The role of western- based scientific, Indigenous and local knowledge in wildlife management and conservation

1. Managers of wildlife are faced with decisions and issues that are increasingly complex, spanning natural and human dimensions (i.e. values, preferences, attitudes). A strong evidence base that includes multiple forms and sources of knowledge is needed to support these complex decisions. However, a growing body of literature demonstrates that environmental managers are far more likely to draw on intuition, past experience or opinion to inform important decisions rather than empirical evidence. 2. We set out to assess how decision- makers and other potential knowledge users (a) perceive, evaluate and use western- based scientific, Indigenous and local knowledge and (b) the extent to which social, political and economic considerations challenge the integration of different


| INTRODUC TI ON
Managers of wildlife (free-ranging, non-domestic animals) are faced with decisions and issues that are increasingly complex (Arlinghaus et al., 2015;Powell, 2020). This is especially true given the ecological crisis that is upon us, which includes pervasive and escalating threats to wildlife populations from a wide range of sources. For example, the UN body, the Intergovernmental Science-Policy Platform on Biodiversity and Ecosystem Services (IPBES), has recently assessed the global extent of this crisis, finding that up to one million species of animals and plants are at risk of extinction in the short term (Balvanera et al. 2020;Díaz et al. 2019;IPBES, 2019aIPBES, , 2019b.
The cumulative and interacting drivers of these changes such as exploitation of organisms, climate change, pollution, invasion of alien species, etc., pose a particular challenge for already complex wildlife management and conservation.
Wildlife management and conservation involves managing wildlife, their habitat and the people who engage and interact with animals and ecosystems. Such efforts require engagement with not only conventional user groups, such as hunters and anglers, but also anyone with a vested interest in a wildlife issue, programme, action or decision. This reflects the need to integrate human dimensions of wildlife into management (i.e. values, decisions, actions, preferences, attitudes) from an array of increasingly diversified actors (i.e. resource-user groups, industry, private landowners, farmers, policymakers, conservation organizations and other stakeholders and rightsholders) with high expectations for involvement in the process (Decker et al., 2012;Riley et al., 2002). Thus, wildlife managers must not only consider the complex biological and ecological context of the decisions they make but also complex political, social and economic circumstances and influences. This vast social and ecological complexity is key for understanding how wildlife managers use evidence and make decisions, and vice versa.
Evidence-based decision-making is seen by many as an important tool for managing social-ecological complexity (Addison et al., 2016;Pullin & Knight, 2003;Sutherland et al., 2004). Evidence is important for both understanding complex interactions, and for politically justifying policy and management decisions (Adams & Sandbrook, 2013;Pielke, 2007). Moreover, the centrality of evidence to making and legitimizing decisions is prompting researchers and practitioners to consider multiple forms and sources of knowledge (Reed et al., 2013;Salafsky et al., 2019;Tengö et al., 2014).
Specifically, in environmental management, there is a growing interest and emphasis to incorporate a broad range of knowledge types including Indigenous and local knowledge (ILK) alongside the foundations of western-based science (Reyes-García & Benyei, 2019; Thompson et al., 2020;Wheeler & Root-Berenstein, 2020).
However, there are indications that the rhetorical popularity of evidence-based management is not matched in practice. For example, numerous studies have shown that environmental managers are far more likely to draw on intuition, past experience or opinion to inform important decisions rather than evidence derived from western-based science (e.g. see review in Kadykalo et al., 2021;Cook et al., 2010;Fabian et al., 2019;Matzek et al., 2014;Pullin et al., 2004;Young & Van Aarde, 2011). This is described as 'evidence complacency' (Sutherland & Wordley, 2017) in which despite the availability of evidence, it is not sought or used to make decisions. Even less is known about the use of more informal and tacit types of knowledge such as ILK. However, Lemieux et al. (2018) recently revealed that in Canada's protected areas organizations, ILK is also valued and used less than personal and institutional experiential knowledge.
These studies suggest that the creation of knowledge and collection of evidence are necessary but not sufficient criteria for enacting evidence-informed decision-making. A growing research field, so-called 'knowledge exchange', has been focusing on how knowledge is exchanged and mobilized, and with whom it is exchanged (Cvitanovic et al., 2016;Fazey et al., 2012;Reed et al., 2014). This literature insistently emphasizes the need for knowledge and knowledge generators to be, or perceived as, salient (relevant and timely), credible and legitimate to enable effective knowledge exchange (see Cash et al., 2003;Cook et al., 2013). However, effective knowledge exchange requires determining, first and foremost, how (sometimes competing) knowledge is used, perceived and evaluated by potential users. While this seems like a logical and obvious objective for those studying knowledge exchange, it has not often been explored empirically (Tengö et al., 2017;. The characteristics of knowledge and knowledge-holders notwithstanding, the use of knowledge is also dependent on the intense 5. We encourage transformative change in wildlife management enabling decisionmakers to draw upon multiple forms of knowledge. This transformative change should include direct involvement of knowledge holders, co-assessment of knowledge and transparency in how (multiple forms of) evidence contribute to decision-making.

