Learning to live with smoke: characterizing wildland fire and prescribed fire smoke risk communication in rural Washington

Extreme smoke events from wildland fires are increasing in frequency and intensity across the country. Risk communication around wildland and prescribed fires is an essential component of both smoke-readiness and resilience. To date, little research has been conducted on how smoke exposure risks can be communicated effectively, especially within the context of rural and tribal communities, who experience a disproportionate burden of smoke risks and impacts. This qualitative study analyzed how tribal and non-tribal communities in the Okanogan River Airshed Emphasis Area (ORAEA) receive and share information about smoke exposure to highlight gaps and communication opportunities for smoke risk communication. The ORAEA is a region of north-central Washington that is frequently blanketed with smoke year-round from wildland fire in the summer, prescribed fire in the fall and spring, and wood burning stoves in the winter. This study was the result of a partnership between the Okanogan River Airshed Partnership, the Natural Resource Division for the Confederated Tribes of the Colville Reservation, the Colville Environmental Trust Air Quality Program, and the University of Washington. The study team conducted seventeen key informant interviews and six focus groups to identify community perceptions of smoke exposure and describe its impact. A thematic analysis of interview and focus group data identified five themes around wildfire smoke risk communication: (1) perception of the health risks of smoke; (2) current ways of sharing information about smoke; (3) trusted sources of information; (4) gaps and communication opportunities; and (5) perceptions of prescribed fire. Based on these themes, we developed a set of six recommended actions. To create effective smoke risk communication that may be applicable to smoke-impacted regions across the country, messaging must address barriers to action, be rooted in community perceptions of risk, and be delivered through trusted channels.


Introduction
Wildfires represent a growing public health threat. In 2021, over 7.1 million acres burned across the United States and hundreds of thousands of people were exposed to dangerous levels of wildfire smoke (National Interagency Fire Center 2021). Historical trends demonstrate that current wildfire severity is due to a variety of factors, including a warming climate and the accumulation of dead fuels caused, in part, by over a century of fire exclusion policies and colonization (North et al 2015). Acknowledging the essential role of fire that has always been part of the landscape of much of the western United States, tribes in this region have practiced burning as part of a holistic land management and cultural practice (Parks and Abatzoglou 2020). European influence vulnerability to environmental hazards. According to the Washington Environmental Health Disparities Map, a cumulative environmental impacts assessment tool, two of three census tracts on the Colville Reservation are considered 'highly impacted' (ranking of '9' or '10') by social vulnerability to hazards, as is the nearby town of Omak (Washington State Department of Health 2020). Smoke is not a new phenomenon in this region, as Native and Indigenous peoples have continuously used fire to manage wildfire-prone landscapes in this area for millennia (Wynecoop et al 2019, Roos et al 2021. Growing development in the wildland-urban interface (WUI) and the increasing frequency of large wildfires in this region, however, mean that more people are being exposed to potentially hazardous levels of smoke than before (Hammer et al 2009). In 2021, the town of Omak, which hosts ORAEA's only permanent government-hosted air quality monitor, recorded 41% of days with moderate air quality or worse (EPA 2021). The frequency and intensity of wildfires are only expected to increase under current climate change projections (Parks and Abatzoglou 2020), and there are concurrent efforts to scale up implementation of prescribed fire on nearby tribal and non-tribal land (Franz 2017).
Due to the unique fire and smoke scenario of the ORAEA, local governments and community-based organizations have taken great steps to prepare and help their community adapt to smoke season. However, there remains a growing urgency to further prepare for and mitigate the potential impacts of the inevitable influx of smoke to this community. To accomplish this, we established a partnership among the Natural Resource Division for the CTCR, the Okanogan River Airshed Partnership (ORAP), and the University of Washington (UW). The overall goal of this project was to learn from the experience and expertise of tribal and non-tribal communities of the ORAEA to identify strengths and barriers to smoke risk communication, and, ultimately, share this with other smoke-impacted communities in the region. We sought to understand which communication channels, past those traditionally used by public health and air quality agencies, resulted in behavior change or modification. The transdisciplinary nature of the research team allowed us to identify and understand challenges within the context of social and economic equity and environmental resilience, and our partnerships with local leaders helped us to reach and connect with the community throughout all phases of the project.

