Exposure to environmental toxins can have marked impacts on brain structure and function, particularly among children and adolescents for whom vulnerability to toxic exposures is the greatest. Herein, we explored whether there were unique and interactive effects of exposures to two ubiquitous toxins, namely indoor radon and ambient outdoor PM2.5, on sensitive subcortical brain morphology in a sample of youth. Sixty-three adolescents ages 12-to-17 years-old underwent 3 T MRI scans from which we measured total bilateral subcortical gray matter volume for each of seven nuclei. Parents of participants also completed a home radon test, and provided a home address from which we derived 5-year average PM2.5 concentrations within a 1-kilometer squared area surrounding their home via spatiotemporal modeling. We computed chronic exposure indices for both PM2.5 and radon based on the measured concentrations multiplied by the duration of time that the child had lived in the current residence. Using hierarchical regression modeling, we found that youth had significantly larger nucleus accumbens (b = 50.473, β =.324, pFDR =.037) and amygdala volumes (b = 114.141, β =.249, pFDR =.047) as a function of increasing home radon exposure. We did not detect any significant associations between subcortical morphology and PM2.5 exposure, nor did we find any interactive effects of radon and PM2.5. These findings suggest an important potential effect of long-term home radon exposure on the morphology of the extended amygdala circuitry, and impress the need for further study into the functional implications of these structural aberrations in youth.

Acute exposure to surface nitrogen dioxide (NO 2 ) poses substantial global health risks. Using machine learning, we generated the first global, daily, 1-km-resolution, gap-free surface NO 2 data from satellite observations for 2018–2022. Surface NO 2 shows strong day-to-day variability, with fluctuations reaching 73% of the global mean. We identify pronounced pollution hotspots and large global inequalities in acute NO 2 exposure: although only 28% of inhabited land exceeds the WHO daily guideline (25 μg m -3 ) at least once per year, these exceedances disproportionately affect more than 61% of the global population and nearly all megacities (98%). When exceedances are aggregated over 7-day and 30-day windows, 77% and 56% of megacities remain exposed. The decline in NO 2 -affected areas outpaces reductions in population exposure, highlighting the challenge of mitigating impacts in densely populated urban centers. Acute NO 2 exposure caused approximately 576,000 (95% CI: 473,000–678,000) premature deaths globally in 2019. COVID-19 strictest lockdowns in 2020 temporarily reduced NO 2 levels, but rebounds occurred in 91% of countries by 2022. These results reveal the widespread and under-recognized burden of acute NO 2 exposure and emphasize the need for high-resolution global monitoring to support effective pollution control.

Exposure to fine particulate matter (PM2.5) in ambient air is recognized as the leading environmental risk factor for mortality. A more comprehensive characterization of its chemical composition is needed for its management and health effects research. We improve estimates of total PM2.5 mass concentration and its chemical composition across North America by developing, optimizing, and applying convolutional neural networks (CNN) with information from satellite-, simulation-, and monitor-based sources to estimate the local bias in monthly geophysical a priori PM2.5 and component concentrations over 2000-2023. Significant long-term agreement is found with traditional 10-fold spatial cross-validation for total PM2.5 (R 2 = 0.82), sulfate (R 2 = 0.98), nitrate (R 2 = 0.93), ammonium (R 2 = 0.94), organic matter (R 2 = 0.83), black carbon (R 2 = 0.78), dust (R 2 = 0.71), and seasalt (R 2 = 0.37). We introduce Buffered Leave Isolated Sites and Clusters Out (BLISCO) spatial cross-validation to evaluate the model extrapolation ability over remote regions, and find that traditional spatial cross-validation may overestimate performance and underrepresent uncertainty due to the spatial autocorrelation of ground monitors. The use of geophysical information from a chemical transport model (GEOS-Chem) significantly increases CNN performance in BLISCO cross-validation, for example, increasing R 2 for NO3 - (0.51 to 0.81) and NH4 + (0.27 to 0.67). We represent spatial uncertainty for PM2.5 and its components based on the statistical results of BLISCO cross-validation by integrating information from both the spatial distribution of ground observations and the variability in predictors space representation, and find that distance from monitor is a key predictor of uncertainty.

