O₃‐NO_<i>x</i>‐VOC sensitivity and NO_<i>x</i>‐VOC indicators in Paris: Results from models and Atmospheric Pollution Over the Paris Area (ESQUIF) measurements

Abstract

A three‐dimensional photochemical model has been used to interpret aircraft measurements from the Atmospheric Pollution Over the Paris Area campaign near Paris, with special attention to measurements that are related to predicted O 3 ‐NO x ‐volatile organic compound (VOC) sensitivity. The model (CHIMERE) includes a representation of ozone formation over Europe and a more detailed spatial representation of the region around Paris. A series of model scenarios were developed with varying wind speeds and emission rates. Comparisons are shown with measured O 3 , total reactive nitrogen (NO y ), summed VOCs, and isoprene. Results show that model NO x ‐VOC sensitivity predictions are correlated with the ratio O 3 /NO y but not with O 3 /peroxyacetyl nitrate. Measured O 3 and NO y on high‐ozone days tends to agree with model values when models predict NO x ‐sensitive or transitional chemistry but not when models predict VOC‐sensitive chemistry. Model values for O 3 /NO y and the O 3 ‐NO y slope are lower than measured values, suggesting the possibility of missing, unmeasured VOCs in the Paris plume. Standard performance tests for ozone models, such as normalized bias, show good agreement between models and measurements, even in cases when significant differences appear in the O 3 ‐NO y correlation. Model predictions shift slightly toward NO x ‐sensitive chemistry when model wind speeds are increased. Isoprene represents 20% of total VOC reactivity–weighted carbon in the center of the Paris plume and 50% in the surrounding rural area during high‐ozone events.