Conceptualizing indicators as boundary objects in integrating Inuit knowledge and western science for marine resource management

Risk and uncertainty over marine resource availability and use in the circumpolar Arctic is an important issue for both conservation managers and local users, who are under increasing pressure due to climate change, increased anthropogenic activity, and both direct and indirect human development stresses for the marine environment. In the context of the Canadian Arctic, the strong economic, social, and cultural dependence in Inuit Nunangat¹ on fisheries exacerbates these concerns. Inuit Nunangat spreads across all four Inuit land claim regions in Canada (Nunatsiavut, Nunavik, Nunavut, and the Inuvialuit Settlement Region), in an area that covers both land and vast areas of water and sea ice. In total, this area amounts to about half of the coastline of Canada and is populated by about three-quarters (72.8%) of the Inuit population (65 025) (Statistics Canada 2017) (Fig. 1).

The complexity in Arctic marine management that comes from the intersection of high resource use and high uncertainty requires decision support tools and frameworks (Berkman and Young 2009; Shephard et al. 2016; Kaiser et al. 2018). Indicators are one type of tool that can inform decision-making for marine resource management at multiple levels to identify and monitor changes over time, and can therefore serve as a barometer of changes in ocean health and socioeconomic welfare derived from marine resource use (Fletcher et al. 2005; Jennings 2005; Rice and Rochet 2005). Additionally, they can inform public decisions and are seen as an influential tool in policy making and often used in public debate given that they are designed in ways that can be understood by resource users and the general public (Vandermeulen 1998; FAO Fishery Resources Division 1999; Boyd and Charles 2006; Raakjær et al. 2007). However, despite their increased usage in marine resource management, their development relies heavily on western science (WS) and largely fails to incorporate other knowledge systems such as Inuit knowledge² (IK). Although there is still progress to be made, IK is becoming prioritized in co-management systems, with more inclusivity still needed in decision-making processes (Berkes et al. 2007; Snook et al. 2019).

Fish and marine mammals are key in the subsistence and wage-based economies of Inuit Nunangat, in addition to their importance in food security (Ford 2009; Hoover et al. 2013; Lysenko and Schott 2019). An ever-growing number of fishery performance indicators, including indicators for marine spatial planning, fisheries management, and broader marine resource governance, have appeared in the literature over the past two decades (see for example Pelletier et al. 2008, Rice and Rochet 2005, Garcia and Staples 2000, Anderson et al. 2015, and Jepson and Colburn 2013). However, their usefulness in informing decision-making and supporting policy is not always clear. Past literature has primarily focused on places where good data are available and have, therefore, been used as a tool to guide management and communicate research to policy makers and the general public and (or) non-specialist audiences (FAO Fishery Resources Division 1999). For the purposes of managing and monitoring marine resources in Inuit Nunangat, where strong cultural ties exist and extensive data sets are not always available, traditional fisheries indicators might not prove adequate or suitable to encompass regional needs and local knowledge. This issue applies to other parts of the world where there is a strong element of Indigenous participation in the use of natural resources (Dickison 1994; Newman and Moller 2005; Jollands and Harmsworth 2007; Moller et al. 2009b; Izurieta et al. 2011). Despite international frameworks for guidance in developing Indigenous-led indicators and efforts to follow progress, the relationships between IK and the use of indicators for monitoring and managing natural resources are not well defined in the scholarly literature and the application to management is still very young (Berkes et al. 2007; TEBTEBBA 2008; IPBES 2017, 2019).