Background: The eye has shown potential as a reliable, readily accessible and clinically relevant site for investigating patients with multiple sclerosis (pwMS). Optical coherence tomography angiography (OCTA) shows promise in revealing new metabolic and vascular elements driving multiple sclerosis (MS) disease pathology. This study aimed to explore correlations between OCTA parameters and clinical characteristics in newly diagnosed relapsing-remitting MS (RRMS) patients. Methods: In this cross-sectional study, forty-one newly diagnosed RRMS patients underwent comprehensive evaluations, including neurological examinations, functional and cognitive tests (9-Hole Peg Test, Montreal Cognitive Assessment), and OCT/OCTA scans. Multiple regression analyses assessed correlations between OCT/OCTA parameters and baseline clinical characteristics. Results: Lower superficial capillary plexus (SCP) vessel density was associated with longer disease duration, higher EDSS scores (visual, pyramidal, cerebellar, ambulation), and impaired 9-Hole Peg Test performance, especially in the non-dominant hand. Higher values of choriocapillaris (CC) flow voids correlated with worse cognitive performance (MoCA). Structural OCT parameters showed limited clinical correlations. Conclusions: OCTA-derived parameters are associated with disability, fine motor function, and cognitive performance in newly diagnosed RRMS patients without prior ON. These findings suggest that retinal vascular alterations may reflect early neurodegenerative processes and provide complementary information beyond structural OCT metrics. OCTA may represent a sensitive, non-invasive imaging biomarker for patient assessment in early MS.

This study investigated the longitudinal progression of retinal structure and microvasculature over 3 years in patients with relapsing–remitting multiple sclerosis (RRMS) using optical coherence tomography (OCT) and OCT angiography (OCTA). It also explored the correlation between these changes and the Expanded Disability Status Scale (EDSS) scores. In this prospective, longitudinal study, we enrolled 66 patients with RRMS without history of optic neuritis and 124 healthy controls. All participants underwent full ophthalmological examination, OCT/OCTA scans, and disability scoring (EDSS) at baseline and after 12 and 24 months. OCT data were analyzed for retinal layer thickness, while OCTA assessed microvascular perfusion in the retinal capillary plexuses and choriocapillaris. Statistical models evaluated yearly rates of change and their association with EDSS scores. The patients with RRMS exhibited 3.6 times faster thinning of the inner plexiform layer (IPL; − 0.47 µm per year, P = 0.001) compared to controls over 3 years. Additionally, superficial retinal capillary layer perfusion density decreased more rapidly at − 0.44

PURPOSE:Multiple sclerosis (MS)-related optic neuritis (ON) causes thinning of inner retinal layers. It remains unclear whether unilateral MSON also affects the unaffected contralateral eye. The purpose of this study was to compare macular retinal layer thicknesses in MS eyes with unilateral optic neuritis (MSON), their unaffected contralateral eyes and MS participants without a history of ON (MSnON). METHODS:This cross-sectional screening study included 101 MSON and 106 MSnON participants. Retinal layer thicknesses were measured using optical coherence tomography scans at the standardized zones of the macula (central circle, inner ring and outer ring) and compared between the groups. RESULTS:The unaffected contralateral non-ON eyes of MSON participants had thinner inner retinal layers including a thinner retinal nerve fibre layer (p values, 0.003-0.009), a thinner ganglion cell layer (p values, <0.001-0.006) and a thinner inner plexiform layer (p values, 0.004-0.012) compared to MS participants without a history of ON. Affected MSON eyes had thinner inner retinal layers compared to both unaffected fellow eyes and to MSnON participants (p < 0.001 for all comparisons). Additionally, in MSON eyes, the inner nuclear layer and outer retina were thicker at the inner and outer rings compared to contralateral eyes (p < 0.001 for all comparisons), but not when compared to the MSnON participants. CONCLUSION:We recommend bilateral examination, OCT imaging and follow-up for MS patients with unilateral acute ON to monitor also the contralateral eyes, which present with thinner inner retina layers than MSnON participants' eyes.

