Long-Term Environmental and Social Monitoring in the Andes: State of the Art, Knowledge Gaps, and Priorities for an Integrated Agenda

Andean social–ecological systems (SES) play a key role in the livelihoods of South American people by conserving biodiversity, providing natural resources, and regulating water supply. Long-term social–ecological monitoring (LTSEM) of Andean SES needs to be coordinated to inform sustainable management and increase resilience. We combined quantitative and qualitative approaches to identify the state of the art, knowledge gaps, and monitoring priorities for a research agenda targeted toward Andean LTSEM. We carried out participatory and transdisciplinary meetings to design a conceptual model of the functioning and monitoring of Andean SES. This was contrasted with the themes and labels of LTSEMs identified through an electronic survey and active searches of bibliographies and montane monitoring networks. Most LTSEM addressed biophysical issues, with a minor fraction addressing social aspects; participatory efforts were very rare. By combining both approaches, we identified research priorities that were grouped into 5 categories. Our main proposals advocate: (1) the development of integrated models of Andean SES to frame a transdisciplinary approach in long-term studies, (2) the coordination of independent LTSEMs to forecast the functioning of Andean SES under environmental change scenarios, (3) the inclusion of external dynamics and drivers on Andean systems, (4) the promotion of science–policy dialogue to attain a more effective governance of mountain SES, and (5) the increase of information accessibility to improve the adaptive management of Andean SES.


Introduction
The coordination of long-term social and environmental monitoring systems should be a core component of any strategy to study social-ecological responses to climate and land-use changes in mountain regions.Ecosystems are essential to the functioning of societies (Isbell et al 2017), and societies in turn influence ecosystems directly (eg through land-use changes) and indirectly (eg through actions that cause climate change; Krausmann et al 2013).Globally, different initiatives focused on mountain regions have highlighted the role that long-term monitoring of socialecological systems (SES) should play in understanding their dynamics (eg Bj€ ornsen Gurung et al 2012; Peralvo and Bustamante 2015;Gleeson et al 2016;Adler et al 2018).Longterm social-ecological monitoring (LTSEM) efforts are expensive and demanding in terms of their institutional governance.Therefore, it is important to coordinate their objectives and logistics across spatial scales, so they can answer relevant questions and inform management policies ranging from the local to the national and continental contexts (Wymann et al 2018;Llambí et al 2019).
Due to their steep topographic environmental gradients and wide spatiotemporal variability, Andean SES are particularly vulnerable to environmental changes (Cuesta et al 2020).Thus, different management challenges arise because Andean SES play a key role in the sustenance and functioning of human societies at local to continental scales (Schoolmeester et al 2016).Around 84 million people live in the Andes, and the welfare of a significant portion of lowland inhabitants depends on Andean ecosystem services (Devenish and Gianella 2012).Andean systems act as biodiversity refuges and are crucial to climate regulation and water provision for lowlands; they also provide food, fiber, and forest resources for local populations (Buytaert et al 2011).A thorough understanding of the functioning of the Andes requires long-term interdisciplinary monitoring efforts integrating approaches that consider multiple spatial and temporal scales (Llambí and Garc es 2021).
A recent research agenda for Andean forest landscapes identified major knowledge gaps around the interactions between social and environmental dynamics (Mathez-Stiefel et al 2017).Other synthesis works have highlighted the importance of coordinating monitoring efforts of Andean SES at a continental scale to inform policies and sustainable management programs, and to design effective conservation strategies (Cuesta et al 2012;Llambí et al 2019).LTSEM, the systematic recording of social and environmental indicators over an extended period (Lindenmayer and Likens 2009), can provide valuable information for understanding the nonlinear responses of complex SES under ongoing climate change, through the use of indicators that, being locally relevant, allow comparison across studies (Maldonado et al 2012;Huggel et al 2015).However, to coordinate Andean monitoring efforts and achieve sustainable management, it is necessary to identify the main information gaps and questions to be addressed from a transdisciplinary perspective.
The aim of this study was to identify knowledge gaps and monitoring priorities that can inform the design of a research agenda for LTSEM to understand the response of Andean SES to climate and land-use changes.Specifically, we: (1) proposed a simple conceptual model integrating the relationships between the processes of social and environmental change and the main stakeholders involved in knowledge generation and management; (2) identified and reviewed available long-term monitoring systems across the region and analyzed their congruence with the conceptual model; and (3) combined quantitative methods and experts' perceptions to identify knowledge gaps, research priorities, and key thematic areas for Andean LTSEM.

