada de 42.27 60.82 37.11 2.06 0 51 abril 52 It is necessary to prevent April frosts 53 [Phy3] Por Todos los Santos la nieve en los 59.79 8.25 90.72 1.03 0 54 altos, por San Andrés la nieve en los 55 pies 56 For All Saints Day (November 1st ), 57 snow in the heights, for San Andrés 58 (November 30th ) snow in the feet 59 [Phy4] Hiela en mayo hasta el día 10, 35.05 41.24 54.64 3.09 1.03 60 alguna que otra vez 61 62 25 63 64 65 It can freeze until May 10th , but 1 rarely 2 [Phy5] Pues que en marzo siempre hiela, 28.87 68.04 28.87 3.09 0 3 que se espere cuanto pueda 4 As in March it always freezes, it is 5 better if the freeze comes late. 6 [Phy6] Niebla de marzo, helada en mayo 14.43 9.28 17.53 1.03 72.16 7 Fog in March, frosts in May 8 [Phy7] Año de nieves, año de bienes 100 86.6 12.37 1.03 0 9 Snowy year, bumper harvest year 10 [Phy8] Hielos en la cruz de mayo siempre 49.48 45.36 51.55 3.09 0 11 hacen daño 12 Frosts in the May 3rd (La Cruz de 13 mayo) are always harmful 14 [Phy9] Lluvia o nieve de febrero, el mejor 35.05 64.95 20.62 3.09 11.34 15 estercolero 16 February rains or snows, the best 17 manure 18 [Phy10] Arbolitos que estáis en flor libraos 23.71 92.78 7.22 0 0 19 de los bruscos de la Encarnación 20 Blooming trees, avoid the freeze of 21 the Encarnación (March 25th ) 22 AVERAGE PHYSICAL 44.54 55.36 34.23 1.96 8.45 23 [Bio1] A higuera sin higos no acuden 19.59 74.23 23.71 0 2.06 24 25 mirlos 26 In a fig tree without figs, there are no 27 common blackbirds (Turdus merula) 28 [Bio2] La noche de San Juan cuaja la 31.96 63.92 16.49 3.09 16.49 29 almendra y la nuez 30 In San Juan’s night (June 23rd), the 31 almond and walnut curd 32 [Bio3] El cuco a 10 ó 15 de abril; si no 48.45 51.55 42.27 0 6.19 33 está, malo o se va a morir 34 If the common cuckoo (Cuculus 35 canorus) has not arrived by April 36 10th or 15th is because it is sick or 37 dead 38 [Bio4] Por San Cipriano castaña en mano 19.59 35.05 60.82 3.09 1.03 39 In San Cipriano (September 16th ), 40 chestnut in hand 41 [Bio5] Al final de abril en flor la vid 41.24 61.86 21.65 3.09 13.4 42 The vine should bloom at the end of 43 April. 44 [Bio6] Entre San Pedro y San Juan las 52.58 83.51 15.46 1.03 0 45 hierbas olores dan 46 In between San Pedro (June 29th ) and 47 San Juan (June 23rd), the herbs smell 48 [Bio7] Los estorninos cien visitas diarias 25.77 42.27 49.48 0 8.25 49 hacen a los olivos pero acaba la 50 aceituna ninguna 51 The common starling (Sturnus 52 vulgaris) visit olive trees a hundred 53 times per day, but only when there 54 are olives in the tree. 55 [Bio8] Cuando vienen los vilanos es 17.53 34.02 26.8 3.09 36.08 56 conclusión del verano 57 With the appearance of the Carduus 58 flower, summer is over 59 [Bio9] Por enero, florece el romero 79.38 56.7 25.77 1.03 16.49 60 In January the Rosemary blooms 61 62 26 63 64 65 [Bio10] Cuando el olivo empieza a cernir se 26.8 18.56 49.48 0 31.96 1 caza la perdiz 2 When the olive trees start to release 3 the flower pollen, it is time to hung 4 partridges. 