Skip to main content
NCBI home page

This is a preprint.

It has not yet been peer reviewed by a journal.

The National Library of Medicine is running a pilot to include preprints that result from research funded by NIH in PMC and PubMed.

ArXiv logoLink to ArXiv
[Preprint]. 2025 May 22:arXiv:2505.16613v1. [Version 1]

Long-term impact of PM2.5 on mortality is exacerbated when wildfire events occur

Federica Spoto 1,2,*, Francesca Dominici 1, Danielle Braun 1,3, Joan A Casey 4,5
PMCID: PMC12136479  PMID: 40470474

Abstract

There is extensive evidence that long-term exposure to all-source PM2.5 increases mortality. However, to date, no study has evaluated whether this effect is exacerbated in the presence of wildfire events. Here, we study 60+ million older US adults and find that wildfire events increase the harmful effects of long-term all-source PM2.5 exposure on mortality, providing a new and realistic conceptualization of wildfire health risks.

There is extensive and consistent evidence that long-term exposure to PM2.5 from all sources has an adverse effect on human health and increases the risk of mortality [1, 2, 3, 4, 5, 6]. This robust evidence base led to a recent revision of the annual standard for the US National Ambient Air Quality Standards (NAAQS) for PM2.5, reducing the allowable threshold from 12 μg/m3 to 9 μg/m3. These more stringent standards are expected to yield significant public health benefits, including the prevention of up to 4,500 premature deaths per year across the US. Moreover, the net economic benefits—accounting for reduced healthcare expenditures, fewer workdays lost, and other health-related gains minus the estimated costs of regulatory compliance—are projected to reach approximately $46 billion by 2032 [7].

The increasing frequency and geographic scope of wildfires and their smoke have increased interest in their health impacts. Climate-induced increases in fuel aridity, historical fire suppression, and the growing human presence in wildland areas have driven increased wildfire activity.[8] Since 2000, an average of 70,025 wildfires burned 7.0 million acres annually in the US, more than twice the average of the 1990s [9, 10]. Beginning in 2016, these events have reversed decades of progress in reducing PM2.5 concentrations in nearly three-quarters of US states [11].

Several studies have reported an association between short- and medium-term exposure to wildfire events and higher rates of emergency department visits, hospitalizations, and consultations for respiratory and cardiovascular problems [12, 13, 14, 15], as well as a higher risk of all-cause mortality [16, 17, 18] and and deaths due to COVID-19 [19]. In these studies, wildfire events were defined either based on the levels of wildfire PM2.5–fine particulate matter originating from wildfire smoke–or all-source PM2.5 during periods when wildfires burned. Recent studies have documented associations between long-term exposure to wildfire PM2.5 and adverse respiratory, mental, and overall health outcomes, including an increased risk of mortality [20, 21, 22].

Despite extensive research, we lack evidence on whether the risk conferred by long-term exposure to all-source PM2.5 becomes amplified when wildfire events occur. Our study addresses the research question ”Do wildfire events exacerbate the adverse mortality effects associated with all-source PM2.5?”

Our study cohort comprises more than 60 million US adults aged 65 years and older, with a total of 17,608,624 recorded deaths from 2007 to 2016. Exposure to PM2.5 is characterized as the annual average concentration of all-source PM2.5 between 2006 and 2015, as utilized by Wu et al. [2], derived from the daily estimates developed by Di et al. [23]. The potential effect modification by wildfire events is defined as the ”number of days with non-zero wildfire PM2.5 per year” [24]. Estimates of daily wildfire PM2.5 are obtained from a previously validated daily model for the contiguous US over the period 2006–2015 [25]. This metric effectively captures both the duration and frequency of exposure to wildfire events in each year. For our analyses, we categorize this effect modifier into three distinct strata based on annual counts of wildfire event days: (1) 0 to 20 non-zero wildfire PM2.5 days per year (approximating the first tertile); (2) 21 to 35 non-zero wildfire PM2.5 days per year (approximating the second tertile); and (3) more than 35 non-zero wildfire PM2.5 days per year. Figure 1 illustrates the spatial and temporal distribution of these strata across the US. Exposure and effect modification variables are aggregated at the ZIP code level and linked to Medicare beneficiaries according to their residential ZIP code and year of assessment. This linkage enables us to associate mortality data with exposure measures and effect modifiers from the previous year.