Despite this underrepresentation of IK indicators in the literature when compared with WS indicators, there is growing recognition both in the literature and in practice on the importance of IK, as laid out by Johnson (1992) and Huntington (2000), for example. Scholars within natural and social sciences have long recognized the benefits to be gained by having multiple knowledge generating systems functioning in parallel, such as understanding climate variability and shifts, health of stocks, resource use, and community vulnerability (Moller et al. 2004, 2009a; Laidler 2006). Building partnerships between science and IK can jointly produce results in support of conservation and prudent management (Giles et al. 2016), alleviate resource management conflicts, and rebuild trust and collaborative relationships among communities where colonial practices have previously led to ill-advised management practices, disempowerment, marginalization and (or) unfairness (Keenan et al. 2018; Held 2019). This recognition of the need to integrate IK is, however, bereft of any comprehensive and practical guidelines that can provide informative tools for effectively engaging local resource users and empowering their role in the decision-making process. Across Canada, a common concern that Indigenous communities face in the consultation processes is the lack of capacity to gain an adequate understanding of technical documents but also the limited time and funding available to them to participate and respond to decision-making processes, discussions, or hearings (Supreme Court of Canada 2017), as well as the fact that decision-making largely takes place in English (Mason 2015; Giles et al. 2016). These are all large barriers in the process of knowledge co-production that hinder meaningful participation of Indigenous communities in regulatory processes. The increasing body of work on the value of IK conservation narratives highlights several overarching areas deserving of further research attention (Berkes et al. 2007; Kaiser et al. 2019). In this paper we explore the role of indicators for communities across Inuit Nunangat, that are unique compared with where much of the indicators work has been developed and applied before. These differences are, to a large extent, grounded in the challenges associated with the geographically remote locations of those communities, resulting in high expenses related to ongoing data collection and monitoring (Nilsson et al. 2013; Adlard et al. 2018), in addition to the presence and importance of IK and difficulties in properly accounting for this knowledge system in indicator development and use in management (Berkes et al. 2007). Meanwhile, taking into account the ongoing rapid shifts in Arctic ecosystems, the need for developing a monitoring framework for marine resource management that is responsive to the values and the changing needs of the communities that are dependent upon those resources has never been greater. Furthermore, we review circumstances where IK has not been included in decision-making and discuss the ramifications of this exclusion. We then highlight, conceptually, how indicators as boundary objects may serve in helping to provide a path forward for incorporating multiple knowledge systems and co-producing knowledge. Indicators can facilitate co-production of knowledge by translating knowledge across boundaries between different knowledge holders, and enabling cross-cultural communication and collaboration for improved marine resource management. Because indicators are often viewed as conceptually simple tools that have the ability of deconstructing complex metrics into easily understood, clear reference points, they can be interpreted in different ways, based on the research question or need for the indicator. This flexibility lends itself useful to multiple parties or knowledge systems.

Although the focus of this paper is in demonstrating where boundaries exist and identifying ways to overcome them through boundary work that bridges WS and IK in support of marine resource management decision-making in the Arctic, the same argument can likely be made for addressing such boundaries between western science with other IK systems.

Despite increasing literature in recent years supporting the importance of integrating IK in resource management (Huntington et al. 2002; Brook et al. 2009; Kaiser et al. 2019), more research effort is needed to shift attention to this issue and specifically to highlight how IK has and could impact further management decisions. All four of the land claim agreements in Inuit Nunangat explicitly reference the need to incorporate IK (termed differently in the different agreements) into fisheries and wildlife decision making (Table 1). Yet several recent high-profile cases exemplify contradictions to this. This analysis does not attempt to provide an exhaustive review of those cases but instead follows selected examples such as the polar bear quotas in Nunavik (Makivik Corporation 2016, 2017), seismic testing in Nunavut (Supreme Court of Canada 2017), and decisions around commercial fisheries allocations in Inuit Nunangat (Snook et al. 2019).

Article 8(j) of the Convention for Biological Diversity requires its parties to “respect, preserve and maintain knowledge, innovations and practices of Indigenous and local communities embodying traditional lifestyles relevant for the conservation and sustainable use of biological diversity and promote their wider application with the approval and involvement of the holders of such knowledge, innovations and practices and encourage the equitable sharing of the benefits arising from the utilization of such knowledge innovations and practices (CBD 2013).” The progress towards achieving the goals set within the Strategic Plan for Biodiversity 2011–2020 and the Aichi Biodiversity Targets are closely monitored. The Working Group on Indicators of the International Indigenous Forum on Biodiversity is assigned with assessing progress towards Article 8(j) goals, and has developed a “community-based monitoring and information system” approach, to operationalize the indicators on the status of traditional knowledge, innovations and practices, and customary sustainable use (CBD 2013). Such policy mandates provide useful frameworks as well as guidance of qualitative nature for fostering the development of inclusive resource management indicators that respect values and aspirations of end-users. Despite international bodies, assessment programs, and frameworks (United Nations 2007; CBD 2013; Thaman et al. 2013; IPBES 2017) that recognize the important role of cultural norms and customs of Indigenous communities in the process of developing resource monitoring and management tools, there still remains an underlying challenge tied to the question of how to put this vision into a set of concrete policies and strategies for the design of indicators.