5 AVERAGE BIOLOGICAL 36.29 52.16 33.2 1.44 13.2 6 TOTAL AVERAGE 54.71 51.13 36.6 3.23 9.04 7 8 667 9 10 668 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 27 63 64 65 669 Table 2 Results of Poisson regression showing sociodemographic correlates of proverb 1 670 recognition (n=97) 2 3 4 (1) (2) (3) (4) 5 All Climatic Physical Biological 6 proverbs proverbs proverbs proverbs 7 8 9 Age 0.0024 0.0030*** 0.0007 0.0031 10 (0.0019) (0.0008) (0.0025) (0.0039) 11 Woman -0.0039 -0.0365*** -0.0536 0.1267 12 (0.0329) (0.0099) (0.0460) (0.1178) 13 Schooling -0.1936 *** 0.0057 -0.4568*** -0.3453*** 14 15 (0.0276) (0.0468) (0.0068) (0.0033) 16 Primary 0.0536 -0.0040 0.1649** 0.0610 17 sector (0.0611) (0.0255) (0.0781) (0.1281) 18 Municipality (excluded Bérchules) 19 Trevélez -0.3419*** -0.0737*** -0.5911*** -0.6607*** 20 21 (0.0122) (0.0181) (0.0033) (0.0004) 22 La Taha -0.2042*** -0.0584*** -0.4167*** -0.2747*** 23 (0.0204) (0.0009) (0.0416) (0.0566) 24 Poqueira -0.1328 *** 0.0347*** -0.2416*** -0.3777*** 25 26 (0.0027) (0.0033) (0.0166) (0.0097) *** 27 Lanjarón -0.3717 -0.0888*** -0.5930*** -0.8040*** 28 (0.0028) (0.0008) (0.0130) (0.0080) 29 _cons 2.9116 *** 1.9629*** 1.9587*** 1.5718*** 30 (0.1711) (0.0927) (0.1893) (0.3212) 31 671 Standard errors in parentheses 32 * 672 p < 0.10, ** p < 0.05, *** p < 0.01 33 34 673 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 28 63 64 65 674 Table 3 Results of Poisson regression showing sociodemographic correlates of proverb 1 675 inaccuracy (n=97) 2 3 4 (1) (2) (3) (4) 5 All proverbs Climatic proverbs Physical Biological proverbs 6 proverbs 7 8 Age 0.0038*** 0.0075*** 0.0112*** -0.0072 9 10 (0.0010) (0.0027) (0.0009) (0.0078) *** 11 Woman -0.1329 -0.2228*** -0.0074 -0.1378*** 12 (0.0179) (0.0345) (0.0884) (0.0027) 13 Schooling -0.0613 -0.0718*** -0.0181 -0.0819 14 15 (0.0828) (0.0263) (0.1602) (0.1308) *** 16 Primary Sector 0.1096 0.2037*** 0.0571 0.0634*** 17 (0.0071) (0.0036) (0.0358) (0.0044) 18 Municipality (excluded Bérchules) 19 Trevélez 0.0007 -0.1945*** -0.3349*** 0.3757*** 20 21 (0.0329) (0.0100) (0.0613) (0.0579) *** 22 La Taha -0.0952 -0.2520*** 0.2100*** -0.2734*** 23 (0.0119) (0.0315) (0.0362) (0.0641) 24 Poqueira 0.1443 *** 0.0329* 0.3384*** 0.0962*** 25 26 (0.0136) (0.0200) (0.0213) (0.0194) *** 27 Lanjarón 0.2328 0.1132*** 0.6986*** -0.3578*** 28 (0.0024) (0.0362) (0.0035) (0.0751) 29 _cons 2.1167 *** 1.0152*** 0.2949** 1.6552*** 30 (0.1202) (0.1757) (0.1415) (0.5651) 31 32 N 97 97 97 97 33 676 Standard errors in parentheses * 34 677 p < 0.10, ** p < 0.05, *** p < 0.01 35 678 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 29 63 64 65 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 7 63 64 65 Responses from Blanca Ramos from the Observatorio de Sierra Nevada to our questions 1 on their level of involvement with local communities and dissemination strategy 2 3 Contesto a tus preguntas: 4 5 6 1. "Participan las comunidades locales en la recolección de datos del Observatorio?" - 7 Con carácter general no, aunque hay una