Wildfires can inject smoke at high altitudes into the atmosphere. The resulting free tropospheric aerosols may affect inference of ground-level fine particulate matter (PM2.5) from satellite aerosol optical depth (AOD), yet the effects of accounting for plume height in this inference are poorly understood. Here, we include in the GEOS-Chem chemical transport model a fire plume height parametrization (GFAS, Global Fire Assimilation System) to examine its effect on PM2.5 inferred from satellite AOD during wildfires over the United States and Canada. Comparison with six years satellite observations of plume height reveals a low bias of a factor 1.7 in the GFAS plume height over evergreen needleleaf forests. We scale the GFAS plume height over evergreen needleleaf forests in GEOS-Chem to better represent the satellite observations, focusing on 2018 and 2020 when large wildfires yield prominent signals. Replacing the default ground-level wildfire emissions in GEOS-Chem with the scaled GFAS vertically distributed emissions reduces the bias between measured PM2.5 and PM2.5 inferred from satellite AOD, and significantly improves the consistency of simulated AOD with sun photometer measurements. Overall, this study signifies the importance of vertically distributing wildfire emissions for the inference of PM2.5 from satellite AOD.

Background Excess health risk estimates of exposure per unit mass concentration of fine particulate matter (PM25) still exhibit a wide range, potentially due to variations in aerosol size and composition. Submicron particulate matter (PM1) was recently reported to exert stronger health impacts than PM25 from studies in China, but an absence of long-term PM1 data in the USA has prohibited such investigations despite a wealth of cohorts. This study aims to fill this data gap and estimate PM1 concentrations over 1998-2022 across the USA. Methods We estimated biweekly gapless ambient PM1 concentrations and their uncertainties at 1 km(2) resolution across the contiguous USA over the 25-year period of 1998-2022, from hybrid estimates of PM25 chemical composition that merged information from satellite retrievals, air quality modelling, and ground-based monitoring. The mass fractions of PM25 components with diameters below 1 mu m were constrained by observations for four major components and from established scientific understanding for the other components. Findings PM1 concentrations exhibited pronounced spatial variation across the contiguous USA with enhancements observed in the east, major urban and industrial areas, and areas affected by wildfires; low concentrations are prevalent over the arid west. The main components of population-weighted mean (PWM) PM1 in 2022 (61 mu g/m(3)) were organic matter (47%), sulphate (22%), nitrate (12%), black carbon (8%), and ammonium (7%). The biweekly PM1 estimates were highly consistent with independent ground-based PM1 measurements (slope=096, R-2=078). The estimated 1-sigma uncertainties of annual mean PM1 for the 25 years over more than 8 million land pixels were less than 20% for 98% of data points, while 03% of the population of the contiguous USA was associated with uncertainties of more than 30% due to wildfires. The PWM PM1 decreased significantly (p<00001) at a rate of -023 mu g/m(3) per year during 1998-2022, accounting for 86% of the overall reduction of PWM PM25; the PWM PM1/PM25 ratio experienced simultaneous decrease (-00013 per year, p<00001). Interpretation The dominance of PM1 in PM25 reduction and the decreasing PM1/PM25 ratio reflect the strong association of PM1 with fossil fuel and other combustion sources and their responses to air quality regulations during the 25-year study period. The gradual coarsening of PM25 calls for increasing urgency to separately assess health impacts of PM1 versus PM25, as supported by the quality of the derived PM1 estimates. Future particulate matter monitoring programmes, health studies, and regulatory deliberations should consider PM1 in addition to PM25.