Methods
We conducted a qualitative study using key informant interviews and focus group discussions. The development of interview and focus group guides used the principles of the Talanoa Dialogue to engage in inclusive, participatory, and transparent conversations. This approach, used to bring countries together to collaborate on climate change, focuses on storytelling and guides conversations based on the three questions 'Where are we now?' , 'Where do we want to go?' and 'How do we get there?' (Presidencies of COP 22 and COP 2017). To answer the first question, 'Where are we now?' , we sought to understand how information about smoke exposure is currently shared within the rural and tribal communities in the ORAEA. To answer, 'Where do we want to go?' , we worked to describe the most effective communication route from trusted source to engaged recipient who is ready and willing to modify their behavior. Finally, the answer to 'How do we get there?' , combined the results collected from the first two questions to define the most effective sources, materials, and communication networks to share information about smoke exposure. Both the interview guide and focus group questions were workshopped with the full study team, including our partners at CTCR and ORAP. Interview and Focus Group questions are available as supplemental data.

Recruitment and data collection
Recruitment for key informant interviews included a combination of purposive and snowball sampling. The full research team worked together to identify a list of stakeholder categories and determine a target number of participants within each category (table 1). CTCR and ORAP co-investigators then identified potential contacts for interviews and facilitated introductions between potential interviewees and the UW project team, who followed up to coordinate and facilitate interviews. Inclusion criteria for key informant interviewees included: (1) residents of the ORAEA; (2) 18 years or older; (3) self-identify as elders, community leaders, and/or tribal/local government staff (including public health and natural resources staff).
Recruitment for focus groups was a combination of purposive sampling and convenience sampling. The inclusion criteria for focus groups participants included residents of ORAEA and 18 years or older. Focus groups were designed to include age as a variable and allow dynamics between participants of the same age group to help to shape the data (Finucane 2008). The first two focus groups were with CTCR employees at the tribal government center and were roughly divided by age, with one group emphasizing younger people under the age of 25 and one group composed of people of mixed ages and generations. A CTCR project co-investigator supported the recruitment and facilitation of these focus groups. A second set of focus groups were held in Omak and were similarly divided by age, with a younger group and a mixed age group. Recruitment for non-tribal focus groups utilized flyers posted around the towns of Omak and Okanogan as well as on several local Facebook groups. A third set of focus groups was held over lunch with tribal elders at a senior meal site on the Colville Reservation.
A total of seventeen key informant interviews were conducted by the UW research team between June and November 2021. Each interview lasted roughly 1 hour and was facilitated remotely over Zoom or phone by a member of the UW research team. Interviews were audio recorded in Zoom or Google Voice and transcribed by professional transcription service TranscribeMe and checked by a member of the UW research team. Focus group discussions took place between October and November 2021. Each discussion lasted around 19 min and was facilitated in-person by a member of the UW research team, and in a few cases, a co-investigator for the project who worked for the CTCR. Focus groups were audio recorded in Audacity and transcribed verbatim by a member of the UW research team into a written document.
All personally identifying information for both interview and focus group data was removed prior to analysis. Additionally, a member of the UW research team took notes at each interview and focus group discussion for later reference. All participants were offered a $50 gift card for their participation. UW Institutional Review Board (IRB) and CTCR Tribal Council approval were acquired before participants were Table 1. Key informant interview categories (summarized in bold), and the target and actual sample frames. This table has been adapted from a manuscript from co-author Wood (Wood et al 2022), describing the role of trusted sources in smoke risk communication in the ORAEA. Adapted from Wood et al (2022). CC BY 4.0.

Stakeholder category Target Actual
People with significant firsthand work and/or volunteer experience in local government leadership

2-3 3
People with significant firsthand work and/or volunteer experience with elders 2-3 3 People with significant firsthand work and/or volunteer experience at or with schools and/or youth organizations

2-3 2
People with significant firsthand work and/or volunteer experience in public health or healthcare

1-2 3
People with significant firsthand work and/or volunteer experience communicating about air quality

1-2 3
People with significant firsthand work and/or volunteer experience in emergency management

1-2 2
People with significant firsthand work and/or volunteer experience in agriculture and/or forestry contacted and all participants provided consent prior to participation. Interviewees over the phone provided verbal consent, and in person focus group participants provided written consent before each discussion.

