Research

INTRODUCTION

Local ecological knowledge (LEK) consists of intimate understandings of local ecosystems and environmental relationships that develop over extended periods of time living in, working with, and observing a region’s environment (definition adapted from Charnley et al. 2007). Indigenous communities globally have tremendous knowledge of their local ecologies derived from direct management, local observations, and intergenerational transmission of knowledge since time immemorial (Berkes 2009, Leenhardt et al. 2017, Brondizio et al. 2021). Other forms of local ecological knowledge arise on shorter time scales from living in and working closely with the land for decades or generations, such as with third and fourth generation farming or fishing families (Mugerwa et al. 2011, Pauli et al. 2016, Martins et al. 2018). In contrast to the fine scale nature of LEK, Earth system models (ESMs) combine current scientific understandings of relationships between Earth’s various physical processes with large-scale observational input derived from tools such as satellite imagery to generate simulations of past, present, and future states of Earth’s atmospheric, terrestrial, and marine processes, usually at global and regional scales. These models are at the heart of climate science research and are utilized by organizations like the Intergovernmental Panel on Climate Change, national meteorological services, and climate adaptation practitioners to help define the scope of global environmental change and identify adaptive actions (Touzé-Peiffer et al. 2020, Huntingford et al. 2023).

ESMs are developed to represent large scale processes and use parameterizations and parameters that are optimized for long space and time scales. Because of this, ESMs generally fail to represent specific local scale dimensions of physical processes and may generate considerable uncertainty, especially at short time intervals (Huntingford et al. 2023). Further, global scale ESMs rarely include the necessary details of biogeochemical and hydrological processes relevant to local scale decision making (McDermid et al. 2017). In contrast, LEK globally represents an immense quantity of intimate landscape and microclimate knowledge (Joa et al. 2018, Martins et al. 2018, Stori et al. 2019). ESMs and LEK, therefore, produce complementary types of knowledge about Earth’s systems and environmental change. Yet LEK has remained undervalued within scientific work, limiting our ability to develop deeper understandings of Earth’s changing environmental processes (Reyes-García et al. 2020). Furthermore, ESMs have been developed apart from the input and perspectives of local communities who could benefit from ESM-generated data, thus hampering usability (Done et al. 2021).

Given the potential benefits of connecting LEK and ESMs, we analyze the opportunities and challenges presented by three different methodological approaches employed by the authors for combining LEK and ESMs in research to generate deeper understandings of environmental processes at multiple scales and produce data outputs that are more relevant and useful for local communities.

LITERATURE REVIEW

Efforts to integrate knowledge generated by Western scientific inquiry with knowledge held by local communities have been increasing since the 1990s when feminist critiques of science effectively argued that scientific knowledge is not superior to, but rather just one of, many forms of situated knowledge (Turnbull 1997). In the intervening decades, a rich and extensive body of scholarship has developed on the potentials, challenges, and best practices for integrating multiple forms of knowledge to achieve better environmental decision making (Huntington 2000, Armitage et al. 2011, Bohensky and Maru 2011, Bohensky et al. 2013, Brunet et al. 2014, Alessa et al. 2016, Behe and Daniel 2018). Environmental research that engages stakeholders early in the process and integrates LEK through participatory modeling activities has been found to improve the usefulness of models for decision making and to improve collaborative understanding and consensus (Voinov and Gaddis 2008, Laniak et al. 2013, Sterling et al. 2019, Eitzel et al. 2021). In addition to improvements in scientific outputs from engaging with Indigenous and local community knowledge, integrating LEK and scientific inquiry can also benefit Indigenous and non-Indigenous local groups by recognizing their management practices and facilitating the flow of resources to these groups (Barber and Jackson 2015).

Alongside these positive aspects of knowledge integration, however, scholars have also demonstrated that different types of knowledge are at times incommensurable, that is, difficult to compare or combine, and that integration is not always beneficial for local communities. For example, drawing from their work with Indigenous Australian populations, Barber and Jackson (2015) argue that despite the positive dimensions of knowledge integration, environmental models are generally so far abstracted from the realities and needs of local communities that they are of very little value to communities, while the participatory process simultaneously requires much of their time and knowledge. Further, and perhaps of even greater importance, when multiple pieces of evidence contradict one another, for example when evidence generated from local knowledge contradicts that which scientific models generate, community knowledge is often expected to conform to, or fit within, scientific knowledge. Wheeler et al. (2020) find that enduring colonial and assimilation policies in the Arctic region prevent a genuinely equitable integration of the two forms of knowledge and instead lead to Indigenous knowledge being forced into science models or being problematically “verified by” science. “Integration” and “incorporation” of Indigenous knowledge into Western science are too often euphemisms for assimilation (Reid et al. 2021).

Indigenous communities have long challenged this assimilating approach to knowledge integration. Instead, many highlight the importance of multiple types of knowledge being held in relation to one another, valuing and utilizing multiple perspectives without trying to mesh them, for example through the frameworks of Two-Eyed Seeing (Bartlett et al. 2012, Mantyka-Pringle et al. 2017, Reid et al. 2021), Plural Coexistence (Howitt and Suchet-Person 2006, Zanotti and Palomino-Schalscha 2016), Two Ways (Muller 2012), and Double-Canoe (Maxwell et al. 2019). Similarly, Tengö et al. (2014) propose a multiple evidence based (MEB) approach to combining complementary forms of knowledge, including scientific, practitioner, Indigenous, and other forms of local knowledge to produce deeper understandings of environmental problems and better decision making. A key tenet of the MEB approach is that each form of knowledge is considered equally important and is evaluated by its own measurements, rather than by the measurements used for evaluating other forms of knowledge. Rather than integrating LEK into scientific knowledge, with an MEB approach, LEK, scientific knowledge, and other forms of complementary knowledge are each brought to bear on a situation, and a decision is made considering each piece of, perhaps contradictory, evidence to have unique value (see also Tengö et al. 2017, 2021).

Bearing in mind the importance of avoiding the assimilating tendencies of data integration, we build on recent work that shows how carefully engaging with holders of LEK early and throughout a study improves the usability of outcomes (Eddy et al. 2017, Bélisle et al. 2018, Stori et al. 2019) and facilitates community investment in and ownership over data products (Eitzel et al. 2021). It has been well documented that the participatory methods needed to engage LEK holders require time, trust building, and creative methods such as storytelling, participatory mapping, and other community-led activities (Ablah et al. 2016, Robinson et al. 2016, Teufel-Shone et al. 2019, Liguori et al. 2021). Our three research cases present different contexts and methodological approaches to combining LEK and ESMs, and therefore provide important insights into how different approaches can produce different outcomes.

METHODS

The three separate research projects examined in this article each use a mix of qualitative and participatory methods to bring together LEK and ESMs. Each project was conducted by several of the authors of this article in interdisciplinary research teams composed of university social scientists, scientists from the National Center for Atmospheric Research (NCAR), and community partners. These research teams were supported by NCAR’s Early Career Faculty Innovators Program, which promotes convergence research between physical and social scientists and local communities. The Innovators Program facilitated opportunities for the co-authors to meet regularly to compare the methods and outcomes of our respective research projects, which allowed us to see important patterns emerging over time. We applied a grounded theory approach to our comparison, sharing specifics about each case through writing and monthly meetings over a period of nine months, and contrasting the methods and outcomes across our projects to allow inductive findings to emerge (Glazer and Strauss 2017).