BACKGROUND:Excess health risk estimates of exposure per unit mass concentration of fine particulate matter (PM2·5) still exhibit a wide range, potentially due to variations in aerosol size and composition. Submicron particulate matter (PM1) was recently reported to exert stronger health impacts than PM2·5 from studies in China, but an absence of long-term PM1 data in the USA has prohibited such investigations despite a wealth of cohorts. This study aims to fill this data gap and estimate PM1 concentrations over 1998-2022 across the USA. METHODS:We estimated biweekly gapless ambient PM1 concentrations and their uncertainties at 1 km2 resolution across the contiguous USA over the 25-year period of 1998-2022, from hybrid estimates of PM2·5 chemical composition that merged information from satellite retrievals, air quality modelling, and ground-based monitoring. The mass fractions of PM2·5 components with diameters below 1 μm were constrained by observations for four major components and from established scientific understanding for the other components. FINDINGS:PM1 concentrations exhibited pronounced spatial variation across the contiguous USA with enhancements observed in the east, major urban and industrial areas, and areas affected by wildfires; low concentrations are prevalent over the arid west. The main components of population-weighted mean (PWM) PM1 in 2022 (6·1 μg/m3) were organic matter (47%), sulphate (22%), nitrate (12%), black carbon (8%), and ammonium (7%). The biweekly PM1 estimates were highly consistent with independent ground-based PM1 measurements (slope=0·96, R2=0·78). The estimated 1-σ uncertainties of annual mean PM1 for the 25 years over more than 8 million land pixels were less than 20% for 98% of data points, while 0·3% of the population of the contiguous USA was associated with uncertainties of more than 30% due to wildfires. The PWM PM1 decreased significantly (p<0·0001) at a rate of -0·23 μg/m3 per year during 1998-2022, accounting for 86% of the overall reduction of PWM PM2·5; the PWM PM1/PM2·5 ratio experienced simultaneous decrease (-0·0013 per year, p<0·0001). INTERPRETATION:The dominance of PM1 in PM2·5 reduction and the decreasing PM1/PM2·5 ratio reflect the strong association of PM1 with fossil fuel and other combustion sources and their responses to air quality regulations during the 25-year study period. The gradual coarsening of PM2·5 calls for increasing urgency to separately assess health impacts of PM1 versus PM2·5, as supported by the quality of the derived PM1 estimates. Future particulate matter monitoring programmes, health studies, and regulatory deliberations should consider PM1 in addition to PM2·5. FUNDING:National Institute of Environmental Health Sciences, National Institutes of Health.

AbstractINTRODUCTIONDiagnostic performance of optical coherence tomography (OCT) to detect Alzheimer's disease (AD) and mild cognitive impairment (MCI) remains limited. We aimed to develop a deep‐learning algorithm using OCT to detect AD and MCI.METHODSWe performed a cross‐sectional study involving 228 Asian participants (173 cases/55 controls) for model development and testing on 68 Asian (52 cases/16 controls) and 85 White (39 cases/46 controls) participants. Features from OCT were used to develop an ensemble trilateral deep‐learning model.RESULTSThe trilateral model significantly outperformed single non‐deep learning models in Asian (area under the curve [AUC] = 0.91 vs. 0.71–0.72, p = 0.022‐0.032) and White (AUC = 0.84 vs. 0.58–0.75, p = 0.056‐ < 0.001) populations. However, its performance was comparable to that of the trilateral statistical model (AUCs similar, p > 0.05).DISCUSSIONBoth multimodal approaches, using deep learning or traditional statistical models, show promise for AD and MCI detection. The choice between these models may depend on computational resources, interpretability preferences, and clinical needs.Highlights A deep‐learning algorithm was developed to detect Alzheimer's disease (AD) and mild cognitive impairment (MCI) using OCT images. The combined model outperformed single OCT parameters in both Asian and White cohorts. The study demonstrates the potential of OCT‐based deep‐learning algorithms for AD and MCI detection.