Figure 1: Wildfire event distribution.

Figure 1:

Open in a new tab

The plot represents the distribution of the three wildfire event day strata over the contiguous US for the years 2006, 2010, and 2015. The three wildfire event day strata are (1) 0 to 20 days with non-zero wildfire PM2.5 per year, (2) 21 to 35 days with non-zero wildfire-PM2.5 per year, and (3) more than 35 days with non-zero wildfire PM2.5 per year.

Within each of the three wildfire event days strata, we estimate the association between annual all-source PM2.5 levels and all-cause mortality counts in the subsequent year for each ZIP code using Poisson regression models allowing for a nonlinear exposure–response relationship between all-source PM2.5 and mortality. These models are also stratified by individual-level characteristics and follow-up year. To control for additional potential confounding, we adjust for 10 ZIP code or county-level covariates, four ZIP code–level meteorological variables, and indicators for geographic region (Northeast, South, Midwest, West) and calendar year [2]. To examine additional potential effect modification by geography or socioeconomic status, we further stratify the cohort by region and ZIP code-level poverty, using 2010 US Census data. ZIP codes are categorized as high or low poverty based on whether 15% or more of the population lives below the federal poverty level. Geographic and poverty-related modifiers, along with potential confounders, were assigned for the year corresponding to the outcome.

Our main parameter of interest is the relative change in all-cause mortality risk associated with deviations from the current National Ambient Air Quality Standard (NAAQS) for annual PM2.5 levels (9 μg/m3), under each strata (0–20, 21–35, and >35 wildfire PM2.5 days per year). During the study period, the mean (SD) level of annual average all-source PM2.5 exposure was 9.0 μg/m3 (2.7 μg/m3) and wildfire PM2.5 was 0.37 μg/m3 (0.34 μg/m3). Table S1 describes the cohort characteristics across the contiguous US during the study period.

We find that wildfire event days modify the relationship between all-source PM2.5 and mortality (Figure 2). Being exposed to a higher number of days with non-zero wildfire PM2.5 during the year was associated with higher mortality risk in the following year, with the highest risk for those exposed to 12 μg/m3 of all-source PM2.5 and experiencing more than 35 days with non-zero wildfire PM2.5 days per year.

Figure 2: Mortality hazard ratio with 95% confidence intervals and all-source PM2.5 distribution.

Figure 2:

Open in a new tab

The plots illustrate the exposure-response curve and distribution of all-source PM2.5 values, ranging from the 1st to the 99st percentile of the entire cohort all-source PM2.5 distribution. The top plots represent the hazard ratio of mortality compared to the current NAAQS for annual all-source PM2.5(9 μg/m3) for the entire cohort and stratifying by the three wildfire days strata. The lower plots show the distribution of all-source PM2.5 for the entire cohort, as well as stratified by the three wildfire day strata.

When further stratifying the cohort by ZIP code poverty levels, we find that the all-source PM2.5-mortality association is further modified, especially at lower levels of annual all-source PM2.5 exposure where the exposure–response relationship is steeper in higher-poverty ZIP codes (Figure 3). This finding indicates that incremental increases in PM2.5 concentrations at relatively low baselines may lead to disproportionately higher mortality risks in socioeconomically disadvantaged communities. At higher levels of all-source PM2.5, we find different risk patterns based on area-level poverty. In lower-poverty ZIP codes, increases in mortality risk are more pronounced with higher numbers of wildfire events. In contrast, in ZIP codes where the poverty rate exceeds 15% we observe a different pattern. The mortality risks associated with all-source PM2.5 exposure modified by wildfire event are comparable for areas experiencing 0–20 and 21–35 wildfire event days annually, with risk curves largely overlapping. However, a marked increase in risk emerges for areas experiencing more than 35 wildfire event days per year, and risk does not appear to plateau at high all-source PM2.5 as in other strata and in the lower-poverty ZIP codes.

