•Long-term air pollution (PM10) exposure was associated with higher fat mass index and lower lean mass index in children and adolescents.•Estimates from longitudinal analyses investigating associations between environmental exposures and changes in body composition over an average of 4 years, were positive for PM10 for certain conditions but not for other exposures.•Railway and road-traffic noise exposures were not associated with body composition, both in cross-sectional and longitudinal analyses. While long-term air pollution and noise exposure has been linked to increasing cardiometabolic disease risk, potential effects on body composition remains unclear. This study aimed to investigate the associations of long-term air pollution, noise and body composition. We used repeated data from the LEAD (Lung, hEart, sociAl, boDy) study conducted in Vienna, Austria. Body mass index (BMI; kg/m2), fat mass index (FMI; z-score), and lean mass index (LMI; z-score) were measured using dual-energy x-ray absorptiometry at the first (t0; 2011-ongoing) and second (t1; 2017-ongoing) examinations. Annual particulate matter (PM10) and nitrogen dioxide (NO2) concentrations were estimated with the GRAMM/GRAL model (2015–2021). Day-evening-night (Lden) and night-time (Lnight) noise levels from transportation were modeled for 2017 following the European Union Directive 2002/49/EC. Exposures were assigned to residential addresses. We performed analyses separately in children/adolescents and adults, using linear mixed-effects models with random participant intercepts and linear regression models for cross-sectional and longitudinal associations, respectively. Models were adjusted for co-exposure, lifestyle and sociodemographics. A total of 19,202 observations (nt0 = 12,717, nt1 = 6,485) from participants aged 6–86 years (mean age at t0 = 41.0 years; 52.9 % female; mean PM10 = 21 µg/m3; mean follow-up time = 4.1 years) were analyzed. Among children and adolescents (age ≤ 18 years at first visit), higher PM10exposure was cross-sectionally associated with higher FMI z-scores (0.09 95 % Confidence Interval (CI): 0.03, 0.16) and lower LMI z-scores (−0.05 95 % CI: −0.10, −0.002) per 1.8 µg/m3. Adults showed similar trends in cross-sectional associations as children, though not reaching statistical significance. We observed no associations for noise exposures. Longitudinal analyses on body composition changes over time yielded positive associations for PM10, but not for other exposures. Air pollution exposure, mainly PM10, was cross-sectionally and longitudinally associated with body composition in children/adolescents and adults. Railway/road-traffic noise exposures showed no associations in both cross-sectional and longitudinal analyses.