In the next three sections of this paper, we detail the context, methods, and outcomes specific to each of our three research cases. The first case examines wildfire evacuee knowledge and behavior during the East Troublesome Fire in 2020 in Colorado, USA using field interviews and elicited responses to a set of 3D data visualizations from the Weather Research and Forecasting (WRF) model (Authors Edgeley, Cannon, Pearse). The second case uses co-generated field maps and elicited responses from a set of time series charts and maps from the Community Earth System Model (CESM) to analyze farmer landscape knowledge and climate adaptive decision making in the Willamette Valley, Oregon, USA (Authors Emard, Cameron, and Lombardozzi). The third case employs focus groups, iterative conversations, and analysis of historical photographs with the Karuk Tribe and its partners to contribute LEK regarding regional hydrology and fire behavior in Karuk Aboriginal Territory, California and Oregon, USA. This local and Karuk knowledge guided the development of a downscaled Community Terrestrial Systems Model (CTSM) for the Klamath Basin region (Authors Wölfle Hazard, Sarna-Wojcicki, Hillman, McCovey, Lombardozzi, and Newman). We chose to compare these three research cases because each project had similar goals of combining LEK and ESMs to improve research, usability, and local resources, yet they employed distinct approaches to accomplish these goals, allowing a unique comparison to emerge.

CASE 1: MODEL VISUALIZATIONS AS A TOOL FOR EXPLORING LEK ABOUT WILDFIRE BEHAVIOR

Context

The East Troublesome Fire burned 78,433 hectares across Grand County, Colorado in 2020, causing widespread evacuations of Grand Lake and adjacent communities. At least 366 homes were destroyed, and two fatalities were reported (Meldrum et al. 2022). Drought conditions, fuel build up, and high wind speeds contributed to extreme fire behavior on 21–22 October 21, pushing the fire 35,245 hectares over a 24-hour period to become the second largest fire in state history at the time (InciWeb 2020). This study was designed to help scientists who model and visualize fires create products that were more accessible to the public. For additional details regarding this methodology, please see Edgeley et al. (2024). Understanding how residents used their own ecological knowledge to interpret fire behavior and made evacuation decisions under these conditions can help inform more streamlined community evacuation and associated planning, tailor emergency communications to local perceptions, and build stronger relationships between citizens and emergency professionals among other benefits (McCaffrey et al. 2018, Edgeley and Paveglio 2019).

Approach

We conducted 36 interviews with 51 participants in August and September of 2022 to explore how local weather and fire behavior, as well as local knowledge about these conditions, influenced decisions to evacuate. Interviewees included residents from communities directly affected by evacuations, local professionals engaged in evacuation, fire suppression, and recovery efforts related to the East Troublesome Fire, and fire weather and behavior scientists. Individuals not engaged in or affected by this fire were not recruited. Potential participants were identified via a combination of theoretical and snowball sampling (Biernacki and Waldorf 1981, Glaser and Strauss 2017) with the intent to develop a representative understanding of local populations and their experiences. Interviewees were invited to recall as much as they could about the fire’s behavior and local conditions that might have influenced it. Once participant recall was exhausted, the interviewers shared model visualizations on an iPad to elicit any additional information associated with the East Troublesome Fire’s behavior. The feedback provided by participants was used to improve the next iteration of model visualizations for a case study of a second fire, generating participatory input that benefitted subsequent participants.

Model visualizations drew on data recorded during the fire event in an effort to mimic the fire’s behavior and progression as closely as possible, although there is an assumed level of uncertainty at smaller scales within the model. Model data was sourced from WRF-FIRE, a module of the Weather Research and Forecasting (WRF) model that simulates wildfire behavior and its interactions with localized environmental and atmospheric conditions (Coen et al. 2013, DeCastro et al. 2022). WRF-FIRE uses varied data resolutions when examining a singular wildfire event, ranging from a 1x1 km grid cell size where the fire was ignited up to coarser 13x13 km grid cell size over the entire fire region. Common uses among fire managers and scientists include exploration of fire behavior and examination of how weather or atmospheric characteristics interact with fire. The East Troublesome Fire simulation incorporated various fire behavior, including spotting and consideration of fuels (for more detail see Frediani et al. 2021 and DeCastro et al. 2022). These data were then imported into VAPOR (Visualization and Analysis Platform for Ocean, Atmosphere, and Solar Researchers), a platform for developing 3D visualizations to turn static modeled data into lifelike video representations (Li et al. 2019). Figure 1 is a still image from the model visualizations for the East Troublesome Fire, which represented fire progression, smoke development, and wind speed and direction from a WRF-FIRE simulation for 21–22 October, the peak period of extreme fire activity during which most study participants were considering whether to evacuate. Visualizations were generated from multiple angles so that participants could select a viewpoint that most closely aligned with where they were situated during the fire.

Outcomes

In this first case, model visualizations proved an invaluable tool for making connections between diverse lived experiences of a fire event and LEK. For example, many interviewees described how wind speed and direction during the East Troublesome Fire captured by model visualizations varied from both daily and seasonal patterns they typically observed in the area. After looking at the visualizations, interviewees were able to identify specific times, locations, and thought processes associated with personal experiences of these sudden shifts and how they prompted reevaluation of personal safety to trigger an evacuation decision. Visualizations also helped interviewees situate their own place-specific experience with the fire within the broader context of the overall fire event. Some were drawn to the model’s depictions of rapid shifts in the fire’s boundaries and how that might be explained based on the interviewee’s familiarity with local topography. Use of visualizations as a cognitive interview tool also invited conversations about systemic factors driving modeled fire behavior, including climate change, bark beetle infestations and relationships to forest health, and observed drought in the months leading up to the East Troublesome Fire, which had raised concerns about risk ahead of this event. Use of visualizations advanced interview depth and scope, allowing more nuanced and contextualized place-based knowledge to emerge, while also providing study participants the time and information to make more personal connections between their own experience and broader environmental phenomenon.

CASE 2: PARTICIPATORY MAPPING AND MODELED DATA TO EXPLORE FARMER’S LANDSCAPE KNOWLEDGE

Context

Farmers are among those who feel the impacts of climate change most directly as warming temperatures and extreme weather events disrupt crop production and threaten farmers’ livelihoods (Altieri and Nicholls 2017). Simultaneously, emerging research shows that farmers can not only adapt and build resilience to these changes but may also contribute to climate change mitigation through practices that sequester carbon in the soil (Altieri and Nicholls 2017, Bayu 2020, Gosnell et al. 2020). Oregon’s Willamette Valley is one area where farmers are coping with warmer, less predictable conditions as the region has experienced warming of 0.5 °C per decade since 1980 with a concurrent decline in snowpack and streamflow (Dalton and Fleishman 2021). The valley is a highly productive agricultural region with over 600,000 hectares planted in 170 different crops representing over 2.3 billion U.S. dollars in market value (ODA 2021).

Approach

Between April and August 2022, we conducted 25 semi-structured interviews with 31 Willamette Valley farmers to understand how farmers pair LEK and climate data in making climate adaptive farming decisions. Thirteen interviewees were multi-generation farmers whose families had been farming the same land for at least 50 years, the other 18 were first-generation farmers. Interviewees were selected through purposive and snowball sampling (Robinson 2014), and the sample was stratified by farm size, farmer approach (e.g., industrial, organic, regenerative), and social demographic. Anyone over 18 years of age who held a decision-making role on a farm producing crops in the Willamette Valley was eligible to participate. We publicized the research opportunity through farmer listservs maintained by Oregon State University Extension and Soil Water and Conservation Districts. Twelve farmers contacted us to participate after hearing about the project in this way. We actively sought an additional 19 participants through cold calling, recruiting at farmer events, and snowball sampling to diversify our sample. We described the project, time required, potential outcomes, and methods for protecting data privacy before asking if farmers would like to participate. For additional details on the methodology of this study, please see Emard et al (2024).

To better understand how farmers’ landscape knowledge shapes their adaptive decision making, we blended two methodological approaches: participatory field mapping and discussions prompted by climate model simulations for the Willamette Valley. Each interview was 1–2 hours long and conducted on the farm. We first walked the farm with the farmer and jointly created a hand drawn map over satellite imagery using the free software Draw Maps on an iPad. With the farmer, we recorded locations and reasons for planting, irrigation, and soil amendment decisions. We also overlaid images taken during the farm walk (see Fig. 2 for an example). This method captured details about how farmers’ ecological knowledge shapes their farm decision making.