Coding and analysis
A codebook for analysis was developed using a combination of inductive and deductive approaches. Codes were developed inductively based on themes identified in notes and transcripts, and deductively organized using the constructs outlined in the Health Belief Model and Theory of Planned Behavior (Rosenstock 1974, Rosenstock et al 1988, LaMorte 2019. Codes and their definitions were memorialized in a codebook. Codes were applied using NVivo qualitative analysis software by two members of the UW research team. Discrepancies in code application were identified and adjudicated using consensus building discussion. Code definitions were updated to improve clarity and enhance codebook reliability. Codes were applied by a single investigator in NVivo qualitative analysis software to the remaining transcripts using the updated codebook. We used the framework method to complete a thematic analysis of the data (Braun and Clarke 2006, Smith and Firth 2011, Maguire and Delahunt 2017, Collaço et al 2021. Data were summarized and plotted to a matrix, organized by Key Informant Interviews and Focus Group Discussions. For each code, data were further synthesized and grouped into tribal, non-tribal, and aggregate data to facilitate comparison and review. As preliminary themes were identified, we compiled and then read and re-read coded excerpts under each theme, revising the theme as needed until themes and sub-themes were clearly defined (Braun and Clarke 2006).
To check for accuracy and cultural relevance in interpretation of data, we integrated opportunities for feedback at multiple points in the analysis process. As co-investigators, CTCR staff and ORAP leadership provided guidance and feedback throughout data collection and analysis. After each interview, a summarized version of notes from each interview was organized by theme and sent to each interviewee for review and feedback. Of seventeen interviews, five sent additional commentary and four others responded affirmatively with no further comment. Additionally, preliminary findings were presented at an ORAP community meeting in December 2021, which multiple interviewees attended, and feedback received was integrated into the final analysis.

Demographics
A total of 18 individuals participated in 17 interviews, including one interview with two participants. Two of the 18 interview participants did not provide demographic information. There were thirty-three participants included in our six focus group discussions. Demographic information for interview and focus group participants are provided in figures 2(a) and (b), respectively.

Themes
In both key informant interviews and focus groups, participants were asked to describe how they perceive the health risks of wildfire and prescribed fire smoke and how information about those health risks is shared throughout the community. Participants were also asked to discuss potential strategies to mitigate those health risks and actions, or resources needed. Results from these questions and others were synthesized into five themes and subthemes summarized in table 2. When not otherwise stated, 'participants' are inclusive of both interviewees and focus group participants.

Theme 1: perception of health impacts of smoke
Participants' perceptions of the health impacts of smoke and the risk of these health impacts were varied and complex. Participants shared concerns about the immediate physical symptoms they had experienced during smoke as well as uncertainty around long-term effects, impacts to mental health, and impacts to their communities and lifestyles. At the same time, however, the competing urgency of damage and destruction from wildfires complicated that perception.
3.3.1. Subtheme 1.1: participants shared concern and uncertainty about the various health and lifestyle impacts of smoke Both tribal and non-tribal participants expressed significant concern about the health impacts of wildfire smoke. Participants shared stories of their own experiences during smoke events, including difficulty breathing; irritation of the eyes, nose, throat, and lungs; exacerbated respiratory illness; or delayed recovery from other illnesses from breathing smoke. They noted uncertainty and concern about the long-term impacts Table 2. Qualitative themes and subthemes related to smoke risk communication. Subtheme labeled with a 'NT' was identified in non-tribal participants only.
Theme Subtheme 1. Perception of the health impacts of smoke Participants shared concern and uncertainty about the various health and lifestyle impacts of smoke.
Compared to fire, smoke is generally considered to be less urgent of a concern.
2. Current ways of sharing information about smoke Most residents of the ORAEA are getting their information primarily from local sources.
Facebook is by far the most used channel for information.
3. Trusted sources of information Local and tribal agencies are generally more trusted than state or federal government agencies.
Political ideology influences which channels people trust (NT).

Gaps and communication opportunities
Staying indoors is the most common strategy for protecting against smoke; however, effectiveness is limited by access to clean indoor air.
Need to shift towards preparation and smoke-readiness in advance of smoke season.