Indicators can also serve as a means of engaging communities in monitoring of their adjacent marine resources and can strengthen communication channels between scientists and communities affected by environmental changes. Note that these perspectives on indicators, the objective, degree of specificity, and ways of monitoring may differ (Berkes et al. 2007). The Tarium Niryutait Marine Protected Area (TNMPA) in the Northwest Territories (Inuvialuit Settlement Region), which is also Canada’s first Arctic MPA, provides an ever-evolving example of how critical, yet complicated, the process of developing and selecting appropriate indicators can be (Fisheries and Oceans Canada 2010; Loseto et al. 2010). Despite an increasing literature that provides evidence on the positive ecological effects of MPAs and their role in strengthening marine ecosystem resilience, as Pendleton et al. (2018) point out, much still remains to be learned about their socio-ecological impacts. In the context of the Canadian Arctic, a relatively data-deficient place compared with other parts of the world, and where environmental change is occurring quickly as a result of climate change, understanding the dynamics of an MPA becomes even more challenging. In addition to that, decision-making for selecting those indicators that can link the human dimension to the ecological effects of the MPA tends to be notoriously difficult, if one takes into account the differences in the knowledge-generating systems of WS and IK, and the complexities of communities whose livelihoods may be heavily dependent upon the resource(s) within the MPA(s).

The development of indicators can be a demanding process given that they are expected to be robust and meaningful to end-users without being too complex or unfocused for policy-makers and users (Potts 2006; Medley et al. 2009; Turnhout 2009). Turnhout (2009) describes “effective” ecological indicators as “boundary objects”, given that they are the result of science-policy boundary work. The concept of “boundary objects”, draws from the field of Sociology of Science and was first introduced by Star and Griesemer (1989) to describe flexibility and adaptability in information, to allow for integrating different viewpoints, maintaining at the same time robustness and integrity. Boundary objects and boundary work are theoretical concepts developed to satisfy, among other things the desire to “analyze the nature of cooperative work in the absence of consensus” (Star 2010, p. 604). It is therefore a convenient framing tool, used in various fields, to refer to artefacts of practice that help promote intercommunal negotiations and knowledge sharing between diverse groups (Brown and Duguid 2001; Sapsed and Salter 2004; Clark et al. 2010; Zurba et al. 2019), as well as facilitate an understanding of alternative meanings among those (Henderson 1991).

White et al. (2010, p. 221) describe boundary objects as hybrid constructs that integrate elements of different worlds “to facilitate the negotiation and exchange of multiple types of knowledge and action”. These different perspectives or understandings, in this case IK and WS, can benefit from a “boundary object” that inhabits the intersection of the different worlds and facilitates interactions between the two (van Pelt et al. 2015). The existence of a boundary object helps create an operating space for the different knowledge holders to share their understanding and interpretations without consensus being a necessary condition (Star and Griesemer 1989; Shackley and Wynne 1996; van Pelt et al. 2015). Prominent examples of proposed applications of boundary objects to bridge the heterogeneity of different knowledge systems and support improved decision-making in marine and coastal environments include coastal zone management (van Enst et al. 2018), uses of coastal and marine protected areas (Bainbridge 2014; Döring and Ratter 2015; Floor et al. 2016), ocean acidification (Dannevig et al. 2019), coastal resilience and hazard-risk mitigation (Becker 2017) as well as adaptation (Shaw et al. 2013). Relevant to broader applications in environmental decision-making, a diverse body of literature highlights the potential of boundary work and boundary objects for improved environmental sustainability and ecosystem outcomes (Clark et al. 2016; Bednarek et al. 2018; Ainscough et al. 2019; Posner and Cvitanovic 2019) with a considerable portion of this scholarship focusing on climate change policies and adaptation (Miller 2001; Adelle 2015; Kirchhoff et al. 2015; Laudien et al. 2019). As evidenced by these applications, the idea of boundary objects leading to more successful environmental policies and their implementation is not new. However, only a handful of studies consider the role of indicators specifically for ecosystem monitoring, resource management, and decision-making (Turnhout et al. 2007; Turnhout 2009; Uehara and Mineo 2017).