Acute exposure to surface nitrogen dioxide (NO2) poses substantial health risks worldwide. Using advanced machine learning techniques, we retrieved the first global, daily, gap-free surface NO2 data at 1-km resolution from downscaled TROPOMI satellite observations since its launch to 2022. Global NO2 exhibited pronounced day-to-day variability, with population-weighted values ranging from 11.8 to 23.7 µg m− 3, representing daily fluctuations of up to 73% relative to the mean level of 16.4 µg m− 3. China (23%) and India (18%) contribute most significantly to global NO2 exposure. Although only 28% of inhabited areas experience daily NO2 concentrations exceeding the World Health Organization’s short-term air quality guideline of 25 µg/m3 for at least one day within the year, these exceedances disproportionately affect over 61% of the global population. This disparity is especially pronounced in densely populated megacities, where 98% experience at least one day of unhealthy air quality per year. The proportion remains high (77% and 56%) when extending the threshold to 7-day and 30-day exceedances, respectively. We estimate that acute NO2 exposure contributed to approximately 576,000 (95% CI: 473,000–678,000) premature deaths globally in 2019, while the chronic mortality burden was roughly five times higher, at 2,688,000 (95% CI: 702,000–3,921,000). The strictest COVID-19 pandemic lockdowns implemented in 2020 caused widespread temporary declines in NO2 levels globally, followed by rebounds as restrictions eased in 91% of countries by 2022; however, 59% of countries have maintained NO2 levels below 2019 pre-pandemic baselines. Our findings underscore the significant yet often under-recognized risks posed by acute NO2 exposure, highlighting the urgent need for comprehensive and detailed global exposure assessment to inform the design of effective public health and environmental policies.

Exposure to ambient fine particulate matter (<2.5µm; PM2.5) increases risk for suboptimal brain aging and dementia. Here we examined exposure associations with regional cortical thickness (CTh) across seven international datasets within the ENIGMA-Environment consortium. Data were collected from 2,086 adults (60% female), ages 18-89 years (M=57.6±14.1); 1,711 were cognitively unimpaired controls and 375 had diagnoses of mild cognitive impairment ( n  = 46), dementia ( n  = 51), obsessive-compulsive disorder ( n  = 158), trichotillomania ( n  = 50), methamphetamine use disorder ( n  = 29), social anxiety disorder ( n  = 21), gambling disorder ( n  = 10), or Parkinson's disease ( n  = 10) (Table 1). Satellite-based estimates of residential PM 2.5 were quantified as 3-year average exposures prior to MRI. We tested exposure associations with FreeSurfer-derived CTh (34 bilateral regions) using linear mixed models, adjusting for age, age 2 , sex, education, race, diagnosis, age-by-sex interactions, with random effects of cohort and scanner. Secondarily, we examined PM 2.5 interactions with age and sex. FDR corrections were applied per analysis. PM 2.5 exposure was not associated with CTh in the whole sample but was inversely associated with CTh in the superior temporal gyrus (STG; p FDR =0.03) in controls. We observed a significant quadratic PM 2.5 effect over age across most regions, with negative associations at young and older ages, and positive associations during midlife (Figure 1). In the whole sample only, sex-by-exposure interactions were significant in five frontal regions. Exposures were inversely associated with CTh in males, but were not associated with CTh in females (Figure 2). Interaction effects could not be explained by PM 2.5 exposure differences (t(2084)=-1.5, p  = 0.130), and although age differed significantly by sex (t(2084)=-2.8, p  = 0.005; male=58.7±14.5, female=56.9±13.8), these differences likely are not meaningful. PM 2.5 was inversely associated with CTh in the STG of controls only. Age-by-exposure interactions revealed a quadratic effect of PM2.5 on CTh across the lifespan, possibly indicating biphasic exposure-related neurodegeneration, where CTh increases with early pathology and declines as neurodegeneration progresses. Discordant sex effects may reflect normal sexual dimorphism that is exacerbated by air pollution, but more work is needed to better understand these results.