K E Y W O R D S
co-assessment, evidence complacency, evidence-based conservation, fish and wildlife management, Indigenous and local knowledge, knowledge-action gap, knowledge exchange, natural resource management political, social or economic considerations faced by potential knowledge users. Some of these considerations may constrain, compromise and interfere with the ability of wildlife managers to make decisions based on evidence. Notably, limited financial resources, lack of staff and inadequate timeframes are significant barriers to knowledge use in both western (Lemieux et al., 2018;Westwood et al., 2017; and Indigenous  environmental management agencies. Indeed, these challenges have already been well documented in our study area, the province of British Columbia in Canada. For example, a survey of 403 provincial government scientists found the majority (71%) said they had witnessed a decrease in research capacity in their ministry and/ or branch over the course of their tenure in BC government; and 68% believe that there are insufficient resources to effectively fill their branch or ministerial mandate (Smith et al., 2017). It is also telling that 57% of government scientists believed that public service cuts compromise the government's ability to use the best available evidence in decision-making, and that 49% believed political interference has compromised their ability to develop laws, policies and programmes based on evidence.
While we delineate, western-based scientific knowledge, Indigenous knowledge and local knowledge for this article, we also recognize that the differences among these knowledge types fall along, at best, a fuzzy spectrum. This delineation is an artificial construct and risks oversimplifying knowledge systems that are diverse, complex and increasingly intertwined. Indeed, ILK Díaz et al., 2015;Tengö et al., 2017) shares many similarities with western-based scientific knowledge (i.e. learning by doing, building and organizing knowledge). However, delineations play a necessary role in facilitating evaluation of knowledge, which occurs primarily within, rather than across knowledge systems (Alexander et al., 2019;Tengö et al., 2014). It is important to consider the differences among these knowledges and their approaches may not be only related to the different data or information themselves but also possibly with different understandings of how management should proceed. Overlooking these differences can hinder collaborative arrangements among western, Indigenous and local governments and communities.
Using western-based scientific knowledge, ILK individually or synergistically can yield complementary insights that can enrich and enhance our collective understanding of the natural world Tengö et al., 2014). ILK can provide valuable information from long-term ecological monitoring to inform conservation goals and adaptive management, especially in data-poor scenarios.
For example, Indigenous knowledge such as current and historical estimates of fish body size and abundance extended baselines for data-poor species such as yelloweye rockfish in British Columbia, Canada (Eckert et al., 2018). Local knowledge from recreational anglers and spear fishers in Galicia, Spain provided valuable data for management on the temporal declines of targeted species such as cephalopods and finfish stocks but also on non-target keystone species such as the poor status of kelp beds which support these coastal ecosystems (Pita et al., 2020).
Although there is an expanding interest from the western governments and institutions in engaging with ILK holders (Simpson, 2004;, there are significant barriers to meaningful inclusion of Indigenous and local views and knowledge. Such issues include, for example, the perceived need to 'validate' ILK with western-based science; the low level of knowledge-holder inclusiveness; the amount of time required to build relationships and gather knowledge; and the blatant disrespect or ignorance of Indigenous rights Reyes-García & Benyei, 2019;Wong et al., 2020). The extraction of Indigenous knowledge, in particular, for use by western decision-makers may be problematic and work to further settler colonialism, especially if subsequent decisions are made without the full involvement, collaboration and consent of the Indigenous and local communities themselves. This can lead to the marginalization, appropriation and commodification of Indigenous knowledge (Simpson, 1999(Simpson, , 2001. The mountainous province of British Columbia (BC), Canada makes for a relevant case to explore the role of well-informed decision-making to enhance social-ecological resilience. BC is rich in natural resources contributing substantially to the local and national economies. It is also home to ethnically and culturally diverse people (Indigenous peoples with traditional and constitutional rights, European and Asian immigrants, engaged resource-user groups).
The province is experiencing rapid biophysical changes to its highly diverse ecosystems impacting tightly linked social-ecological systems. Climate-driven hydrological changes (e.g. increased summer freshwater temperatures, more hypoxic lakes; reduced snowpack; earlier onsets of spring snowmelt) are a primary concern (Healey, 2011;Islam et al., 2017;Zwiers et al., 2011). Meanwhile, BC's boreal forests have suffered extreme wildfire seasons and a severe mountain pine beetle Dendroctonous ponderosae Hopkins outbreak (Dhar et al., 2016;Kirchmeier-Young et al., 2019). These vulnerable habitats support charismatic at-risk wildlife species of major significance to Indigenous and non-Indigenous people alike such as cutthroat Oncorhynchus clarkii and bull trout Salvelinus confluentus, and mountain caribou Rangifer tarandus. Moreover, they support social, cultural and economic well-being of BC's people in the form of, for example, water supply, subsistence and recreational fisheries and hunting, wild foods, sense of place, cultural identity and heritage.
The goal of this paper is to provide an initial assessment of the extent to which Indigenous, local and western-based scientific knowledge are incorporated into wildlife management in BC.
We used semi-structured interviews to assess how these different knowledges are (a) perceived, evaluated and used by potential knowledge users-Indigenous governments, parliamentary governments and stakeholders and (b) the extent to which social, political and economic considerations challenge the integration of evidence into decision-making. In (a), perceptions and evaluations apply to the knowledge generators (i.e. holders) by extension also. While we have attempted to have a representative dataset, we acknowledge our data are biased overwhelmingly to parliamentary government or NGO decision-makers and this assessment is therefore most reflective of how these different knowledges are used by non-Indigenous and non-local decision-makers. We explore this using the case of managed rainbow trout Oncorhynchus mykiss fisheries in BC. This effort to gain a better understanding of how knowledge is perceived, evaluated and used will hopefully lead to a stronger and more diverse evidence base and in turn, more informed decision-making based on multiple ways of knowing.