Using the framework in Bowker and Star (2000) for classifications (and standards), Turnhout (2009) refers to effectiveness of ecological indicators in policy and management as having the flexibility required to encompass the needs of those who develop the indicator as well as those who use it to modify it accordingly to shifting circumstances. Using two case studies, Turnhout (2009) provides evidence that ecological indicators function as “boundary objects” only in the existence of shared values and preferences, which implies that culture and social context should touch a common basis. This can be seen as a limitation in the interpretation of flexibility for boundary objects.

In support of the notion of ecological indicators as boundary objects that connect the domains of science and policy (Turnhout 2009), we further argue that in socioecological contexts the boundaries between science and policy may be even more flexible and subject to more negotiation between communities and policy-makers than simply a consideration of their effectiveness in the ecological sense. Stronger ties across not only research fields, but also regional rightsholders and policy-makers, is the premise for an effective transdisciplinary approach to governing social-ecological systems, including fisheries management. Socioecological indicators for marine resource management are expected to result from boundary work across disciplines to enhance possibilities for successful resource management (Palmer 2001), and we argue here across knowledge systems, such that knowledge integration or knowledge co-production with marine-dependent communities can be realized. Glaeser et al. (2009) identify large knowledge gaps in socioecological research and point out as particularly problematic the limited use of interdisciplinary theories and discourses in research for coastal and marine ecosystems. Among other things, they suggest designing indicators that can adequately measure progress in sustainable development and resource management as a way to address these methodological deficits in socioecological research.

In the following sections, we outline case studies to show that indicators grounded in WS and IK, can cross disciplinary and knowledge boundaries, and can serve as a powerful tool for fisheries management and management of marine resources more broadly.

As an emerging literature has shown, fisheries management in communities with a strong Indigenous element typically face more challenges compared with fisheries in other parts of the world, including a greater vulnerability in the face of institutional changes such as rights privatization (McCormack 2010; Carothers and Chambers 2012) as well as complex trade-offs that come along with shifting institutions posing threats to communal values (Plagányi et al. 2013). A prominent example of such institutional shifts that occurred through rights privatization is the introduction of Individual Transferable Quotas (ITQs) in New Zealand that do not align well with the Māori values of communal sense of property (McCormack 2010). In Alaska, Indigenous fishers have challenged exclusionary fishing policies and the implementation of ITQs for violating their rights (Carothers 2011; Carothers and Chambers 2012). Indigenous communities overall tend to be marginalized from decision-making in fisheries management (Richmond 2013), an effect of colonialism that has affected Indigenous Peoples across the world in very similar negative ways. Zurba et al. (2019) describe this marginalization as a “structural injustice” that is fueled by the lack of understanding across communication barriers that exist at the boundaries between groups.

Drawing on Ostrom’s seminal framework on institutions (Ostrom 1990, 2005; Crawford and Ostrom 1995), Zurba et al. (2019) lay a solid foundation for the importance of boundary work when working with Indigenous communities and introduce place-based research and governance as an essential element for overcoming barriers and enhancing effective collaborations between communities and western science. Specifically, Zurba et al. (2019) describe “Research Communities of Practice” as an indispensable tool for bringing those groups closer. Boundary objects have been used in the past in successful ways among Indigenous communities to bring to surface different perspectives, build a common understanding and communicate messages across boundaries, as well as to promote discussions and facilitate relationship building (see Zurba et al. (2019) for a review of such case studies).