Methods
To identify knowledge gaps and research priorities for Andean LTSEM, we combined quantitative methods with expert knowledge.We built a conceptual model to be contrasted with existing LTSEM in the region in terms of the addressed topics (climate and hydrology, environmental quality, fauna, vegetation and disturbances, land-use and land-cover change, social-environmental, and social analyses).We conducted 2 virtual interdisciplinary meetings (IER 2021a(IER , 2021b) ) to collect the opinions of experts and knowledge demands of key Andean stakeholders.The first meeting was a discussion panel designed to promote the dialogue between scientists and decision-makers, share monitoring experiences, and identify challenges to integrating Andean LTSEM systems.The outputs of this meeting, which gathered 53 actors from all the Andean countries, the United States, Switzerland, and the UK, were synthesized into a conceptual model describing the processes of knowledge generation on the dynamics of Andean SES.The description and labeling (Figure 1) of the main relationships identified in the model were used to classify existing Andean LTSEM systems.We also conducted an immersive session that gathered researchers from natural and social sciences and decision-makers from public and private sectors from all Andean countries.Five of the 17 participants with a range of backgrounds combined the conceptual model with their knowledge and experience of the region to identify the main constraints to implementing efficient LTSEMs and to integrating them into decision processes.
To analyze the state of the art of Andean LTSEM, we compiled a database of existing efforts, including an electronic survey, a literature search, and a compilation of active research networks (see Appendixes S1, S2, and S3, Supplemental material, https://doi.org/10.1659/mrd.2022.00018.S1).We surveyed Andean LTSEM systems through an open questionnaire implemented in Google Forms and distributed through email to 425 experts and decision-makers known to be involved in research, monitoring, or policy design in the Andes.Besides providing feedback about the nature of their monitoring systems, experts also indicated key publications.To complement these publications, we performed a literature search using Google Scholar to identify published articles and reports based on Andean LTSEM.The results of this search, driven through different word combinations (see Appendix S2, Supplemental material, https://doi.org/10.1659/mrd.2022.00018.S1), constituted a representative sample of key issues considered in Andean LTSEM literature.We also compiled and systematized LTSEM platforms, networks, observatories programs, and projects in the region (ie existing networks), identified through diverse mechanisms (eg our research experience, expert opinion, the electronic survey, and the literature review).We compiled and mapped the geographic locations of the study sites reported in the electronic survey and publications.Each LTSEM system (ie record) was assigned to one or more topics and labels identified in the conceptual model.To assess agreement between the model and the LTSEM database, we quantified the frequency of records corresponding to each label.