Perception of prescribed fire
Poor forest management and fire exclusion (deliberate prevention of natural or human caused fire) are perceived as exacerbating wildfires.
Prescribed fire is viewed as relatively safe and beneficial.
While prescribed fire is appreciated for its role in mitigating wildfire risk, smoke impacts to communities need to be acknowledged.
of smoke exposure and what the future health implications of prolonged, repeated exposure to high levels of smoke might be. Additionally, the exacerbation of the health impacts of smoke with the co-occurrence of multiple disasters, such as extreme heat and the COVID-19 pandemic was discussed. Participants described the difficulties of staying inside, closing windows, and wearing masks to protect themselves from smoke during the summer of 2021's heat wave. Two out of five focus groups and three interviewees mentioned COVID-19 and its associated economic impacts have strained community resources, such as shelters for people experiencing homelessness, hospital beds, and other resources to meet basic community needs.
PTSD is often the label that gets slapped on it. I don't know that anyone's really carefully looked at that. But it feels very fitting that because of that ongoing repeated trauma of having to deal with wildfires and wildfire smoke-they're difficult to tease apart. Air Quality Interviewee, Non-Tribal In every focus group, and 13 out of 17 interviews, participants expressed the greatest concern about other people in their communities that they viewed as especially vulnerable to the impacts of smoke: elders and older adults, children, people who are unhoused, outdoor workers, and people with asthma, COPD, or other respiratory conditions. Participants were especially concerned about the impact of staying inside on children and what future summers may look like for young people. There was also explicit concern expressed about firefighters, as many participants had fought fires themselves, or have family members who fight fires. In addition to the physical health impacts of smoke, many participants expressed concern about the mental health impacts of living through prolonged smoke events. Participants noted that the smell of smoke can incite fear, anxiety, and other feelings associated with those experiences. They also shared how smoke had impacted their lifestyles, including limitations on their ability to exercise, recreate, and spend time outdoors, contributing to a sense of restlessness, feeling 'cooped up' and other mental health impacts. Tribal participants expressed concern about impacts on cultural activities, like the ability to spend time outside and gather huckleberries and other first foods and medicines, as well as the impacts of fires on the ecosystems on which these practices depend.
3.3.2. Subtheme 1.2: compared to fire, smoke is generally considered to be less urgent of a concern Four of six focus groups and 14 of 17 interviewees noted the growing intensity and frequency of large wildfires and smoke over the past 5-10 years, noting the 2014, 2015, and 2020 fire seasons as particularly destructive. Across all conversations with elders and people who had grown up in the ORAEA, it was noted 43 times that while there have always been wildfires in the area, they had never experienced fires like they have over the past decade. Participants described how smoke seemed to be 'getting worse' over time, with longer, more intense smoke episodes. However, although smoke is growing as a community concern, at the same time, smoke was generally viewed as less immediate and threatening of a concern than the destruction or damage caused by wildfire.
I just think the biggest thing is that-not that we don't think smoke is a bad thing -but I just feel like if it's between saving your home and staying near the smoky area or saving your home and trying to do a fire line, most people will choose to just risk their health and try to save their property. Youth

Focus Group Participant, Tribal
When asked about the priority of smoke as a community concern, eleven participants responded that they perceive it ranked highly during periods of intense or prolonged smoke. That concern, however, fades quickly when the smoke dissipates. Participants explained that this is due, in part, to competing priorities in the community that are seen as more urgent or that happen year-round. Participants also explained that because smoke is so pervasive during the summer months, people do not want to talk about it or think about it after it fades.
Sixty quotes were recorded that described participant's sense of lack of control around smoke-that there's not much that can be done about it, and that there is a need to continue with their daily lives. Many people in the ORAEA either work outside, in forestry or agriculture, or spend significant time outside to maintain their lifestyles and livelihoods, taking care of livestock or other animals, or looking after their property. Some participants noted that feeling a lack of control and wanting to maintain normalcy can act as a barrier to taking protective action to mitigate the impacts of smoke. Additionally, in non-tribal interviews and focus groups, many people referenced the perceived cultural attitudes around 'toughness' and resilience associated with a rural lifestyle as relevant to how people respond to smoke. In describing communities in the ORAEA, many participants used words like 'self-sufficient' and 'independent' , and said that worrying about smoke or health, or taking precautions to protect one's health may be viewed as being not tough enough to withstand the smoke. Many people in the ORAEA have worked or volunteered as fire fighters, where exposure to high levels of smoke is common and familiar, further influencing their perception of the health risk of smoke.

Theme 2: current ways of sharing smoke information
Participants shared that they sought information from a variety of sources, but preferred information received through CTCR, local government, and informal networks, to sources from outside of the region. Facebook was used by many to share information quickly with a lot of people.
3.4.1. Subtheme 2.1: most residents of the ORAEA are getting their information primarily from local sources Participants overwhelmingly (every interviewee and at least one mention in every focus group) described local and community-based channels as their main sources of information on smoke and the health impacts of smoke. Tribal participants primarily described receiving information through the various CTCR departments, specifically looking for information from the Environmental Trust Department and Mt. Tolman Fire Center. Other sources of information for tribal participants included tribal broadcast emails, which are sent to all CTCR employees; community information boards; local and tribal news; friends and family; weather apps; and social media. For nontribal participants, sources of information tended to be less centralized, and included local government agencies such as conservation districts, schools, and fire districts; community-based organizations; friends and family; local news and media; and social media.
The messaging, if it's tailored correctly, and if it's local, tied into and accurate for these various communities, and put out by the tribes, it's generally well accepted. Healthcare Interviewee, Tribal All focus groups, and six interviewees emphasized the strength of informal networks-groups of friends, family, and neighbors-within communities and the potential of leveraging these word-of-mouth communication and community networks to reach a wide range of people with smoke risk communication messaging. These networks are effective at reaching people living in more remote, rural areas in the ORAEA or who are living off the-grid and are not connected to internet or phone service. Several participants described a sense of community in the ORAEA, where people look out for one another, and neighbors will often go out of their way to share information. Anecdotes about having conversations with friends and family about smoke and fires were shared by at least three people in every focus group.
Twenty participants (interviewees and withing focus groups) shared challenges for communicating in the ORAEA, including remoteness, lack of bilingual information, and poor internet and phone connectivity in the region, which in some cases are exacerbated by the destruction of signal towers during wildfires. To accommodate this, CTCR departments and local non-tribal agencies use a variety of strategies to reach community members, including physically posting information on community boards; posting on social media; leveraging local radio, TV, and newspapers; and partnering with other local groups, such as ORAP or Clean Air Methow, a community-based organization in a nearby area focused on air quality.

