ARTICLE https://doi.org/10.1038/s43247-023-01164-y OPEN Indigenous Peoples and local communities report ongoing and widespread climate change impacts on local social-ecological systems Victoria Reyes-García 1,2 ✉, David García-del-Amo 2, Santiago Álvarez-Fernández 2, Petra Benyei2,3, Laura Calvet-Mir 2,4, André B. Junqueira 2, Vanesse Labeyrie5,6, Xiaoyue LI2, Sara Miñarro2, Vincent Porcher 2,7, Anna Porcuna-Ferrer 2,5, Anna Schlingmann2, Christoph Schunko 8, Ramin Soleymani2, Adrien Tofighi-Niaki2, Mariam Abazeri9, Emmanuel M. N. A. N. Attoh10,11, Ayansina Ayanlade12,13, Julia Vieira Da Cunha Ávila 14, Daniel Babai15, Rodrigo C. Bulamah16,17, Joao Campos-Silva18, Rosario Carmona 19,20, Julián Caviedes 2,20,21, Rumbidzayi Chakauya22, 1234567890():,; Mouna Chambon2, Zhuo Chen23, Fasco Chengula 24,25, Esther Conde26, Aida Cuní-Sanchez27,28, Christophe Demichelis 29, Evgeniya Dudina30, Álvaro Fernández-Llamazares 2,31, Eranga K. Galappaththi 32, Claudia Geffner-Fuenmayor 2,33, Drew Gerkey34, Marcos Glauser 35, Eric Hirsch36, Tomás Huanca26, José Tomás Ibarra 20,21, Andrea E. Izquierdo 37,38, Leneisja Junsberg 39, Marisa Lanker40, Yolanda López-Maldonado 41, Juliette Mariel5,6, Giulia Mattalia2, Mohamed D. Miara42,43, Miquel Torrents-Ticó 31,44, Maedeh Salimi45, Aibek Samakov46, Reinmar Seidler 47,48, Victoria Sharakhmatova 49,50, Uttam Babu Shrestha51, Alpy Sharma52, Priyatma Singh53, Tungalag Ulambayar 54, Rihan Wu55 & Ibrahim S. Zakari56 The effects of climate change depend on specific local circumstances, posing a challenge for worldwide research to comprehensively encompass the diverse impacts on various local social-ecological systems. Here we use a place-specific but cross-culturally comparable protocol to document climate change indicators and impacts as locally experienced and analyze their distribution. We collected first-hand data in 48 sites inhabited by Indigenous Peoples and local communities and covering all climate zones and nature-dependent liveli- hoods. We documented 1,661 site-agreed reports of change corresponding to 369 indicators. Reports of change vary according to climate zone and livelihood activity. We provide com- pelling evidence that climate change impacts on Indigenous Peoples and local communities are ongoing, tangible, widespread, and affect multiple elements of their social-ecological systems. Beyond potentially informing contextualized adaptation plans, our results show that local reports could help identify economic and non-economic loss and damage related to climate change impacts suffered by Indigenous Peoples and local communities. A full list of author affiliations appears at the end of the paper. COMMUNICATIONS EARTH & ENVIRONMENT | (2024)5:29 | https://doi.org/10.1038/s43247-023-01164-y | www.nature.com/commsenv 1 ARTICLE COMMUNICATIONS EARTH & ENVIRONMENT | https://doi.org/10.1038/s43247-023-01164-y C limate change is a global phenomenon, but its impacts on observations of climate change and its impacts as locally experi- biophysical and sociocultural systems are governed by enced and examines their distribution across climate zones and local conditions1,2. Modeling work shows that some world livelihood activities. We provide compelling evidence that climate regions will experience higher temperature increase than others, change impacts on Indigenous Peoples and local communities are and that some regions will receive more rainfall while others will ongoing, tangible, widespread, and affect multiple elements of experience more frequent droughts3. Regional changes will also their social-ecological systems and that reports of change vary have differentiated impacts on human societies, with nature- according to climate zone and livelihood activity. Our analysis dependent communities being more prone to be directly affected deliberately adopts a global quantitative perspective, oriented to by, e.g., droughts or other extreme weather events, than com- leverage the inclusion of Indigenous Peoples and local commu- munities less immediately dependent on nature for their nities and their knowledge into climate change research and livelihood4,5. In addition, climate change may be differently policy. This work relies on the wealth of qualitative information perceived and experienced across social groups, as societies have collected across case studies and published elsewhere24. We developed unique cultural and epistemological frameworks for emphasize the importance of working with both quantitative and understanding and interacting with the world6. Consequently, a qualitative data to identify common trends as well as idiosyn- comprehensive understanding of climate change and its impacts crasies – both of which are important to understand global trends should consider not only spatial heterogeneity, but also variations and the myriad of climate change impacts across the world25. derived from how climate change and its impacts are experienced and understood by people affected by them. Current research streams face different challenges in presenting Results a global picture of climate change and its impacts. On the one We selected 48 field sites across all inhabited continents. Sites hand, research based on instrumental measurements, while fun- covered four climate zones of the Köppen-Geiger climate damental to our comprehension of the global impacts of climate classification26 (i.e., tropical, arid, temperate, and snow/polar), change on physical and biological systems3,7, falls short in pro- with a slight predominance of tropical climates sites (n = 16) and viding a comprehensive view of how climate change and its the lowest number in snow/polar climates (n = 7) (Fig. 1; Sup- impacts are locally experienced. Predictive models constructed plementary Table 1). All the sites, except the Tsimane’ Indigenous using instrumental measurements8 lose accuracy and precision territory (Bolivia, Site_ID 24), are in areas with increasing average when downscaled, particularly in areas with scarce ground data9. decadal temperature (see Methods), most of them (n = 20) For example, since meteorological stations are sparsely dis- experiencing a medium average increase (i.e., 0.15 to 0.3 °C/ tributed in areas inhabited by Indigenous Peoples and local decade). The predominant livelihood activities in studied sites communities10, predictions for such areas have low accuracy11,12. were agropastoralism (n = 16), agriculture (n = 13), and fishing In addition to their coarse resolution, predictive models focus on (n = 12). Pastoralism (n = 4) and other nature-dependent liveli- a few variables, typically defined by the scientific community, hoods (e.g., hunting and non-timber forest product collection, often overlooking locally important phenomena or concerns. For hereafter ‘other NRD’; n = 3) were less common. example, modeling work forecasting climate change impacts on agriculture focuses on the world’s most common cereal crops (e.g., wheat, maize, rice)13–15, neglecting crops critical to food Reports of climate change and its impacts. Across the 48 sites, sovereignty, economic security, and cultural identity in many we documented 1,661 observations of climate change and its local food systems16. Finally, instrumental measures are unable to impacts, although the number of citations largely varied across capture the diverse and complex ways in which people under- sites (Supplementary Fig. 1). For each site, only observations for stand, experience, and relate to the environment, which limits which there was agreement across focus group discussions were their suitability for risk analysis and adaptation