Governance in socio-environmental research: an analysis of multi-stakeholder cooperation mechanisms in two research laboratories in Yucatan, Mexico

ABSTRACT This article analyzes the governance of socio-environmental research in two laboratories in Yucatan, Mexico, LANRESC and LENERSE, aiming to explore the operation of cooperation mechanisms involving multi-stakeholder interaction processes that support the generation of knowledge and the development of strategies to solve complex socio-environmental issues. The analytical-conceptual framework was that of complex associative systems, a governance approach focused on understanding associative efforts among heterogeneous stakeholders seeking to reach a common objective. The present research was based on two case studies from a comparative perspective. We found that both experiences have involved the development of relatively stable networks adapted to meet their objectives. However, the less conventional LANRESC research scheme has succeeded in producing more innovative structures and mechanisms to operate as non-scientific stakeholders and be recognized as a national laboratory. For their part, the mechanisms of research organization at LENERSE, being more traditional and technocentric in nature, have not been able to fully establish links with social and business actors to participate in the regional energy transition. In the case of LENERSE, the traditional and technocentric evaluation mechanisms in place have inhibited the establishment of links with social and business actors to influence energy transitioning in the region. In both cases, betting on inter- and trans-discipline will be essential for improving their contributions to solving problems such as climate change or oil depletion.


Introduction
Socio-environmental research aims to produce knowledge to understand and solve complex issues arising from the aggravation of the global environmental crisis.Climate change, loss of biodiversity, depletion of water reserves, excessive energy demand, and food insecurity are some of the main problems that demand attention through knowledge-based actions.This requires new institutional support for research through a more inclusive inter-and transdisciplinary approach incorporating local knowledge systems and paving the way for the collaborative design of solutions among diverse stakeholders, including scientific communities and experts, governments, funding organizations, and civil society.
In this framework, the governance concept applied to social-environmental research becomes relevant as an analytical approach that allows examining the interaction between governmental and non-governmental stakeholders in a territory; assessing the extent to which they build capacities to reach agreements and consensuses aiming to meet common objectives, while building shared visions of the territory through multidirectional (horizontal and vertical) cooperation between various actors (Bagnasco and Le Galés 2000).Collaboration between stakeholders allows managers to incorporate diverse types of knowledge into a process and understand how stakeholders might respond to a political decision (Newig, Günther, and Pahl-Wostl 2010).
Examining the organization and functioning conditions of the processes mentioned above is essential to improving not only their comprehension, but also their outcomes in terms of the effectiveness of the application of knowledge in solving specific socialenvironmental issues.In this sense, several authors have explored these issues by applying concepts such as "territorial governance" (Martínez Salvador and Martínez Salvador 2021; Torres-Salcido 2008); "governance networks" (Newig, Günther, and Pahl-Wostl 2010;Rhodes 2007); and "collaborative governance" (Ulibarry 2019); these authors agree on the need to integrate multiple actors in these processes and, above all, promote a greater participation and agency in decision-making that, ideally, translates into environmental policies built in a more participatory way.
In this context, this article explores a perspective that has been shown to be suitable for analyzing governance processes involving multi-stakeholder partnerships and more robust cooperative mechanisms, named complex associative systems (CASs).CASs are a conceptual public action network model that seeks to understand the combined efforts resulting from the integration of actors within a governance network, understood as a relatively stable structure of social relationships between different actors interacting in an interdependent manner around issues or topics of common interest, in accordance with formal or informal rules (Newig, Günther, and Pahl-Wostl 2010;Klijn, Koppenjan, and Termeer 1995).A network of this type achieves the CASs status when it achieves a significant degree of autonomy and interdependence among its members (Luna and Chávez 2014).This means that the actors involved have their own identity, ownership, and resources, which give them independence, but who require cooperating with others to achieve certain objectives (Puga 2021).This leads to tensions that stimulate the development of rules and strategies to achieve network sustainability through a dynamic process.
This article outlines the main results and findings of research that, based on the CASs approach, analyzed governance processes and multi-stakeholder cooperation mechanisms deployed at the Laboratorio Nacional de Resiliencia Costera (National Coastal Resilience Laboratory; LANRESC) and the Laboratorio de Energías Renovables del Sureste (Renewable Energy Laboratory of the Southeast; LENERSE), both located in the state of Yucatan, Mexico.
Importantly, the Yucatan Peninsula is a critical region from a socio-environmental standpoint, mainly because of the heterogeneity of the regional ecosystems, since it is part of the Mesoamerican reef system and the Mesoamerican biological corridor.It harbors 12 million hectares of forests and the largest tropical forest north of the Amazon in Latin America, i.e. the Sian Kaan-Calakmul Natural Corridor, one of the most important carbon reservoirs worldwide after the Amazon.At the same time, the Yucatan peninsula is highly vulnerable to current and potential impacts of climate change.It is located within the subtropical line of high incidence of hurricanes and other climatic anomalies, making it a hotspot of socio-environmental problems linked to climate change involving human settlements; agricultural, forestry, and livestock raising; tourism; and availability of water and energy resources; among others, which also increase social inequity (Soares and Gutiérrez 2011;García et al. 2007).
The establishment of the laboratories is based on this territorial context; both were developed and funded with public resourcesmainly from the federal government and, to a lesser extent, from the state government, which in the past decade has set an agenda of scientific and technological policy that considered socio-environmental issues with a specific territorial expression as top priorities.In this sense, both LANRESC and LENERSE can be conceived as scientific policy instruments since they materialize the state government's intervention to address these priorities by building new knowledge production capabilities, expressed in new physical infrastructure and above all, in the establishment of linked research environments involving the exchange of diverse resources, information, data, and practices: not only between scientists, but also with non-scientific actors.The ultimate goal is to address socio-environmental issues related to climate change in coastal zones of Yucatan through LANSREC, as well as issues related to renewable energies and energy transition in the region through LENERSE.
In this way, a case study from a comparative perspective was carried out, with the premise that both laboratories have set inter-institutional arrangements and configured associative schemes between various stakeholders for the development of inter-and transdisciplinary research with relatively successful results in terms of the understanding and impact of key issues.Nevertheless LANRESC has achieved a more plural and diverse multi-actor organization than LENERSE.Therefore, the research question that guided the present study was how the cooperation mechanisms between the actors involved in both laboratories have progressed in LANRESC and LENERSE to meet their primary objective, and the results achieved in each experience from these different configuration modalities.Particularly, the objective was to determine how these two networks function, both in the political dimension (leadership and decision-making) and in the social dimension (network cohesion), to identify and analyze the results in terms of functional performance (achievement of the objectives set) and associative performance (learnings and collaboration capacities developed), particularly for the inter-and transdisciplinary research.
This article is structured in four sections.The first section outlines our conceptual and analytical frameworks; the second, the methodological strategy.The third, analyzes the conditions shown by the two case studies as examples of CASs.The fourth, discusses the development of the cooperation mechanisms in both experiences to achieve sustainability and meet their objectives, as well as the results produced.Finally the conclusions of the study are outlined.