After cogenerating the farm maps, we shared time series data generated from the Community Earth System Model, version 2, Large Ensemble (CESM2-LE) indicating past, current, and future climate trends for the Willamette Valley with the interviewees (see Fig. 3 for an example). The CESM is a global scale, coupled atmosphere-ocean-sea-ice-land model that simulates the Earth’s physical processes over hundreds of years, in our case using a one-degree grid cell size. Given the shape of the Earth, the number of kilometers in a degree varies between the equator and poles, but in the Willamette Valley, a one-degree cell is approximately 110 km x 80 km. Because of its coarse spatial resolution and long climate timescales, this model is not frequently used by individuals for everyday decision making. However, it is relevant to agriculturalists because the Community Land Model portion of the coupled CESM2-LE incorporates significant detail on crop types, the nitrogen cycle, and farmer practices when compared to similar models (McDermid et al. 2017, Kawamiya et al. 2020). Therefore, the potential for integrative knowledge generation between CESM and farmers’ ecological knowledge may be higher than with other ESMs.

The specific CESM data used in this project were generated through a four-month, iterative process with the research team that involved producing sample data from CESM2-LE, discussing the data’s relevance to farmers with input from Oregon State University (OSU) extension service coordinators and personal farmer contacts in the Willamette Valley, and making necessary changes. The goal of this iterative process was to generate data that related to the experiences and concerns of farmers in the region. The final charts were shared with farmers during interviews and used as the basis for a discussion of climate trends in the region. We asked about how well the data reflected the farmers’ experiences, the extent to which the data would be likely to influence farmers’ decisions, and how the data could be made more useful. We discussed uncertainty inherent in ESM outputs and how uncertainty was represented in the data through shaded areas that represented a wide range of possible projected futures generated by the model. This study was instigated by the researchers and aimed to help modelers understand how climate model outputs could be made more relevant and usable for farmers (see Emard et al. 2024 for more details). Although the instigation for knowledge integration came from the researchers’ side, Willamette Valley farmers and OSU extension service were consulted at all stages of the research, from the selection of ESM sample data to participatory mapping to recommendations for future ESM development.

Outcomes

During mapping, farmers showed that they have deep intimate knowledge of their farms’ landscapes and microclimates. In response to the modeled data, farmers indicated that their landscape knowledge allows them to adapt the larger scale data generated by the CESM2-LE to the specific ways it may impact their farms. To assist with overcoming the scale mismatch between data from ESMs and LEK, farmers desired the ability to interact with the data via an app or website and tailor projections to their farm’s specific context by inputting their own local scale farm conditions, such as soil types, vegetation covers, slope, and aspect. Relatedly, many interviewees expressed a need for their practices, including residue management, cover cropping, and tillage levels, to be reflected in the models so that they could see to what extent those practices shape future projections of soil moisture, organic matter, and nitrogen levels, which are factors that drive their decision making. These findings are of interest to scientists who are deciding the next steps of ESM development, and the approach used in the study may be employed to support the implementation of a collaborative approach to ESM development

CASE 3: INTEGRATING KARUK AND LOCAL KNOWLEDGE OF LAND COVER AND FIRE BEHAVIOR WITH LAND SURFACE MODEL REPRESENTATIONS

Context

The Karuk Tribe, whose ancestral territory is in the middle stretch of the Klamath River Basin in Northern California and Southern Oregon, are working to repair lands and water bodies degraded by settler-colonial resource extraction and threatened by climate change stressors. Karuk people have long used cultural fire to promote cultural use species and habitats, manage streamflow, and reduce late-summer stream temperature. Today, the Karuk Tribe is returning cultural fire regimes to the landscape to restore ecosystems, revitalize cultural use species, and buffer the impacts of climate change-induced drought, flooding, pest infestations, and catastrophic/high intensity fire. Compared to the current fire suppression era, fire return intervals were shorter under Karuk pre-colonial management, ranging from 1 to 3 years near villages to 15 to 20 years in upslope areas managed for hunting. This management created a mosaic of vegetation types that reduced wildfire intensity and area (Knight et al. 2022, Greenler et al. 2024). However, few ESMs have explicitly evaluated the effects of restoring “good fire” (cultural fire and prescribed fire) at the landscape scale (David et al. 2018). Such management approaches can potentially benefit riverine species, and also mitigate the impacts of increased wildfire expected in western North America as climate change intensifies (David et al. 2018, Grantham et al. 2018). In upland forests where fire suppression and timber industry practices have increased forest density, re-introducing good fire is increasingly recognized as crucial to reducing catastrophic wildfire, and also may increase dry season baseflows (Bales et al. 2011, David et al. 2018, Karuk Tribe 2019, Long et al. 2020, Ma et al. 2020).

Approach

Our research team, co-led by Karuk ceremonial leaders and cultural practitioners (Authors Hillman and McCovey), set out to create “good fire” scenarios as inputs to an NCAR land model, the Community Terrestrial Systems Model (CTSM), within a cultural context group made up of Native and local natural resource managers in the Mid-Klamath Basin.^[1] The participatory modeling approach developed over eight virtual meetings of the university and NCAR teams, augmented by two in-person meetings of the Karuk and University of Washington teams. Data sovereignty and reciprocity were negotiated through a Tribal process. In June 2022, the research team convened two workshops with Karuk resource managers and local government and NGO representatives. The workshop goals were to present model examples to local experts, to see whether they wanted us to proceed with modeling, and to elicit local and Indigenous knowledge of land cover change, both in the past and under a future “maximum prescribed fire” scenario.

For this collaboration, we used CTSM, an overarching framework for land-modeling at NCAR, of which a specific instantiation of CTSM, the Community Land Model (CLM) version 5, resides within the CESM2. CTSM is the result of decades of land-model development at NCAR and elsewhere (Bonan 1996, Lawrence et al. 2020) and represents much of the complex set of land-atmosphere exchanges, soil, ecological, hydrological, cryosphere, and biogeochemistry processes found in the Earth’s natural processes, as well as some aspects of human engineered systems such as urban, irrigation, and simple reservoir parameterizations for Earth’s land areas (e.g. Lawrence et al. 2019, 2020). Because CTSM/CLM is developed primarily for use on climate timescales and within the CESM modeling system, it is typically run at coarse resolutions as described in Case 2 and needs in-depth configuration for community-centric applications. Given this, the NCAR team downscaled the model to 1 km grid cells and clipped to the boundary of the Klamath Basin.

The participatory workshops were centered in Karuk cultural protocols for knowledge sharing, including an opening by ceremonial leader Leaf Hillman and Tribal oversight and data sovereignty procedures communicated to participants. The workshops focused on how local knowledge and priorities can inform model scenario generation. To generate past, present, and future land cover and hydrology datasets, we convened four breakout groups focused on (1) identifying landscape patterns and features in a 1944 aerial photograph; (2) distilling local knowledge of important plant species into plant functional types (PFTs); (3) developing a timeline of floods and drought periods to ground truth the hydrological model; and (4) identifying high-elevation meadows that could be restored to store snowmelt and release cool water downstream (see Figs. 4 and 5 for more detail). The future land cover scenario uses the Western Klamath Restoration Partnership’s vision to scale-up forest thinning and prescribed and cultural fire throughout the western Klamath to test how streamflow responds. The workshop included discussion of the overall utility of the model to Tribal and local resource management and climate adaptation initiatives.