Conceptual model
The conceptual model graphically represents how different stakeholders are articulated in the management of Andean SES in a context of global environmental change, and the related flows of information relevant to LTSEM (Figure 1).It includes 4 types of elements: systems, stakeholders (ie actors), processes, and their interactions, which were labeled and contrasted with existing LTSEM.This model is a conceptual tool with which to explore knowledge generation and governance of Andean SES and the equitable distribution of decision-making power, as it includes the main mechanisms that involve actors in research and knowledge production and its application to informed decision-making (Ariza-Montobbio and Cuvi 2020).
Andean social-ecological systems: Since societies depend on functioning ecosystems, and there is almost no ecosystem free from anthropic influence, it is essential to consider SES as a single unit of analysis (Petrosillo et al 2015), highlighting the complex interactions among biophysical and social components (Llambí and Llambí 2001;Pulver et al 2018).LTSEM of Andean systems depends on the identification of sensitive indicators and variables that are easy to measure and summarize the state of the SES (Turnhout et al 2007;Maggino and Zumbo 2012).For example, land-use and landcover changes are key indicators of the effects of society on ecosystems (Potschin 2009); moreover, the provision of ecosystem services should be assessed with indicators quantifying their demand by society (Burkhard et al 2012).
Stakeholders: Three groups of actors interact with SES and are involved in their conceptualization, research, and monitoring (Young et al 2006).These groups, with different spheres of action (public and private) and scales (local to global), conceptualize and influence the system differently (Villamor et al 2014).
Academia: Scientists usually describe systems through the quantification of indicators (Carpenter et al 2005), with a clear distinction between biophysical and social spheres.However, transdisciplinary studies are increasingly deemed to be better sources for informing decisionmaking processes (Mathez-Stiefel et al 2017).Thus, scientists are increasingly interacting with other stakeholders, considering alternative visions of SES (eg Teng€ o et al 2014), and promoting research codesign, which confers a greater decision power to the community (Moser 2016).Moreover, transdisciplinary research can facilitate the integration of scientific knowledge, governmental requirements, and community demands and perspectives (Holzer et al 2018).Governments: Governments have traditionally been responsible for managing SES through legislation, policies, and so forth (Ebbesson 2010).However, in recent years, the increasing complexity of SES has fostered the implementation of participatory and adaptive management (Dupuits 2021), in which governments and society define goals and policies.These are assessed on the basis of expected results and observed data (Vos et al 2000).Community involvement in LTSEM increases the transparency and usability of data, favoring effective governance and comanagement (Armitage et al 2009).Community: Community includes groups with diverse visions of SES and knowledge demands (eg local and Indigenous communities, private sector, nongovernmental organizations; Teng€ o et al 2014).These groups are increasingly involved in research design (Moser 2016) and participatory monitoring (Staddon et al 2014), which is facilitated by automatized data collection (eg Conrad and Hilchey 2011).With increased access to information, Andean communities can be involved in participatory management and local policy design and decision-making (Dupuits et al 2022).
Processes: Andean SES are influenced by external factors such as climate change, telecouplings, and international legislation (Young et al 2006).The widespread manifestations of climate change (Parmesan 2006) are magnified in Andean SES because of the steep gradients and vulnerable social-ecosystems (Carey 2010).Telecouplings, the exchange of material, money, information, and people between distant systems (Hull and Liu 2018), increasingly influence Andean dynamics (eg farmers' decisions are influenced by global food markets; Llambí and Llambí 2001;Zimmerer et al 2018).Andean SES are also influenced by international legislation (ie agreements that regulate global relations and standards) driven by supranational organizations and conventions (eg United Nations Framework Convention on Climate Change, Intergovernmental Panel on Climate Change, Intergovernmental Science-Policy Platform on Biodiversity and Ecosystem Services), intended to coordinate policies and minimize environmental injustices through the use of scientific information (Galaz et al 2012;Biermann et al 2016).