Subtheme 2.2: Facebook is by far the most used channel for information
Twenty-eight quotes were recorded of participants sharing that social media-in particular, Facebook-is used to share information quickly throughout the community by CTCR and other local government and community agencies. Several participants described the wide use of private resident-run Facebook groups, such as 'Okanogan Highlands Fire Watch LLC,' which post crowd-sourced fire and smoke information from the area. Participants also noted that community members often re-share information posted in groups, which then get shared with a wider audience.

The community is really good about getting stuff out quick and fast that needs to go out and have, again, social media pages. It seems like everything's-not everything, but most things are done by social media. Leadership Interviewee, Tribal
Seven participants also raised concerns about the threat of misinformation on social media. They expressed worry that what is shared is not always accurate and also mentioned misinformation in reference to COVID-19 and vaccines. Generally, participants described feeling that information shared by tribal and local agencies was accurate, as was information posted in fire watch Facebook groups, which are moderated by community volunteers.

Theme 3: trusted sources of information
While participant responses varied, 32 participants (inclusive of interviews and focus groups) perceived local sources of information, such as tribal or local government, or informal networks, to be more trusted than the state or federal government. In non-tribal interviews and focus groups, seven participants described how political ideology influences which sources people in their communities trusted.

Subtheme 3.1: local and tribal agencies are generally more trusted than state or federal government
When asked about which sources of information they trust most, participants stated a preference for local sources over information shared directly from state or federal government agencies. In part, this is determined by ORAEA residents' preferences for information that is locally relevant and accurately reflects the experiences of residents of the ORAEA. Additionally, participants stated their tendency to trust people and agencies with whom ORAEA residents have relationships and past experiences. For tribal participants, this is also related to a preference for information delivered through CTCR channels, which are viewed as looking out for the best interests of tribal members and are tailored to be culturally and contextually relevant. Several tribal participants described specifically seeking information from the CTCR Environmental Trust Department's Communications team, as well as Public Information Officers at the Mt. Tolman Fire Center, which are viewed as respected authorities on fire and smoke in the community. Both tribal and non-tribal participants identified firefighters and fire departments as well trusted sources in the community, due to their authority on the topic of fires and their presence within communities.

Definitely people prefer local. I will say county-wide there is a general distrust of state and federal agencies as a whole. Healthcare Interviewee, Non-Tribal
For both tribal and non-tribal participants, trustworthiness was reported to be determined by the perceived credibility of the source, the quality of information, and the relationship with the source. Within these categories, tribal participants identified authority and expertise of the source, accuracy and timeliness of the information, and local and tribal relevance as most important when evaluating trustworthiness. Non-tribal participants identified perceived political neutrality and transparency, authority, and respect as most important. Meanwhile, participants reported negative experiences with specific sources. They reported perceptions of these sources as untrustworthy and characterized them as having a hidden agenda, a lack of transparency, and sharing poor quality information that was not relevant or accurate.
Fifteen participants (including interviewees and focus groups) described the importance of personal relationships in determining trust within small, rural communities like those in the ORAEA. Several participants described how public figures within the community, such as local sheriffs or fire chiefs who live in the area and are well-known by many people, are perceived to be more trusted because of their earned respect and the perception that they are less likely to have a hidden agenda.

Subtheme 3.2: political ideology influences which channels people trust (non-tribal)
Five non-tribal interviewees and two focus group participants shared the perception that political ideology plays a significant role in how people in the ORAEA determine which sources are trustworthy. Each of these participants described Okanogan County, WA as a majority-conservative area, or 'red county' , and that government or media entities that are not perceived as aligned with residents' worldviews and experiences or are viewed as having a political agenda may be less trusted.
Federal government, least, and state government, it's-we're a very red county and so whoever's in power in the legislature, if they're not red, they're not trusted at all. State agencies, there's a lack of trust for state agencies as well. Leadership Interviewee, Non-Tribal Twelve non-tribal participants also highlighted the parallels between communication about COVID-19 and communication about the health impacts of smoke related to the political polarization of protective actions like masking and vaccinations. Of these twelve, several participants discussed the complication of community perceptions of public health agencies because of their association with COVID-19 for certain populations within the ORAEA, though they also emphasized that they imagined other populations viewed public health agencies as trusted sources of health information. Three participants also noted that wearing masks during smoke events has been more common over the past few years with the normalization of masking for COVID-19.