AbstractMany chemical transport models treat mineral dust as spherical. Solar backscatter retrievals of trace gases (e.g., OMI and TROPOMI) implicitly treat mineral dust as spherical. The impact of the morphology of mineral dust particles is studied to assess its implications for global chemical transport model (GEOS‐Chem) simulations and solar backscatter trace gas retrievals at ultraviolet and visible (UV‐Vis) wavelengths. We investigate how the morphology of mineral dust particles affects the simulated dust aerosol optical depth; surface area, reaction, and diffusion parameters for heterogeneous chemistry; phase function, and scattering weights for air mass factor (AMF) calculations used in solar backscatter retrievals. We use a mixture of various aspect ratios of spheroids to model the dust optical properties and a combination of shape and porosity to model the surface area, reaction, and diffusion parameters. We find that assuming spherical particles can introduce size‐dependent and wavelength‐dependent errors of up to 14% in simulated dust extinction efficiency with corresponding error in simulated dust optical depth typically within 5%. We find that use of spheroids rather than spheres increases forward scattered radiance and decreases backward scattering that in turn decrease the sensitivity of solar backscatter retrievals of NO2 to aerosols by factors of 2.0–2.5. We develop and apply a theoretical framework based on porosity and surface fractal dimension with corresponding increase in the reactive uptake coefficient driven by increased surface area and species reactivity. Differences are large enough to warrant consideration of dust non‐sphericity for chemical transport models and UV‐Vis trace gas retrievals.

Ambient fine particulate matter (PM2.5) is the leading global environmental determinant of mortality. However, large gaps exist in ground-based PM2.5 monitoring. Satellite remote sensing of aerosol optical depth (AOD) offers information to help fill these gaps worldwide when augmented with a modeled PM2.5–AOD relationship. This study aims to understand the spatial pattern and driving factors of this relationship by examining η (PM2.5AOD) using both observations and modeling. A global observational estimate of η for the year 2019 is inferred from 6870 ground-based PM2.5 measurement sites and satellite-retrieved AOD. The global chemical transport model GEOS-Chem, in its high-performance configuration (GCHP), is used to interpret the observed spatial pattern of annual mean η. Measurements and the GCHP simulation consistently identify a global population-weighted mean η value of 96–98 µg m−3, with regional values ranging from 59.8 µg m−3 in North America to more than 190 µg m−3 in Africa. The highest η value is found in arid regions, where aerosols are less hygroscopic due to mineral dust, followed by regions strongly influenced by surface aerosol sources. Relatively low η values are found over regions distant from strong aerosol sources. The spatial correlation of observed η values with meteorological fields, aerosol vertical profiles, and aerosol chemical composition reveals that spatial variation in η is strongly influenced by aerosol composition and aerosol vertical profiles. Sensitivity tests with globally uniform parameters quantify the effects of aerosol composition and aerosol vertical profiles on spatial variability in η, exhibiting a population-weighted mean difference in aerosol composition of 12.3 µg m−3, which reflects the determinant effects of composition on aerosol hygroscopicity and aerosol optical properties, and a population-weighted mean difference in the aerosol vertical profile of 8.4 µg m−3, which reflects spatial variation in the column–surface relationship.

Global fine particulate matter (PM2.5) assessment is impeded by a paucity of monitors. We improve estimation of the global distribution of PM2.5 concentrations by developing, optimizing, and applying a convolutional neural network with information from satellite-, simulation-, and monitor-based sources to predict the local bias in monthly geophysical a priori PM2.5 concentrations over 1998-2019. We develop a loss function that incorporates geophysical a priori estimates and apply it in model training to address the unrealistic results produced by mean-square-error loss functions in regions with few monitors. We introduce novel spatial cross-validation for air quality to examine the importance of considering spatial properties. We address the sharp decline in deep learning model performance in regions distant from monitors by incorporating the geophysical a priori PM2.5. The resultant monthly PM2.5 estimates are highly consistent with spatial cross-validation PM2.5 concentrations from monitors globally and regionally. We withheld 10% to 99% of monitors for testing to evaluate the sensitivity and robustness of model performance to the density of ground-based monitors. The model incorporating the geophysical a priori PM2.5 concentrations remains highly consistent with observations globally even under extreme conditions (e.g., 1% for training, R2 = 0.73), while the model without exhibits weaker performance (1% for training, R2 = 0.51).