actividad concreta del Observatorio, el 8 seguimiento de las mariposas diurnas, que se desarrolla con una importantísima 9 10 participación de un grupito de voluntarios que son residentes en el entorno geográfico 11 próximo a Sierra Nevada. Todos los años realizan transectos cada 15 días desde que las 12 mariposas inician los vuelos (desde marzo-abril) hasta que llega el otoño y dejan de 13 estar activas por el frío, allá por septiembre-octubre. Al final del invierno o principios 14 de la primavera organizamos una reunión en la que exponen sus resultados y 15 16 aprovechamos para invitar a algún profesor o experto que nos dé una charla magistral. 17 Dura un día completo y es de lo más gratificante, pues todos aprendemos mucho de 18 un tema que nos apasiona. Este año la celebraremos el día 14 de marzo y nos 19 acompañará una representante de un programa internacional de seguimiento de 20 mariposas con el que Sierra Nevada colabora desde hace años. El resto de los datos del 21 Observatorio que están inlcuidos en el Manual de Metodologías (que seguramente 22 23 conoces) los levantan los técnicos y auxiliares de la Agencia de Medio Ambiente y Agua 24 (la empresa pública medioambiental de la Junta de Andalucía) y nuestros Agentes y 25 Celadores, muchos de los cuales son oriundos de la zona. También los científicos de las 26 Universidades de Granada y Almería contribuyen recabando datos y especialmente 27 con sus análisis y modelos. 28 29 2. "Son las campañas de investigación visibles para las comunidades locales?"Los 30 resultados de investigación son objeto de gran interés por parte de los medios de 31 comunicación locales (sobre todo Granada Hoy e Ideal de Granada), que con 32 frecuencia sacan dobles páginas a color sobre el cambio global, las especies de fauna y 33 flora más sobresalientes o sobre temas de conservación de la naturaleza. Otra fuente 34 de divulgación es la difusión de trabajos como "La Huella del Cambio Global". Es 35 36 difundida entre los componentes del Consejo de Participación del Espacio Natural de 37 Sierra Nevada, compuesto por representantes de las administraciones con 38 competencias en la zona, incluyendo las locales (Ayuntamientos y Diputaciones) y a 39 todo aquel que nos lo pide. La difusión ya dentro de sus respectivos ámbitos ya es muy 40 variable, como puede suponer. La página web del Observatorio dejó de funcionar bien 41 42 hace algunos años pero se está trabajando activamente por recuperarla. Ya se puede 43 acceder a algunos contenidos (http://obsnev.es/) y es de suponer que con el paso del 44 tiempo mejore su accesibilidad y la información que contenga. También hay mucha 45 conexión en las redes sociales, aunque de este tema yo tengo poca información 46 porque me salen granos solo de oir hablar de ellas (las odio). A lo mejor otras personas 47 te pueden informar mejor que yo. Otra cosa que hacemos con frecuencia es ir a 48 49 Colegios de Infantil y Primaria y a Institutos de Bachillerato a contar nuestra 50 experiencia de gestión, y aprovechamos