planning6,17. For considered (see Methods). Across sites, observations of changes in instance, instrumental measurements might capture changes in elements of the atmospheric system, i.e., indicators of climate rainfall patterns but miss crucial relationships between climate change, were most frequently cited (accounting for 46.4% of all change awareness, sensitivity, and vulnerability18. observations; Fig. 2), with observations of changes in the sub- On the other hand, the growing body of research relying on systems precipitation (20.4%) and temperature (13.0%) being grounded information from Indigenous Peoples and local com- predominant. Observations of changes in the physical and life munities, while successfully emphasizing the idiosyncrasies of systems, i.e., climate change impacts, were less cited. In particular, climate change impacts, does not provide globally comparable climate change impacts in the physical system were the least information, thus hampering our ability to discern general trends frequently cited (19.6%), with observations of changes in the and patterns and to upscale local research results to global climate subsystem freshwater (9.1%) being the most common observa- change research19,20. Following calls from the 2014 Inter- tions in this system. Changes in the life system accounted for governmental Panel on Climate Change (IPCC) report to develop about one third of all the observations (33.9%), with observations the evidence base for the potential contribution of local knowl- of changes in the subsystems plant cultivation (9.9%) and ter- edge to climate change research21, the last decade has seen an restrial flora (7.9%) being the most predominant. increase in the number of works examining the intricate knowl- Site-agreed observations of climate change and its impacts were edge systems that allow Indigenous Peoples and local commu- classified in 369 local indicators of climate change impacts nities to identify changes in natural systems, the drivers of such (LICCI), corresponding to 94 indicators of climate change (i.e., changes, and the local responses to them10,22. These works pro- reports of change in elements of the atmospheric system) and 275 vide rich, multi-sited, qualitative, and place-based information on indicators of climate change impacts (i.e., reports of change in locally experienced climate impacts, although they have limited elements of the physical and the life systems) (see Methods). capacity to discern general trends because they do not rely on a The most frequently reported indicators of climate change common strategy to gather globally comparable information. include changes in mean temperature (45 citations, 2.7%) and Here, we present the results of a large, globally coordinated changes in mean temperature in a season (34 citations, 2.0%), study including 48 Indigenous Peoples and local communities on whereas the most frequently reported indicators of climate all inhabited continents. The study, which used a place-specific change impacts include changes in crop productivity (52 but cross-culturally comparable research protocol23, documents citations, 3.1% of the 1,661 observations documented) and 2 COMMUNICATIONS EARTH & ENVIRONMENT | (2024)5:29 | https://doi.org/10.1038/s43247-023-01164-y | www.nature.com/commsenv COMMUNICATIONS EARTH & ENVIRONMENT | https://doi.org/10.1038/s43247-023-01164-y ARTICLE Fig. 1 Site distribution across climate zones. Site labels correspond to Site_ID in Supplementary Table 1, which also contains site characterization. Dot size indicates the number of local indicators of climate change impacts reported in a site (see Online Methods). Climate zones were adapted from the Köppen- Geiger climate classification26, differentiating tropical, arid, temperate, snow and polar climates. Fig. 2 Percentage of observations of climate change and its impacts on the physical and life system (n = 1,661 site-agreed observations, n = 48 sites). Subsystems in black refer to the atmospheric system, in blue to the physical system, and in green to the life system. changes in the abundance or density of wild plant or fungi species observations and explanations of change across sites. For (33 citations, 2.0%). instance, Dagomba-Gur agriculturalists (Kumbungu, Ghana, This classification conceals the complex, nuanced, and Site_ID 23) report a general temperature increase which they integrated nature of the reported observations of climate change attribute to a warmer Harmattan (dry wind blowing from the and local explanations of its impacts. For example, reports Sahara), whereas farmers in Chiloé (Chile, Site_ID 8), while also classified as changes in mean temperature conceal very different reporting a general temperature increase, attribute it to a COMMUNICATIONS EARTH & ENVIRONMENT | (2024)5:29 | https://doi.org/10.1038/s43247-023-01164-y | www.nature.com/commsenv 3 ARTICLE COMMUNICATIONS EARTH & ENVIRONMENT | https://doi.org/10.1038/s43247-023-01164-y precipitation and streamflow decrease. In a similar way, Bassari average number of indicators of climate change/site referring to farmers (Bassari Country, Senegal, Site_ID 2) report changes in changes in temperature was higher than in predominantly the productivity of sorghum long-cycle landraces, which they pastoralist sites. Additionally, the average number of LICCIs/site attribute to the shortening of precipitation cycles, whereas Csángó referring to impacts in freshwater and soils was also higher than farmers (Gyimes, Romania, Site_ID 16) report declining potatoes where agropastoralism predominates. For example, Akha farmers yields due to pests’ infestations, which they relate to warmer (Nannuoshan, China, Site_ID 31) report decreasing levels of winters that allow pests to breed over winter. water in mountain creeks and rivers and Gurung farmers (Laprak, Nepal, Site_ID 24) report increased soil erosion due to heavier rains during the monsoon season. Distribution of reports of climate change and its impacts. Where fishing predominates, we found higher average number While our classification masks nuanced observations of changes, of indicators of climate change/site referring to changes in air it also allows for a global analysis of the distribution of reports of masses (relative to sites where agriculture, agropastoralism, or local climate change and its impacts. The number of site-agreed pastoralism predominate), higher number of impacts in oceans observations of change (see Methods) showed large variation and seas (relative to agropastoralism and other NRD), and higher across sites (avg. = 34.9 LICCIs/site, sd = 18.8, min = 9, max = number of impacts in marine ecosystems (relative to agriculture). 