10th International Aerosol Conference
September 2 - September 7, 2018
America's Center Convention Complex
St. Louis, Missouri, USA

Abstract View


Overprediction of Fine Aerosol Nitrate by Chemical Transport Models: The Role of Nighttime Chemistry and Mixing

Maria Zakoura, SPYROS PANDIS, University of Patras

     Abstract Number: 326
     Working Group: Aerosol Modeling

Abstract
Fine particulate nitrate is often over-predicted by chemical transport models (CTMs) in areas of the Eastern US and Europe. This overprediction is especially severe during a few nights during the summer or other photochemically active periods, when CTMs predict unreasonably high nitric acid production rates. We test the hypothesis that the main reason of this nighttime aerosol nitrate overprediction is the coarse grid resolution used that cannot capture the local phenomena that take place in the plume of major sources. The CTM PMCAMx is used to simulate a summer period over the Eastern US taking advantage of the dense fine nitrate measurement networks (STN and IMPROVE) in that area.
The base case simulation with coarse resolution (36x36 km) often predicted high nighttime nitrate production rates, thus leading to widespread overprediction of aerosol nitrate levels both during the night but also during the next day. We show that this overprediction is due to the artificial mixing by the model of NOx-rich plumes from major point and area sources with the background atmosphere. The bias for PM2.5 nitrate decreased by 65% when the grid resolution was increased to 4x4 km. However, this improvement comes with a significant increase in the computational cost of the simulation.
The ability of a Plume-in-Grid (PiG) approach to increase the accuracy of the model with a small increase in computational cost was also investigated. The PiG sub-model was applied for the gas-phase chemistry only for the major NOx point sources in the Eastern US during the same period. Different CTM grid resolutions were used. The results suggest that the PiG approach can be a computationally efficient method to reduce the nitrate bias in CTMs.

Despite all the above improvements, systematic discrepancies remain between the model predictions and observations. Other potential improvements in both the simulation of nitric acid partitioning, nighttime chemistry, nitric acid removal, and the emissions of ammonia are discussed.