la ocasión para contarles los resultados más 51 interesantes de los trabajos de Observatorio y en general de las investigaciones 52 científicas que se realizan aquí. También recibimos cada año un número variable 53 (entre 3 y 10) de alumnos que realizan sus prácticas de Grado o de Máster con 54 55 nosotros. Se les incorpora a los equipos de segumiento con toma de datos en campo y 56 se les familiariza con la búsqueda de información, los instrumentos de ordenación del 57 espacio, la legislación ce conservación de la biodiverisdad, etc. 58 3. "La data que se recoge se difunde ampliamente entre las comunidades?" Los datos 59 como tales no se difunden ampliamente. Se difunden mucho más los resultados de los 60 61 62 8 63 64 65 análisis de los datos. No obstante, el repositorio de datos es público puesto que se 1 financia con fondos públicos y eso lo tenemos muy claro. Están disponibles para quien 2 los solicite, pero a demanda. También estamos divulgando las series de datos 3 mediante lo que se conoce como "data papers", es decir se publican en revistas 4 5 formato Open Access el conjunto de datos en bruto sobre temas específicos durante 6 un período amplio de tiempo. Por ejemplo, acabamos de enviar un data paper sobre 7 los datos de abundancia, distribución, estructura poblacional y ecología de la cabra 8 montés (Capra pyrenaica), que es exclusivamente propia de la Península Ibérica, en 9 Sierra Nevada, donde la especie presenta la población más abundante y mas diversa 10 11 genéticamente de toda su área de distribución. Por ejemplo si un investigador de Irán 12 quiere hacer un trabajo comparativo entre las cabras de su país y las de España, podrá 13 hacerlo contando con los datos de Sierra Nevada sin moverse de su silla. Finalmente te 14 señalo que hay datos que por ser sensibles no se pueden divulgar así como así, por 15 ejemplo la localización exacta de rapaces o de especies sensibles, pues esa 16 información las puede poner en peligro. Hay mecanismos para que esa información 17 18 sensible sea manejada por gente que no la va a usar para fines inconfesables. 19 20 Sobre lo que preguntas relativo a si "el contacto con el observatorio pueda influir en la 21 percepción local del cambio climático", francamente no tengo argumentos para decirte 22 23 si si o si no. A nosotros nos llegan muchos comentarios sobre artículos que la gente lee 24 con mucho interés en el periódico y otros medios de comunicación y también nos piden 25 información adicional. Pero de ahí a "influir en la percepción local de cambio 26 climático", ya es más difícil de medir. Mi impresión es que poco a poco, con lo que se 27 oye en la tele y en la radio, con lo que se ve en prensa, con lo que los propios 28 29 ciudadanos perciben en carne propia y con las consecuencias devastadoras de los cada 30 vez más frecuentes eventos climáticos extremos, la gente está convencida de que 31 efectivamente el clima está cambiando. Otra cosa es que sean conscientes de que ellos 32 mismos contribuyen con sus hábitos de consumo y en consecuencia que estén 33 34 dispuestos a cambiarlos. Yo creo que es una bola que ya se ha puesto en movimiento y 35 que ya es imparable. 