84). However, the frequency distribution across systems and For example, iTaukei fishers (Ba, Fiji, Site_ID 1) report an subsystems of site-agreed observations of change (Fig. 3, column increase in the intensity of tropical cyclones, Catalan fishers ‘Total’) resembles the overall distribution of observations (Fig. 2, (Costa Brava, Spain, Site_ID 11) report changes in coastal storms subsystems). with fast winds and strong waves, and Swahili fishers (South We tested whether site-agreed observations of change were coast, Kenya, Site_ID 21) report an increase in sea water differently distributed across climate zones using a Poisson temperature during the Kaskazi (wind) season resulting in general linear model followed by a series of post-hoc tests (Fig. 3). increased shell mortality. The average number of indicators of climate change/site referring Where pastoralism predominates, the average number of to changes in precipitation, temperature, air masses, and seasons LICCIs/site referring to impacts in pastures and grasslands and were not statistically different across climate zones. However, in land cover change and land degradation was higher than in sites sites in tropical climates, the average number of indicators/site with any other livelihood activity. For example, Kolla- referring to changes in freshwater was higher than in sites in Atacameños pastoralists (Puna, Argentina, Site_ID 36) report a snow/polar climates. Grounded observations exemplify the many decrease in pasture productivity related to decreasing rainfall and ways in which changes in the freshwater subsystem are observed Mongolian pastoralists (Mu Us Desert, Ordos, China, Site_ID 32) in tropical zones. For example, ribeirinhos from the Juruá River report the increased presence of non-edible grasses which (Brazil, Site_ID 7) reported changes in the river dynamics, displace edible pastures under drier conditions. including changes in flood duration, height, and sedimentation Finally, in sites where other NRD livelihoods predominate, the patterns; farmers in Coastal-Vedda (East region, Sri Lanka, average number of LICCIs/site referring to impacts in ice and Site_ID 14) reported changes in groundwater quality and lower snow was higher than in all other sites, although the difference water levels in village tanks; and agriculturalists in Milot (Haiti, was not statistically significant. In fact, this high average seems to Site_ID 28) reported the drying of lakes, ponds and wells, be largely driven by reports from Inuit (Baffin Island, Canada, increasing sediments in rivers, and larger but more erratic floods. Site_ID 34), who reported numerous impacts, such as slower In arid and in snow/polar climates, the average number of freezing and faster melting of sea ice, more ice break-ups, and LICCIs/site referring to impacts in pastures and grasslands was changing thickness of the sea ice. significantly higher than in tropical climates, while the average In a final analysis, we used a multivariate approach to number of LICCIs/site referring to impacts in plant cultivation investigate how the assemblage of LICCIs reported in a site, was lower than in temperate climates. including the type of LICCI and the frequency in which they are Finally, in snow/polar and temperate climates, the average reported (hereafter LICCI composition, see Methods) relates to number of LICCIs/site referring to impacts in ice and snow was the site (i) climate zone, (ii) change in decadal temperature, and higher than in arid and tropical climates, where ice and snow are (iii) main livelihood activity. We found that LICCI composition unusual, and the average number of LICCIs/site referring to was significantly related with both the climate zone where a site is impacts in terrestrial fauna was also higher in snow/polar and located (NPMANOVA F3,47 = 1.8, p < .01, Supplementary temperate climates than in tropical climates. Table 2) and the site’s main livelihood activity (NPMANOVA We followed a similar approach to test whether reports of F4,47 = 1.2, p = .04), but not to the change in decadal tempera- change were differently distributed across different levels of ture. In other words, the distribution of LICCIs reported in a site exposure to change in decadal temperature (Supplementary is related with both the site’s climate zone and the way people use Fig. 2). The average number of LICCIs/site referring to impacts in the resources of their environment through their main livelihood pastures and grasslands was higher in sites experiencing high activity. decadal temperature increases (>0.3 °C/decade) than in sites experiencing medium (0.15 to 0.3 °C/decade) or mild (0 to 0.15 °C/decade) temperature increases. Also, the average number Discussion of LICCIs/site referring to land cover change and land The results of this large global effort add to the growing evidence degradation was higher in sites experiencing high decadal that Indigenous Peoples and local communities observe climate temperature increases than in sites experiencing medium decadal change and that the impacts of climate change on them are temperature increases. ongoing, tangible, and pervasive1. They also suggest that, despite We found important differences when using the same idiosyncrasies, there are important patterns in how Indigenous approach to test differences in reports of climate change and its Peoples and local communities experience and report climate impacts across site’s main livelihood activity (Fig. 4). Where change and its impacts. agriculture predominates, the average number of indicators of Before discussing these findings, we examine three important climate change/site referring to changes in precipitation was caveats of our work. First, our method cannot discern whether higher than where other NRD livelihoods predominate and the reported impacts can be fully attributed to climate change, as 4 COMMUNICATIONS EARTH & ENVIRONMENT | (2024)5:29 | https://doi.org/10.1038/s43247-023-01164-y | www.nature.com/commsenv COMMUNICATIONS EARTH & ENVIRONMENT | https://doi.org/10.1038/s43247-023-01164-y ARTICLE Fig. 3 Average number of LICCIs/site, by climate zone. Subsystems in black refer to the atmospheric system, in blue to the physical system, and in green to the life system. The error bars represent the standard error of the mean (SEM). * at the end of a column indicates a statistically significant difference (p < 0.05) in the average number of LICCIs/site in the subsystem between the climate zone indicated in the header and the snow/polar climate (category of comparison) in a Poisson general linear model. The error bars represent the standard error of the mean. Fig. 4 Average number of LICCI per site, by main livelihood activity. Subsystems in black refer to the atmospheric system, in blue to the physical system, and in green to the life system. The error bars represent the standard error of the mean (SEM). * at the end of a column indicates a statistically significant difference (p < 0.05) in the site average number of LICCI in the subsystem between the livelihood activity indicated in the header and agriculture (category of comparison) in a Poisson general linear model. The error bars represent the standard error of the mean. complexity and confounding drivers of environmental change but rather they often exacerbate situations of environmental (e.g., land-use change or overextraction) make attribution change resulting from colonialism, racism, inequality, and difficult27,28. While recognizing that reported impacts might not environmental injustice29,30. Indeed, research increasingly always be driven by climate change, we also argue that, for emphasizes that environmental change reflects the outcome of Indigenous Peoples and local communities, climate change multiple drivers of change and synergistic relations among impacts do not occur in isolation from other drivers of change, them31,32, for which interventions might be more effective if they COMMUNICATIONS EARTH & ENVIRONMENT | (2024)5:29 | https://doi.org/10.1038/s43247-023-01164-y | www.nature.com/commsenv 5 ARTICLE COMMUNICATIONS EARTH & ENVIRONMENT | https://doi.org/10.1038/s43247-023-01164-y consider both social context and climate change interactions with We conclude with three policy recommendations derived from other drivers of change. A second important limitation of our our work. First, Indigenous Peoples and local communities can work refers to the fact that our sample of sites is small, geo- expand, enrich, and deepen understandings of climate