Governance processes in socio-environmental research: an analyticalconceptual framework
This section outlines the analytical framework that guides this study, which represents a novel approach for understanding governance processes in the laboratories analyzed.Methodologies to explore cooperative knowledge production in socio-environmental research are combined, as well as approaches on governance, particularly CASs, to explain the dynamics of cooperation and stable partnership between diverse social stakeholders.

Socio-Environmental research: new logics and challenges for the coproduction of knowledge
Given the magnitude of the current environmental crisis and its implications for the sustainability of the Earth system and its inhabitants, social-environmental research is projected as one of the most dynamic areas of research and knowledge production on the global scientific agenda.In recent decades, the urgency of building a new paradigm of environmental research that helps to face this crisis and achieve sustainability has been underscored (Turner et al. 2016).
In this sense, several authors agree that the new paradigm should respond to the complexity of social-environmental issues, which require not only interdisciplinary but, more importantly, transdisciplinary research, through cooperative knowledge production approaches involving public and private stakeholders well beyond the scientific field.
The "Post-Normal Science" concept put forward by Funtowicz and Ravetz (1999) is one of the earliest theoretical approaches underlining the need to address social-environmental issues from novel research approaches.The complex nature of these issueswhich entail uncertain factors, disputed values, and urgent political decision-makingmakes it hard to foresee all the consequences and risks involved in these decisions.According to these authors, the canonical methods of knowledge production are insufficient and, instead, scientific problem-solving strategies should be articulated customized to particular contexts.It is there that postnormal science emerges as a methodology that brings together elements from scientific research and political and social management, considering the perspectives of the various stakeholders involved.To this end, the incorporation of participatory mechanisms is proposed to expand current knowledge about environmental systems and improve political decision-making within an "extended community of peers" (Funtowicz and Ravetz 1999).
Given the above, research focused on social-environmental systems supposes a shift in the conception of the world, which could even represent an epistemological revolution sensu Kuhn, which also involves the complexity and uncertainty concepts intrinsic to this type of systems (Giannuzzo 2010).This complexity is not only associated with a larger number of variables to be analyzed, but also with an interdisciplinary design with closer involvement in research.For its part, uncertainty derives from multiple sources: lack of data, inaccurate data, limited understanding of the system, among others.Therefore, risk is a constant that shall be assumed as inherent of these processes, but it should not discourage actions (Giannuzzo 2010).
The new paradigm, then, implies challenges related to new cognitive dynamics and forms of organization of social-environmental research guided by interdiscipline and transdiscipline principles; this requires the formation of work teams to develop conceptual and methodological frameworks in line with these principles (Moldenhaouer and Johnson 2016).At the same time, the challenge is to build a greater capacity to respond to the needs of society when choosing research priorities; set explicit objectives to produce knowledge that is useful and applicable to social-environmental issues; work coordinately with government actors and agencies to promote the use of evidence and research results in decision-making, among other challenges of enormous importance (Turner et al. 2016).
In response to these challenges, novel frameworks for social-environmental research have been proposed, such as open science that advocates participatory and trans-disciplinary research through the interaction of stakeholders with diverse interests, targeting intersectoral collaboration between academia and society (Kondo et al. 2019;Moldenhaouer and Johnson 2016).Under this approach, research experts and social stakeholders, such as government agencies, industries, NGOs, and members of civil society adopt coleadership roles in decision-making through research agendas designed for the coordinated production of knowledge and the co dissemination of results.
Despite the growing trend to put these new approaches and logics into practice, social-environmental research does not always translate into solutions and improvements.Often times, there is disarticulation between environmental issues, the research that studies them, and the actions necessary to tackle them (Giannuzzo 2010).These mismatches reveal coordination failures that can be understood from the perspective of governance, which offers an analytical framework to examine the interaction and cooperation processes between the multiple actors that incorporate heterogeneous values, concerns, intents, and purposes to build applied knowledge facing a social-environmental problem.When these cooperative interactions are organized around shared objectives and projected over time, they become a more stable partnership or alliance that allows to jointly build solutions for issues based on the integration of knowledge, achieving their application with transformative results to manage specific social-environmental issues.
Thus, the concept of governance applied to social-environmental research also takes on new significance and relevance, and can be understood as a democratizing principle.In the scientific field, this would imply recognizing and integrating into research processes the plurality of social and economic demands, political interests, technical capacities, and sociocultural factors that are involved in a social-environmental issue.
In this sense, an approach that has proved useful for analyzing governance processes that involve more stable multi-stakeholder partnerships is CASs, which is detailed in the following section.

