A diagnostic intercomparison of modeled ozone dry deposition over North America and Europe using AQMEII4 regional-scale simulations

This study analyzes ozone (O₃) dry deposition fluxes and velocities (V_d) from 12 regional-scale simulations that were performed over North America and Europe in Phase 4 of the Air Quality Model Evaluation International Initiative (AQMEII4). AQMEII4 collected grid-aggregated and land-use-specific (LU-specific) O₃ V_d and effective conductances and fluxes for the four major dry deposition pathways. Consistent with recent findings in the AQMEII4 point model intercomparison study, analysis of the grid-aggregated fields shows that grid models with a similar V_d can exhibit significant differences in the absolute and relative contributions of the different depositional pathways. Analysis of LU-specific V_d and effective conductances reveals a general increase in model spread compared to grid-aggregated values. This indicates that an analysis of only grid-aggregated deposition diagnostics can mask process-specific differences that exist between schemes. An analysis of AQMEII4 LU distributions across models revealed substantial differences in the spatial patterns and abundance of certain LU categories over both domains, especially for non-forest partially vegetated categories such as agricultural areas, shrubland, and grassland. We demonstrate that these differences can significantly contribute to or even drive differences in LU-specific dry deposition fluxes which can increase the variability in model predictions of ozone. Two recommendations for future deposition-focused modeling studies emerging from the AQMEII4 analyses presented here are to (1) routinely generate diagnostic outputs to advance a process-based understanding of modeled deposition and support impact analyses and (2) recognize the importance of documenting and analyzing the representation of LU across models and work towards harmonizing this aspect when using air quality grid models and model ensembles for deposition analyses.