change and graphically biased, and potentially imbalanced as there is some its impacts on local social-ecological system. This knowledge can overlap between climate zones and livelihoods. Despite this lim- contribute to the formulation of feasible and locally relevant itation, this research constitutes the largest global effort to com- adaptation plans and interventions37–39. Our results show that pile, harmonize, and categorize local observations of climate reports of climate change and its impacts are dependent not only change and its impacts by Indigenous Peoples and local com- on climate zones, but also on main livelihood activities, for which munities. Finally, we also acknowledge challenges common in such reports could help made adaptation and mitigation plans research with Indigenous Peoples and local communities, more locally meaningful. While several mechanisms have been including overgeneralizations and misinterpretations deriving created to engage Indigenous Peoples’ and local community from the transformation of locally meaningful information to fit representatives and delegates in climate policy, they mostly have a into scientific classifications32. While acknowledging ontological global character (e.g., the Local Communities and Indigenous differences and positionalities, we argue that the current climatic Peoples Platform under the UNFCCC40 or the Local and Indi- emergency requires exploring multiple forms of collaboration genous Knowledge Systems (LINKS) program at UNESCO). between knowledge systems aiming towards similar pragmatic Parallel actions at the local and national levels are strongly nee- goals33, even if collaboration is based on categorical ded, with mechanisms to increase coordination across scales. equivocations34. Second, as discussions on the need to address loss and damage to In contrast with narratives analyzing future changes in climate achieve climate justice are gaining recognition in the academic and potential impacts under different scenarios, results from this literature41,42 and in climate negotiations (e.g., Decision 2/CP.27, work emphasize that Indigenous Peoples and local communities UNFCCC. Conference of Parties), our work provides a metho- report a myriad of indicators of climate change as well as ongoing dology that allows both the identification of climate change environmental impacts, which they (fully or partially) attribute to impacts at the local level and the cross-cultural comparison of climate change. Changes reported stem from people’s close reported impacts. This methodology has been applied to various interactions with the environment and include richly described, tools, including the OpenTEK platform and Oblo, both of which nuanced, and locally contextualized changes in many elements of have served to help assess local climate change impacts, either by the atmospheric systems and their impacts in the physical and the researchers, citizen scientists, or communities themselves. In this life systems. In that sense, the 369 local indicators of climate sense, the methodology has the capacity to be applied to existing change and its impacts documented here, which range from or new tools and adapted in a way that would inform about the indicators of changes in rainfall or fog, to how these changes specific economic and non-economic loss and damage related to impact freshwater availability, soil humidity, abundance climate change impacts on local social-ecological systems43. By and phenology of wild and cultivated plants and terrestrial combining the methodology with a climate impact valuation and aquatic animals, or increases in crop pests, reflect the detailed process, as defined either by communities or regional stake- understanding of changes captured by Indigenous Peoples and holders, the work has the potential for providing empirical basis local communities. Our findings stress the importance of for parties to discuss appropriate compensatory measures. This empirically grounded research to obtain a complete picture of would contribute to a broader recognition of the diverse histories climate change and its impacts. They also provide strong of social, environmental, and ecological harm while opening just empirical support to perspectives that emphasize the need to pathways into uncertain futures. Finally, considering that the enlarge our research focus to account for impacts that change the foundations of Indigenous and local knowledge systems are being interwoven relationships between humans and nature at the local compromised and disrupted in many places44, there is a greater scale35 as well as for diverse perceptions of which impacts need than ever before to support Indigenous Peoples and local matter36. communities in mobilizing their knowledge basis to adapt to In addition to presenting a more complete picture of climate ongoing impacts and contribute to global mitigation efforts. Our change that includes indicators and impacts as locally experi- compilation of ethnographic data documenting diverse Indigen- enced, the work presented here also allows to identify general ous and local expertize about present and future climate change patterns on how Indigenous Peoples and local communities impacts and adaptation plans could inform elite-driven debates perceive them. For example, our global analysis shows the pre- about loss and damage, CO2 emissions mitigation, and the turn dominance of reports of changes in precipitation, which represent to renewables. Developing locally specific data ultimately serves about 20% of all reports, and their impact on elements of the life the goal of reducing harm to Indigenous Peoples and local system. The finding could be read in terms of the direct and communities that comes from both acute climate impacts and cascading impacts of changes in the water cycle on local social- top-down climate adaptations. ecological systems32. Results of the global analysis also show that Indigenous Peoples and local communities reports of climate change and its impacts are largely contingent on the climate zone Online Methods where they live and on their main livelihood activity. For This research was carried out within the framework of a research example, changes in precipitation and impacts on freshwater project aiming to bring contextualized information from Indi- systems were more frequently reported in sites where agriculture genous Peoples and local communities to climate change research predominates, whereas changes in air masses and impacts on by creating a research network that coordinated data collection19. marine ecosystems were more frequently reported in sites where The research network included people with different nationalities, fishing predominates and impacts in pastures and grasslands and disciplinary backgrounds, and positionalities, who recognized the land cover were more frequent where pastoralism predominates. need to increase the transferability, integration, and scalability of This finding provides empirical basis to the previously untested information from Indigenous Peoples and local communities into hypothesis that the way people interact with the environment -in climate research using standardized methods for the collection this case through livelihood activities- is an important predictor and coding of locally relevant but cross-culturally comparable of the changes they observe. data. We are aware that our framing of local climate change and 6 COMMUNICATIONS EARTH & ENVIRONMENT | (2024)5:29 | https://doi.org/10.1038/s43247-023-01164-y | www.nature.com/commsenv COMMUNICATIONS EARTH & ENVIRONMENT | https://doi.org/10.1038/s43247-023-01164-y ARTICLE its impacts is largely influenced by Western epistemic traditions precipitation, temperature, air masses, and seasonality are and that, therefore, the indicators used are situated and partial. subsystems within the atmospheric system). The third level We understand that Indigenous Peoples and local communities corresponded to the impacted element (e.g., mean temperature, might assess, conceptualize, and weigh local manifestations of temperature extremes, seasonal temperature, and sunshine climate change and its impacts differently. intensity in the temperature subsystem), and the last level to The Ethics Committee of the Universitat Autònoma de Bar- the indicator23. celona approved the research protocol used in this project In each site, we organized 3–5 focus groups to discuss (CEEAH 4781). Before data collection started, we obtained per- observations and indicators for which semi-structured interviews mits from local authorities in each site to conduct research, as resulted in contradictory or unclear information. In this work, we well as the Free Prior Informed Consent of all participants. only consider indicators of climate change and indicators of Where necessary, we also obtained authorizations from national climate change impacts agreed upon focus groups discussions. In ethics committees. Following open science principles, the proto- sites where we could not conduct focus group discussions due to col for the research project is openly available23 and the team has COVID restrictions, we conducted additional semi-structured made available a series of online training materials to facilitate the interviews to further discuss indicators for which we had found use of the protocol (https://www.licci.eu/resources/) so here we contradictory or unclear information. We considered that there focus on explaining the methodological aspects that directly relate was agreement on an indicator when (i) there were no to this work. contradictions in reporting a given observation of change in semi-structured interviews or (ii) participants agreed with the reported change in focus group discussions. Data collection. We collected first-hand data in 48 field sites with In each site, the list of indicators of climate change impacts predominantly nature-dependent livelihoods during 2019–2022. used in the analysis reflects changes for which participants agreed A site was defined as a group of villages or households showing that are -at least partly- driven by changes in temperature, relative environmental and socio-cultural homogeneity. We precipitation, air masses, or seasons, and that reflect temporal selected sites to maximize variability in climate zones and nature- depth (i.e., the change refers to “compared to your youth”). The dependent livelihoods. participatory and iterative nature of the data collection process In each site, we conducted semi-structured interviews with helped identify climate change indicators and impacts that reflect 15–25 key participants to document climate change, understood the collective intergenerational knowledge held and owned by as changes in elements of the atmospheric systems, and its Indigenous Peoples and local communities living in close impacts on the environment. Our interviews focused on changes interaction with the local environment. registered over decades, and noting the direction (e.g., increase or Each site was assigned to a climate zone using the site location decrease) and the driver(s) of reported changes. We started by and the Köppen-Geiger climate classification26. We differentiated asking, “Compared to your youth, what changes have you noticed between sites located in tropical, arid, temperate, and snow/polar in the environment?” and continued asking, “Have you noticed climates. We combined sites in the snow and polar climates any other change?” Once the participant stopped mentioning because we only had one site in the polar climate. For each site, changes, we directly asked about changes in temperature, we calculated the decadal temperature change using estimates precipitation, wind, seasons, soils, water, and wild and domestic reported by the IPCC45 and constructed using the CRU-TS animals or plants. For each change mentioned, we asked about dataset that spans 1901 to 201846. We then categorized sites into the perceived driver(s) of the change (i.e., “Why do you think this high positive (>0.3 °C/decade), medium positive (0.15, 0.3 °C/ change happened?”). If the participant responded with another decade), mild positive (0, 0.15 °C/decade), and negative (<0 °C/ environmental change, we asked again for the perceived driver(s) decade) average decadal temperature change. of the latest mentioned change, until the respondent did not We used information gathered through open ended and semi- identify any additional driver. A full description of the protocol is structured interviews to characterize the main livelihood activity publicly available23. As observations of changes and its drivers of the communities we worked with. Most communities in this came in contextualized narratives, they often included multiple study included households relying in complex livelihood drivers. For example, the observation that “Plants “cook” on the strategies which often combine different capabilities, assets, and ground. The ground is hot, it rains, and then it is hot again, and activities to make a living47. While recognizing this complexity, this “cooks” the plants” made by Tsimane’ informants (Bolivia, for the analysis presented here, we only focus on the main Site_ID 24) was classified as a report of “changes in precipita- livelihood activity, differentiating among sites where the pre- tion”, “changes in soil temperature”, and “changes in crop dominant livelihood activity was agropastoralism, agriculture, mortality rates”. fishing, pastoralism, and other nature dependent livelihoods (e.g., In each site, we created a list of local indicators of climate hunting, gathering, non-timber forest products collection). We change and its impacts (LICCIs) that were locally agreed upon. note that our classification reflects the current economic activity The list includes reports of change in elements of the atmospheric of the sites, which might differ from group’s identity and from system, considered as ‘indicators of climate change’, and reports previous classifications of some of the groups in the sample. of change in elements of the physical and the life systems associated with those, considered as ‘indicators of climate change impacts’. As our original question referred to environmental Data analysis. We started the analysis by providing a description changes, our first step was to remove observations of environ- of the local indicators of climate change and its impacts (i.e., mental changes that participants did not consider to be at least LICCIs) reported in the full sample (n = 48 sites). We then used partly driven by changes in temperature, precipitation, wind, or the site as a unit of analysis to explore whether the agreed reports seasons. We classified the remaining observations into indicators. of indicators of climate change and its impacts were differently Verbatim observations depicting the same change were classified distributed across (i) climate zones, (ii) decadal temperature into the same indicator. Indicators were organized using a system change, and (iii) main livelihood activities. We did so by using a with four hierarchical levels. The two highest levels corresponded set of general linear models (GLM) with Poisson distribution