Governance in socio-environmental research: the CASs approach
From the above discussion, governance is a government approach that can be conceived at the micro scale.Accordingly, we analyze multi-stakeholder coordination required in large social-environmental research projects.In this article, governance will be understood as "a more inclusive and cooperative approach intended for the solution of common issues by various actors, not necessarily government actors" (Luna and Chávez 2014, 189); indeed, it frequently involves "mixed public-private networks" (Mayntz 1995, 83).As stated above, studies addressing social-environmental topics have explored concepts such as "territorial governance" (Martínez Salvador and Martínez Salvador 2021; Torres-Salcido 2008); "governance networks" (Newig, Günther, and Pahl-Wostl 2010;Rhodes 2007), and "collaborative governance" (Ulibarry 2019).The latter has been frequently addressed, based on the assumption that decisions regarding an environmental issue are made in inter-organizational bodies through deliberative and consensus processes allowing effective management of scientific and administrative uncertainties, and incorporating different values and types of knowledge (Ulibarry 2019). 1 recent proposal for analyzing governance processes in social-environmental research which we consider suggestive because it is aligned with these approachesis the CASs approach of Luna and Velasco (2017).This proposal is based on previous theoretical frameworks, such as collective action (Ostrom 2014), political networks (Messner 1999), or governance networks (Rhodes 2007;Newig, Günther, and Pahl-Wostl 2010;Rosenau 2004).The authors of CASs take up these approaches, and those of other authors, to understand the origin, structure, and dynamics of the cooperation processes through different analytical criteria.These are then systematically organized in a conceptual model that supports an empirical approach to experiences of multi-stakeholder partnership such as those studied here (refer to Figure 1).
Thus, this approach analyzes the factors that facilitate interactions and the mechanisms that explain how stable partnerships produce mutual benefits through the exchanges between mutually complementary actors.2Representing a step forward that exceeds other approaches focused solely on exploring the interactions between diverse stakeholders, CASs consider networks as entities in their own right, where cooperation and conflict conditions coexist, to provide solutions to a common issue (Luna and Velasco 2017).Additionally, CASs are expressions of a territorial setting, which, although capable of entailing different geographic scales, also covers aspects such as culture and identity (formal and informal) institutional frameworks, and social and political priorities relative to specific territories; and which, within the research governance framework addressed here, are oriented to address particular social-environmental issues in spatial and temporal terms.
This approach considers that partnerships achieve a greater or lesser degree of formalization, adopting a set of emerging characteristics and properties derived from creative collaboration.Thus, CASs are characterized by the convergence of multiple initiatives from heterogeneous stakeholders who build reciprocal relationships to reach a collective objective (Steward and Conway 1996).These are associations where actors with particular identities, interests, and loyalties coexist (individual autonomy), but with a significant degree of interdependence to meet their objectives, which leads to a variable balance in the dynamics of the network (Warren 2001).The interaction between stakeholders and organizations from different territorial entities leads to stress resulting from the heterogeneity and aims to enforce individual autonomies, on the one hand, and the need to cooperate, on the other.
When these partnerships advance to a certain degree of stability, they achieve systemic autonomy, i.e. they are not subordinate to any particular organization and, therefore, can make decisions and define objectives governed by their own rules, from a flexible and horizontal hierarchical structure adapted to achieve sustainability (complex associative leadership) (DeRue 2011; Ahrne and Brunsson 2005).In turn, cohesion is warranted through a horizontal decision-making (March 1997) framework driven by deliberation, negotiation, and consensus, thus helping to retain the participating stakeholders and preserve their autonomy and identity (Warren 2001;Brunsson and Olsen 1998;Schmitter 2001).The integration of these networks also requires building trust which fosters cooperation when rules are weak (Burt and Knez 1996;Ostrom 2014), and effective communication which involves translation to establish a common language.Finally, the evaluation of CASs is based on functional performance and associative performance.These delimit both the achievement of CASs objectives and the interactive capacities that increase and expand public action networks, reaching the legitimacy required to influence the policy area of interest.
This analytical perspective is adapted in this article to explain how collaborative processes occur in practice, regarding the development of the two social-environmental research projects analyzed herein, which, from postnormal science, open science, or inter-and transdiscipline approaches, require the setting of partnerships between different groups of scientists, and between these and sectors of the government, civil society, citizen organizations, or companies, aiming to form extended communities of peers.

Methodological strategy of this research
Our work is based on qualitative exploratory research using a comparative case study approach, taking two laboratories, LANRESC and LENERSE, as research loci to address the questions highlighted in the introduction to this article.
The comparison was based on the Most Similar systems approach (Sartori 1999), according to which the two elements to be compared are similar in all but one criteria.The similar criteria set for selecting the two cases were: (a) they should be collaborative projects involving several academic institutions along with non-academic actors (government, civil society, or businesses); (b) they should have been conceived as scientific and technological policy instruments to build new knowledge production capacities considering multi-actor perspectives; (c) they should have succeeded in establishing medium-or long-term associative structures across their members, and (d) they should have been created in the Yucatan peninsula, focusing on the study of socio-environmental issues within that territorial context.As a differentiating criterion, one laboratory should have a more diverse and plural participation of actors than the other, to identify the effect of these different multi-actor configuration modalities on the generation of cooperation mechanisms and the results derived from them.
The study was based on the analytical perspective of CASs and its intersections with the postnormal science framework, to explain the operation of networks of academic researchers in the political and social arenas and the results produced in terms of the impact of research findings on political decision-making or the application of knowledge to specific solutions.The research work reported here considered most of the analytical criteria of the CASs conceptual model to obtain an overview that would support the comparative analysis of the two case studies.
The empirical work consisted of semi-structured interviews with key stakeholders seeking to obtain different perspectives that would triangulate the information.Thus, in the case of LANRESC, we interviewed scientists from founding academic institutions and members of the laboratory, the state government, and one non-government organization.In the case of LENERSE, we interviewed technologists from two founding academic institutions and members of the laboratory, the state government, and one company.To note, interviews in the academic sector were conducted with project coordinators or managers and with participating researchers.A total of 14 interviews were conducted, seven for each laboratory.
Based on the CASs model, the script designed for the interviews was structured into 4 sections and 20 questions.Some of these questions are shown in Table 1.The interviews allowed obtaining extensive information on each experience, covering the different analytical criteria of CASs until a theoretical saturation was achieved using the data obtained.Due to COVID-19 restrictions these interviews were conducted remotely.In parallel, a literature survey was conducted through the collection of official and hemerographic documents to contrast and supplement the information provided by interviewees.
Interviews were recorded and then fully transcribed.The resulting data were encoded using the software Atlas.ti,using a common set of codes for both cases to facilitate a comparative analysis (refer to Table 2).Including open coding from interviewees' narratives (see Figure 2) and, subsequently, axial coding (Coffey and Atkinson 2005).Due to the confidentiality of the information provided, interviewees remain anonymous and are referenced by means of a consecutive code.The list of interviews and the dates on which they were conducted are indicated in the literature references.