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 9 63 64 65 Poisson regression results using i) age as continuous variable, and ii) education as a set 1 of 5 categories (0=no school, 1= primary, 2=secondary, 3=technical education, 2 3 4=university degree). 4 5 Knowledge 7 Poisson, correlates of Proveb Knowledge 8 (1) (2) (3) (4) 9 prov1_tot prov1_clim prov1_phy prov1_bio 10 main 11 age 0.0017 0.0015* 0.0014 0.0033 12 (0.0020) (0.0009) (0.0028) (0.0040) 13 female -0.0067 -0.0367*** -0.0664 0.1383 14 (0.0342) (0.0073) (0.0491) (0.1141) 15 0.education 0.0000 0.0000 0.0000 0.0000 16 (.) (.) (.) (.) 17 1.education -0.1766*** 0.0182 -0.4463*** -0.3023*** 18 (0.0269) (0.0447) (0.0170) (0.0075) 19 2.education -0.2409*** -0.0304 -0.5154*** -0.4147*** 20 (0.0145) (0.0393) (0.0106) (0.0658) 21 3.education -0.2628*** -0.2845*** -0.2888*** -0.1940 22 (0.0853) (0.0599) (0.0989) (0.1425) 23 4.education -0.2050*** 0.0601* -0.2515*** -0.8900*** 24 (0.0180) (0.0342) (0.0011) (0.0206) 25 activity 0.0482 -0.0278 0.1860** 0.0718 26 (0.0705) (0.0322) (0.0894) (0.1446) 27 1.municipality 0.0000 0.0000 0.0000 0.0000 28 (.) (.) (.) (.) 29 2.municipality -0.3283*** -0.0626*** -0.5811*** -0.6386*** 30 (0.0133) (0.0183) (0.0099) (0.0062) 31 3.municipality -0.1909*** -0.0392*** -0.4107*** -0.2762*** 32 (0.0210) (0.0013) (0.0474) (0.0523) 33 4.municipality -0.1200*** 0.0486*** -0.2284*** -0.3688*** 34 (0.0011) (0.0028) (0.0203) (0.0002) 35 36 5.municipality -0.3596*** -0.0756*** -0.5793*** -0.8017*** 37 (0.0000) (0.0012) (0.0147) (0.0025) _cons 2.9518*** 2.0646*** 1.8971*** 1.5411*** 38 39 (0.1835) (0.1008) (0.2149) (0.3418) 40 N 97 97 97 97 41 Standard errors in parentheses * 42 p < 0.10, ** p < 0.05, *** p < 0.01 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 10 63 64 65 Poisson regressions using i) age a dichotomous variable (retired=0 if age<65, and 1 retired=1 if age>=65) and ii) education in 5 categories 2 3 4 Knowledge 5 Poisson, correlates of Proveb Knowledge 6 (1) (2) (3) (4) 7 prov1_tot prov1_clim prov1_phy prov1_bio 8 main 9 retired -0.1291*** -0.0400*** -0.2216*** -0.2447 *** 10 (0.0038) (0.0022) (0.0038) (0.0016) 11 female -0.0078 -0.0361*** -0.0688 0.1329 12 (0.0337) (0.0070) (0.0472) (0.1119) 13 0.education 0.0000 0.0000 0.0000 0.0000 14 (.) (.) (.) (.) 15 1.education -0.2293*** -0.0023 -0.5353*** -0.4090*** 16 (0.0177) (0.0406) (0.0334) (0.0152) 17 2.education -0.2961*** -0.0576** -0.5900*** -0.5302*** 18 (0.0144) (0.0260) (0.0320) (0.1236) 19 3.education -0.4187*** -0.3607*** -0.5128*** -0.4936*** 20 (0.0221) (0.0333) (0.0022) (0.0121) 21 4.education -0.2247*** 0.0410* -0.2688*** -0.9095*** 22 (0.0042) (0.0226) (0.0304) (0.0508) 23 activity -0.0307 -0.0647*** 0.0722 -0.0691 24 (0.0428) (0.0199) (0.0480) (0.0912) 25 1.municipality 0.0000 0.0000 0.0000 0.0000 26 (.) (.) (.) (.) 