Characterization of monitoring efforts and identification of knowledge gaps
We analyzed 239 records of Andean LTSEM systems that resulted from 65 responses to the electronic survey, 124 published studies (from a total of 2443 publications) that passed scrutiny to ensure they addressed Andean LTSEM, and 50 active networks including Andean LTSEM systems (see Appendixes S1, S2, and S3, respectively, Supplemental material, https://doi.org/10.1659/mrd.2022.00018.S1).Each record was assigned one or more labels from the conceptual model, which allowed the fit between expected and existing monitoring efforts to be identified (Figure 1).We acknowledge that some monitoring efforts may be overrepresented (eg literature records or networks reported in the electronic survey).However, rather than performing a formal numerical analysis, we aimed to identify general patterns and research gaps, depicting the most infrequent (eg unpublished) labels within our conceptual model (eg those involving community-based approaches).The inclusion of all the available sources of information increased the likelihood of detecting these gaps.In addition, to visually assess the geographic representativeness of our study and of the LTSEM efforts in the region, we mapped every record (survey and published studies with known study area; Figure 2).
Tabulating the labels assigned to the 239 LTSEM records and their graphical representation in the model (Figure 1) allowed us to identify research gaps around Andean SES dynamics.While most Andean LTSEM systems are related to environmental issues (235 records considered some environmental aspect, and 162 records did not consider any social aspect at all), with vegetation (n ¼ 106) and climate (n ¼ 100) being the most studied topics, social aspects were underrepresented.The effect of climate change was considered in 24 LTSEM systems, and international legislation was explicit in only 3 of these systems.Socialenvironmental studies were mainly focused on environmental quality (eg water and air conditions and urban solid wastes, n ¼ 28), ecosystem services (n ¼ 25; eg Fern ández-Turiel et al 2003), or the quantification of landuse and land-cover changes (n ¼ 24), and social aspects were usually disregarded (eg studies monitor environmental quality but do not explicitly quantify its effects on human health; Gramsch et al 2006).Only 4 of the identified LTSEMs focused solely on the social sciences (eg L ópez-Sandoval and Maldonado 2019).
The disproportion between biophysical and social monitoring efforts is likely explained by methodological issues; some biophysical indicators have been used to summarize the state of an environmental system for a long time (eg rainfall, temperature).By contrast, the complexity and spatial variability of social systems and rapid technological and economic changes complicate the use of single indicators through time and space.Some additional aspects may contribute to the low representation of social aspects: (1) long-term social studies are seldom integrated into formal monitoring networks; (2) most social monitoring efforts are part of official statistics and programs such as national censuses, which were not detected explicitly in our search, although they are important sources of information for social-ecological research; and (3) social research is still infrequently integrated explicitly into social-environmental initiatives.
We found that community participation is scarce in the design, implementation, and operation of Andean LTSEM systems.Only one LTSEM, led by ProYungas Foundation in collaboration with governments and private companies, implemented codesigned research.This is also 1 of the 2 records that involved adaptive management (the other one is a risk management project in Colombia; Alzate Buitrago 2010).Participatory monitoring was more frequent in the implementation of Andean LTSEM; local communities participated in data collection for 9 LTSEM systems, but they were not involved in data analysis or interpretation.Most examples of community participation involved hydrological monitoring (eg Saavedra Daza et al 2020; Ulloa et al 2021) and took advantage of cell phone applications and webbased data collection (eg World Water Monitoring Challenge or the Litterati platform for surveying urban solid waste).Collaboration with local communities is deemed essential for successful social-environmental monitoring, as it promotes knowledge appropriation and increases the social relevance and applicability of research.However, the implementation of community participation poses big challenges, such as creating an effective dialogue between technical quantification and community knowledge.Participatory monitoring seems to be more likely when interests are transparent and homogeneous (eg private sector with clearly formulated goals or when some resource [eg water] is essential for the whole group) and when the participatory methods are accessible to local stakeholders (eg Litterati citizen science).
FIGURE 2 Geographic distribution of monitoring systems and temporal studies in the Andean region detected through electronic surveys and literature search.More than one location could be reported by an electronic survey.
Although academics are frequently consulted by governments, and LTSEM systems depend on public funds, none of the records presented effective comanagement between academia and government; it is possible that the lack of common expectations, language, and time frames is a source of mutual distrust (Karam-Gemael et al 2018).Policymakers usually need rapid technical responses, while scientists are interested in novel and abstract problems, which are not always linked with social demands.Adaptive management (reported in 2 records; a single publication of Dallmeier et al 2002), considered an efficient tool to deal with uncertainty, implies a continuous reassessment of policies by contrasting expected and observed results (which need to be monitored).Most of the time, policymakers must address pressing problems and have no time or resources to monitor the impacts of implemented policies or revisit past problems that are assumed to be solved (eg crisis of water scarcity).This may raise a problem in coordinating and implementing effective adaptive management and policy implementation across scales (see Dupuits et al 2022).Even though policies are usually implemented in single geographic systems, their results could provide information across multiple spatiotemporal scales, a benefit that is usually beyond the scope of short-term policymaking.In this context, strengthening communication between policymakers and academics (eg comanagement of information platforms, explicit social-environmental science-policy dialogues) is key to reinforcing monitoring networks and regional political platforms and actions (eg within the framework of the Andean Mountain Initiative and other continental policy forums).
Collaborative networks promote the use of indicators in LTSEM, facilitating comparison and integration of different case studies (eg Carilla et al 2013;Cuesta et al 2020;Malizia et al 2020) and different areas of knowledge (Jimenez et al 2021).However, some constraints may arise in the selection of comparable variables because most LTSEM systems choose indicators based on specific social-ecological problems (eg changes in plant diversity, water pollution).Biophysical sciences have identified indicators that are easy to measure (eg the normalized difference vegetation index [NDVI] indicates vegetation activity, and the abundance of some organisms is deemed to be a good proxy of the health of an ecosystem; Linares Arias et al 2019).By contrast, social-environmental and social indicators enabling the comparison between sites are difficult to identify due to the diversity of social actors and conditions between Andean countries and landscapes (Bustamante et al 2014).The lack of core social indicators and networks focused on social issues reduces the opportunity to perform integrative analyses and create regional maps with adequate resolution.Although social studies might be underrepresented in our search, there are at least 2 objective aspects that make the integration of social studies more difficult.On the one hand, social issues are more complex than biophysical systems and present multiple nonlinear responses, so simple models incapable of extrapolation are generally less useful.On the other hand, the narrative approach, widely used in social areas, permits accurate characterization of complex systems, but their inclusion in quantitative comparisons and metaanalyses, particularly across wide spatial scales, is difficult.
Global changes affecting Andean SES highlight the importance of LTSEM because they generate nonanalogous conditions precluding the extrapolation of results.While most LTSEM systems only mention the potential effects of climate change on the Andean SES, some LTSEM systems assess these effects on environmental dynamics (eg Gloria or RBA networks; Morales et al 2015;Carilla et al 2018;Fadrique et al 2018).In contrast, we found no assessments of telecoupling effects, even though they are considered to be relevant drivers of Andean land-use changes (Aide et al 2019).International agreements advocate knowledge-based policies to handle problems that cannot be addressed by local governments (eg Aichi Biodiversity Targets; O'Connor et al 2015), and some LTSEM systems are considered to inform these policies: At least 3 networks were involved in the transference of knowledge to inform international agreements (Global Change in Mountain Regions or GLOCHAMORE, World Glacier Monitoring Service, and Global Cryosphere Watch, from the United Nations Educational, Scientific and Cultural Organization and World Meteorological Organization, which are specialized agencies of the United Nations).Information from LTSEM in the Andean region has informed international reports aimed at decision-makers such as those of the Intergovernmental Panel on Climate Change (eg Adler et al 2022).Recent efforts have also aimed to integrate LTSEM results into policy briefs and facilitate communication mechanisms of regional policy platforms such as the Andean Mountain Initiative (eg see iam-andes.org).