Outcomes

Three key observations that arose from discussions at the workshops demonstrate how LEK can improve ESMs’ representations of local conditions relevant to communities and allow for more nuanced study of fire impacts using ESMs. First, although several participants said that a model that integrated climate, fire, and hydrology would be helpful in planning restoration, they felt that the representation of fire would need to change to account for local knowledge of fire behavior. The workshop participants had extensive knowledge about how fire had historically burned across the landscape, as well as fuel loading conditions, topography and wind patterns likely to drive fire spread. Because the CTSM representation was developed for models at much coarser spatial scales, and treats each grid cell’s fire occurrence probabilistically, CTSM could not represent fire intensity or spread, which could be added into the model using LEK instead. Second, many participants thought that the current CTSM fire model would not capture key aspects of fire ecology, such as the different effects of low and moderate intensity fires on culturally important plants, which could be contributed through LEK, and which would improve the model’s representation of ecological conditions and recovery after a fire. Third, the way the model represents plants, as “plant functional types” (PFTs), could not capture the devastating effects of plantation forestry and Douglas fir (Pseudotsuga menziesii) encroachment driven by fire suppression to protect tree crops, as it lumped Douglas fir plantations together with other culturally significant and fire-adapted species such as Sugar Pine (Pinus lambertiana) into the Needleleaf Evergreen Tree PFT. Participants wanted to adapt the PFT classifications to distinguish between invasive/noxious species and culturally significant hardwood, shrub, and herbaceous species used for food, fiber, medicine, and regalia. For example, cultural practitioner Kathy McCovey argued that Douglas fir is “an agricultural commodity-tree farm. Fir plantations need to be called out as an invasive/noxious species because of their impact on the regional landscape ecology and hydrology.” These insights were shared with NCAR and university researchers modeling fire behavior in CTSM.

DISCUSSION

The three case studies represent different contexts and approaches for engaging complimentary forms of knowledge in environmental modeling research. Cases varied considerably in duration and level of local input, ranging from iterative discussions about model design over multiple years with Karuk partners to one-off reflective reviews of pre-determined models after the East Troublesome Fire. Given these differences, comparing the three cases produces important observations that contribute to the scholarship on combining ESMs and LEK.

First, given the high time and resource commitments that knowledge integration necessitates from all contributing parties, as well as the assimilating tendencies of Western science, prior work has questioned the extent to which integration is beneficial as opposed to engaging multiple distinct types of knowledge separately when making decisions (Barber and Jackson 2015, Wheeler et a. 2020, Tengö et al. 2021). Our cases indicate that although integrative research can be complex, it produces notable benefits, namely, the generation of richer qualitative data, more complex accounts of environmental processes, and improved ESM capabilities (see Table 1). In all three of our cases, combining LEK and ESM during the research process rather than at the data analysis stage led to improved understandings and more usable outcomes for communities. This was especially visible in the Karuk context, where the collaboration arose from long-time, community engaged work between the university research team and the Karuk Department of Natural Resources and drew on insights and data generated from prior collaborations. This highlights the importance of supporting long-term community-university collaborations that span multiple grants and projects when data integration is a goal.

Second, a unique contribution of this article to the literature on integrating ESMs and LEK is the ability to compare and contrast varying approaches to integration. Although studies have indicated that engagement should occur early and continually through the process (Eddy et al. 2017, Eitzel et al. 2021, Bélisle et al. 2018, Stori et al. 2019), our comparison of three cases allows us to offer a more nuanced examination of the ways that timing and type of engagement shapes outcomes. We found that sharing ESM data during qualitative data collection, as was done in the East Troublesome and Willamette Valley cases, generates deeper and more nuanced understandings of LEK by the scientific community and of scientific data by the community (see Table 1, row 1). We also found that, in all three cases regardless of the timing and method of participatory approach used, LEK effectively identifies gaps and errors in ESM-generated data (see Table 1, row 2). Additionally, our cases show that LEK can inform the design and development of ESMs themselves, although we note important distinctions between the three cases in this area (see Table 1, row 3). When modeled data is produced prior to participant engagement, like in the East Troublesome case, participant suggestions for model improvement were focused on adding relevant LEK that could be provided alongside the data output, rather than ideas for changing the design of the model itself. In contrast, in the case with the Karuk Tribe, the local community was involved in selecting parameters, parametrizations, and other elements before any ESM data was generated. The early and extensive involvement, over multiple meetings, of the Karuk people resulted in recommendations for fundamental changes to how the ESM operates, including the incorporation of new plant functional types and fire fuel loads that are not currently represented in the ESM. This outcome was quite distinct from that of the East Troublesome Fire case. The case with Willamette Valley farmers represents a “middle-of-the-road” approach. Farmers were given ESM data already generated by the research team, but they were asked to make requests and recommendations for future iterations and were given the opportunity to see and comment on those iterations. In this case, we found that farmers did provide ideas for changing the ESM design, although not to the extent that was generated by the Karuk case.

Third, the comparison across our three cases contributes nuanced understandings of the challenges that accompany work that combines LEK and ESMs. As outlined in Table 2, our cases confirm several challenges discussed in the literature including the importance of long periods of time to build collaborations that facilitate knowledge integration (Cornell et al. 2013, Done et al. 2021), mismatches in the temporal and spatial scale at which LEK and ESMs function (Obermeister 2019, Wheeler et al. 2020), the need to protect data sovereignty and privacy and avoid exploitative data collection methods (Latulippe and Klenk 2020), and the issues of epistemic authority when understandings of what knowledge is and counts for differ (Barbour and Jackson 2015, Tengö et al, 2014, 2017, 2021, Wheeler 2020, Reid et al. 2021). Although these challenges were significant across all three approaches, we noticed that the challenges became more pronounced as we moved toward a more deeply integrated process of data generation. That is, when community members were involved from the beginning of the project in shaping and developing the ESM configuration and its outputs, as they were in the Karuk case, the time required for the projects increased. Further, when we asked participants to contribute to developing the model rather than simply interpreting the modeled data, the scalar mismatches felt even more pronounced. For example, although farmers in the Willamette Valley case found the grid spacing to be coarse, they were generally comfortable mentally adjusting the data to the specifics of their farm’s microclimate (Emard et al. 2024), whereas Karuk participants emphasized the difficulty of translating their local knowledge about sites and stand or patch dynamics to the resolution of the model for informing model design. As our cases moved toward deeper integration, as with the Karuk Tribe, the question of data sovereignty also became more pertinent given the deeper level of LEK being contributed to the research than in the first two cases. Discussions about data sovereignty were initiated early and covered by Tribal research oversight protocols, and in this case, Karuk partners are co-authors on this and other research outputs. Although the deeply integrated nature of participation in the Karuk case was one contributing factor to these challenges being more pronounced in this case, we also note other important factors differentiating the Karuk case, including that it is the only one of the three that is a collaboration with Indigenous Peoples and the only one to build from a longer collaboration. Further, it was the only one of our cases in which the local community initiated the knowledge exchange and where the researchers and local community had already been working together for years. This was a key difference between the cases, and the fact that the Karuk community initiated and desired to lead knowledge exchange workshops allowed for deeper integration and overcoming the above challenges in ways that the East Troublesome and Willamette Valley cases did not achieve.

A final important contribution to the literature is that in each of our cases, the research teams encountered epistemic tensions among different actors involved in the projects, as has been well documented by other scholars working at the intersection of LEK and modeled data (e.g., Wheeler et al. 2020, Tengö et al. 2021). We noted that epistemic tensions were not solely between scientists and community members but also among various members of communities themselves and even between different academic collaborators. For example, in the Willamette Valley case, farmers held a range of sometimes conflicting beliefs about climate and how it could be known and represented, while in the Karuk case, local fire managers, hydrologists, NGO staff, and cultural practitioners held a range of perspectives on how to know and study the environment. Rather than attempting to reconcile or integrate different ways of knowing among collaborators, our teams explored these epistemic tensions without trying to resolve them, following principles of Two-Eyed Seeing, Double-Canoe, Ethical Space, and related approaches (Howitt and Suchet-Person 2006, Bartlett et al. 2012, Muller 2012, Zanotti and Palomino-Schalscha 2016, Mantyka-Pringle et al. 2017, Maxwell et al. 2019, Reid et al. 2021, Nikolakis and Hotte 2022). For example, rather than trying to reach precise agreement on climate change and its impacts in the Willamette Valley case, we collected various farmers’ perspectives on how the data mirrored or differed from their observations and carried that information to modelers for them to incorporate into evaluation of the model. Our goal was to better understand limitations of the modeled data from local perspectives and share recommendations for modelers to account for local dynamics in future model iterations.