Figure 3: Mortality hazard ratio with 95% confidence intervals and all-source PM2.5 distribution stratifying by poverty level.

Figure 3:

Open in a new tab

The plots illustrate the exposure-response curve and distribution of all-source PM2.5 values, ranging from the 1st to the 99st percentile of the entire cohort all-source PM2.5 distribution. The top plots represent the hazard ratio of mortality compared to the current NAAQS for annual all-source PM2.5(9 μg/m3) stratifying by poverty level. The lower plots represent the distribution of all-source PM2.5 for the two poverty-based strata.

We then evaluated region differences by stratifying the overall all-source PM2.5-mortality association by census region. We observed varying shapes of the exposure-response curve by census region without considering wildfire event day effect modification. When adding the number of wildfire days per year as an effect modifier, the results remained variable by region (Figure S1).

In this study, we find that experiencing more days of non-zero wildfire PM2.5 amplifies the association between long-term exposure to all-source PM2.5 and all-cause mortality in a cohort of >60 million older US adults, providing a new and realistic conceptualization of wildfire health risks.

A. Online Methods

A.1. Study population

Our study cohort comprises 60,999,431 Medicare beneficiaries 65 years and older for the period 2007–2016 in the contiguous US. Medicare claims data is obtained from the Centers for Medicare and Medicaid Services and includes individual-level data on age, sex, race/ethnicity, date of death, and residential ZIP code. We use residential ZIP codes to link exposure, effect modifiers, and area-level covariate information for each patient. Each Medicare beneficiary is assigned a unique patient identifier, which allows for longitudinal tracking. Cohort inclusion began in 2007 for those already enrolled prior to January 1, 2007 or upon their enrollment after 2007. Each enrollee is monitored annually until death or the end of the study period on December 31, 2016.

A.2. Exposure assessment

For the period 2006–2015, we consider the annual average of all-source PM2.5 concentrations at the ZIP code level as the exposure derived from daily estimates by Di et al. [23]. These daily estimates were obtained from an ensemble prediction model [23]. We then obtained ZIP code level estimates by averaging grid cells with centroids inside each ZIP code. For P.O. Box-only ZIP codes, all-source PM2.5 concentrations were assigned based on the nearest grid cell. Finally, annual concentrations were obtained by averaging the daily levels. We assign exposure based on the Medicare enrollee’s residential ZIP code in each calendar year, linking the prior year’s exposure to the current year’s mortality outcome.

A.3. Potential effect modifiers

For the period 2006–2015, we consider the number of days with non-zero wildfire PM2.5 per year as a potential effect modifier of the all-source PM2.5-mortality association. We obtain daily levels of wildfire PM2.5 at 10-km2 grid resolution from Childs et al. [25]. We overlay the 10-km2 grid and the ZIP code boundaries and use area-weighting to obtain daily ZIP code level wildfire PM2.5 estimates. For P.O. Box-only ZIP codes, wildfire PM2.5 concentrations are assigned based on the nearest non-P.O. Box ZIP code. We then use the daily ZIP code estimates of wildfire PM2.5 to compute the annual number of non-zero wildfire PM2.5 days for each ZIP code. Following [24], we count the number of days where wildfire PM2.5 concentrations exceed zero per year in each ZIP code. Lastly, we categorize the annual number of non-zero wildfire PM2.5 days in three strata: (1) 0 to 20 non-zero wildfire PM2.5 days per year (approximating the first tertile); (2) 21 to 35 non-zero wildfire PM2.5 days per year (approximating the second tertile); and (3) more than 35 non-zero wildfire PM2.5 days per year. We assign wildfire event days based on the Medicare enrollee’s residential ZIP code in each calendar year, linking the prior year’s number of days with non-zero wildfire PM2.5 to the current year’s mortality.

In addition, we examined two additional potential effect modifiers: ZIP code–level poverty and US Census region. ZIP code–level poverty was dichotomized, classifying participants based on whether they resided in ZIP codes where 15% or more of the population lived below the poverty threshold, versus those in areas with less than 15% of residents living in poverty. The US Census region was categorized into four geographically distinct strata: Northeast, Midwest, South, and West. Geographic and poverty-related modifiers, along with potential confounders, were assigned for the corresponding year of the mortality outcome.