Research priorities for the development of an agenda for integrated social-environmental monitoring in the Andes and conclusions
Andean LTSEM systems constitute the best source of information to understand SES responses to unprecedented global changes and to increase their resilience.We combined a conceptual model representing interactions within Andean SES with an active search of LTSEM systems (through an electronic survey, literature review, and search of existing networks) to characterize monitoring efforts in terms of the topics addressed and the actors involved.This characterization of existing studies is necessary to coordinate efforts to fulfill existing knowledge gaps, facilitate information sharing, summarize complex processes at different spatial and temporal scales, and inform local social-environmental policies and international agreements.
The combination of our quantitative approach with the experts' opinions compiled from 2 interdisciplinary workshops allowed us to identify some research priorities for Andean LTSEM.We classified them into 5 categories (Table 1).( 1) System conceptualization: The characterization of Andean SES, identifying main stakeholders, economic activities, ecological elements, and their interactions, should form the basis for analytical classifications of Andean SES (eg social-ecological land systems; Zarb á et al 2022).This classification, based on unified indicators and conceptual approaches, will permit more accurate integration of SES knowledge and the establishment of LTSEM at representative learning sites.(2) System functioning: The integration of systematic social-ecological information gathered through unified protocols with alternative proxies to fill geographic gaps (eg derived from remote sensing) and modeling approaches will allow a comprehensive understanding of Andean SES.The integration of TABLE 1 Research priorities for integrated monitoring of Andean SES identified from the perception of experts representing academia, government, and private sectors of different Andean countries (compiled during 2 interdisciplinary workshops), and from the analysis of different information sources on the state of the art of environmental and social monitoring along the Andes (surveys, literature, and networks) and the ways in which they are linked with the proposed conceptual model.Each priority was assigned with a recommendation type: ''What'' refers to the content of monitoring, ''How'' refers to the strategy, and ''Why'' refers to the purpose of the monitoring.(Table continued  will strengthen existing research networks, foster knowledge coproduction, support science-policy dialogue, and facilitate science-based governance.Making coherently gathered information accessible is necessary for regional synthesis analyses (eg Malizia et al 2020), which are key pieces in understanding complex systems.In summary, the foundation and the point of departure for the consolidation of the integrated monitoring agenda advocated here should be a reconceptualization of Andean SES, focusing on social-ecological interactions and considering the views of different stakeholders.It should also involve the integration of different disciplines and spatiotemporal scales.The identification of unified indicators leading to regional comparisons and providing relevant information for decision-making at multiple spatial scales remains a central challenge for LTSEM in the Andes.Research networks and long-term learning sites will play key roles in generating field information, measuring integrative indicators, and performing analytical syntheses (combining observational, experimental, and remote-sensing data), which can better inform science-based policies.Moreover, the participation of diverse stakeholders in knowledge production and system management is essential to strengthen governance and to promote the resilience of Identify the needs and demands for financial resources and cofinancial strategies for socialecological research and monitoring for the coming years from both the public and private sectors (companies, research centers, universities, among others) at national and regional levels.