27 2.municipality -0.3564*** -0.0723*** -0.6355*** -0.6920*** 28 (0.0122) (0.0178) (0.0151) (0.0050) 29 3.municipality -0.1358*** -0.0108 -0.3365*** -0.1713*** 30 (0.0041) (0.0091) (0.0101) (0.0031) 31 4.municipality -0.1036*** 0.0613*** -0.2208*** -0.3429*** 32 (0.0124) (0.0092) (0.0043) (0.0247) 33 5.municipality -0.3297*** -0.0571*** -0.5478*** -0.7406*** 34 (0.0190) (0.0092) (0.0114) (0.0449) 35 _cons 3.1870*** 2.2051*** 2.1904*** 1.9873*** 36 (0.0410) (0.0384) (0.0084) (0.0546) 37 N 97 97 97 97 38 Standard errors in parentheses 39 * p < 0.10, ** p < 0.05, *** p < 0.01 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 11 63 64 65 Poisson regression results using i) age as continuous variable, and ii) education as a set 1 of 5 categories (0=no school, 1= primary, 2=secondary, 3=technical education, 2 3 4=university degree). 4 5 Accuracy 6 Poisson, correlates of Proveb non-accurate, from all proverbs in the list 7 (1) (2) (3) (4) 8 prov_fail3_t prov_fail3_cl prov_fail3_p prov_fail3_b 9 ot im hy io 10 main 11 age 0.0031*** 0.0072* 0.0099*** -0.0083 12 (0.0002) (0.0042) (0.0007) (0.0073) 13 female -0.1183*** -0.2090*** 0.0090 -0.1210*** 14 (0.0158) (0.0414) (0.0851) (0.0106) 15 0.education 0.0000 0.0000 0.0000 0.0000 16 (.) (.) (.) (.) 17 1.education -0.0816 -0.1194*** -0.0157 -0.0885 18 (0.1000) (0.0276) (0.1788) (0.1753) 19 2.education 0.0083 0.0536 0.0120 -0.0439 20 (0.0159) (0.0786) (0.0871) (0.0371) 21 3.education -0.3741*** -0.4537*** -0.6499*** -0.2818 22 (0.0418) (0.1272) (0.0779) (0.2275) 23 4.education -0.1812*** -0.0535 -0.4255*** -0.2973* 24 (0.0450) (0.0717) (0.1308) (0.1717) 25 activity 0.0817*** 0.1706*** 0.0230 0.0384** 26 (0.0040) (0.0001) (0.0392) (0.0181) 27 1.municipality 0.0000 0.0000 0.0000 0.0000 28 (.) (.) (.) (.) 29 2.municipality -0.0184 -0.2331*** -0.3371*** 0.3674*** 30 (0.0493) (0.0068) (0.0801) (0.0963) 31 3.municipality -0.1053*** -0.2699*** 0.2017*** -0.2753*** 32 (0.0331) (0.0464) (0.0615) (0.0337) 33 4.municipality 0.1251*** -0.0006 0.3217*** 0.0881*** 34 (0.0341) (0.0333) (0.0461) (0.0120) 35 5.municipality 0.2146*** 0.0832 0.6821*** -0.3663*** 36 (0.0217) (0.0550) (0.0224) (0.0403) 37 _cons 2.1873*** 1.0679*** 0.4134*** 1.7493*** 38 (0.0486) (0.2768) (0.0494) (0.5685) 39 N 97 97 97 97 40 Standard errors in parentheses 41 * p < 0.10, ** p < 0.05, *** p < 0.01 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 12 63 64 65 Poisson regressions using i) age a dichotomous variable (retired=0 if age<65, and 1 retired=1 if age>=65) and ii) education in 5 categories 2 Poisson, correlates of Proveb non-accurate, from all proverbs in the list 3 (1) (2) (3) (4) 4 prov_fail3_t prov_fail3_cl prov_fail3_p prov_fail3_b 5 ot im hy io 6 main 7 retired 0.1331*** 0.1808* 0.2934*** -0.0857 8 (0.0116) (0.1068) (0.0407) (0.1743) 9 10 female -0.1145*** -0.2031*** 0.0160 -0.1312 *** (0.0191) (0.0333) (0.0965) (0.0029) 11 12 0.education 0.0000 0.0000 0.0000 0.0000 13 (.) (.) (.) (.) 