We are cognizant of the risk of reifying colonial and exploitative knowledge structures in any attempt to engage LEK from the position of a colonial scientific institution and realize that all best practices may fall short, though they are crucial to employ regardless (de Leeuw et al. 2012). There may be times when ESMs simply cannot serve the needs of local communities. Nevertheless, the findings from our cases support those of Eitzel et al. (2021) who found that even though colonial power dynamics have not been completely overcome, integration holds significant potential for producing knowledge that is relevant to and usable by local communities and which local communities feel ownership over. Carefully employing methods designed to reduce colonial power dynamics and engage ethically, including free, prior, and informed consent; ethical space; and OCAP (ownership, control, access, and position) are critical to collaborative projects (Tauri 2018, Nikolakis and Hotte 2022, Konczi and Bill 2024). Additionally, beginning with the research interests of local communities and continually reshaping the project around their choices, as described in our third case, was critical to achieving results that the community felt were meaningful and valuable.

Future directions for ESM development informed by LEK

Although it was clear from our cases that ESM outputs will not always be relevant to communities, several ideas emerged from our projects for ways to increase ESM data relevance and usability. First, more work needs to be done to match model generated data with specific on-the-ground applications, such as adaptation training, hazard incident response, or post-disaster analysis. Data from ESMs that our participants indicated would be useful to them included fire spread patterns connected to different weather conditions in the East Troublesome case, projected decadal changes in soil moisture in the Willamette Valley case, and the multi-scalar (temporal and spatial) impact of good fire regimes on local hydrology in the Karuk case. But participants highlighted that these data would be more usable if they were placed into decision-making contexts, such as incident response protocols, adaptation plans, or resource trainings. Second, in some cases, new model components will need to be developed. For example, in our cases we found a need for models that depict fire spread and behavior and ecological transition. The Functionally Assembled Terrestrial Ecosystem Simulator (FATES) model emerged in the Klamath case as a potential candidate for future efforts but would need to be evaluated in consultation with community members to see whether it would fit their needs (Fisher et al. 2018, Koven et al. 2020).

Third, from our experiences, it is important to develop models adjacent to social data collection, rather than as two distinctly separate efforts, which may lend itself to more locally relevant model design and outputs. This would mark a departure from current approaches in which data is generated from ESMs to answer questions posed by scientists and then distributed to local community members as a final step (Done et al. 2021). Fourth, we encourage more research into how LEK can be collected at a scale that can be integrated into regional and global scale models. For example, can local ecological observations about precipitation patterns, fire scars, and vegetation cover be collected using photovoice or online surveys? How can we collect participant submissions of geolocated soil moisture tests, soil types, or organic matter content in an accessible way? How can these crowd-sourced data be applied as inputs and parameters in ESMs going forward? And how might this be done while ensuring data sovereignty and internal validity of the local knowledge system for Indigenous and other place-based communities who provide this data? All such future research should be conducted following best practices of free, prior, and informed consent; ethical space; and data sovereignty principles to ensure ethical engagement with Indigenous and local communities (Tauri 2018, Nikolakis and Hotte 2022, Konczi and Bill 2024). Research into these issues will facilitate additional methods for merging ESMs and LEK to better understand Earth’s future climate system and shape informed management decisions.

CONCLUSIONS

This paper provides three examples of methodological approaches that integrate LEK with Earth system modeling outputs. These cases indicate significant opportunities for work that connects LEK and ESMs to generate deeper and more nuanced data about environmental change, reveal knowledge gaps in ESMs, and inform future model development. Further, the participatory approaches employed contribute to knowledge exchange between scientific and local communities that can be used to develop improved policies supporting local community-led adaptation planning. Despite these opportunities, the cases also confirm significant challenges to this type of work including the time requirements to accomplish collaborative work, temporal and spatial scale mismatches, and concerns for data sovereignty and privacy for community members. In addition, the cases raised broader questions about the overall usability of this data for local communities and decision makers, even if the above challenges were to be addressed. By comparing the three cases, we were able to identify how these challenges become more acute as one moves toward deeper integration of types of knowledge compared with holding the knowledge in parallel. The comparison also shows the ways that length of time spent building relationships is key to the level of integration one can achieve while simultaneously respecting epistemic differences and protecting data sovereignty. Even when taking the differences between the cases into account, all three cases demonstrate that bridging ESMs and LEK with participatory approaches can help build deeper and more nuanced data sets, support holders of LEK by providing them with usable data and data tools, and improve ESMs by applying the insights of LEK to the development and configuration of the model itself to produce more representative simulations.

__________

^[1] This collaboration arose from long-time, community engaged work between the University of Washington research team and the Karuk Department of Natural Resources (co-author Sarna-Wojcicki’s 15 years and co-author Wölfle Hazard’s 8 years of working in partnership with the Karuk people). Wölfle Hazard instigated this project based on Karuk cultural practitioner Hillman’s desire to better understand fire and streamflow management implications of managed fire and prescribed fire policy shifts. Wölfle Hazard then co-developed the modeling proposal with Newman of NCAR.

LITERATURE CITED

Ablah, E., J. Brown, B. Carroll, and T. Bronleewe. 2016. A community-based participatory research approach to identifying environmental concerns. Journal of Environmental Health 79(5):14-19.

Alessa, L., A. Kliskey, J. Gamble, M. Fidel, G. Beaujean, and J. Gosz. 2016. The role of Indigenous science and local knowledge in integrated observing systems: moving toward adaptive capacity indices and early warning systems. Sustainability Science 11:91-102. https://doi.org/10.1007/s11625-015-0295-7

Altieri, M. A., and C. I. Nicholls. 2017. The adaptation and mitigation potential of traditional agriculture in a changing climate. Climatic Change 140(1):33-45. https://doi.org/10.1007/s10584-013-0909-y

Armitage, D., F. Berkes, A. Dale, E. Kocho-Schellenberg, and E. Patton. 2011. Co-management and the co-production of knowledge: learning to adapt in Canada’s Arctic. Global Environmental Change 21:995-1004. https://doi.org/10.1016/j.gloenvcha.2011.04.006

Bales, R. C., J. J. Battles, Y. Chen, M. H. Conklin, E. Holst, K. L. O’Hara, P. Saksa, and W. Stewart. 2011. Forests and water in the Sierra Nevada: Sierra Nevada watershed ecosystem enhancement project. Report number 11.1. Sierra Nevada Research Institute, University of California, Merced, USA.

Barber, M., and S. Jackson. 2015. “Knowledge making”: issues in modelling local and indigenous ecological knowledge. Human Ecology 43(1):119-130. https://doi.org/10.1007/s10745-015-9726-4

Bartlett, C., M. Marshall, and A. Marshall. 2012. Two-eyed seeing and other lessons learned within a co-learning journey of bringing together Indigenous and mainstream knowledges and ways of knowing. Journal of Environmental Studies and Sciences 2(4):331-340. https://doi.org/10.1007/s13412-012-0086-8

Bayu, T. 2020. Review on contribution of integrated soil fertility management for climate change mitigation and agricultural sustainability. Cogent Environmental Science 6(1):1823631. https://doi.org/10.1080/23311843.2020.1823631

Behe, C., and R. Daniel. 2018. Indigenous knowledge and the co-production of knowledge process: creating a holistic understanding of Arctic change. Bulletin of the American Meteorological Society 99(8):S160-S161.