4.6
Develop research and funding strategies to enhance replicability and comparability of social and environmental research at a continental scale that can inform global policies.
Why 5. Information accessibility 5.1 Promote regional integration of existing networks with different scopes and approaches (social, environmental), based on a comprehensive action plan that involves cooperation mechanisms, promotion of learning sites, integration of scales, capacity building, and identification of key indicators to guide the various monitoring efforts.
How 5.2 Systematize available regional and multiscale information (eg through regional comparative analyses and social-ecological indicator systems accessible to different stakeholders through virtual platforms). How

5.3
Strengthen collaborative regional long-term monitoring systems and efforts to integrate environmental information into regional-scale analyses, which are widely communicated (eg via scientific publications, integrative web portals, policy briefs, science-policy dialogs) and incorporated into social-ecosystem management.

How, why
Andean SES to global changes that will increasingly influence these systems.

A C K N O W L E D G M E N T S
We thank the 2 reviewers and Editor Susanne Wymann von Dach for their very valuable suggestions and comments.Special thanks go to all the experts and decision-makers who participated in the workshops and provided us with their insights on the challenges for integrated monitoring in the Andes.We also want to acknowledge the financial support of the Adaptation at Altitude and Andean Forest Programmes (SDC) and the Scientific and Technological Research Fund of Argentina (FONCyT, PICT 2020 03355).
O P E N P E E R R E V I E W on next page.) (5) Information accessibility: The growing availability of data and information (eg via integrated social-ecological indicator platforms; see Adaptaci ón en las Alturas et al n.d.)

TABLE 1
Continued.(Firstpart of Table1on previous page.)Strengthen regional, national, and local capacities in planning, execution, and management of research, monitoring, and technological development projects in social-environmental matters, including innovative strategies for promoting citizen science through the use of new technologies (eg cell phone applications).Generate joint research agendas, in the medium and long term, involving the public and private sectors, local communities, and academia, promoting research and monitoring codesign and comanagement.Promote the use of environmental and social monitoring information for the design of more inclusive, equitable, and sustainable governance systems adapted to local realities.
This article was reviewed by Jere L. Gilles and Elizabeth Jim enez.The peer review process for all MountainAgenda articles is open.In shaping target knowledge, values are explicitly at stake.The open review process offers authors and reviewers the opportunity to engage in a discussion about these values.