Application of WRF-chem for predicting air quality resulting from the formation of photochemical compounds in a subtropical urban environment

Modeling ozone (O 3 ) concentrations in urban areas is challenging due to interactions between emissions inventory, meteorology, and atmospheric chemistry. This study evaluates the performance of WRF-Chem in simulating O 3 , NOx, SO 2 , and Ox (= NO 2 + O 3 ) in the Metropolitan Region of Rio de Janeiro (MRRJ), representing the first assessment of WRF-Chem against observations from the INEA automatic monitoring network. In three contrasting periods, the modeled hourly concentrations show moderate agreement with observations (R ≈ 0.6 for NOx, 0.5 for SO 2 , and 0.7 for O₃). O 3 is overestimated during the day (≈ 20 ppb) and underestimated at night (≈ 10 ppb). NO 2 is overestimated at night (≈ 20 ppb) and SO 2 (≈ 25 ppb). Biases are persistent: there is daytime overestimation of O 3 and Ox and nighttime underestimation. These discrepancies result mainly from the model's treatment of emissions inventory, particularly volatile organic compounds (VOCs) and NOx, which are important for the formation of O 3 through photochemical reactions. The results indicate the need to improve the representation of total emission magnitudes, spatial allocation, and diurnal variability of anthropogenic sources, particularly VOCs and NOx, which strongly influence ozone production in the region. Spatial patterns suggest that uncertainties in VOC emissions, combined with urban meteorology (e.g., sea/land-breeze cycles, boundary-layer evolution), may contribute to O 3 overestimation in industrial corridors such as Santa Cruz/RJ and Volta Redonda/RJ; however, this interpretation should be considered a hypothesis consistent with model limitations. For air quality management, the findings suggest priority actions: refining the magnitude, spatial allocation, and diurnal variability of anthropogenic emissions (especially VOC and NOx). These measures can guide contingency plans and O 3 reduction strategies in the state of Rio de Janeiro. This research provides essential guidance for policymakers to implement emission controls and devise effective contingency plans for pollution episodes in tropical metropolitan areas. • The WRF-Chem model shows moderate agreement with observed O3, NOx, and SO2 concentrations. • Daytime O3 is overestimated, while nighttime O3 and Ox are underestimated due to emissions. • Urban/industrial site is an O3 hotspot due to VOC emissions, highlighting the need for a refined inventory.