Theme 4: gaps and communication opportunities
Participants described existing gaps in access to resources or tools with which to mitigate the health impacts of wildfire smoke, as well as opportunities to address these gaps. Many of these opportunities centered around access to clean indoor air and smoke-readiness in advance of wildfire seasons and smoke.

Subtheme 4.1: staying indoors is the most common strategy for protecting against smoke; however, effectiveness is limited by access to clean indoor air
While participants shared a variety of strategies for minimizing the impacts of smoke for themselves and their families, the most common strategy (coded in 45 quotes) was staying indoors on smoky days and limiting the amount of time spent outdoors. Participants described the challenges of weatherizing homes to be smoke proof, as many of the homes in the ORAEA are older or do not seal completely, limiting the effectiveness of staying inside and cleaning indoor air. Some participants shared that they would run portable air cleaners (PACs) or do-it-yourself box fan filters to clean indoor air, while others face fans out of windows to try to blow smoke out of their homes. Participants also noted that the high cost of PACs, masks, and new filters for heating, ventillation and air conditioning systems may restrict community members' abilities to clean indoor air. Additionally, many people in the ORAEA do not have air conditioning and rely on opening windows to cool their homes.

And it's tough because a lot of folks don't have air conditioning necessarily, so the standard way you cool your house down is to open the windows at night. And then, of course, if it's smoky, you're going to infuse your house with smoke too. Forestry Interviewee, Non-Tribal
Thirteen participants also described barriers to staying indoors, such as the need to leave to run errands, go to work, and take care of animals or property outside. Participants expressed concern, especially for agricultural and forestry workers, who have no choice but to work outdoors, and are routinely exposed to high levels of smoke. Participants also shared the difficulties of keeping children indoors for long periods of time because of smoke. Several participants shared their concerns about the mental and physical health implications of limiting outdoor activity.

I left my house … and fortunately someone went out and watched it for me and then it started to catch fire, my deck started to burn. They hosed it down. Maybe if they didn't do that I would've lost it. Elder Focus Group Participant, Tribal
In times of extreme or prolonged periods of intense smoke, tribal participants described how CTCR would temporarily evacuate elders to areas with less smoke, sometimes as far as Spokane, which is several hours away. While this was successful in mitigating smoke exposure for elders, who may have health conditions that place them at higher risk of health impacts from smoke, participants described how these experiences were intensely stressful for elders or those who have limited mobility, who may be worried about damage to their homes or property during fires and burglary while they are away. In focus groups with tribal elders, several participants shared stories of evacuating during the Chuweah Creek Fire and the emotional toll of leaving behind their homes and animals.

Subtheme 4.2: there is a need to shift towards preparation and smoke-readiness in advance of smoke season
Participants emphasized the need to shift from emergency communication around fires to preparation in advance of smoke season, including ongoing education on the health impacts of smoke and how to protect one's health. ORAP and the CTCR Environmental Trust Department were credited by some participants with having helped advance conversations around preparation through their 'Smoke Ready Week' and work to integrate smoke readiness into tribal and county planning. Seventeen participants described the need for tools to support personal and organizational risk assessment and preparation.
Participants also described a need for timely, locally accurate air quality information, including the ability to predict smoke levels and dispersion during wildfire events. One participant described how a lack of weather and air quality information specific to their area can undermine public trust in smoke messaging, as the information available is not always accurate.

Resources would be the ability to preplan. Instruction on preplanning. The ability to self-assess their risk of remaining in an area that is currently impacted by wildfire smoke. Healthcare Interviewee, Tribal
Three participants noted the need to reduce non-wildfire sources of smoke in the ORAEA to limit the cumulative year-round exposure to fine particulate matter and other air pollutants. Participants described efforts to reduce agricultural and yard waste burning through public chipping and composting efforts, which have been led by local governments and ORAP partners. Additionally, participants described programs through the CTCR Environmental Trust Department to exchange wood stoves in homes, which are many ORAEA residents' primary form of heating, to newer, more efficient models to reduce indoor air pollution.

Theme 5: perception of prescribed fire
Participants stressed the importance of fire prevention and forest management, including prescribed fire. All but one participant expressed some sense of appreciation for the benefits of prescribed fire, with regard to improving forest health, regenerating ecosystems, and mitigating wildfires. Some, however, cautioned against smoke impacts to communities from prescribed fire and emphasized the importance of taking action to minimize those risks.