LANRESC and LENERSE as CASs
This section describes the creation and development of each of the laboratories analyzed, and the distinctive features of each as CASs.

Origin and development of the laboratories analyzed
This article discusses LANRESC, and LENERSE, as case studies.Both laboratories have been established within the framework of a regional scientific and technological policy agenda that sought the development of knowledge production capacities by strengthening the infrastructure and the links of cooperation networks between academic actors and governments, the private sector, and non-government organizations (NGOs) to address issues, demands, and opportunities relevant to the region.Along this line and through institutional arrangements and mixed funding between the federal and state governments, these laboratories were conceived as scientific policy instruments, i.e. as mechanisms to materialize state intervention in a complex field (Isuani 2012), expressed in this case as the need to address some pressing socio-environmental issues in the region through the production and application of new knowledge.
Together with other initiatives, such as the creation of Parque Científico y Tecnológico de Yucatán (Yucatan Science and Technology Park) in 2009, LANRESC and LENERSE can be considered policy instruments functioning as management tools (Isuani 2012) by deploying an agglutinating organizational framework that facilitates the activities needed to meet scientific policy objectives.In this way, the laboratories function as organizations that allow coordination for applied socio-environmental research from a territorial perspective.Under this logic, LANRESC is an organizational instrument focused on producing scientific and technological knowledge on the conservation and adaptation of coastal socio-ecosystems, while LENERSE focuses on the development of technological solutions in renewable energies.Table 3 summarizes the main characteristics of both laboratories.
LANRESC is an initiative of researchers from the UNAM, Sisal campus, Yucatan.This collective project was launched from CONACYT's Call for National Laboratories (AI13-12-07-21) with the participation of the three academic areas of UNAM in Sisal: Instituto de Ingeniería (Institute of Engineering; II), Faculty of Sciences, and Faculty of Chemistry, the Centro de Cambio Global y la Sustentabilidad (Center for Global Change and Sustainability; CCGS), Tabasco, and the Instituto Tecnológico de Sonora (Institute of Technology of Sonora; ITSON).Additional institutions that later joined the project were the Centro de Investigación y de Estudios Avanzados (Center for Research and Advanced Studies; CINVES-TAV), campus Merida, of the Instituto Politécnico Nacional (National Polytechnic Institute; IPN) (LANRESC 2021b).LANRESC is, therefore, an initiative sponsored by scientists from Yucatan and Southeast Mexico, coordinated by the II (UNAM), launched with the aim of producing applied science to guide decision-making.It emerged from the interest to carry out applied science that could be used for decision-making in environmental policy; in other words, it seeks to produce postnormal science.This project has been shaped over time into a national laboratory encompassing the analysis of several coastal socio-ecosystems (AI8,12-07-21; AI10,29-06-21).
LANRESC was initially delineated as a project to address coastal issues in different regions of the country, focusing on resilience to natural or anthropogenic disturbances (IA13,02-06-21).At the beginning it focused primarily on physical and ecological issues of coastal areas, but has gradually promoted an interdisciplinary dialogue with the social sciences to broaden the understanding of the phenomenon addressed.For this reason, the Instituto de Investigaciones Económicas (Institute of Economic Research; IIE) of UNAM joined the project.It currently comprises seven partner institutions (four from UNAM) and six participating institutions that do not contribute concurrent funds but collaborate in the development of this initiative (AI12,03-08-21).The LANRESC collaboration network has also been extended to incorporate other institutions across the country, with 96 additional academic scientists actively participating as of 2020 (LANRESC 2020).In addition, links have been established with regional government agencies, NGOs, and coastal populations as active actors in the project, seeking to form an "extended community of peers" from the perspective of postnormal science.
The project participants set a first stage between 2015 and 2017, in which time the laboratory was formally launched, the infrastructure of partner institutions was strengthened, and collaboration links were set between scientists (AI13,12-07-21).However, from 2017 to date, greater synergy has been achieved to advance from collaborations between individual partners to a more comprehensive project.Consequently, initiatives such as "report cards" started being produced. 4Coastal observatories have also been launched as a project for the historical and standardized recording of the resilience of several coastal socio-ecosystems. 5eparately, LENERSE is an initiative that was launched prior to the existence of national laboratories.It is hence considered a pioneering case for having gathered, since 2009, research groups of institutions in Yucatan that worked on renewable energies.These links were expanded to the states of Campeche, Quintana Roo, and Tabasco, to boost the scientific and technological capacities in this region of Mexico through collaboration in joint research projects, use of common infrastructure, and training of human resources (Smith 2012).LENERSE originated from the interaction between researchers of the Centro de Investigación Científica de Yucatán (Center for Scientific Research of Yucatan; CICY), and CINVESTAV, to which the following institutions later joined: Faculty of Engineering of the Universidad Autónoma de Yucatán (Autonomous University of Yucatan; UADY), Instituto Tecnológico de Cancún (Technological Institute of Cancun), Universidad Autónoma del Carmen (UNACAR), Universidad de Quintana Roo (University of Quintana Roo; UQROO), Instituto Tecnológico de Campeche (Technological Institute of Campeche), and Universidad Juárez Autónoma de Tabasco (Autonomous University of Tabasco; UJAT, which no longer participates) (Smith 2010).The laboratory is coordinated by CICY and comprises six peninsular institutions.
This project was preceded by interaction and linkage dynamics between stakeholders of the academic sector and the regional government, which has fostered a governance environment resulting in the formation of the Sistema de Investigación, Innovación y Desarrollo Tecnológico de Yucatán (Research, Innovation, and Technological Development System of Yucatan; SIIDETEY), created by decree in 2008.LENERSE is also a collective and inter-institutional research project preceded by other collaborative initiatives such as the dissemination of post-graduate studies, joint libraries, and SIIDETEY itself, which were already interested in generating cooperative science and technology policies and actions in the state of Yucatan (AI5,. The development of this research laboratory took place in two stages.The first was financed by the Fondo Institucional de Fomento Regional para el Desarrollo Científico, Tecnológico y de Innovación (Institutional Fund of Regional Promotion for Scientific, Technological, and Innovation Development; FORDECYT) at CONACYT (2009-2013) (AI1,28-07-20; AI4,18-08-20).The second was developed with funding from a project of the Secretary of Energy (SENER) and CONACYT for institutional strengthening seeking energy sustainability (2017 to date) (AI7-24-11-21).LENERSE has evolved from the early stages when collaboration mechanisms were formalized and instrumental and post-graduate infrastructure was created, to a second stage aiming to consolidate the links with public and private sectors and establish the laboratory as a national reference (LENERSE 2017).In this sense, this experience is also based on postnormal science principles, acknowledging the need to generate useful and applicable knowledge about technological solutions in renewable energies and seek the coordination with actors in the government and productive sectors to influence the energy transition process at the local level.