A.4. Potential confounders

In order to address confounding bias from community-level factors, we use information from various sources, including ZIP code-level socioeconomic status (SES) indicators from the 2000 and 2010 Census and the 2005–2012 American Community Surveys (ACS) and county-level information from the Centers for Disease Control and Prevention’s Behavioral Risk Factor Surveillance System (BRFSS). We spatially align all data with postal ZIP codes to match Medicare resolution. This includes: (i) two county-level variables: average body mass index and smoking rate; (ii) eight ZIP census variables: proportion of Hispanic and black residents, median household income, median home value, proportion of residents in poverty, proportion of residents with a high school diploma, population density, and proportion of residents that own their house; and (iii) four ZIP code-level meteorological variables from Gridmet [26, 27]: summer (June to September) and winter (December to February) average maximum daily temperature and relative humidity. The ZIP code–level meteorological variables are obtained using area-weighted aggregations based on daily temperature and humidity data on 4-km2 gridded rasters from Gridmet via Google Earth Engine. Finally, we create categorical variables for the four geographic US census regions (Northeast, South, Midwest, and West), and calendar year (2007–2016), to help account for any residual or unmeasured spatial and temporal confounding. This set of confounders corresponds to the ones used in [2].

A.5. Data linkage

Mortality data are available at the ZIP code level. ZIP codes that have incomplete data or no Medicare patients are excluded from the analysis (n = 9,882). The analysis includes 34,444 ZIP codes (∼80% of US ZIP codes) with complete data on outcomes, exposures, effect modifiers, and confounders.

A.6. Statistical analysis

We conduct an analysis to examine whether the number of wildfire event days serves as an effect modifier of the all-source PM2.5–mortality association.

To assess this potential effect modification, we implement a stratified Poisson regression approach, where the study cohort is divided into three strata based on the number of days with >0 μg/m3 of wildfire PM2.5 per year: (1) 0–20 days, (2) 21–35 days, and (3) more than 35 days. Within each wildfire day stratum, we independently estimate the exposure–response relationship between annual average all-source PM2.5 levels and mortality. This stratified design allows us to directly compare the hazard ratios for PM2.5 across different levels of wildfire days.

The Poisson model is specified for each stratum using ZIP code–level death counts by calendar year and follow-up year as the outcome, with annual average all-source PM2.5 as a time-varying exposure and the log of person-years included as an offset term. We incorporate natural spline terms to model non-linear associations with PM2.5. Models are adjusted for 10 ZIP code or county-level socioeconomic and demographic covariates, four ZIP code–level meteorological variables, as well as fixed effects for geographic region (Northeast, South, Midwest, West) and calendar year to account for geographic and temporal heterogeneity. To control for individual-level variability in baseline mortality risk, each model is stratified by four individual-level characteristics: (i) a 5-year category of age at entry (65 to 69, 70 to 74, 75 to 79, 80 to 84, 85 to 89, 90 to 94, 95 to 99, and above 100 years of age), (ii) race/ethnicity (Hispanic, Native American and Alaska Native, Non-Hispanic Asian, Non-Hispanic Black, Non-Hispanic White, and other [e.g., multi-racial and missing]), (iii) sex (male or female), and (iv) an indicator variable for Medicaid eligibility (a surrogate for individual-level SES). For more details, see [2].

For each wildfire strata, the Poisson model is expressed as:

logE[death counts]∼nsall-source PM2.5,df+area-level risk factors+meteorological variables+year+region+offset(log[person-years])+strata(age, race, sex, Medicaid status, follow-up year).

The degrees of freedom (df) for the natural spline terms in each model were selected independently, using the Akaike Information Criterion (AIC). The df that minimized the AIC was chosen for each wildfire stratum. Table S2 summarizes the optimal degrees of freedom selected in the Poisson models.