| The case
The near-ubiquitous range of rainbow trout in BC is managed through a complex combination of provincial and federal government agencies and processes. Indigenous communities and governments also manage rainbow trout in certain territories. Additionally, stakeholders (e.g. academic researchers, non-profit organizations, private consultants, resource user groups) are also important actors in the conservation and management of rainbow trout in BC.

Today, many Indigenous communities and governments manage
Indigenous and non-Indigenous recreational and subsistence fisheries on their reserve lands and beyond. Over the vast majority of BC, colonial immigrants settled on land for which Indigenous title had not been ceded or negotiated. The Dominion of Canada and provincial governments undertook a joint process in BC that imposed a very small reserve for each of the many First Nation (Indigenous) communities (Harris, 2008). The small reserve allotment process (only slightly more than one-third of 1% of the land areas in the province) was designed on the erroneous assumption that Indigenous peoples in the province were primarily fishing peoples and did not require In BC, myriad stakeholders wield great influence on the conservation and management of rainbow trout. Many of the lakes within BC do not naturally support fish populations due to a lack of spawning habitat or other limitations, meaning that populations are maintained by annual stocking. The Freshwater Fisheries Society of British Columbia (FFSBC; https://www.gofis hbc.com) is a private non-profit organization contracted by the provincial natural resources ministry to deliver the provincial fish stocking programme and to offer a range of conservation services (i.e. outreach activities, education) to protect wild fish by diverting recreational angler pressure to hatcheryraised fish. BC Hydro (https://www.bchyd ro.com), a province-owned electric utility, is a unique actor that has a major water and land footprint in BC. Wildlife mitigation (e.g. fish spawning/bird nesting/ migratory bird habitat protection; fish salvage) and monitoring is conducted by BC Hydro in watersheds impacted by their dams, primarily the Kootenay, Columbia and Peace rivers. Local environmental non-governmental organizations (ENGO) also play an important role as stakeholders and advisors to provincial and regional wildlife conservation and management. For example, BC Wildlife Federation www.steel heads ociety.org) also work towards issues on fisheries conservation and quality fishing opportunities. Many private environmental consultants are contracted throughout the province by federal, provincial and Indigenous governments to conduct research on wildlife issues. Many academic researchers in BC and even across North America also work closely with federal, provincial, Indigenous governments and FFSBC-in research partnerships and collaborations on wildlife and habitat issues or simply act as advice providers.
Retired provincial government employees are also important actors, providing a unique perspective on wildlife issues and many are still active within this area as, for example, members or employees in ENGOs described above, fishing guides, issue advocates, informal government advisors.

| ME THODS
This research is exploratory, aimed at investigating and categorizing how decision-makers and other potential knowledge users involved in the conservation and management of wildlife within British Columbia view and use various types of knowledge (i.e. Indigenous, local and western-based scientific). As such, this research is intended to be primarily descriptive, and hypothesisgenerating rather than hypothesis-testing. The data reported in this article were collected as part of a broader study entitled 'Sustaining Freshwater Recreational Fisheries in a Changing Environment' that aims to develop conservation genomic tools and policy recommendations to help manage and preserve the genetic diversity of rainbow trout. The study objective is to support and sustain healthy populations of rainbow trout and the recreational fishery that depends on them.
Befitting exploratory research, we developed and employed an interview schedule using open-ended questions (Axinn & Pearce, 2006;Creswell, 2014;Young et al., 2018). Open-ended questions allowed a wide range of respondents to explain their positions, priorities and opinions freely. It also allows them to be precise in their answers, providing hard to obtain and sensitive information on evidence use and decision-making processes. The set of questions analysed in this article are provided in Table 1 and a copy of the interview schedules (questionnaires) are provided in the supporting information. This study was conducted in accordance to the University of Ottawa Research Ethics Board (File Number: 02-18-08). All participants gave informed consent to participate in the study. A copy of the consent form is in the supporting information. We performed a pilot interview after ethical clearance that showed no issues. Qualitative data were transcribed from audio to text using Trint (https://trint.com) and analysed using NVivo 12 software (QSR International Pty Ltd., 2018). For open-ended responses, a three-step inductive coding process was applied to qualitative data (Thomas, 2006). First, responses were read to identify keywords, which became a list of potential codes. Similar potential codes were then grouped into themes. Responses were read a second time and sorted under these themes to provide a measure of their prevalence.  Therefore, the responses in this article are most specific to fisheries management but are described throughout under the broader term 'wildlife management and conservation'. Interviews lasted between 18 min and 2 hr, depending on the level of detail provided by the respondent.

| RE SULTS
The results are organized by the order of questions in Table 1.
Respondent sources and illustrative quotations which support our results are provided in Appendix A (Supporting Information) and linked as citations (end-noted superscripted numbers).