to to the system (i.e., atmospheric, physical, and life systems) and systematically compare the site average number of LICCI cited in the subsystem in which the change was observed (e.g., a subsystem with a selected category of comparison (i.e., snow/ COMMUNICATIONS EARTH & ENVIRONMENT | (2024)5:29 | https://doi.org/10.1038/s43247-023-01164-y | www.nature.com/commsenv 7 ARTICLE COMMUNICATIONS EARTH & ENVIRONMENT | https://doi.org/10.1038/s43247-023-01164-y polar for climate zones; high positive decadal temperature change; 8. Armstrong, M. S., Kiem, A. S. & Vance, T. R. Comparing instrumental, and agriculture for main livelihood activities). When the GLM palaeoclimate, and projected rainfall data: Implications for water resources assessing the average LICCIs in a subsystem showed significant management and hydrological modelling. J. Hydrol. Reg. Stud. 31, 100728 (2020). relationships with an explanatory variable, pairwise comparisons 9. Maraun, D. & Widmann, M. Statistical downscaling and bias correction for between the different levels of the variable were carried out with a climate research. Stat. Downscale. Bias Correc. Clim. Res. https://doi.org/10. Tukey’s honest significance test. 1017/9781107588783 (2017). In our final analysis, we explored the relative importance of the 10. Savo, V. et al. Observations of climate change among subsistence-oriented selected variables in explaining the variation in number and type communities around the world. Nat. Clim. Chang. 6, 462–473 (2016). 11. Fernández-Llamazares, Á. et al. An empirically tested overlap between of LICCI reported at each site. Drawing on methods from indigenous and scientific knowledge of a changing climate in Bolivian community ecology and species composition analysis48, we Amazonia. Reg. Environ. Chang. 17, 184–198 (2017). created a variable that captured the number and identity of 12. Tumenjargal, S. et al. Variability and change of climate extremes from LICCIs reported at a site, named site LICCI composition. To indigenous herder knowledge and at meteorological stations across central create the variable, we constructed a dissimilarity matrix using the Mongolia. Front. Earth Sci.14, 286–297 (2020). 13. Challinor, A. J. et al. A meta-analysis of crop yield under climate change and Bray-Curtis dissimilarity index, calculated at the subsystem level adaptation. Nat. Clim. Chang. 4, 287–291 (2014). and including only impacts on the physical and the life system. 14. Jägermeyr, J. et al. Climate impacts on global agriculture emerge earlier in new The new variable considered both the number and the type of generation of climate and crop models. Nat. Food 2, 873–885 (2021). LICCI and compared them between sites. We then explored 15. Zhao, C. et al. Temperature increase reduces global yields of major crops in whether variation in site LICCI composition was associated with four independent estimates. Proc. Natl Acad. Sci. USA 114, 9326–9331 (2017). 16. Chivenge, P., Mabhaudhi, T., Modi, A. T. & Mafongoya, P. The potential role climate zone, decadal temperature change, and main livelihood of neglected and underutilised crop species as future crops under water scarce activities by using a non-parametric multivariate analysis of conditions in sub-saharan Africa. Int. J. Environ.Res. Public Health 12, variance (NPMANOVA). Multivariate analysis was performed 5685–5711 (2015). using the vegan package49 in R version 4.2.150. GLM analyses 17. Hiwasaki, L. & Luna, E. Syamsidik & Shaw, R. Process for integrating local were run using the stats package49 and tukey post-hoc tests were and indigenous knowledge with science for hydro-meteorological disaster risk reduction and climate change adaptation in coastal and small island run with the emmeans package50. communities. Int. J. Disast. Risk Reduc.10, 15–27 (2014). 18. Kosmowski, F. & Leblois, A. & Sultan, B. Perceptions of recent rainfall Reporting Summary. Further information on research design is changes in Niger: a comparison between climate-sensitive and non-climate available in the Nature Portfolio Reporting Summary linked to sensitive households. Clim. Chang. 135, 227–241 (2016). 19. Reyes-García, V. et al. A collaborative approach to bring insights from local this article. observations of climate change impacts into global climate change research. Curr. Opin. Environ. Sustain. 39, 1–8 (2019). 20. IPCC. Climate change 2023: synthesis report. A Report of the Data availability Intergovernmental Panel on Climate Change. Contribution of Working Groups The datasets generated analyzed during the current study are available in the Dataverse I, II and III to the Sixth Assessment Report of the Intergovernmental Panel on repository, https://doi.org/10.34810/data878. Climate Change. (2023). 21. IPCC. Climate change 2014: synthesis report. Contribution of Working Groups I, II, and III to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. https://www.ipcc.ch/report/ar5/syr/ (2014). Code availability 22. Reyes-García, V. et al. Local indicators of climate change: the potential This work did not use any custom computer code or algorithm. contribution of local knowledge to climate research. Wiley Interdiscip. Rev. Clim. Chang. 7, 109–124 (2016). 23. Reyes-García, V. et al. Local indicators of climate change impacts described by Received: 1 May 2023; Accepted: 6 December 2023; indigenous peoples and local communities: Study protocol. PLoS One 18, e0279847 (2023). 24. Reyes-García, V. et al. Routledge Handbook of Climate Change Impacts on Indigenous Peoples and Local Communities. https://www.routledge.com/ Routledge-Handbook-of-Climate-Change-Impacts-on-Indigenous-Peoples- and/Reyes-Garcia/p/book/9781032412139 (2024). References 25. Reyes-García, V. Beyond artificial academic debates: for a diverse, inclusive, 1. IPCC. Climate change 2022: impacts, adaptation, and vulnerability. Contribution and impactful ethnobiology and ethnomedicine. J. Ethnobiol. Ethnomed. 19, of Working Group II to the Sixth Assessment Report of the Intergovernmental 1–4 (2023). Panel on Climate Change (Cambridge University Press, 2022). 26. Beck, H. E. et al. Present and future Köppen-Geiger climate classification 2. Seddon, A. W. R., Macias-Fauria, M., Long, P. R., Benz, D. & Willis, K. J. maps at 1-km resolution. Sci. Data 5, 1–12 (2018). Sensitivity of global terrestrial ecosystems to climate variability. Nature 531, 27. Hansen, G., Stone, D., Auffhammer, M., Huggel, C. & Cramer, W. Linking 229–232 (2016). local impacts to changes in climate: a guide to attribution. Reg. Environ. 3. Costello, M. J. et al. Cross-chapter paper 1: biodiversity hotspots. in Climate Chang. 16, 527–541 (2016). Change 2022: Impacts, Adaptation, and Vulnerability. Contribution of 28. Rosenzweig, C. & Neofotis, P. Detection and attribution of anthropogenic Working Group II to the Sixth Assessment Report of the Intergovernmental climate change impacts. Wiley Interdiscip. Rev. Clim. Chang. 4, 121–150 Panel on Climate Change 2nd edn, Vol. 3 (eds. Pörtner, H.-O.) Ch. 2123–2161 (2013). (Cambridge University Press, 2022). 29. Bee, B. A., Rice, J. & Trauger, A. A feminist approach to climate change 4. Barnes, J. et al. Contribution of anthropology to the study of climate change. governance: everyday and intimate politics. Geogr. Compass 9, 339–350 Nat. Clim. Chang. 3, 541–544 (2013). (2015). 5. Cissé, G. et al. Health, wellbeing, and the changing structure of communities. 30. Hulme, M. Geographical work at the boundaries of climate change published. in Climate Change 2022: Impacts, Adaptation, and Vulnerability. Contribution Transac. Inst. Brit. Geograph. 