The Canadian government was among the first to develop environmental indicators in the late 1980s (Hammond et al. 1995), which were later recognized during the United Nations Conference on Environment and Development in 1992 as an important tool for informed decision-making. The Canadian Environmental Sustainability Indicators program, a recently launched initiative, in line with the Sustainable Development Goals that tracks Canada’s environmental performance, lists among its long-terms goals “Healthy Coasts and Oceans” (Environment and Climate Change Canada 2019, p. 54). Fisheries and Oceans Canada (DFO), the federal department responsible for monitoring the performance, recognizes the need to use indicators for marine resource management. This is reflected in management frameworks for commercial fisheries such as in Integrated Fisheries Management Plans (IFMPs) that list utilization of indicators for monitoring stock and fishery change, meeting conservation objectives, developing monitoring plans (which include the socioeconomic dimension) and assessing effectiveness, but also assisting managers and informing the general public. Yet, despite the recognized need to develop indicators and monitor their performance throughout time for fisheries management, this literature is still in its infancy, with very few studies in the indicators literature focusing on the Canadian Arctic or other regions of the world with similar socioeconomic characteristics.

The ISR beluga case study can serve as evidence of indicators operating as boundary objects, whereby WS and IK holders collaborate in a transparent manner from different sides of the boundary for an improved common understanding of beluga health. Co-produced indicators strengthen the effect of the boundary work, when both WS and IK knowledge systems contribute to a common indicator. Ideally, the development of indicators that can function as boundary objects begins with identification of the most critical shared questions of interest to local communities, researchers, rightsholders, and managers to ensure the indicator is informative. Early involvement of all groups in identifying the needs and challenges for the development of indicators is critical for making sure that indicators have a valuable informative potential and utility. However, as identified, existing programs have the capacity to re-evaluate goals and find new ways to integrate IK or co-produce new knowledge to ensure better understanding across knowledge systems.

The literature on conducting research among Indigenous Peoples puts a great emphasis on methods such as participatory modelling, Indigenous co-management, co-production of knowledge, and generally inclusive methods to ensure that local communities are well-engaged in the decision-making process (Drawson et al. 2017). Participatory processes that engage communities are key in promoting sustainable long-term resource use (Coombes et al. 2012; Zurba et al. 2019). In recent years, scholars mindful of past colonial research practices that were exclusionary, marginalizing, and disempowering (Leeuw et al. 2012; Smith 2012; Zurba et al. 2019), have put significant effort into meaningfully engaging Indigenous communities through participatory and community-based research methodologies in ways that allow for Indigenous self-determination (Castleden et al. 2012, 2017; Smith 2012; Zurba et al. 2019). These efforts have created a shift in research methodologies to aim at having Indigenous Peoples at the forefront of research conducted in their lands and having them drive the research questions, needs, and practices (Davidson-Hunt and Michael O’Flaherty 2007; Castleden et al. 2017; Graeme and Mandawe 2017). Such long-term joint research efforts that involve Indigenous communities and WS knowledge holders have strengthened co-learning and have contributed to building relationships that are key for knowledge co-production (Country et al. 2016; Castleden et al. 2017; Moore and Hauser 2019). However, the notion of community involvement is in many cases viewed as rhetoric on behalf of the communities themselves; a feeling that has been created largely due to the ways in which research has been conducted in the past and also unexpected uses and communication of the research results (Natcher et al. 2005). Despite the recent efforts to advance research practices, there is still criticism of what WS researchers perceive as participatory, community-empowering research in what makes a genuine partnership with marginalized and disenfranchised participants who may have limited capacity to engage in research, planning, and resource management (Matunga 2013; Coombes et al. 2014; Zurba et al. 2019). Western knowledge can, therefore, easily be discredited as uncertain if the conditions for equitable collaborations are not in place. Several Indigenous scholars, however, are writing prolifically on the empowerment of Indigenous communities in the research process, and these success stories should not be minimized (see for example Tuck 2009 and Tuck and Yang 2012).