| Indigenous knowledge
Indigenous knowledge plays a distinct role in the work for 51% of respondents 1 (i.e. they use it on a day-to-day basis), a minimal or limited role for 9% of respondents 2 and no distinct role for 22% of respondents 3 (Figure 1). For 23% of respondents, Indigenous knowledge was openly described as important, even critical, to their work. 4 The use of Indigenous knowledge was categorized in multiple different ways (Table 3). Many respondents cited statutory and legal (and funding agency) obligations to engage Indigenous peoples in decision-making and consider Indigenous knowledge more strongly. 5 Although, as some respondents noted, much of these obligations are just 'paid lip service', lacking action-'it's very early days for us around that now'.

TA B L E 2
Affiliations of the 65 participants and 96 non-participants (who were contacted but did not participate because they (a) did not respond to our request, or (b) declined to participate), grouped as members from natural resource management branches of Indigenous governments, and parliamentary governments, as well as stakeholders Confidentiality of Indigenous knowledge and a lack of trust may be a (by)-product of (perceived) insufficient legal protection of intellectual property. 7 It may also be due to a perceived concern of further losing constitutional (i.e. hunting and fishing) rights.
Several respondents (from the provincial natural resources ministry, ENGOs, the province-owned electric utility) acknowledged that the role of Indigenous knowledge is less than it should be, and a better job should be done in reflecting Indigenous values and knowledge in modern wildlife decision-making. 8

| Local knowledge
Local knowledge plays a role in the work of 59% of respondents 9 and a minor or minimal role for 3% of respondents 10 (Figure 1). In all, 12 respondents (19%) were definitive on the importance of local knowledge to their work, relying heavily upon it. 11 Descriptions of local knowledge use by parliamentary government and stakeholder respondents overwhelmingly focused on the value of local knowledge from local communities (e.g. resource users groups) identifying blind spots (

| The interface between Indigenous and local knowledge and management
Indigenous and local knowledge are grouped here not to conflate them, or to suggest they are interchangeable, but to best discuss the findings given these types of knowledge or information have not traditionally been the foundation or explicitly accounted for within the frameworks of western (i.e. parliamentary-governed) wildlife management.
Engaging ILK and then incorporating and reflecting it into modern wildlife management is being attempted or discussed 17 (by 45%), but actually doing it in practice is characterized as a challenge. 18 To summarize respondent perceptions, ILK is 'not really straightforward' and is difficult to understand, translate and assess. Additionally, while Indigenous governance systems are highly diverse-they may follow a very different decision-making process which may be very specific, consensus-driven and therefore TA B L E 3 Indigenous, local and western scientific knowledge used as evidence in the work of n = 65 respondents. Raw counts (and %) are number of respondents making a mention to the corresponding use of evidence. Respondent sources and illustrative quotations which support evidence use are provided in Appendix A and linked as citations (end-noted superscripted numbers) When it comes to reconciling ILK with wildlife management, twoway dialogue 25 focused on respectfully unpacking party interests and long-term goals was recommended for building trustful relationships. 26

| Western scientific knowledge
Respondents were near-unanimous that western scientific knowledge plays a role in their work and decision-making 30 (98%; Figure 1). Western scientific knowledge is used by the majority of respondents (69%) to guide decisions, priorities and management actions ( Western scientific knowledge used in decision-making is primarily sourced from 'in-house' 'evidence-producers' 37 (49%; e.g. stock and lake assessments, 38 monitoring programmes, 39 longterm experiments, academic partnerships) and 'external' secondary sources 40 (39%; e.g. peer-reviewed journals and publications, books, information from government management agencies in other jurisdictions). Several respondents clarified that they were not statutory decision-makers, and their roles were entirely about producing western scientific knowledge, and in cases, also providing advice (e.g. briefing notes) for decision-makers or stakeholders. 41

| The diminishing role of evidence in the decision-making process
While evidence clearly has a considerable role in wildlife management and conservation-more than 40% of all respondents, namely, According to these accounts, evidence in decision-making is limited by increased political and socio-economic interference (28%), 45 decreased institutional resources and capacity (9%) 46 and science integrity (9%) 47 (Figure 2).