33, 5–11 (2008). of Working Group II to the Sixth Assessment Report of the Intergovernmental 31. Junqueira, A. B. et al. Interactions between climate change and infrastructure Panel on Climate Change 2nd edn, Vol. 3 (eds. Pörtner, H.-O.) Ch. 1041–1170 projects in changing water resources: an ethnobiological perspective from the (Cambridge University Press, 2022). Daasanach, Kenya. J. Ethnobiol. 41, 78962 (2021). 6. Owen, G. What makes climate change adaptation effective? a systematic 32. Pörtner, H. O. et al. Overcoming the coupled climate and biodiversity crises review of the literature. Glob. Environ. Chang. 62, 102071 (2020). and their societal impacts. Science 380, 1979 (2023). 7. Doblas-Reyes, F. J. et al. Linking global to regional climate change. in Climate 33. Orlove, B. et al. Placing diverse knowledge systems at the core of Change 2021: The Physical Science Basis. Contribution of Working Group I to transformative climate research. Ambio 52, 1431–1447 (2023). the Sixth Assessment Report of the Intergovernmental Panel on Climate Change 34. Viveiros De Castro, E. Perspectival anthropology and the method of 2nd edn, Vol.3 (eds. Masson-Delmotte, V.) Ch. 1363–1512 (Cambridge controlled equivocation. J. Soc. Anthropol. Lowland South Am. 2, 77695 University Press, 2021). (2004). 8 COMMUNICATIONS EARTH & ENVIRONMENT | (2024)5:29 | https://doi.org/10.1038/s43247-023-01164-y | www.nature.com/commsenv COMMUNICATIONS EARTH & ENVIRONMENT | https://doi.org/10.1038/s43247-023-01164-y ARTICLE 35. Bond, M. O., Anderson, B. J., Henare, T. H. A. & Wehi, P. M. Effects of RSoleymani: Methodology, Writing—review and editing. ATN: Project administration, climatically shifting species distributions on biocultural relationships. People Writing—review and editing. MA: Investigation, Writing—review and editing. EA: Nat. 1, 87–102 (2019). Investigation, Writing—review and editing. AA: Investigation, Writing—review and 36. Castree, N. et al. Changing the intellectual climate. Nat. Clim. Chang. 4, editing. JA: Investigation, Writing—review and editing. DB: Investigation, Writing— 763–768 (2014). review and editing. CB: Investigation, Writing—review and editing. JCS: Investigation, 37. Bauer, A. M. & Ellis, E. C. The Anthropocene divide: obscuring understanding Writing—review and editing. RC: Investigation, Writing—review and editing. JC: of social-environmental change. Curr. Anthropol. 59, 209–227 (2018). Investigation, Writing—review and editing. RC: Investigation, Writing—review and 38. Ford, J. D. et al. Including indigenous knowledge and experience in IPCC editing. MC: Investigation, Writing—review and editing. ZC: Investigation, Writing— assessment reports. Nat. Clim. Chang. 6, 349–353 (2016). review and editing. FC: Investigation, Writing—review and editing. EC: Investigation, 39. Mcneeley, S. M. & Lazrus, H. The cultural theory of risk for climate change Writing—review and editing. ACS: Investigation, Writing—review and editing. CD: adaptation. Weath Clim. Soc. 6, 506–519 (2014). Investigation, Writing—review and editing. ED: Investigation, Writing—review and 40. Shawoo, Z. & Thornton, T. F. The UN local communities and Indigenous editing. AFL: Investigation, Writing—review and editing. EG: Investigation, Writing— peoples’ platform: a traditional ecological knowledge‐based evaluation. WIREs review and editing. CGF: Investigation, Writing—review and editing. DG: Investigation, Clim. Chang. https://doi.org/10.1002/wcc.575 (2019). Writing—review and editing. MG: Investigation, Writing—review and editing. EH: 41. Mcnamara, K. E. & Jackson, G. Loss and damage: a review of the literature and Investigation, Writing—review and editing. TH: Investigation, Writing—review and directions for future research. Wiley Interdiscip. Rev. Clim. Chang. 10, e564 editing. JTI: Investigation, Writing—review and editing. AI: Investigation, Writing— (2019). review and editing. LJ: Investigation, Writing—review and editing. ML: Investigation, 42. Whyte, K. Too late for indigenous climate justice: ecological and relational Writing—review and editing. YLM: Investigation, Writing—review and editing. JM: tipping points. Wiley Interdiscip. Rev. Clim. Chang. 11, e603 (2020). Investigation, Writing—review and editing. GM: Investigation, Writing—review and 43. Serdeczny, O., Waters, E. B. & Chan, S. Non-economic loss and damage in the editing. MM: Investigation, Writing—review and editing. MTT: Investigation, Writing— context of climate change: understanding the challenges. Clim. Chang. 77, review and editing. MS: Investigation, Writing—review and editing. ASamakov: Inves- 89114 (2016). tigation, Writing—review and editing. RSeidler: Investigation, Writing—review and 44. Fernández-Llamazares, Á. et al. Scientists’ warning to humanity on threats to editing. VS: Investigation, Writing—review and editing. US: Investigation, Writing— indigenous and local knowledge systems. J. Ethnobiol. 41, 144–169 (2021). review and editing. ASharma: Investigation, Writing—review and editing. PS: Investi- 45. Iturbide, M. et al. Repository supporting the implementation of FAIR gation, Writing—review and editing. TU: Investigation, Writing—review and editing. principles in the IPCC-WG1 Atlas. Zenodo https://doi.org/10.5281/zenodo. RW: Investigation, Writing—review and editing. IZ: Investigation, Writing—review and 3691645 (2021). editing. 46. Harris, I., Osborn, T. J., Jones, P. & Lister, D. Version 4 of the CRU TS monthly high-resolution gridded multivariate climate dataset. Sci. Data 7, 1–18 (2020). Competing interests The authors declare no competing interests. 47. Scoones, I. Sustainable rural livelihoods: a framework for analysis. Polit. Ecol. 72, 58561 (1998). 48. Legendre, P. & Legendre, L. Numerical Ecology Vol. 990, 78625 (Elsevier, Additional information 2012). Supplementary information The online version contains supplementary material 49. Oksanen J. et al. Community Ecology Package https://rdrr.io/cran/vegan/ available at https://doi.org/10.1038/s43247-023-01164-y. (2022). 50. R Core Team, R. The R Project for Statistical Computing https://www.r- Correspondence and requests for materials should be addressed to Victoria Reyes- project.org/ (2022). García. Peer review information Communications Earth & Environment thanks Susan Chiblow, Acknowledgements Julio Postigo, and the other, anonymous, reviewer(s) for their contribution to the peer We thank all the people and communities who participated in this research for sharing review of this work. Primary Handling Editors: Niheer Dasandi, Clare Davis, Heike their time and wisdom with us. The collective knowledge that constitutes the basis for the Langenberg. A peer review file is available. analyses presented here belongs to the 48 communities with whom we have worked. We would also like to thank the students, interns, administrative staff, and advisors who have Reprints and permission information is available at http://www.nature.com/reprints participated at different stages of the LICCI project. Research leading to this paper has received funding from the European Research Council under an ERC Consolidator Grant Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in (FP7-771056-LICCI). This work contributes to the “María de Maeztu” Programme for published maps and institutional affiliations. Units of Excellence of the Spanish Ministry of Science and Innovation (CEX2019- 000940-M). JC and JTI acknowledge ANID (Fondecyt 1200291, FONDAP 15110006, and PIA/BASAL FB0002 and PFB210018). AEI acknowledge FONCYT (PICT2018-04228). Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, Author contributions adaptation, distribution and reproduction in any medium or format, as long as you give VRG: Conceptualization, Data analysis, Funding acquisition, Methodology, Project appropriate credit to the original author(s) and the source, provide a link to the Creative administration, Supervision, Writing—original draft, DGA: Methodology, Project Commons licence, and indicate if changes were made. The images or other third party administration, Supervision, Writing—review and editing. SAF: Methodology, Data material in this article are included in the article’s Creative Commons licence, unless analysis, Writing—review and editing. PB: Methodology, Project administration, indicated otherwise in a credit line to the material. If material is not included in the Supervision, Writing—review and editing. LCM: Project administration, Supervision, article’s Creative Commons licence and your intended use is not permitted by statutory Writing—review and editing. AJ: Investigation, Methodology, Project administration, regulation or exceeds the permitted use, you will need to obtain permission directly from Supervision, Writing—review and editing. VL: Methodology, Supervision, Writing— the copyright holder. To view a copy of this licence, visit http://creativecommons.org/ review and editing. XL: Methodology, Project administration, Supervision, Writing— licenses/by/4.0/. review and editing. SM: Methodology, Supervision, Writing—review and editing. VP: Investigation, Methodology, Project administration, Writing—review and editing. APF: Investigation, Methodology, Project administration, Supervision, Writing—review and © The Author(s) 2024 editing. ASchlingmann: Methodology, Project administration, Supervision, Writing— review and editing. CS: Methodology, Supervision, Writing—review and editing. 1 Institució Catalana de Recerca i Estudis Avançats (ICREA), 08010 Barcelona, Spain. 2Institut de Ciència i Tecnologia Ambientals, Universitat Autònoma de Barcelona (ICTA-UAB), 08193 Cerdanyola del Vallès, Barcelona, Spain. 3Instituto de Economía, Geografía y Demografía, Consejo Superior de Investigaciones Científicas, Madrid, Spain. 4Institut d’Estudis Regionals i Metropolitans de Barcelona (IERMB), Universitat Autònoma de Barcelona, 08193 Cerdanyola del Vallès, Spain. 5CIRAD, UMR SENS, F-34398 Montpellier, France. 6SENS, Univ Montpellier, CIRAD, IRD, UPVM, Montpellier, France. 7UMR SENS IRD Montpellier, Montpellier, France. 8Institute of Organic Farming, Department of Sustainable Agricultural Systems, University of Natural Resources and Life Sciences, 1180 Vienna, Austria. 9Rosenstiel School of Marine, Atmospheric, and Earth Science, University of Miami, Miami, USA. 10Water Systems and Global Change Group, Wageningen University and Research, COMMUNICATIONS EARTH & ENVIRONMENT | (2024)5:29 | https://doi.org/10.1038/s43247-023-01164-y | www.nature.com/commsenv 9 ARTICLE COMMUNICATIONS EARTH & ENVIRONMENT | https://doi.org/10.1038/s43247-023-01164-y Wageningen, Netherlands. 11Water Risk to Development and Resilient, International Water Management Institute, Colombo, Sri Lanka. 12 Department of Geography, Obafemi Awolowo University, Ile-Ife, Nigeria. 13Department of Geography and Regional Research, University of Vienna, 1010 Vienna, Austria. 14Mamirauá Sustainable Development Institute, Tefé, Amazonas, Brazil. 15Research Centre for the Humanities, Institute of Ethnology, H-1097 Budapest, Hungary. 16Federal University of São Paulo (Unifesp), São Paulo, Brazil. 17Rio de Janeiro State University (UERJ), Rio de Janeiro, Brazil. 18Instituto Juruá, 69083-300 Manaus, Brazil. 19Center for Integrated Disaster Risk Management (CIGIDEN), Pontificia Universidad Católica de Chile, Santiago, Chile. 20ECOS (Ecosystem-Complexity-Society) Co-Laboratory, Center for Local Development (CEDEL), Center for Intercultural and Indigenous Research (CIIR) & Center of Applied Ecology and Sustainability (CAPES), Pontificia Universidad Católica de Chile, Villarrica, Chile. 21Cape Horn International Center (CHIC), Universidad de Magallanes, Puerto Williams, Chile. 22College of Agriculture and Environmental Science (CAES), University of South Africa, Florida, 1710 Johannesburg, South Africa. 23Faculty of Social Sciences, University of Helsinki, Helsinki, Finland. 24Institute of Resource Assessment, University of Dar es Salaam, 16103 Dar es Salaam, Tanzania. 25 School of Life and Health Sciences, University of Roehampton London, Roehampton Lane, London SW15 5PJ, United Kingdom. 26Centro Boliviano de Desarrollo Socio Integral (CBIDSI), San Borja, Beni, Bolivia. 27Department of Environment and Geography, York Institute for Tropical Ecosystems, University of York, York, United Kingdom. 28Department of International Environmental and Development Studies (NORAGRIC), Norwegian University of Life Sciences, Ås, Norway. 29IRD GET UMR 5563, Universite de Toulouse, Toulouse, France. 30Dickinson College, 28 N College St, Carlisle, PA 17013, USA. 31Helsinki Institute of Sustainability Science (HELSUS), Faculty of Biological and Environmental Science, University of Helsinki, FI-00014 Helsinki, Finland. 32Department of Geography, Virginia Polytechnic Institute and State University (Virginia Tech), Blacksburg, USA. 33School of Sustainability, Arizona State University, Tempe, AZ, USA. 34Department of Anthropology, School of Language, Culture and Society, Oregon State University, Corvallis, Oregon 97331, USA. 35Iniciativa Amotocodie, Asunción, Paraguay. 36Department of Earth and Environment, Franklin & Marshall College, Lancaster, PA 17603, USA. 37Instituto Multidisciplinario de Biología Vegetal (IMBiV), CONICET- Universidad Nacional de Córdoba, Córdoba, Argentina. 38Facultad de Ciencias Naturales e Instituto M. Lillo, Universidad Nacional de Tucumán, de Tucumán, Argentina. 39Nordregio, Nordic centre for regional development and planning, SE-111 86 Stockholm, Sweden. 40The Nelson Institute for Environmental Studies, University of Wisconsin-Madison, Madison, WI, USA. 41Indigenous Science, CP, 97000 Mérida, Yucatán, México. 42 Department of Nature and Life Sciences, Faculty of Nature and Life Sciences; Ibn Khaldoun University, Tiaret, Algeria. 43Laboratory of Agro- Biotechnology and Nutrition Research in Semi-Arid Areas, Ibn Khaldoun University, Tiaret, Algeria. 44Global Change and Conservation (GCC), Organismal and Evolutionary Biology Research Programme, Faculty of Biological and Environmental Sciences, University of Helsinki, FI-00014 Helsinki, Finland. 45Centre for Sustainable Development and Environment (Cenesta), Tehran, Iran. 46Eberhard Karls Universität Tübingen, Tübingen, Germany. 47Department of Biology, University of Massachusetts Boston, Boston, USA. 48ATREE-USA, Belmont, USA. 49Arctic, Remote, and Cold Territories, Interdisciplinary Center (ARCTICenter), Department of Geography, University of Northern Iowa, 50614 Cedar Falls, USA. 50 Russian Foreign Trade Academy Far East Branch, 683003 Petropavlovsk-Kamchatsky, Russia. 51Global Institute for Interdisciplinary Studies, Kathmandu, Nepal. 52Department of Environmental Technology, CSIR- Institute of Himalayan Bioresource Technology Palampur, 176061 Palampur, HP, India. 53School of Science and Technology, University of Fiji, Saweni, Lautoka, Fiji. 54Zoological Society of London, Mongolia Representative Office, Ulaanbaatar 14201, Mongolia. 55Department of Sociology and Anthropology, Peking University, Beijing, China. 56 Department of Mathematics and Computer Science, Abdou Moumouni University, 10896 Niamey, Niger. ✉email: Victoria.reyes@uab.cat 10 COMMUNICATIONS EARTH & ENVIRONMENT | (2024)5:29 | https://doi.org/10.1038/s43247-023-01164-y | www.nature.com/commsenv