14 1.education -0.0601 -0.1077*** 0.0067 -0.0689 15 (0.1009) (0.0293) (0.1802) (0.1855) 16 2.education 0.0079 0.0211 -0.0147 0.0269** 17 (0.0152) (0.0519) (0.0956) (0.0118) 18 3.education -0.3651*** -0.5427*** -0.7264*** -0.1075 19 (0.0395) (0.0751) (0.0653) (0.1370) 20 4.education -0.2173*** -0.1379*** -0.5215*** -0.1638*** 21 (0.0489) (0.0201) (0.1397) (0.0585) 22 activity 0.1092*** 0.1732*** 0.0488 0.0978*** 23 (0.0119) (0.0069) (0.0634) (0.0119) 24 1.municipality 0.0000 0.0000 0.0000 0.0000 25 (.) (.) (.) (.) 26 2.municipality -0.0025 -0.2201*** -0.3121*** 0.3693*** 27 (0.0488) (0.0133) (0.0758) (0.1093) 28 3.municipality -0.1039*** -0.2338*** 0.2391*** -0.3502*** 29 (0.0336) (0.0262) (0.0647) (0.0042) 30 4.municipality 0.1363*** 0.0329** 0.3603*** 0.0382 31 (0.0333) (0.0136) (0.0428) (0.0497) 32 5.municipality 0.2326*** 0.1343*** 0.7483*** -0.4362*** 33 (0.0188) (0.0232) (0.0133) (0.0112) 34 _cons 2.2937*** 1.4252*** 0.8635*** 1.2341*** 35 (0.0534) (0.0672) (0.0565) (0.1966) 36 N 97 97 97 97 37 Standard errors in parentheses 38 * p < 0.10, ** p < 0.05, *** p < 0.01 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 13 63 64 65 Accuracy 1 Poisson, correlates of Proveb non-accurate, from all proverbs in the list 2 (1) (2) (3) (4) 3 4 prov_fail3 prov_fail3 prov_fail3 prov_fail3 5 _tot _clim _phy _bio 6 main 7 retired 0.1331*** 0.1808* 0.2934*** -0.0857 8 (0.0116) (0.1068) (0.0407) (0.1743) 9 10 female -0.1145*** -0.2031*** 0.0160 -0.1312*** 11 (0.0191) (0.0333) (0.0965) (0.0029) 12 0.education 0.0000 0.0000 0.0000 0.0000 13 (.) (.) (.) (.) 14 15 1.education -0.0601 -0.1077*** 0.0067 -0.0689 16 (0.1009) (0.0293) (0.1802) (0.1855) 17 2.education 0.0079 0.0211 -0.0147 0.0269** 18 (0.0152) (0.0519) (0.0956) (0.0118) 19 3.education -0.3651*** -0.5427*** -0.7264*** -0.1075 20 21 (0.0395) (0.0751) (0.0653) (0.1370) 22 4.education -0.2173*** -0.1379*** -0.5215*** -0.1638*** 23 (0.0489) (0.0201) (0.1397) (0.0585) 24 activity 0.1092*** 0.1732*** 0.0488 0.0978*** 25 26 (0.0119) (0.0069) (0.0634) (0.0119) 27 1.municipal 0.0000 0.0000 0.0000 0.0000 28 ity 29 (.) (.) (.) (.) 30 2.municipal -0.0025 -0.2201*** -0.3121*** 0.3693*** 31 32 ity 33 (0.0488) (0.0133) (0.0758) (0.1093) *** *** *** 34 3.municipal -0.1039 -0.2338 0.2391 -0.3502*** 35 ity 36 (0.0336) (0.0262) (0.0647) (0.0042) 37 *** ** *** 38 4.municipal 0.1363 0.0329 0.3603 0.0382 39 ity 40 (0.0333) (0.0136) (0.0428) (0.0497) 41 5.municipal 0.2326*** 0.1343*** 0.7483*** -0.4362*** 42 43 ity 44 (0.0188) (0.0232) (0.0133) (0.0112) *** *** *** 45 _cons 2.2937 1.4252 0.8635 1.2341*** 46 (0.0534) (0.0672) (0.0565) (0.1966) 47 48 N 97 97 97 97 49 Standard errors in parentheses * 50 p < 0.10, ** p < 0.05, *** p < 0.01 51 52 53 54 55 56 57 58 59 60 61 62 14 63 64 65 Figure Click here to ccess/downlo d;Figure;Fig1.png .t Study Area Spa n.Andaluc a & Sierra Nevada · Sampled Areas • NationalPark Natural Park Ravine Borders ., ...... l. 1- -1 :"'"""·'";r , ,.,., -i .......,,_ ...'.,.. ' ...',.,. ...'..,. o S '"""" 10 15 20 km Authors' Response to Reviewers' Comments Click here to access/download Electronic Supplementary Material Supplementary Material_13Feb20.docx