In addition, we consider two additional potential effect modifiers: ZIP code–level poverty and US census region. ZIP code–level poverty is operationalized as a binary variable, dividing the cohort into two groups: those living in ZIP codes where 15% or more of residents live below the federal poverty level, and those living in ZIP codes with less than 15% living in poverty. Census region is categorized into four geographic strata: Northeast, Midwest, South, and West.

We apply the previously defined Poisson regression model across all three wildfire exposure strata within each level of these secondary effect modifiers. That is, the model is fit separately for each wildfire exposure group within each poverty group and each census region. When stratifying by poverty or region, we exclude poverty or region as a covariate from the corresponding models to avoid redundancy and ensure appropriate interpretation of the stratum-specific estimates.

Similar to the main model, the degrees of freedom for the natural spline terms in each model were determined independently. This was achieved by minimizing the Akaike Information Criterion (AIC) for each combination of wildfire category and poverty stratum, as well as for each combination of wildfire category and region. The optimal degrees of freedom selected for the Poisson models in the stratified analyses are presented in Table S2.

Effect estimates are presented as hazard ratios comparing mortality risk at varying levels of annual average all-source PM2.5 to the current US National Ambient Air Quality Standard (NAAQS) of 9 μg/m3. These hazard ratios are estimated separately within each wildfire exposure stratum, providing insight into whether the mortality risk per unit PM2.5 varies depending on wildfire day levels.

B. Code availability

R code implementing the analyses can be found at https://github.com/NSAPH-Projects/totalPM-smokePM-mortality.

Supplementary Material

Supplement 1

Acknowledgments

JA Casey received funding from the National Institute on Aging (R01AG071024).

