BACKGROUND AND AIM Pulmonary tuberculosis (TB) is a persistent public health concern, with 1.3 million attributable deaths and 10 million new cases recorded annually worldwide. Low- and middle-income countries account for the greatest proportion of TB cases globally (>95%), where some of the highest levels of air pollution occur. Epidemiological studies of an association, however, have been limited in number. We conducted a small-area study in Ningxia, a low socioeconomic and high TB-burden area in rural China, using TB-registry data (2005 – 2017). METHODS We used TB registry data to determine if annual average concentrations of remote-sensing based estimates of ambient air pollution (particulate matter <2.5 µm; [PM2.5], nitrogen dioxide [NO2] ozone [O3]) were associated with township-level (n=358) TB notification rates. Pollutant effects on age-standardised TB- notification rates (as Incidence Rate Ratios [IRR]) at township-level were estimated using Generalised Estimating Equations. RESULTS Median concentrations of PM2.5, NO2, and O3 were 42 µg/m3 (IQR: 10 µg/m3), 15 ppb (IQR: 4 ppb) and 56 ppb (IQR: 1 ppb), respectively. In single pollutant models adjusted for socio-economic covariates, an interquartile range (IQR) increase (10 µg/m3) in PM2.5 was significantly associated with an increase in TB notification incidence rates (IRR: 1.35; 95% CI: 1.27–1.41). Comparable effects were observed per IQR (4 ppb) increase in NO2 (IRR: 1.23; 95% CI: 1.19–1.28). Ground-level ozone was not associated with TB-notification rate in any models. The observed effects remained consistent in multi-pollutant models, and when adjusted for indicators of household crowding, solid fuel use and remoteness. CONCLUSIONS These findings may suggest that improving air quality may have population-level health benefits in socio-economically disadvantaged, TB-endemic settings. KEYWORDS Tuberculosis, TB, respiratory disease, China, outdoor air pollution