Heterogeneity, autonomy, and interdependence in the laboratories analyzed
LANRESC and LENERSE have a collective origin, showing that both were conceived to establish interactive processes from the beginning or even before the formation of these research laboratories.Thus, to create LANRESC, the researchers of the Institute of Engineering (II; UNAM) sought partnerships with their peers in other higher education institutions (HEIs), such as CCGS and ITSON (AI13,12-07-21; AI8,12-07-21), since the national laboratories concept stems from the idea of forming inter-institutional projects that allow strengthen scientific and technological capabilities in different regions of the country (ININ 2021), and the call required at least three associated academic institutions (AI9,22-06-21).As mentioned above, LENERSE is a pioneering initiative based on a process of regional governance in the scientific and technological sectors (Giraldo 2018).It is a collective structure based on previous research capacities generated around renewable energies (AI5,31-08-20).In both initiatives, the federal and regional governments (especially Yucatan's) are also key partners in the development of these projects, because they have been important financing and information sources.
At the same time, the fact that these projects are multi-stakeholder fosters the heterogeneity of members that, in both cases, do not necessarily occur among individual researchers, but at institutional and regional levels regarding asymmetries in scientific and technological capabilities.Accordingly, in the case of LANRESC, there is an evident superiority in the infrastructure available to UNAM, compared to the other associated institutions (AI10,AI9,.This disparity has also been reflected in the availability or lack of resources for the development of activities such as the "report cards."Asymmetries are also apparent in the availability of coastal resilience data; for example, while there are historical records spanning many years in Yucatan, the data are still very limited in Tabasco (AI9,22-06-21).
In the case of LENERSE, there are also asymmetries between the most prominent Yucatan research groups relative to those of Quintana Roo and Campeche.In this case, there is evidence of a higher development of CICY or CINVESTAV stemming from the development of LENERSE, compared to institutions in the other states .Another aspect worth noting is the differences in the operation rules between education centers and research institutions, sometimes dissimilar and contradictory, which have posed a challenge in the administrative and financial operation of projects (IA7,.It is important to note that the more plural and participatory processes in socioenvironmental research involve important heterogeneities, not only among actors from different sectors, but even among members within the academic sector, thus adding complexity to these associative structures.
Heterogeneity is a challenge for maintaining the cohesion of collaborative networks, but, is simultaneously an opportunity for groups and academic institutions to work in synergy (AI5,.Additionally cross-sectoral differences are evident.In the case of LANRESC, government stakeholders such as the Secretariat for Sustainable Development of Yucatan, or Civil Society Organizations (CSOs) such as Pronatura, have scientific staff that easily dialog with the academic sector.However, the available time and resources are different, for example, in terms of the accessible data and the period of time during which these have been obtained (AI11,AI14,.As for LENERSE, one entrepreneur stressed that the objectives of the scientific sector are ambitious and long-term, thus being hard to match with the priorities of the productive sector, which requires specific results over shorter time spans (AI6,.This leads to stress in the governance of socio-environmental research arising from different objectives about a common issue, such as the consequences of climate change. This heterogeneity coexists with the autonomy of laboratory members as entities in their own right.Thus, a high degree of autonomy characterizes both LANRESC and LENERSE, evident for individual researchers as well as for each participating institution, within the commitments adopted in relation to the projects that provide funding (AI1,28-07-20; AI12,03-08-21).Autonomy also allows proposing and conducting research within the network itself and, in LENERSE in particular, has facilitated the creation of post graduate programs promoted by this laboratory, but managed by each of the associated HEIs (AI1,28-07-20, AI10,29-06-21).
However, in socio-environmental research contexts, the need to promote collaboration among scientists to share knowledge through interdisciplinary schemes and solve issues that could not be addressed by individual sectors entails reaching a balance between autonomy and interdependence.In the case of LANRESC, an example is the participation of scientists specialized in water quality from CINVESTAV, who joined coastal engineering experts to strengthen the study of ecosystem health (AI13,12-07-21; AI10,29-06-21); IIE researchers (UNAM) also participated in research on socio-economic issues aiming to understand the resilience process (AI8-12-07-21), to which researchers from IIE (UNAM) joined, aiming to contribute to the understanding of the resilience concept from a socioeconomic perspective.In LENERSE, this aspect is illustrated by the integration of technologist experts in different types of energy, working on the development of technologies for the production of solar cells based on cadmium telluride customized to tropical environments (AI5,AI3,.
In addition, there is interdependence with non-scientific members of these initiatives.In both laboratories there is a marked interdependence between the academy and the federal government regarding the resources required for the development of projects and, in turn, of socio-environmental research policies issued by the government and implemented by HEIs (AI13,AI9,AI3,.In this context, both laboratories become policy instruments for generating new scientific and technological capabilities on a regional scale.Conversely, in LANRESC, CSOs, and regional governments have contributed data from different communities that were not elsewhere available to scientists; in turn, these organizations lack the scientific infrastructure required for data analysis that is available to researchers.The end result is a mutually beneficial exchange of resources and competences (AI11,AI14,. Finally, as entities with independence in terms of resources, decision-making, and affiliations, the systemic autonomy of the laboratories analyzed is expressed in the construction of a unique organizational scheme that differs from that of the associated HEIs and non-scientific actors.This collective autonomy demonstrates innovation abilities to organize scientific processes and links with other stakeholders, in both experiences.The LANRESC structure has evolved into a more functional scheme (AI13,12-07-21) organized according to five areas: 1 -Executive Council, 2 -Operational Coordination, 3 -Coastal Observer Coordinators, 4 -Coordination of Thematic Groups, and 5-Coordination of the Geographic Information System (LANRESC 2020).For its part, LENERSE has a structure based on thematic groups organized into six sub-projects: 1 -Project Coordination, 2 -Solar Energy, 3 -Wind Energy, 4 -Hydrogen, 5 -Cells, and 6 -Hybrid Systems (LENERSE 2017), each defining its own activities and the products involved.
Although a more horizontal structure operating around a central coordination is evident in both laboratories, the establishment of an Executive Council with representation of each of the associated HEIs in LANRESC (AI12,03-08-21) involves a more participatory and representative decision-making scheme.Besides, operational coordination is also a novel figure of LANRESC that has facilitated the efficient communication and liaison with government agencies, civil society, and coastal communities (AI13,12-07-21).These aspects will be further discussed in the following section.