Close collaboration and early consultations with communities can assist researchers in the process of understanding what are the indicators that the communities are more likely to use and benefit from, and can better set up multiple knowledge systems to support different levels of decision-making in the long-run. Ideally local actors shall be engaged early in the process of shaping the research agenda together with the researchers or the institutions that fund the research to make sure that there is no effort wasted in developing indicators that no one will ever use. Such a process is also likely to increase acceptability of the management measures proposed, in a way similar to citizen or community science where community members are directly involved in the research. Marine resource social scientists have gone as far as discussing the positive effects of community involvement in building trust that centralized decision-making authorities “do the right thing” even in cases where community views and the information that comes from those are not actually taken into account in the decision-making process (van Putten 2018).

When it comes to Indigenous communities, science should be conducted and communicated in ways that can accommodate cultural norms and customs so as to not only build trust in the results produced but also to ensure that IK and WS can jointly produce sustainable ecological management (Moller et al. 2009b, 2009b). Such an approach, in addition to being more effective for conservation management than WS alone (that has in the past been conducted in isolation), also holds the potential of replicating observations in different places and over time as IK can offer perspectives built through long-term practical experiences or what Moller et al. (2004) describes as “diachronic”. This also allows for the possibility to identify rare events and details about the resource at fine-scale (Moller et al. 2004); examples include voluntary record-keeping or harvesting diaries such as those used for the Atlantic Walrus in Nunavut’s community-based monitoring program and for participatory mapping in Nunatsiavut through the Imappivut project (Nunatsiavut Government 2017). In contrast, the “synchronic” (Moller et al. 2004) nature of WS often lacks the possibility of being replicated at multiple spatial and temporal scales, often due to funding, time, or other limitations. The two systems of knowledge can, therefore, nicely complement each other and also have the potential of providing an overall improved understanding of the local marine ecosystem, which results in opportunities for better monitoring and management.

In the process of co-developing indicators, however, to avoid what Natcher et al. (2005) describe as “research for the sake of research”, it is important to ensure that the necessary structures are in place so that there is a continuum in monitoring of the marine resources of interest. Transferring the necessary skills and knowledge produced in the co-development process is essential for community members to see value in the indicators developed, and if deemed necessary, to continue and further develop those indicators according to their future needs.

The conceptual framework in Fig. 2 highlights indicators as boundary objects that are enhanced by a co-production process, which effectively brings together knowledge systems with co-management boards, resource managers, and Inuit marine resource users. The social, economic, and ecological indicators co-produced by these actors are a result of collectively designing the research agenda by prioritizing local needs, co-conducting research through active community participation, co-learning in the research process, and finally, co-producing knowledge. This process allows users to monitor adjacent marine resources of importance to their livelihoods and strengthens their ability to participate in meaningful ways in the decision-making process. The co-production of socio-ecological and economic indicators through such research partnerships with communities can help alleviate structural injustices such as the ones described earlier where local knowledge was lacking from the decision-making process and also help break any communications “barriers” between the aforementioned groups (see Zurba et al. 2019).

Despite the benefits in operationalizing IK in the development and use of indicators, it is still considered a challenging task (CBD 2013). Community-based monitoring and information systems (CBMIS) are a bottom-up approach used for the development of indicators that contributes to the continuing efforts to utilize IK in combination with newly available technologies for decision-making and planning (CBD 2013). CBMIS enable diverse knowledge holders to effectively engage in monitoring, analyzing, and also reporting (Tengö et al. 2014). Additionally, through facilitating better understanding and communication channels, they have the potential to offer communities useful insights into scientific methodologies that can lead to improved co-production of indicators.

However, we show that the development of indicators meaningful to communities is dependent on incorporating community views at an early stage of their design, which has not been the case so far across Inuit Nunangat. This points at the need to adopt bottom-up approaches, such as CBMIS (Tengö et al. 2014). The indicator literature for economic and social well-being, however, highlights that top-down approaches confer the advantage of providing consistent estimates of indices across space and, therefore, enable comparisons in trends across communities given that the same variables and methodologies are used (Sharpe 1999). Although this is true from a strictly efficiency and economies of scales lens, it is also too narrow in recognizing the differentiated needs of Inuit communities and capturing the complexity of local marine resource use and its contribution to economic and social well-being. This perspective is tied to the inception of indicators as tools that were designed to span national to global scales (Tengö et al. 2014). Comparisons across regions can be useful for monitoring regional trends (Mitchell and Parkins 2011). Caution is needed, though, against drawing strong inferences for the drivers of trends and maintaining those indicators sensitive to local priorities and meaningful to local users and stakeholders (Mitchell and Parkins 2011; Tengö et al. 2014). Top-down management approaches driven by aggregating trends and lacking regional input are seen as part of the colonial legacy, and reduce trust and acceptability of the indicators. Such approaches prevent indicators from serving as tools that cross boundaries between different knowledge systems and limit the understanding of IK, which leads to missing out on the value of its contribution to improved marine resource monitoring, planning and management.