| Political and socio-economic interference
In the perspectives of many respondents, wildlife decisions may begin as evidence-based but are prone to becoming influenced by social, political and economic factors (e.g. values, ideology). In other words, management actions and policy decisions may deviate from

F I G U R E 2
The diminishing role of evidence in provincial wildlife policy and practice due to increased political and socioeconomic interference; reduced science integrity; and reduced institutional resources and capacity. Measured by percentage for n = 65 total respondents which specifically mentioned limitations to evidence-based decision-making r evidence. This is by no means a recent phenomenon but was described as more likely to occur today than it was some 10-25 years ago. Moreover, it was perceived that the higher within an organizational hierarchy evidence is considered in decision-making (e.g. at the Director or Deputy-Ministerial level), the more likely that evidence will be diluted. However, resource management decisionmaking was also frankly portrayed as a lot more complex today than yesteryear due to for example, a much more knowledgeable

| What is 'reliable' or 'unreliable' knowledge?
Appendix C (Supporting Information) presents the thematic codescriteria-associated with 'reliable' and 'unreliable' knowledge along with the number of respondents making mention of each theme.
Starting with reliability, a substantial minority of all groups mentioned the importance of factual corroboration of knowledge claims (e.g. by pictures, data, etc.; Table C.1). All groups also cited the importance of repeatability and reproducibility to demonstrate consistency, sound research design and methods, and peer-reviewed knowledge or information. Similar numbers across all groups also cite the scientific method, reputation-especially trustworthi-