References

  • [1].Josey Kevin P, Delaney Scott W, Wu Xiao, Nethery Rachel C, DeSouza Priyanka, Braun Danielle, and Dominici Francesca. Air pollution and mortality at the intersection of race and social class. New England Journal of Medicine, 388(15):1396–1404, 2023. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [2].Wu X., Braun D., Schwartz J., Kioumourtzoglou M. A., and Dominici F.. Evaluating the impact of long-term exposure to fine particulate matter on mortality among the elderly. Science Advances, 6(29):eaba5692, 2020. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [3].Di Qian, Wang Yan, Zanobetti Antonella, Wang Yun, Koutrakis Petros, Choirat Christine, Dominici Francesca, and Schwartz Joel D. Air pollution and mortality in the medicare population. New England Journal of Medicine, 376(26):2513–2522, 2017. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [4].Moon Jeongmin, Kim Ejin, Jang Hyemin, Song Insung, Kwon Dohoon, Kang Cinoo, Oh Jieun, Park Jinah, Kim Ayoung, Choi Moonjung, et al. Long-term exposure to pm2. 5 and mortality: a national health insurance cohort study. International Journal of Epidemiology, 53(6):dyae140, 2024. [DOI] [PubMed] [Google Scholar]
  • [5].Brunekreef Bert, Strak Maciej, Chen Jie, Andersen Zorana J , Atkinson Richard, Bauwelinck Mariska, Bellander Tom, Boutron Marie-Christine, Brandt Jørgen, Carey Iain, et al. Mortality and morbidity effects of long-term exposure to low-level pm2. 5, bc, no2, and o3: an analysis of european cohorts in the elapse project. Research Reports: Health Effects Institute, 2021, 2021. [PMC free article] [PubMed]
  • [6].Pun Vivian C , Kazemiparkouhi Fatemeh, Manjourides Justin, and Suh Helen H. Long-term pm2. 5 exposure and respiratory, cancer, and cardiovascular mortality in older us adults. American journal of epidemiology, 186(8):961–969, 2017. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [7].United States Environmental Protection Agency. Final rule to strengthen the national air quality health standard for particulate matter fact sheet. https://www.epa.gov/system/files/documents/2024-02/pm-naaqs-overview.pdf, 2024.
  • [8].Parks Sean Aand Abatzoglou John T. Warmer and drier fire seasons contribute to increases in area burned at high severity in western us forests from 1985 to 2017. Geophysical Research Letters, 47(22):e2020GL089858, 2020. [Google Scholar]
  • [9].United States Environmental Protection Agency. Climate change indicators: Wildfires. https://www.epa.gov/climate-indicators/climate-change-indicators-wildfires, 2023. Accessed: 2025-02-18.
  • [10].National Interagency Fire Center. Fire information: Statistics. https://www.nifc.gov/fire-information/statistics, 2025. Accessed: 2025-02-18.
  • [11].Burke Marshall, Childs Marissa L., de la Cuesta Brandon, Qiu Minghao, Li Jessica, Gould Carlos F., Heft-Neal Sam, and Wara Michael. The contribution of wildfire to pm2.5 trends in the usa. Nature, 622(7984):761–766, September 2023. [DOI] [PubMed] [Google Scholar]
  • [12].Chen Hao, Samet James M., Bromberg Philip A., and Tong Haiyan. Cardiovascular health impacts of wildfire smoke exposure. Particle and Fibre Toxicology, 18(1), January 2021. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [13].Groot Emily, Caturay Alexa, Khan Yasmin, and Copes Ray. A systematic review of the health impacts of occupational exposure to wildland fires. International Journal of Occupational Medicine and Environmental Health, March 2019. [DOI] [PubMed]
  • [14].To Patricia, Eboreime Ejemai, and Agyapong Vincent I. O.. The impact of wildfires on mental health: A scoping review. Behavioral Sciences, 11(9):126, September 2021. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [15].Reid Colleen E. Focus on wildfires: impacts on health in the context of climate change. Environmental Research: Health, 2(4):040301, September 2024. [Google Scholar]
  • [16].Chen Gongbo, Guo Yuming, Yue Xu, Tong Shilu, Gasparrini Antonio, Bell Michelle L, Armstrong Ben, Schwartz Joel, Jaakkola Jouni J K, Zanobetti Antonella, Lavigne Eric, Nascimento Saldiva Paulo Hilario, Kan Haidong, Royé Dominic, Milojevic Ai, Overcenco Ala, Urban Aleš, Schneider Alexandra, Entezari Alireza, Vicedo-Cabrera Ana Maria, Zeka Ariana, Tobias Aurelio, Nunes Baltazar, Alahmad Barrak, Forsberg Bertil, Pan Shih-Chun, Íñiguez Carmen, Ameling Caroline, De la Cruz Valencia César, Åström Christofer, Houthuijs Danny, Do Van Dung, Samoli Evangelia, Mayvaneh Fatemeh, Sera Francesco, Carrasco-Escobar Gabriel, Lei Yadong, Orru Hans, Kim Ho, Holobaca Iulian-Horia, Kyselý Jan, Paulo Teixeira João, Madureira Joana, Katsouyanni Klea, Hurtado-Díaz Magali, Maasikmets Marek, Ragettli Martina S, Hashizume Masahiro, Stafoggia Massimo, Pascal Mathilde, Scortichini Matteo, Stagliorio Coêlho Micheline de Sousa Zanotti, Ortega Nicolás Valdés, IRyti Niilo R , Scovronick Noah, Matus Patricia, Goodman Patrick, Garland Rebecca M, Abrutzky Rosana, Garcia Samuel Osorio, Rao Shilpa, Fratianni Simona, Dang Tran Ngoc, Colistro Valentina, Huber Veronika, Lee Whanhee, Seposo Xerxes, Honda Yasushi, Guo Yue Leon, Ye Tingting, Yu Wenhua, Abramson Michael J, Samet Jonathan M, and Li Shanshan. Mortality risk attributable to wildfire-related pm2·5 pollution: a global time series study in 749 locations. The Lancet Planetary Health, 5(9):e579–e587, September 2021. [DOI] [PubMed] [Google Scholar]