Additionally, as opposed to the argument in Sharpe (1999) that the geographical dimension has little effect in the development of an index, several examples across the Arctic provide evidence that it deserves more attention. Compared with communities in more southern latitudes, marine resource management in Inuit communities represents additional challenges owing to their geographically challenged locations. Additionally, while the role of these communities in driving climate change, including anthropogenic activity affecting the marine environment, has historically been lower compared to the rest of the world, they appear more vulnerable and exposed to climate shifts and other changes in oceanographic variables. In this context it is important to acknowledge the motivation for developing the indicator, which goes beyond advocacy purposes or academic interests only. The co-production process shall be driven by local needs and accommodate local norms and customs to ensure that they can be trusted and have an impact on marine resource management.

Engaging in boundary work and creating boundary objects, as we have described earlier in this paper for the development of socioecological and economic indicators, can be a time-consuming process that may result in challenges to those engaged in the co-development process. These may range from requirements of research institutions regarding the ways in which one can engage in discussions with Indigenous communities, to requirements for research ethics from the communities themselves (Zurba et al. 2019). Additionally, although the idea of integrating IK is widely recurrent in the literature, it is the assertion of the authors that the inclusion of such perspectives may prove challenging owing to conditions attached to funding within academic institutions, for example conditions around data ownership, funding cycles, and short turn-around times for proposals. These issues can be a major roadblock to building long-term relationships with communities that are an essential element for the co-production of indicators.

Larsen et al. (2010) bring forward the critical issue of data availability, which is a major barrier for the development of reliable WS indicators. As described earlier, harvesting for fish is an important part of the subsistence economy across Inuit Nunangat. In many cases, though, such as in the case of Arctic char fishing for subsistence or recreational purposes, there may be no monitoring in place for the quantities harvested or the locations where the harvest takes places. This makes it difficult to know if key WS indicators (i.e., spawning locations, recruitment rates, environmental changes in key habitats, and harvest rates) are changing, thus signaling potential stress to subsistence needs. Earlier attempts to develop measures of this mixed economy, where subsistence harvesting plays an equally important role to wage employment, have highlighted the intertwined nature of culture and economy in the Arctic (Poppel et al. 2007; Poppel and Kruse 2009). Those attempts, although very useful in providing a better understanding and measuring the living conditions in the Arctic in a systematic way, have not managed to consistently document indicators across all Arctic regions, which to a large extent includes the Canadian Arctic.

Recognizing the limitations of the data available for developing WS indicators is key to enable better informed and more responsive decision-making as well as for improving the quality of the indicators and their monitoring across time. As MacMillan et al. (2019) rightly point out, for the case of beluga indicators in the TNMPA, it is essential to be able to monitor the area using indicators that can capture both the current state but also potential ecosystem changes (see also Rice and Rochet 2005 and Vandermeulen and Cobb 2004). Long-term monitoring data sets, such as those available for the Eastern Beaufort Sea Beluga whale, can help build an understanding of whether the metrics used are appropriate for monitoring the health of the species, taking into account also its spatial distribution (MacMillan et al. 2019), and how to further advance those metrics using IK (Ostertag et al. 2018). IK holders often have a better awareness of their local ecosystem and are able to identify more and diverse signs of change promptly. The role of IK is essential in co-producing indicators that can capture the type, scale, and spatial dimension of changes quickly enough and provide an understanding of systems or drivers outside the conventional WS indicator framework. This includes focusing on dimensions that WS has historically ignored, such as ways in which beluga whales’ health impacts human health, a potential future indicator for both long-term beluga and human health.