Subtheme 5.1: poor forest management and fire exclusion are perceived as exacerbating wildfires
When asked what potential solutions they see for mitigating the health impacts of smoke on communities, four of six focus groups and nine interviewees described a need to reduce out-of-control wildfires through fire prevention and robust forest management, including prescribed fire and thinning. Many participants shared the perspective that a buildup of fuels and poor forest health are exacerbating the effects of wildfires and that a combination of methods is needed to mitigate the impacts of wildfires and smoke on communities in the ORAEA.

If you can control burn, that's fine for a reason. I think if they clean the forest a little bit more, they might not have such a terrible problem with fires. Elder Care Worker Interviewee, Non-Tribal
Tribal participants described a long history of burning as a cultural practice, using fire to manage landscapes and regenerate ecosystems and food systems at a large scale before the arrival of European settlers. One participant shared an anecdote about how, while burning was outlawed by the United States government's policy of fire exclusion, tribal members continued to ignite fires in secret, out of recognition of the importance of fire on landscapes. Participants emphasized that tribes, like CTCR, continue to practice burning informed by traditional knowledge, but are restricted in their ability to do so at an effective scale by 'red tape' from federal and state agencies and limited capacity to implement.
3.7.2. Subtheme 5.2: prescribed fire is viewed as relatively safe and beneficial All but one participant described their perception of prescribed fire as a relatively safe practice in which the benefits outweigh the risks. Generally, participants described their perception of prescribed fire as more manageable and implemented under ideal conditions so as not to produce as much smoke or smoke that would intrude on nearby communities. Several participants described the permitting process as helping to ensure burning happens at times when the risk of escape is lowest; however, others described this process as restrictive and not allowing enough time to adequately plan or prepare after notification.
I think you guys probably are seeing the tension between prescribed fire and wildfire smoke and kind of it's a tough one because, yeah, prescribed fire puts smoke in the air and at the same time, it's one of our more effective tools. So trying to find the happy medium is always going to be tricky. Forestry Interviewee, non-tribal Participants also described several perceived benefits of prescribed fire, including mitigating the severity of wildfires and smoke by removing dead material and fuels, as well as improving forest health by clearing out diseased or dying trees. Tribal participants shared other benefits of regular fire, such as the regeneration of ecosystems and plants used for food and medicine, including huckleberries, which require some disturbance to regrow, and clearing areas to attract wildlife and game.
3.7.3. Subtheme 5.3: while prescribed fire is accepted as useful for its role in mitigating wildfire risk, smoke impacts to communities need to be acknowledged While many participants viewed prescribed fire as generally beneficial, and not as harmful to health, several others were wary of the potential impacts of smoke from prescribed fires in addition to wildfires. Several participants shared concern about the impacts of cumulative and repeated smoke exposure of prescribed fire in the shoulder seasons of wildfire season, since both wildfire and prescribed fire smoke release particulates into the air. Participants also shared concern about the mental health impacts of continued smoke after a heavy wildfire season and the perception that residents may feel weary or resentful of smoke from prescribed fires in the fall after experiencing wildfire smoke during the summer.

[Prescribed fire] is a tool that also has the potential to have detriments to the population that's right there, and even though the tool might be good for the forest, we don't necessarily have control over the smoke. Air Quality Interviewee, Non-Tribal
To mitigate the negative impacts of prescribed fires on communities, participants suggested bolstering communication and preparation in advance of prescribed fires, similar to preparation in advance of smoke season. Additionally, one participant shared how improved communication from a federal agency implementing prescribed fire has helped improve the perception of both the agency and prescribed fire in the community.