| D ISCUSS I ON
Contrary to studies that suggest evidence-based conservation and and results of other authors (e.g. Ban et al., 2018;Eckert et al., 2018;Pita et al., 2020;Reed et al., 2013). This knowledge then presumably helps wildlife managers oppose shifting baseline syndrome reducing the potential for overexploitation of nature. Clearly, questions that involve values (e.g. Should trout be introduced in this lake where they are currently and/or historically not present?) benefit from ILK.
We found some evidence for management considering values such as harvest preferences and cultural importance. However, the extent to which management asks such questions to ILK holders is unclear from this data and should be an area of future work.
While respondents were generally willing and interested (and in some cases, required) to increase engagement with ILK, challenges pertaining to knowledge evaluation and use were observed. Namely, a lack of trust, hesitancy to share knowledge (particularly from Indigenous communities), difficulties in assessing reliability and difficulties discerning knowledge from advocacy, that is, 'agency capture' (i.e. undue influence on agency decision-making by special interest groups; Artelle et al., 2018).
Concerningly, regardless of knowledge type, our findings point to a diminishing role of evidence in final decisions concerning wildlife management and conservation. In other words, evidence appears to be an important consideration (as revealed by our results) but is often outweighed by other considerations, contrasting evidencebased decision-making. So, while respondents in our interviews rely heavily on multiple forms of knowledge to inform their decisions, their day-to-day decisions are generally at levels of governance that are not responsible for final decisions that concern wildlife management and conservation. In other words, the majority of participants are not at the top of the hierarchy of the organizations in which the work. We attempted to include statutory decision-makers (e.g. Deputy Ministers, Directors, and Section Heads-see 1.1 The Case) responsible for such final decisions as participants, though such people are few, as is the number of representatives of this group who participated in interviews. Hence, our results support the idea that evidence may form the basis of a decision but is often eclipsed by other, perhaps more economically or politically pressing, considerations (e.g. Artelle et al., 2018;Morrison-Saunders & Bailey, 2003).
We find this in our case despite claims that decisions from statutory decision-makers in BC are grounded using an evidence-based standard (Artelle et al., 2018;Government of British Columbia, 2017).
Like others have found (e.g. Lemieux et al., 2018;Nguyen et al., 2018;, mobilizing knowledge in support of wildlife management and conservation is, in parliamentary governments at least, limited by a decrease in research capacity (time, staff and financial resources) and institutional knowledge integrity.
A recent history of austerity at the federal and provincial levels of government is in part, likely culpable (Smith et al., 2017;Westwood et al., 2017). This may partially explain why many parliamentary government respondents attributed the heavy use of local knowledge to identifying blind spots, providing warning signs of potential crises, thus informing adaptive management.
In our case study, increased socio-economic and political interference strongly corresponded to the diminishing role of evidence in wildlife management and conservation. It is tempting to associate the lack of evidence-based decision-making in wildlife management with the prevailing political climate. However, Artelle (2019, p. 38) suggests these 'cracks' 'run far deeper than ephemeral political cycles' and therefore should not be treated as a temporary phenomenon. Due to capacity, socio-economic and political constraints, parliamentary (and perhaps Indigenous) government natural resource managers may not be empowered to use knowledge, regardless of type, even if it is available.