Governance of research in LANRESC and LENERSE from a comparative perspective
Both LANRESC and LENERSE initiatives are associative efforts that have built a multi-stakeholder collaboration network sustained by scientific actors interacting with government, civil society, and business stakeholders to achieve a greater impact in the solution of the socio-environmental issues addressed.This section discusses the operation of research governance processes in relation to their cooperative mechanisms and the results achieved in terms of functional and associative performance.

Cooperation mechanisms in LANRESC and LENERSE
From the CASs perspective, governance is analyzed through factors that foster cohesion from political (leadership and representation, decision-making) and social (trust and communication) perspectives.Thus, collaborative projects are largely supported by effective leadership.In the case of LANRESC, the existence of an Executive Council with representatives of each partner institution has shaped a collective decision-making scheme, which, in addition, facilitates communication with members of each institution (AI12,03-08-21).According to interview testimonials, these representatives are selected for being the persons of their institution with the greatest interest and commitment to participate in the lab, which has resulted in the congregation of multiple leaderships (AI12,03-08-21).In addition, representation is meant to be rotational, which, although not fully formalized, has occurred in practice.However, an aspect to strengthen is the participation of a larger number of social scientists, in both the laboratory in general and in the Executive Council, for teams to be truly interdisciplinary (AI8,12-07-21).The structure of LANRESC, as a CASs, has allowed setting up a decision-making process based on annual activity schedules, which, according to some interviewees, are discussed between members through the representatives of each institution (AI12,03-08-21).It is the opinion of some interviewees, however, that the Executive Council scheme of LANRESC has led to the loss of horizontality, since decisions were made through the direct participation of members when the network was much smaller (AI9,22-06-21).
In addition, this decision-making scheme is solely in the hands of academic actors, hence limiting the establishment of shared leadership with regional governments, civil society, or business actors in the design of a research agenda for the joint production of knowledge aiming to solve socio-environmental issues of interest for this laboratory.This hinders non-academic groups' interests from being included, thus weakening the impact of LANRESC products and actions on environmental policy decision-making, as detailed below.
LENERSE has no such executive council, so the leadership is assumed within the overall coordination of the project.This has led to representativeness issues in certain decisions and, therefore, in leadership; although there is a head in charge of the administration of each project, no rotation is in place (AI1,28-07-20).This has resulted in tensions between collaborators, but, above all, in LENERSE's activities focused on the deliverables committed to projects (AI3,18-08-20).Furthermore, the fact that representativeness is relatively unstructured seemingly leads to a more horizontal decision-making process.According to one interviewee, all researchers have been consulted on the laboratory's most relevant decisions while, and simultaneously, there is autonomy in the particular decisions made by each research group (AI1,28-07-20).However, similar to LANRESC, the main decisions are the responsibility of the academic sector, discouraging the possibility of conducting linkage actions involving companies and government agencies, which would allow for positioning this laboratory as a more active player in the energy transition.
Following the conceptual model of the CASs, another key factor that maintains cohesion in a multi-stakeholder and collaborative project is trust among its members; otherwise, mistrust leads to fragmentation.In the two cases analyzed, trust lies on personal relationships, some of which have been built prior to the creation of these laboratories.For one interviewee, collaboration improves if it has affinity with the people it is going to work with, not only in academic but also in personal terms (AI8,12-07-21).In both experiences, friendship between scientists is a factor that boosted the start-up of these laboratories (AI4,18-08-20).In the academic environment particularly, trust is also based on prestige (personal and institutional), that is, the acknowledgment of the quality of the work of peers.This has favored the willingness to establish inter-institutional partnershipsfor instance, with UNAM in the case of LANRESC, or with CICY and CINVESTAV in the case of LENERSE (AI9,AI5,AI4,. Finally, with regard to communication and translation processes, LANRESC displays an efficient design involving both internal communication through the Operational Coordination and connection with other stakeholders based on strategies that facilitate the dissemination and communication of science, for example, report cards, webinars, and a newsletter (AI12,03-08-21; AI8,12-07-21).Nonetheless, another type of heterogeneity resulting in the participation of investigators with expertise in different areas of knowledge involves the challenge of interdisciplinary translation between researchers, which represents an obstacle for establishing a common language around the resilience concept (AI9,22-06-21).For one interviewee, interacting with colleagues from other disciplines has always been a little cumbersome (AI12,03-08-21).This condition produces uncertainty and implies a major challenge when undertaking the inter-and transdisciplinary processes required in socio-environmental research.
Something of the sort happens in LENERSE, where, despite the interaction of expert scientists in renewable energies, there were also testimonies about communication issues between research groups, constrained by the pressure to comply with the deliverables committed in projects (AI3,.But beyond academic boundaries, in this experience, communication strategies with external stakeholders, entrepreneurs in particular, also need to be strengthened.According to one interviewee, researchers are trained to develop science and technology, and may need a spokesperson who can ground and see both worlds (AI1,28-07-20).Other works (Giraldo and Zárate 2022) have pointed out the importance of social sciences in these translation processes as a mediator between "hard" sciences and non-scientific groups.Therefore, generating interface mechanisms to achieve this liaison between scientific approaches and other social, economic, and political processes will be key to achieve greater dissemination and application of the technological advances in renewable energies produced in this laboratory, as discussed below.