Bélisle, A. C., H. Asselin, P. LeBlanc, and S. Gauthier. 2018. Local knowledge in ecological modeling. Ecology and Society 23(2):14. https://doi.org/10.5751/ES-09949-230214

Berkes, F. 2009. Indigenous ways of knowing and the study of environmental change. Journal of the Royal Society of New Zealand 39(4):151-156. https://doi.org/10.1080/03014220909510568

Biernacki, P., and D. Waldorf. 1981. Snowball sampling: problems and techniques of chain referral sampling. Sociological Methods & Research 10(2):141-163. https://doi.org/10.1177/004912418101000205

Bohensky, E. L., J. R. A. Butler, and J. Davies. 2013. Integrating Indigenous ecological knowledge and science in natural resource management: perspectives from Australia. Ecology and Society 18(3):20. https://doi.org/10.5751/ES-05846-180320

Bohensky, E., and Y. Maru. 2011. Indigenous knowledge, science, and resilience: What have we learned from a decade of international literature on “integration”? Ecology and Society 16(4):6. https://doi.org/10.5751/ES-04342-160406

Bonan, G. B. 1996. A land surface model (LSM version 1.0) for ecological, hydrological, and atmospheric studies: technical description and user’s guide. NCAR Technical Note NCAR/TN-417+STR, National Center for Atmospheric Research, Boulder, Colorado, USA.

Brondizio, E. S., Y. Aumeeruddy-Thomas, P. Bates, J. Carino, A. Fernandez-Llamazares, M. F. Ferrari, K. Galvin, V. Reyes-Garcıa, P. McElwee, Z. Molnár, et al. 2021. Locally based, regionally manifested, and globally relevant: Indigenous and local knowledge, values, and practices for nature. Annual Review of Environment and Resources 46(1):481-509. https://doi.org/10.1146/annurev-environ-012220-012127

Brunet, N. D., G. M. Hickey, and M. M. Humphries. 2014. The evolution of local participation and the mode of knowledge production in Arctic research. Ecology and Society 19(2):69. https://doi.org/10.5751/ES-06641-190269

Charnley, S., A. P. Fischer, and E. T. Jones. 2007. Integrating traditional and local ecological knowledge into forest biodiversity conservation in the Pacific Northwest. Forest Ecological Management 246:14-28. https://doi.org/10.1016/j.foreco.2007.03.047

Coen, J. L., M. Cameron, J. Michalakes, E. G. Patton, P. J. Riggan, and K. M. Yedinak. 2013. WRF-Fire: coupled weather-wildland fire modeling with the weather research and forecasting model. Journal of Applied Meteorology and Climatology 52(1):16-38. https://doi.org/10.1175/JAMC-D-12-023.1

Cornell, S., F. Berkhout, W. Tuinstra, J. D. Tàbara, J. Jäger, I. Chabay, B. de Wit, R. Langlais, D. Mills, P. Moll, I. M. Otto, A. Petersen, C. Pohl, and L. van Kerkhoff. 2013. Opening up knowledge systems for better responses to global environmental change. Environmental Science and Policy 28:60-70. https://doi.org/10.1016/j.envsci.2012.11.008

Dalton, M., and E. Fleishman, editors. 2021. Fifth Oregon climate assessment. Oregon Climate Change Research Institute, Oregon State University, Corvallis, Oregon, USA. https://doi.org/10.5399/osu/1160

David, A. T., J. E. Asarian, and F. K. Lake. 2018. Wildfire smoke cools summer river and stream water temperatures. Water Resources Research 54(10):7273-7290. https://doi.org/10.1029/2018WR022964

de Leeuw, S. D., E. S. Cameron, and M. L. Greenwood. 2012. Participatory and community-based research, Indigenous geographies, and the spaces of friendship: a critical engagement. Canadian Geographer 56(2):180-194. https://doi.org/10.1111/j.1541-0064.2012.00434.x

DeCastro, A. L., T. W. Juliano, B. Kosović, H. Ebrahimian, and J. K. Balch. 2022. A computationally efficient method for updating fuel inputs for wildfire behavior models using sentinel imagery and random forest classification. Remote Sensing 14(6):1447. https://doi.org/10.3390/rs14061447

Done, J. M., R. E. Morss, H. Lazrus, E. Towler, M. R. Tye, M. Ge, T. Das, A. Munéver, J. Hewitt, and J. Hoeting. 2021. Toward usable predictive climate information at decadal timescales. One Earth 4(9):1297-1309. https://doi.org/10.1016/j.oneear.2021.08.013

Eddy, I. M., S. E. Gergel, N. C. Coops, G. M. Henebry, J. Levine, H. Zerriffi, and E. Shibkov. 2017. Integrating remote sensing and local ecological knowledge to monitor rangeland dynamics. Ecological Indicators 82:106-116 https://doi.org/10.1016/j.ecolind.2017.06.033

Edgeley, C. M., W. H. Cannon, S. Pearse, B. Kosović, G. Pfister, and R. Kumar. 2024. Five social and ethical considerations for using wildfire visualizations as a communication tool. Fire Ecology 20:45. https://doi.org/10.1186/s42408-024-00278-8

Edgeley, C. M., and T. B. Paveglio. 2019. Exploring influences on intended evacuation behaviors during wildfire: What roles for pre-fire actions and event-based cues? International Journal of Disaster Risk Reduction 37:101182. https://doi.org/10.1016/j.ijdrr.2019.101182

Eitzel, M. V., J. Solera, E. M. Hove, K. B. Wilson, A. M. Ndlovu, D. Ndlovu, A. Changarara, A. Ndlovu, K. Neves, A. Chirindira, O. Omoju, A. C. Fisher, and A. Veski. 2021. Assessing the potential of participatory modeling for decolonial restoration of an agro-pastoral system in rural Zimbabwe. Citizen Science: Theory and Practice 6(1):2. https://doi.org/10.5334/cstp.339

Emard, K., O. Cameron, W. R. Wieder, D. L. Lombardozzi, R. Morss, and N. Sobhani. 2024. Integrating farmers’ perspectives into Earth system model development: interviews with end users in the Willamette Valley, Oregon to guide actionable science. Weather, Climate, and Society 16(3):453-465. https://doi.org/10.1175/WCAS-D-23-0066.1

Fisher, R. A., C. D. Koven, W. R. L. Anderegg, B. O. Christoffersen, M. C. Dietze, C. E. Farrior, J. A. Holm, G. C. Hurtt, R. G. Knox, P. J. Lawrence, J. W. Lichstein, M. Longo, A. M. Matheny, D. Medvigy, H. C. Muller-Landau, T. L. Powell, S. P. Serbin, H. Sato, J. K. Shuman, B. Smith, A. T. Trugman, T. Viskari, H. Verbeeck, E. Weng, C. Xu, X. Xu, T. Zhang, and P. R. Moorcroft. 2018. Vegetation demographics in Earth System Models: a review of progress and priorities. Global Change Biology 24:35-54. https://doi.org/10.1111/gcb.13910

Frediani, M., T. W. Juliano, A. DeCastro, B. Kosović, and J. Knievel. 2021. A fire-spotting parameterization coupled with the WRF-fire model. Earth and Space Science Open Archive 12:1-13. https://doi.org/10.1002/essoar.10506771.1

Glaser, B. G., and A. L. Strauss. 2017. Discovery of grounded theory: strategies for qualitative research. Routledge, New York, New York, USA. https://doi.org/10.4324/9780203793206

Gosnell, H., S. Charnley, and P. Stanley. 2020. Climate change mitigation as a co-benefit of regenerative ranching: insights from Australia and the United States. Interface Focus 10(5):2020027. https://doi.org/10.1098/rsfs.2020.0027