As we have found here, natural resource management agencies may be perceived as reactionary regulatory bodies, increasingly distanced from the generation and use of evidence. For example, cuts to the public service in BC have resulted in much of the public interest science normally done by the province outsourced to 'qualified professionals' hired by industry and project proponents with little to no oversight (Smith et al., 2017)-putting into question the role of evidence in the public's interest. Concerning the evidence itself, it cannot be simply assumed that there is a dearth of knowledge and that generation of more knowledge, regardless of type is better, benefitting decision-makers in wildlife management and conservation.
As recognized by Lemieux et al. (2018), in capacity-poor organizational settings (the case for many wildlife management agencies) information overload presents a paradox. Increases in information may further stress already limited human and financial capital as staff try to distil the relevant and credible information they need, thus overwhelming management and decision-making processes.
To overcome a lack of effective knowledge exchange, evidence synthesis (e.g. systematic reviews, systematic maps) is frequently endorsed as a logical solution to deliver relevant, accessible and timely information to encumbered environmental decision-makers (see Cook et al., 2017;Dicks et al., 2014;Pullin et al., 2016;Pullin & Knight, 2001). We recognize that evidence synthesis alone is likely not enough to improve the use of knowledge, and that more is required (e.g. knowledge brokers; Segan et al., 2011) to develop the knowledge mediation sphere (Nguyen et al., 2017). Yet, evidence synthesis is a tangible step to amplify and foster multiple forms and sources of knowledge, as well as strengthen partnerships between knowledge producers and decision-makers.
In theory, a benefit of evidence synthesis is that it can draw upon diverse knowledge sources and disciplines in a cohesive manner to comprehensively inform issues on a given matter. However, evidence synthesis has traditionally focused on knowledge from western-based, especially natural sciences (Wheeler & Root-Bernstein, 2020 sists. This will involve moving away from knowledge integration and knowledge co-production to a knowledge coevolution framework (Chapman & Schott, 2000). In such a framework, distinct knowledge systems are bridged and strengthened to generate new understandings while considering the normative impacts of western science and empowering local knowledge holders.
Challenges related to assessing the reliability of ILK from western decision-makers suggest a deep tension about ILK use in western frameworks. This might however be expected for non-Indigenous people given Indigenous knowledge is outside of their own knowledge system. The prevailing perception is that ILK needs to be validated or verified by western scientific knowledge to be useful (Needham et al., 2020;Wheeler & Root-Bernstein, 2020).
In addition to extracting ILK without the full involvement, collaboration, and consent of Indigenous and local knowledge holders, the desire to validate ILK furthers a lack of trust. It may also lead to the marginalization, appropriation and commodification of knowledge.
As revealed here and elsewhere (see Ainsworth et al., 2020;Ban et al., 2018;Huntington, 2000;Reed et al., 2013 employed, open-ended questions in interviews also presents limitations. The interviewer, consciously or otherwise, may influence the direction of interviewee responses through underlying personal biases or preconceptions. However, the benefits in enabling us to collect sensitive data from a wide range of practitioners and providing flexible space for interviewees to explain their positions, priorities and opinions freely and precisely was why this method was chosen over, a survey questionnaire, for example (see Young et al., 2018).