Discussion
Across the themes identified from coding interviews there were not broad differences between the tribal and non-tribal participants and their views on smoke risk communication. However, differences were detected in some of the minutia of each theme. When asked how ORAEA community members perceive the health risks of wildfire smoke, the concept of 'cultural toughness' as described in Theme 1.2 was only described by non-tribal participants. There was also a clear separation between tribal and non-tribal participants in their description of trusted sources of information. Tribal members spoke to valuing accuracy, timeliness, authority, and local-/tribal-relevance, whereas non-tribal participants focused on perceived political neutrality, authority, respect, and transparency. Along these same lines as described in Theme 3.2, non-tribal participants shared that political ideology influences which channels people trust. On the topic of prescribed burning, although all but one interviewee viewed the process as beneficial, only tribal participants discussed the suppression of traditional burning practices and the benefit for regeneration of ecosystems and forest health.
Effective and timely smoke risk communication is essential as fire and smoke seasons continue to worsen and the number of communities and people-both tribal and non-tribal-affected by smoke continues to increase. Results from this project demonstrate that preparedness for smoke exposure varies across age and demographic groups, and that individuals feel an uncertainty surrounding the health impacts of smoke exposure and how much they should be concerned. When it comes to avoiding smoke, our findings indicate that currently recommended exposure reduction strategies are often hard to build into daily life, and others are financially unrealistic for certain populations. Furthermore, our participants described that they and members of their community preferred receiving smoke risk information from social media, tribal or local governments, and word of mouth, compared to sources outside of the region. experiences (Larsson et al 2006, Boyd andFurgal 2019). Potential partners for this type of messaging identified by participants in our study are local and tribal firefighters and fire departments, who are generally perceived as trusted members of the community and who experience high exposure to smoke and high rates of health impacts themselves (Paveglio et al 2015, Adetona et al 2016. Our participants acknowledged that it is no longer possible to think about fire and smoke as seasonal. Our final recommendations consequently suggest designing risk communications campaigns with this in mind: emphasize smoke readiness and preparation (Recommendation 5) and Address and mitigate increasingly frequent and severe wildfires (Recommendation 6). Not only is there a need for year-round fire mitigation efforts, but also a need for year-round planning for increased smoke levels. Smoke readiness strategies must be tailored to the specific organization or household, as everyone has differing levels of risk and capacity to prepare (United States Environmental Protection Agency 2022). A key component of this is increasing access to relevant, timely information for decision-making. Evidence suggests that forest management strategies, including thinning, pile burning, prescribed fire, and managed fire, may reduce the intensity of wildfires and reduce overall smoke output in certain landscapes and environments (Martinson andOmi 2013, Keane 2015). These strategies, however, with the exception of thinning, produce additional smoke that can impact the health of nearby communities (Liu et al 2017). In recognition of this, there is a need for enhanced communication before, during, and after prescribed fire, managed fire, or pile burning; and investment in smoke readiness to mitigate the impacts of smoke from these sources on vulnerable community members.
Although this study focused on residents of the ORAEA and Western Colville Reservation communities, these results may be translatable to other rural and tribal smoke-impacted communities across the western United States. Each of the recommendations above are essential to creating a smoke-ready community. The ORAEA already has several organizations and dedicated individuals working diligently to protect their community during and beyond fire season. Most smoke-impacted communities do not have this. The lessons learned from the ORAEA, which has already taken steps toward smoke-readiness could be a launching point for other rural and tribal smoke-impacted communities. The recommendations listed could each be taken as action items by community-based, public health, or governmental organizations to move toward improved smoke risk communication and the creation of smoke-ready and climate resilient communities.

Strengths and limitations
The ORAEA has been hit consistently and hard by smoke from wildfires over the past 10+ years. These exposures have taught community members and leaders meaningful lessons when it comes to preparedness. Using purposive and snowball sampling, we intentionally selected participants whose expertise and lived experience could uniquely shed light on the challenges and opportunities to communicate smoke risk. Our transdisciplinary, cross-sector research team, including local practitioners and tribal representatives, has led to a variety of dissemination opportunities with partners responsible for risk communication throughout the region. For example, our results have been shared with and enthusiastically received by the Pacific Northwest US Forest Service AirFire Research team, leaders at the CTCR and the Intertribal Timber Council, and members of the U.S. Department of Agriculture's Wildland Fire Mitigation and Management Commission.
While we attempted to provide a sample of perspectives from community members, community leaders, and public agency staff from around the ORAEA, our results are not necessarily representative of other rural and tribal communities and regions in the Pacific Northwest. Moreover, our purposive and snowball sampling are non-random sampling techniques prone to sampling bias. Finally, while, there is a significant agricultural worker population in the ORAEA that was underrepresented in our data; possible reasons for this include the exclusive use of English in recruitment and data collection and a lack of targeted outreach to agricultural workers.

Conclusion
Smoke season is becoming longer and more intense each year, and work is needed to prepare communities as more people are impacted. Heavily smoke-impacted communities such as the ORAEA can provide insight into effective and lacking smoke risk communication strategies. Through key informant interviews and focus group discussions with stakeholders and community members, our project found there is a need to bolster knowledge on both the health effects of smoke exposure and the protective resources that are available to mitigate these exposures. We also found that local sources of information are more trusted, and necessary for effective messaging. Our results, and the positive reception that our recommendations have received across the region, as well as the example set by the ORAEA and Western Colville reservation, demonstrate the potential for findings from this and similar research to improve preparedness and risk communications in affected communities throughout and beyond smoke season.

Data availability statement
All data that support the findings of this study are included within the article (and any supplementary information files).