  • [17].Ye Tingting, Xu Rongbin, Yue Xu, Chen Gongbo, Yu Pei, Coêlho Micheline S. Z. S., Saldiva Paulo H. N., Abramson Michael J., Guo Yuming, and Li Shanshan. Short-term exposure to wildfire-related pm2.5 increases mortality risks and burdens in brazil. Nature Communications, 13(1), December 2022. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [18].Doubleday Annie, Schulte Jill, Sheppard Lianne, Kadlec Matt, Dhammapala Ranil, Fox Julie, and Isaksen Tania Busch. Mortality associated with wildfire smoke exposure in washington state, 2006–2017: a case-crossover study. Environmental Health, 19(1), January 2020. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [19].Zhou Xiaodan, Josey Kevin, Kamareddine Leila, Caine Miah C., Liu Tianjia, Mickley Loretta J., Cooper Matthew, and Dominici Francesca. Excess of covid-19 cases and deaths due to fine particulate matter exposure during the 2020 wildfires in the united states. Science Advances, 7(33), August 2021. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [20].Connolly Rachel, Marlier Miriam E., Garcia-Gonzales Diane A., Wilkins Joseph, Su Jason, Bekker Claire, Jung Jihoon, Bonilla Eimy, Burnett Richard T., Zhu Yifang, and Jerrett Michael. Mortality attributable to pm 2.5 from wildland fires in california from 2008 to 2018. Science Advances, 10(23), June 2024. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [21].Ma Yiqun, Zang Emma, Liu Yang, Wei Jing, Lu Yuan, Krumholz Harlan M., Bell Michelle L., and Chen Kai. Long-term exposure to wildland fire smoke pm 2.5 and mortality in the contiguous united states. Proceedings of the National Academy of Sciences, 121(40), September 2024. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [22].Wei Jing, Wang Jun, Li Zhanqing, Kondragunta Shobha, Anenberg Susan, Wang Yi, Zhang Huanxin, Diner David, Hand Jenny, Lyapustin Alexei, Kahn Ralph, Colarco Peter, da Silva Arlindo, and Ichoku Charles. Long-term mortality burden trends attributed to black carbon and pm2·5 from wildfire emissions across the continental usa from 2000 to 2020: a deep learning modelling study. The Lancet Planetary Health, 7(12):e963–e975, December 2023. [DOI] [PubMed] [Google Scholar]
  • [23].Di Qian, Amini Heresh, Shi Liuhua, Kloog Itai, Silvern Rachel, Kelly James, Sabath M. Benjamin, Choirat Christine, Koutrakis Petros, Lyapustin Alexei, Wang Yujie, Mickley Loretta J., and Schwartz Joel. An ensemble-based model of pm2.5 concentration across the contiguous united states with high spatiotemporal resolution. Environment International, 130:104909, 2019. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [24].Casey Joan A., Kioumourtzoglou Marianthi-Anna, Padula Amy, González David J. X., Elser Holly, Aguilera Rosana, Northrop Alexander J., Tartof Sara Y., Mayeda Elizabeth Rose, Braun Danielle, Dominici Francesca, Eisen Ellen A., Morello-Frosch Rachel, and Benmarhnia Tarik. Measuring long-term exposure to wildfire pm 2.5 in california: Time-varying inequities in environmental burden. Proceedings of the National Academy of Sciences, 121(8), February 2024. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [25].Childs Marissa L, Li Jessica, Wen Jeffrey, Heft-Neal Sam, Driscoll Anne, Wang Sherrie, Gould Carlos F, Qiu Minghao, Burney Jennifer, and Burke Marshall. Daily local-level estimates of ambient wildfire smoke pm2. 5 for the contiguous us. Environmental Science & Technology, 56(19):13607–13621, 2022. [DOI] [PubMed] [Google Scholar]
  • [26].Abatzoglou John T. Development of gridded surface meteorological data for ecological applications and modelling. International journal of climatology, 33(1):121–131, 2013. [Google Scholar]
  • [27].Gorelick Noel, Hancher Matt, Dixon Mike, Ilyushchenko Simon, Thau David, and Moore Rebecca. Google earth engine: Planetary-scale geospatial analysis for everyone. Remote sensing of Environment, 202:18–27, 2017. [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplement 1
NIHPP2505.16613v1-supplement-1.pdf (700.4KB, pdf)

Articles from ArXiv are provided here courtesy of arXiv

RESOURCES