| Conclusions
Wildlife management decisions are highly meaningful, supporting the conservation of biodiversity, habitats and ecosystem services but they are extremely complex. Evidence is an important source for informing decisions under such extreme social-ecological complexity. Our results suggest that gaps between generated knowledge and knowledge users (Bertuol-Garcia et al., 2018;Toomey et al., 2017) may not be as pervasive or expansive as described in some contexts. In our case, Indigenous governments, parliamentary governments and stakeholders use multiple forms of knowledge in decision-making but rely heavily on internal (institutional) knowledge. However, despite agreement that local knowledge, and especially Indigenous knowledge, can yield significant benefits for wildlife management and conservation, it is generally under-utilized in comparison to western scientific knowledge, or personal and institutional experience or opinion.
Concerningly, underlying the use of knowledge is a perception of the diminishing role of evidence in decisions concerning wildlife management and conservation. Interview respondents associated this move away from evidence-informed decision-making with decreases in institutional resources and capacity, but especially with increases in socio-economic and political interference which outweigh evidence.
This research generates further questions. We have assessed how wildlife managers evaluate knowledge, but how they procure it in organizational cultures with capacity shortages and information overload is also important. Whether potential knowledge users perceive claims as more knowledge-based or more advocacy-based and the factors which predict this outcome would benefit evidencebased management and conservation. It would be important in any follow-up work to distinguish how different types of knowledge might be more or less helpful in answering questions that mix empirical data and values (e.g. what is the sustainable level of fish harvest for this lake?), what that information would be, and how it would be used. This was a particular gap observed in this work. Furthermore, empirical investigations of co-assessing knowledge and applying the 'Two-Eyed Seeing' approach are needed to assess their effectiveness and limitations in wildlife management contexts (Reid et al., 2020;.
For wildlife management to be truly adaptive and effective, drawing on the full complement of evidence to develop a holistic and collective understanding of the natural world seems desirable.
Thus, more is needed to improve the use of evidence. Particularly, we emphasize the need for knowledge brokers; standards, guidelines and practices for ILK generation and synthesis developed by knowledge holders; and collaborations and partnerships between and within western science, Indigenous and local communities which embrace knowledge coevolution (Chapman & Schott, 2000). We encourage transformative changes in wildlife management towards direct involvement of knowledge holders, co-assessment of knowledge and transparency in how (multiple forms of) evidence contribute to decision-making. These changes also pertain to organizational cultures so that wildlife managers are motivated and enabled to apply multiple forms of knowing to advance decisions that yield co-beneficial management and conservation outcomes for both people and nature. We believe this can help overcome a lack of trust, hesitancy to share knowledge, difficulties in assessing reliability and difficulties discerning knowledge from advocacy. Urness, Carmen Gudino and Danny Glassman who provided support in transcribing data. We thank the Editor, Associate Editor and two anonymous reviewers whose comments on previous versions of this manuscript greatly improved this article.

CO N FLI C T O F I NTE R E S T
We declare that we have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

AUTH O R S ' CO NTR I B UTI O N S
A.N.K., S.J.C. and N.Y. conceived the ideas and designed the methodology; A.N.K. collected and analysed the data; A.N.K. led the writing of the manuscript. All authors contributed critically to the drafts and gave final approval for publication.

DATA AVA I L A B I L I T Y S TAT E M E N T
There are no linked research datasets for this submission because the data that have been used were collected from human participants in confidentiality and anonymity under our certificate of ethics approval: University of Ottawa Research Ethics Board (File Number: 02-18-08). Archiving the data openly may compromise the confidentiality and anonymity of human participants. However, raw counts and references made by respondent groups are openly available in Appendix A and the thematic codebook in the Supporting Information.