Functional and associative results
As already discussed, both laboratories have produced a number of cooperation mechanisms to achieve their objectives, i.e. to attain satisfactory functional performance.The main objective pursued by each laboratory from the beginning was the formation of an academic team blending together different lines of research around a socio-environmental problem.It is thus noted that a remarkable achievement of both initiatives is to have established a collective project integrating the previous competences of the participating institutions and researchers (AI12, AI5,.Another initial objective in the two laboratories was to provide the research groups with better infrastructure and strengthen human resource training (AI1,28-07-20; AI13,12-07-21).In LANRESC, training has been provided to students through research projects linked to the laboratory and a social service program (LANRESC 2020).In LENERSE, this objective has been met through post-graduate courses created in the associated HEIs, which has fostered closer ties between the researchers participating in inter-institutional tutorial committees (AI1,28-07-20).
These early objectives focused on academic interests, have evolved into others with greater scope.In the case of LANRESC, a central goal has been the construction of an interdisciplinary team for the understanding of resilience.This team first involved environmental sciences and engineering (AI13,12-07-21) and now is pursuing the participation of social sciences.The latter continue having a minor representation (AI8,12-07-21; AI10,29-06-21), but are considered a cornerstone for transitioning from the traditional study of "ecosystems" to the more novel "socio-ecosystems" approach (AI9,22-06-21).In this regard, the interviewees agreed that a multidisciplinary structure has been achieved so far, although it has not yet moved forward to interdiscipline (AI8,.This goal is sought by the convergence of different research topics such as water pollution, beach erosion, or fishing practices (AI12,03-08-21).In LENERSE, although an interdisciplinary scheme has not been explicitly proposed, there is multidisciplinary integration around different technologies for energy generation such as solar, wind, or hydrogen, to which hybrid systems were recently added (LENERSE 2017).
One aspect to highlight in both laboratories is the importance of focusing on problem solving (AI13,12-07-21; AI1,28-07-20), which necessarily implies its linkage with non-scientific stakeholders.One of the key instruments of LANRESC is "report cards," based on an international experience that has been relatively successful with regard to the link with regional government agencies and civil society stakeholders who have participated in several workshops on the construction and dissemination of report cards (AI11,13-08-21; AI14,18-08-21) contributing different types of data and experiences about coastal resilience.Although these products have been used by external actors, according to interview testimonies, it has not yet been demonstrated that they have influenced environmental policy decisions (IA8,, so that this experience should continue developing additional strategies to produce postnormal science. Nonetheless, it is interesting to find that report cards have contributed to the associative performance of LANRESC as they were based on collective competences acquired through an ecological management exercise in Yucatan (POETSI).Currently, this tool is being used in other regions of the country, starting with the seven coastal observatories mentioned above (AI10,29-06-21; AI8,12-07-21).This initiative aims to further consolidate LANRESC as a national laboratory, proposing the recording of physical and natural variables in specific socio-economic contexts, an objective that demands forming a more inter-and transdisciplinary research team (AI10,29-06-21; AI13,12-07-21).For its part, LENERSE has also moved forward to consider new objectives, such as the generation of patents on technological developments for the production of renewable energies (AI1,28-07-20) aiming to achieve the link with private companies to manufacture and market these innovations (AI3,.This goal still represents a challenge that implies a long intellectual property process (AI7,24-11-21) and the transition to science and technology evaluation schemes that promote this type of activities (AI2,06-08-20).
Another result that demonstrates functional performance in both laboratories is the area of scientific publications.In LANRESC, these have transitioned from being particular products of the participating researchers (e.g.journal papers, books, manuals, theses) to collective products that point to inter-and trans-disciplinary approaches.The interviews highlighted the publication of a book on coastal resilience that seeks to build a common language around this concept to facilitate communication among participating researchers (AI8,12-07-21; AI12,03-08-21).In this sense, another product is a glossary of terms to strengthen the translation process to facilitate the dialog with non-scientific stakeholders in a trans-disciplinary research framework (LANRESC 2020).In LENERSE, productivity is reflected mainly in scientific articles and patents produced from a more technocentric logic that characterizes the dynamics of technological development of renewable energies (AI2,06-08-20).
An important achievement of LANRESC in this area is the development of information dissemination mechanisms (e.g.webinars, newsletters, infographics for children).These are signs of the interest in maintaining the cohesion of its network and increasing its associative performance to influence environmental education and policy processes (AI10,29-06-21; AI12,03-08-21).Another valuable exercise is the publication of the 2020 Annual Management Report, which reflects the network's ability to contribute to the transparency of the activities ahead.Such results illustrate the commitment of this laboratory to become a key player in the political arena.LENERSE also conducts dissemination activities that, however, have not been translated into dissemination processes with other external stakeholders.
In general, it is remarkable that both LANRESC and LENERSE have earned a position in the areas of coastal resilience and renewable energies, respectively, at regional and country levels.In both cases, this result has been achieved thanks to the cohesion of their networks, which have sought to grow to increase their social contribution.In this regard, LANRESC, by adopting an unconventional research scheme (AI13,12-07-21), has surpassed the traditional objectives of scientific production to promote multiple initiatives that have strengthened the links with federal and state government agencies, CSOs, businesses, and communities, with whom it exchanges resources to expand the measurement of the resilience phenomenon.This, to develop an extended community of peers and exchange resources to expand the measurement of resilience.However, there is a major challenge in forming an inter-and trans-disciplinary associative structure to achieve further integration between areas and diverse types of knowledge, aiming to more directly influence public decision-making.
LENERSE, being a longer-term initiative, has achieved a stable partnership with strong links with regional governments, which relies on this network for guidance and advisory on the implementation of large-scale renewable energy projects (AI5,31-08-20).A similar partnership has also been achieved with regional business stakeholders, although the exchange is still relatively incipient.The configuration of a more traditional research model in this case restrains the expansion of its network to involve business and social stakeholders, since these links are not part of the commitments agreed in the projects financed (AI2,06-08-20).As a result, this approach has inhibited better associative performance to play a more active role in political processes relating to energy transitioning.This condition also reveals the constraints imposed by the financial instruments created to support these initiatives.