It is also important to recognize that despite an increasing literature aimed at emphasizing the importance of IK in the long-term monitoring and management of marine resources, weighing the role and the use of this knowledge in practice remains a challenge. Fedirechuk et al. (2008) highlight in their Guide for impact assessments for the ISR that it is important to be able to explicitly describe how the knowledge is used in the analysis so as to allow for potentially critical information gaps and limitations in the scientific approach to come to surface — a practice that is often absent in the western scientific culture. To that end, indicators can also be used to measure the process of knowledge integration in the co-management approach.

The challenge we take with this paper is to conceptualize the notion of indicators as boundary objects suitable for integrating IK and WS in geographically challenged areas, drawing on, and synthesizing the relevant literature and reviewing case studies. The Beaufort Sea beluga case provides a solid example of how indicators developed for management can be grounded in local knowledge and the practice of harvesters. Effective integration and utilization of IK remains a compelling priority in achieving the goals of marine resource management plans such as IFMPs as well as for meeting political commitments such as those of “reconciliation”. But even more so, incorporation of IK into marine management is mandated in the various land claim agreements, and is a necessary (but not sufficient) condition for achieving self-determination in Inuit Nunangat. The development of indicators that meet these objectives requires direct involvement of communities in the research process as well as in formulating the research agenda to enable supporting regional management and providing a meaningful tool to local users.

This paper touches upon a few of the diverse issues facing Inuit fisheries. The central finding of our analysis suggests that the co-production of indicators for marine resource use and management, in close collaboration with the communities that have a direct interest in the resource, can serve as a powerful tool in support of decision-making, monitoring, and the design of policies that seek to maximize community well-being. Conceptualizing those indicators as “boundary objects” helps provide a better understanding of how bringing together and utilizing diverse knowledge systems can lead to not only better and more inclusive monitoring of marine resources but also acceptability of policies through better integration of Indigenous perspectives and harvesting rights in management. The case studies from communities in the Canadian Arctic, discussed in this paper, provide a concrete example of how important such transdisciplinary endeavors between the communities themselves, natural and social scientists, and practitioners, are in developing tools such as indicators for monitoring and long-term sustainable management of marine resources. However, while we stress the importance of these endeavors, there are very few clear, strong examples occurring in coastal and marine research across communities, especially those with strong social, cultural, and economic dependence on adjacent marine resources. More work is, therefore, needed to better formally integrate the viewpoints of local users, science, and management.

This research was undertaken thanks in part to funding from the Canada First Research Excellence Fund, through the Ocean Frontier Institute (M.K. and M.B.) and ArcticNet (C.H. and M.B.). The authors thank the Nunavut Wildlife Management Board and the Torngat Secretariat, especially Jason Akearok, Amber Giles, Jamie Snook, and Aaron Dale, for conversations over the past two years that influenced this work, as well as Rachael Cadman for providing feedback, and the Nunatsiavut Government for assisting with translation into Labrador Inuttitut.

Qaujisartuliriniujuq piliriangulilaurtuq qujannamiik kiinaujatigut ikajulaurtut kiinaujatigut Kanatami Sivulliutitaujut Qaujisartulirinirmit Piunirpaanginnit Kinaujatigut Ikajurpaktut ammalu ukiurtartumiut Imavilirinirmi Ilinniarvik (M.K. ammalu M.B.) ammalu ukiurtarturmiut ArcticNet (C.H. ammalu M.B.). Titirartiujut qujalirtiartut Nunavutmi Uumajulirijitjuat Katimajingit ammalu Umajulirijikkunginnit katimajiujut ammalu sunatuinnattianik nunami imarmiluunniit pilirijiujunu, piluartumi Jason Akearok, Amber Giles, Jamie Snook, ammalu Aaron Dale, uqallaqatigijunnalaurakku araagu marruulirtuuk taanna piliriarijaujunnalirmat, taanalu Rachael Cadman tusaumatittitanginnarninganut, ammalu Nunattiavut Gavamanga ikajuttialauqmat inuktituuliqtitijunnaqataulluni Lapatua uqausingatitut.