Background: Combustion-related nitrogen dioxide (NO2) air pollution is associated with pediatric asthma incidence. We estimated global surface NO2 concentrations consistent with the Global Burden of Disease Study for 1990-2019 at 1km resolution, and concentrations and attributable pediatric asthma incidence trends in 13,189 cities from 2000-2019. Methods: We scaled an existing surface annual average NO2 concentrations dataset for 2010-2012 from a land use regression model (based on 5,220 NO2 monitors in 58 countries and land use variables) to other years using NO2 column densities from satellite and reanalysis datasets. We applied these concentrations to epidemiologically-derived concentration-response factors, population, and baseline asthma rates to estimate NO2-attributable pediatric asthma incidence. Findings: We estimated that 1.85 million (95% uncertainty interval: 0.93 – 2.8 million) new pediatric asthma cases were attributable to NO2 globally in 2019, two-thirds of which occurred in urban areas. The fraction of pediatric asthma incidence that is attributable to NO2 in urban areas declined from 20% in 2000 to 16% in 2019. Urban attributable fractions dropped in High-income (-41%), Latin America/Caribbean (-16%), Central Europe, Eastern Europe, and Central Asia (-13%), and Southeast Asia, East Asia, and Oceania (-6%), and rose in South Asia (+23%), Sub-Saharan Africa (+11%), and North Africa and Middle East (+5%) regions. The importance of NO2 concentrations, pediatric population size, and asthma incidence rates in driving these changes differs regionally. Interpretation: Despite improvements in some regions, combustion-related NO2 pollution continues to be an important contributor to pediatric asthma incidence globally, particularly in cities. Funding: Health Effects Institute, NASA

The World Health Organization (WHO) recently reduced its health guideline for Nitrogen dioxide (NO 2) to annual and 24-hr means of 10 µg/m 3 (5.3 ppb) and 25 µg/m 3 (13.3 ppb). NO 2 is a criteria air pollutant that varies spatiotemporally at fine resolutions due to its relatively short lifetime (~hours) and current models have limited ability to capture this variation. To advance global exposure estimates, we created a daily global land use regression (LUR) model with 50 x 50 m 2 spatial resolution using 5.7 million daily air monitor averages collected from 8,250 monitor locations. In cross-validation, the model captured 47%, 59%, and 63% of daily, monthly, and annual global NO 2 variation. Daily, monthly, and annual root mean square error were 6.8, 5.0, and 4.4 ppb and absolute bias were 46%, 30%, and 21%, respectively. The final model has 11 variables, including road density and built environments with fine (30 m or less) spatial resolution and meteorological and satellite data with daily temporal resolution. Major roads and satellite-based estimates of NO 2 were consistently the strongest predictors in all regions. Daily model estimates from 2005-2019 are available 1 and can be used for global risk assessments and health studies, particularly in countries without NO 2 monitoring. Short synopsis: This is the first global NO 2 model with daily temporal and 50m spatial resolution, valuable for capturing NO 2 variation.