Conclusions
This work aimed to understand the operation of governance processes in socio-environmental research in the cases of LANRESC and LENERSE.These research laboratories have coordinated multi-stakeholder schemes to produce knowledge and propose solutions in contexts of complexity, uncertainty, and dynamism, such as climate change and its impact on coastal areas or the depletion of fossil energy resources.
The analytical perspective of CASs made it possible to understand the functioning of both initiatives, which are based on relatively stable multi-stakeholder networks and organized around common objectives, with mid-and long-term projections.Thus, our empirical findings showed that both cases emerged as collective initiatives established by heterogeneous stakeholders in terms of resources, rules, and competences, even demonstrating regional asymmetries in alliances of this type.These heterogeneities illustrate the complexity involved in socio-environmental research projects seeking to create an expanded community of peers, with obvious differences not only in the values and interests between the actors from different social sectors, but also within academic communities that have sought to create interdisciplinary organizations to respond more effectively to the phenomena addressed by them.
In turn, it was found that the participating members have individual autonomy to make decisions, manage their resources, and maintain their identity, but have promoted interdependence with other network members to achieve goals that cannot be reached by individual entities.This interplay has led to a changing balance, which means that none of these two laboratories is subordinated to any other institution.Therefore, they have systemic autonomy and flexibility (adaptation over time) to achieve the sustainability of their association and produce learning processes regarding how to achieve their objectives more effectively ( functional and associative performance).
We found that the differences in the composition of both initiatives involve different configurations and governance processes.The present work revealed that LANRESC has achieved greater integration with government agencies and civil society than LENERSE, which has maintained its interaction with the state government and, to a lesser extent, with regional enterprises.This composition stems from the same conditions imposed on the financing instruments that give rise to such undertakings, demonstrating greater flexibility and innovation in the National Laboratories policy that gave rise to LANRESC, related to the CONACYT-SENER funds that have supported the development of LENERSE in a second phase.In turn, better performance in the governance processes depends on factors such as leadership to coordinate collective efforts in accordance with the goals set, the possibility of building an environment of trust that facilitates exchanges between stakeholders, and the achievement of effective communication characterized by a common language.LANRESC has been more successful than LENERSE in these areas, which in both cases need to be strengthened.
In both experiences, it is evident that the decision-making mechanisms remain concentrated in the academy, limiting a greater social impact of the results that these laboratories seek to achieve.We also identified deficiencies in the communication and translation mechanisms across researchers from different areas of knowledge.In this respect, it is worth highlighting that in both cases, the socio-environmental research processes have been led by researchers from natural sciences, physics, and engineering, with minimal or no participation of researchers from the social sciences who may function as mediators with other social players in aspects such as public policy, environmental education, and the social appropriation of knowledge.This aspect may represent a challenge that, beyond the cases analyzed herein, is a limiting factor in the incidence of academic research addressing socio-environmental topics, where transdisciplinary processes are an elusive goal.
Nonetheless, one aspect that deserves recognition in the LANRESC experience has been the ability to become organized using unconventional scientific research arrangements.In this context, a key aspect of the results achieved has been the ability to organize less conventional arrangements, as evidenced in LANRESC, an experience in which members recognize that their productivity assessment logic is not guided by traditional academic metrics.This has allowed the inclusion of criteria that are closer to the values and interests of non-scientific stakeholders, leading to creative and innovative adaptations; as a result, non-scientific stakeholders are more closely identified with the actions and products generated.For its part, LENERSE has participated in a more technology-driven context in the technological development of renewable energies, which responds to the approach of policies about technological development in this field, constrained to delivering the products committed in the projects that provide financial support.This conditioning has limited their participation and influence in decisions on energy policy or the social massification of renewable energies.
In short, both cases reveal important challenges in shaping research structures within inter-disciplinaryeven trans-disciplinaryframeworks, which is currently more a vision than an objective achieved.It is essential to promote interactive and multi-sectoral schemes, especially through public policies in science and technology, that encourage the development of regional scientific policy instruments with greater intersectoral articulation (for example, with portfolios such as environment or energy), in order to foster the co-production of higher quality and more useful knowledge for urgent and comprehensive attention to the various socio-environmental problems, in territorial contexts.

Disclosure statement
No potential conflict of interest was reported by the author(s).

Figure 2 .
Figure 2. Cloud of words from an interview.Source: Obtained from Atlas.ti from transcribed interviews.

Table 1 .
Example of guiding questions.Characteristics 1. Please describe the process of delineating this project 2. What is the primary objective of the laboratory?Has it evolved over time? 3. What is its organizational structure?Is it functional?Properties 4. What are the key drivers of your group for collaborating in the laboratory? 5. What links have been set with civil society and the business sector?6.How can you succeed in influencing political and social decisions?Integration mechanisms 7. How are strategic decisions made regarding the operation of the laboratory?8. Are the communication modalities in place appropriate for decision-making?Evaluation 9. What are the main achievements and obstacles facing the laboratory?10.Do you consider that the research has been carried out within an inter-and transdisciplinary framework?Source: Own elaboration. 3

Table 2 .
Coding of interviews.