Grantham, T., A. Laird, L. Quinn-Davidson, D. Sarna-Wojcicki, J. Stackhouse, A. Stubblefield, A. Torregrosa, and Y. Valachovic. 2018. North Coast Region Report. California’s Fourth Climate Change Assessment. Publication number: SUM-CCC4A-2018-001. https://www.energy.ca.gov/sites/default/files/2019-11/Reg_Report-SUM-CCCA4-2018-001_NorthCoast_ADA.pdf

Greenler, S. M., F. K. Lake, W. Tripp, K. McCovey, A. Tripp, L. G. Hillman, C. J. Dunn, S. J. Prichard, P. F. Hessburg, W. Harling, and J. D. Bailey. 2024. Blending Indigenous and western science: quantifying cultural burning impacts in Karuk Aboriginal Territory. Ecological Applications 34(4):e2973. https://doi.org/10.1002/eap.2973

Howitt, R., and S. Suchet-Pearson. 2006. Rethinking the building blocks: ontological pluralism and the idea of ‘management’. Geografiska Annaler 88(3):323-335 https://doi.org/10.1111/j.1468-0459.2006.00225.x

Huntingford, C., P. M. Cox, M. S. Williamson, J. J. Clarke, and P. D. L. Ritchie. 2023. Emergent constraints for the climate system as effective parameters of bulk differential equations. Earth System Dynamics 14:433-442. https://doi.org/10.5194/esd-14-433-2023

Huntington, H. P. 2000. Using traditional ecological knowledge in science: methods and applications. Ecological Applications 10(5):1270-1274. https://doi.org/10.1890/1051-0761(2000)010[1270:UTEKIS]2.0.CO;2

InciWeb. 2020. East Troublesome Fire Information. https://inciweb.nwcg.gov/incident/7242/

Joa, B., G. Winkel, and E. Primmer. 2018. The unknown known: a review of local ecological knowledge in relation to forest biodiversity conservation. Land Use Policy 79:520-530. https://doi.org/10.1016/j.landusepol.2018.09.001

Karuk Tribe. 2019. Karuk Climate Adaptation Plan. Karuk Tribe, California, USA. https://karuktribeclimatechangeprojects.files.wordpress.com/2019/10/reduced-size_final-karuk-climate-adaptation-plan.pdf

Kawamiya, M., T. Hajima, K. Tachiiri, S. Watanabe, and T. Yokohata. 2020. Two decades of Earth system modeling with an emphasis on Model for Interdisciplinary Research on Climate (MIROC). Progress in Earth and Planetary Science 7:64. https://doi.org/10.1186/s40645-020-00369-5

Knight, C. A., L. Anderson, M. J. Bunting, M. Champagne, R. M. Clayburn, J. N. Crawford, A. Klimaszewski-Patterson, E. E. Knapp, F. K. Lake, S. A. Mensing, et al. 2022. Land management explains major trends in forest structure and composition over the last millennium in California’s Klamath Mountains. Proceedings of the National Academy of Sciences 119(12):e2116264119. https://doi.org/10.1073/pnas.2116264119

Konczi, A. E., and L. Bill. 2024. Advancing First Nations principles of OCAP®. Pages 37-39 in G. Garvey, editor. Indigenous and tribal peoples and cancer. Springer Nature, Cham, Switzerland. https://doi.org/10.1007/978-3-031-56806-0_8

Koven, C. D., R. G. Knox, R. A. Fisher, J. Q. Chambers, B. O. Christoffersen, S. J. Davies, M. Detto, M. C. Dietze, B. Faybishenko, J. Holm, M. Huang, M. Kovenock, L. M. Kueppers, G. Lemieux, E. Massoud, N. G. McDowell, H. C. Muller-Landau, J. F. Needham, R. J. Norby, T. Powell, A. Rogers, S. P. Serbin, J. K. Shuman, A. L. S. Swann, C. Varadharajan, A. P. Walker, S. J. Wright, and C. Xu. 2020. Benchmarking and parameter sensitivity of physiological and vegetation dynamics using the Functionally Assembled Terrestrial Ecosystem Simulator (FATES) at Barro Colorado Island, Panama. Biogeosciences 17:3017-3044. https://doi.org/10.5194/bg-17-3017-2020

Laniak, G. F., G. Olchin, J. Goodall, A. Voinov, M. Hill, P. Glynn, G. Whelan, G. Geller, N. Quinn, M. Blind, S. Peckham, S. Reaney, N. Gaber, R. Kennedy, and A. Hughes. 2013. Integrated environmental modeling: a vision and roadmap for the future. Environmental Modelling and Software 39:3-23. https://doi.org/10.1016/j.envsoft.2012.09.006

Latulippe, N., and N. Klenk. 2020. Making room and moving over: knowledge co-production, Indigenous knowledge sovereignty and the politics of global environmental change decision-making. Current Opinion in Environmental Sustainability 42:7-14. https://doi.org/10.1016/j.cosust.2019.10.010

Lawrence, D. M., R. A. Fisher, C. D. Koven, K. W. Oleson, S. C. Swenson, G. Bonan, N. Collier, B. Ghimire, L. van Kampenhout, D. Kennedy, et al. 2019. The Community Land Model version 5: description of new features, benchmarking, and impact of forcing uncertainty. Journal of Advances in Modeling Earth Systems 11:4245-4287. https://doi.org/10.1029/2018MS001583

Lawrence, D. M., R. A. Fisher, C. D. Koven, K. W. Oleson, S. C. Swenson, M. Vertenstein, B. Andre, G. Bonan, B. Ghimire, L. van Kampenhout, et al. 2020. Technical description of version 5.0 of the Community Land Model (CLM). National Center for Atmospheric Research, Boulder, Colorado, USA. https://www2.cesm.ucar.edu/models/cesm2/land/CLM50_Tech_Note.pdf

Leenhardt, P., V. Stelzenmüller, N. Pascal, W. N. Probst, A. Aubanel, T. Bambridge, M. Charles, E. Clua, F. Féral, B. Quinquis, B. Salvat, and J. Claudet. 2017. Exploring social-ecological dynamics of a coral reef resource system using participatory modeling and empirical data. Marine Policy 78:90-97. https://doi.org/10.1016/j.marpol.2017.01.014

Li, S., S. Jaroszynski, S. Pearse, L. Orf, and J. Clyne. 2019. Vapor: a visualization package tailored to analyze simulation data in earth system science. Atmosphere 10(9):488. https://doi.org/10.3390/atmos10090488

Liguori, A., L. McEwen, J. Blake, and M. Wilson. 2021. Towards ‘creative participatory science’: exploring future scenarios through specialist drought science and community storytelling. Frontiers in Environmental Science 8:589856. https://doi.org/10.3389/fenvs.2020.589856

Long, J. W., F. K. Lake, R. W. Goode, and B. M. Burnette. 2020. How traditional tribal perspectives influence ecosystem restoration. Ecopsychology 12(2):71-82. https://doi.org/10.1089/eco.2019.0055

Ma, Q., R. C. Bales, J. Rungee, M. H. Conklin, B. M. Collins, and M. L. Goulden. 2020. Wildfire controls on evapotranspiration in California’s Sierra Nevada. Journal of Hydrology 590:125364. https://doi.org/10.1016/j.jhydrol.2020.125364

Mantyka-Pringle, C. S., T. D. Jardine, L. Bradford, L. Bharadwaj, A. P. Kythreotis, J. Fresque-Baxter, E. Kelly G. Somers, L. E. Doig, P. D. Jones, K. E. Lindenschmidt, and the Slave River and Delta Partnership. 2017. Bridging science and traditional knowledge to assess cumulative impacts of stressors on ecosystem health. Environment International 102:125-137. https://doi.org/10.1016/j.envint.2017.02.008

Martins, I. M., R. P. Medeiros, M. Di Domenico, and N. Hanazaki. 2018. What fishers’ local ecological knowledge can reveal about the changes in exploited fish catches. Fisheries Research 198:109-116. https://doi.org/10.1016/j.fishres.2017.10.008