Background/aimsTo investigate whether compensating retinal nerve fibre layer (RNFL) thickness measurements for demographic and anatomical ocular factors can strengthen the structure–function relationship in patients with glaucoma.Methods600 eyes from 412 patients with glaucoma (mean deviation of the visual field (MD VF) −6.53±5.55 dB) were included in this cross-sectional study. Participants underwent standard automated perimetry and spectral-domain optical coherence tomography imaging (Cirrus; Carl Zeiss Meditec). Compensated RNFL thickness was computed considering age, refractive error, optic disc parameters and retinal vessel density. The relationship between MD VF and RNFL thickness measurements, with or without demographic and anatomical compensation, was evaluated sectorally and focally.ResultsThe superior arcuate sector exhibited the highest correlation between measured RNFL and MD VF, with a correlation of 0.49 (95% CI 0.37 to 0.59). Applying the compensated RNFL data increased the correlation substantially to 0.62 (95% CI 0.52 to 0.70; p<0.001). Only 61% of the VF locations showed a significant relationship (Spearman’s correlation of at least 0.30) between structural and functional aspects using measured RNFL data, and this increased to 78% with compensated RNFL measurements. In the 10°–20° VF region, the slope below the breakpoint for compensated RNFL thickness demonstrated a more robust correlation (slope=1.66±0.18 µm/dB; p<0.001) than measured RNFL (slope=0.27±0.67 µm/dB; p=0.688).ConclusionCompensated RNFL data improve the correlation between RNFL measurements and VF parameters. This indicates that creating structure-to-function maps that consider anatomical variances may aid in identifying localised structural and functional loss in glaucoma.

Accurate representation of the hourly variation in the NO2-column-to-surface relationship is essential for interpreting geostationary observations of NO2 columns. Previous research indicated inconsistencies in this hourly variation. This study employs the high-performance configuration of the GEOS-Chem model (GCHP) to analyze daytime hourly NO2 total columns and surface concentrations during summer. We use measurements from globally distributed Pandora sun photometers and aircraft observations over the United States. We correct Pandora total NO2 vertical columns for (1) hourly variations in effective temperature driven by vertically resolved contributions to the total column and (2) changes in local solar time along the Pandora line of sight. These corrections increase the total NO2 columns by 5–6 × 1014 molec. cm−2 at 09:00 and 18:00 across all sites. Fine-scale simulations from GHCP (∼12 km) reduce the normalized bias (NB) against Pandora total NO2 columns from 19 % to 10 % and against aircraft measurements from 25 % to 13 % in Maryland, Texas, and Colorado. Similar reductions are observed in NO2 columns over the eastern US (17 % to 9 %), the western US (22 % to 14 %), Europe (24 % to 15 %), and Asia (29 % to 21 %) when compared to 55 km simulations. Our analysis attributes the weaker hourly variability in the total NO2 column to (1) hourly variations in column effective temperature, (2) local solar time changes along the Pandora line of sight, and (3) differences in hourly NO2 variability from different atmospheric layers, with the lowest 500 m exhibiting greater variability, while the dominant residual column above 500 m exhibits weaker variability.

Air quality management benefits from an in-depth understanding of the emissions associated with, and composition of, local PM2.5 concentrations. Here, we investigate the changing role of biomass burning emissions to North American PM2.5 exposure by combining multiple satellite-, ground-, and simulation-based data sets biweekly at a 0.01° × 0.01° resolution from 2000 to 2022. We also developed a Buffered Leave Cluster Out (BLeCO) method to address autocorrelation and computational cost in cross-validation. Biomass burning emissions contribute an increasingly large fraction to PM2.5 exposure in the United States and Canada, with national annual population-weighted mean contributions increasing from 0.4 μg/m3 (3-5%) in 2000-2004 to 0.8-0.9 μg/m3 (9-14%) by 2019-2022, led by western North American 2019-2022 annual contributions of 1.4-1.9 μg/m3 (15-27%) and maximum seasonal contributions of 3.3-5.5 μg/m3 (29-49%). Other components such as nonbiomass burning Organic Matter (OM) and nitrate can be regionally as (or more) important, albeit with distinct seasonal variability. The contribution of total OM to PM2.5 exposure in the United States in 2016-2022 is 42.2%, comparable to all other anthropogenically sourced components combined. Comparison of BLeCO and random 10-fold cross-validation suggests that random 10-fold cross-validation may significantly underrepresent true uncertainty for total PM2.5 concentrations due to the clustered nature of PM2.5 ground-based monitoring.