Maxwell, K. H., K. Ratana, K. K. Davies, C. Taiapa, and S. Awatere. 2019. Navigating towards marine co-management with Indigenous communities on-board the Waka-Taurua. Marine Policy 111:103722. https://doi.org/10.1016/j.marpol.2019.103722

McCaffrey, S., R. Wilson, and A. Konar. 2018. Should I stay or should I go now? Or should I wait and see? Influences on wildfire evacuation decisions. Risk Analysis 38(7):1390-1404. https://doi.org/10.1111/risa.12944

McDermid, S. S., L. O. Mearns, and A. C. Ruane. 2017. Representing agriculture in Earth System Models: approaches and priorities for development. Journal of Advanced Modeling Earth Systems 9:2230-2265. https://doi.org/10.1002/2016MS000749

Meldrum, J. R., C. M. Barth, J. B. Goolsby, S. K. Olson, A. C. Gosey, J. White, H. Brenkert-Smith, P. A. Champ, and J. Gomez. 2022. Parcel-level risk affects wildfire outcomes: insights from pre-fire rapid assessment data for homes destroyed in 2020 East Troublesome Fire. Fire 5(1):24. https://doi.org/10.3390/fire5010024

Mugerwa, S., M. Nyangito, J. Nderitu, C. Bakuneta, D. Mpairwe, and E. Zziwa. 2011. Farmers’ ethnoecological knowledge of the termite problem in semi-arid Nakasongola. African Journal of Agricultural Research 6:3183-3191.

Muller, S. 2012. ‘Two ways’: bringing indigenous and nonindigenous knowledges together. Pages 59-79 in J. K. Weir, editor. Country, native title and ecology. ANU E Press, Canberra, Australia. https://doi.org/10.22459/CNTE.03.2012.04

Nikolakis, W., and N. Hotte. 2022. Implementing “ethical space”: an exploratory study of Indigenous‐conservation partnerships. Conservation Science and Practice 4(1):e580. https://doi.org/10.1111/csp2.580

Obermeister, N. 2019. Local knowledge, global ambitions: IPBES and the advent of multi-scale models and scenarios. Sustainability Science 14(3): 843-856. https://doi.org/10.1007/s11625-018-0616-8

Oregon Department of Agriculture (ODA). 2021. Oregon State Board of Agriculture 2021 Report. Oregon Department of Agriculture, Salem, Oregon, USA. https://www.oregon.gov/oda/shared/documents/publications/administration/boardreport.pdf

Pauli, N., L. K. Abbott, S. Negrete-Yankelevich, and P. Andrés. 2016. Farmers’ knowledge and use of soil fauna in agriculture: a worldwide review. Ecology and Society 21(3):19. https://doi.org/10.5751/ES-08597-210319

Reid, A. J., L. E. Eckert, J. Lane, N. Young, S. G. Hinch, C. T. Darimont, S. J. Cooke, N. C. Ban, and A. Marshall. 2021. “Two-Eyed Seeing”: an Indigenous framework to transform fisheries research and management. Fish and Fisheries 22(2):243-261. https://doi.org/10.1111/faf.12516

Reyes-García, V., A. Fernández-Llamazares, D. García-del-Amo, and M. Cabeza. 2020. Operationalizing local ecological knowledge in climate change research: challenges and opportunities of citizen science. Pages 183-197 in M. Welch-Devine, A. Sourdril, and B. J. Burke, editors. Changing climate, changing worlds: local knowledge and the challenges of social and ecological change. Springer, New York, New York, USA. https://doi.org/10.1007/978-3-030-37312-2_9

Robinson, C. J., K. Maclean, R. Hill, E. Bock, and P. Rist. 2016. Participatory mapping to negotiate indigenous knowledge used to assess environmental risk. Sustainability Science 11(1):115-126. https://doi.org/10.1007/s11625-015-0292-x

Robinson, O. C. 2014. Sampling in interview-based qualitative research: a theoretical and practical guide. Qualitative Research in Psychology 11(1):25-41. https://doi.org/10.1080/14780887.2013.801543

Sterling, E. J., M. Zellner, K. E. Jenni, K. Leong, P. D. Glynn, T. K. Bendor, P. Bommel, K. Hubacek, A. J. Jetter, R. Jordan, L. S. Olabisi, M. Paolisso, and S. Gray. 2019. Try, try again: lessons learned from success and failure in participatory modeling. Elementa: Science of the Anthropocene 7:9. https://doi.org/10.1525/elementa.347

Stori, F. T., C. M. Peres, A. Turra, and R. L. Pressey. 2019. Traditional ecological knowledge supports ecosystem-based management in disturbed coastal marine social-ecological systems. Frontiers in Marine Science 6:571. https://doi.org/10.3389/fmars.2019.00571

Tauri, J. M. 2018. Research ethics, informed consent and the disempowerment of First Nation peoples. Research Ethics 14(3):1-14. https://doi.org/10.1177/1747016117739935

Tengö, M., B. J. Austin, F. Danielsen, and Á. Fernández-Llamazares. 2021. Creating synergies between citizen science and Indigenous and local knowledge. BioScience 71(5):503-518. https://doi.org/10.1093/biosci/biab023

Tengö, M., E. S. Brondizio, T. Elmqvist, P. Malmer, and M. Spierenburg. 2014. Connecting diverse knowledge systems for enhanced ecosystem governance: the multiple evidence base approach. Ambio 43:579-591. https://doi.org/10.1007/s13280-014-0501-3

Tengö, M., R. Hill, P. Malmer, C. M. Raymond, M. Spierenburg, F. Danielsen, T. Elmqvist, and C. Folke. 2017. Weaving knowledge systems in IPBES, CBD and beyond: lessons learned for sustainability. Current Opinion in Environmental Sustainability 26-27:17-25. https://doi.org/10.1016/j.cosust.2016.12.005

Teufel-Shone, N. I., A. L. Schwartz, L. J. Hardy, H. D. De Heer, H. J. Williamson, D. J. Dunn, K. Polingyumptewa, and C. Chief. 2019. Supporting new community-based participatory research partnerships. International Journal of Environmental Research and Public Health 16(1):44. https://doi.org/10.3390/ijerph16010044

Touzé-Peiffer, L., A. Barberousse, and H. Le Treut. 2020. The coupled model intercomparison project: history, uses, and structural effects on climate research. Climate Change 11(4):e648. https://doi.org/10.1002/wcc.648

Turnbull, D. 1997. Reframing science and other local knowledge traditions. Futures 29:551-562. https://doi.org/10.1016/S0016-3287(97)00030-X

Visualization & Analysis Systems Technologies. 2020. Visualization and analysis platform for ocean, atmosphere, and solar researchers: (VAPOR version 3.2. 0).

Voinov, A., and E. J. B. Gaddis. 2008. Lessons for successful participatory watershed modeling: a perspective from modeling practitioners. Ecological Modelling 216(2):197-207. https://doi.org/10.1016/j.ecolmodel.2008.03.010

Wheeler, H. C., F. Danielson, M. Fidel, V. Hausner, T. Horstkotte, N. Johnson, O. Lee, N. Mukherjee, A. Amos, H. Ashthorn, O. Ballari, C. Behe, K. Breton-Honeyman, G.-B. Retter, V. Buschman, P. Jakobsen, F. Johnson, B. Lyberth, J. A. Parrott, M. Pogodaev, R. Sulyandziga, and N. Vronski. 2020. The need for transformative changes in the use of Indigenous knowledge along with science for environmental decision making in the Arctic. People and Nature 2(3):544-556. https://doi.org/10.1002/pan3.10131

Zanotti, L., and M. Palomino-Schalscha. 2016. Taking different ways of knowing seriously: cross-cultural work as translations and multiplicity. Sustainability Science 11(1):139-152. https://doi.org/10.1007/s11625-015-0312-x