Importance of dry deposition parameterization choice in global simulations of surface ozone

摘要: Abstract. Dry deposition is a major sink of tropospheric ozone. Increasing evidence has shown that ozone dry actively links meteorology and hydrology with air quality. However, there little systematic investigation on the performance different dry deposition parameterizations at global scale how parameterization choice can impact surface simulations. Here, we present results of the first global, multidecadal modelling evaluation velocity ( vd ) using multiple deposition parameterizations. We model velocities over 1982–2011 using four are representative of current approaches in modelling. use consistent assimilated meteorology, land cover, satellite-derived leaf area index (LAI) across all four, such differences simulated entirely due to structures or assumptions about types treated each. In addition, use the sensitivity predicted by chemical transport model estimate mean variability deposition velocity ozone. Our estimated values from different parameterizations evaluated against field observations, while performance varies considerably cover types, our suggest that none universally better than others. Discrepancy simulated among is estimated cause 2 5 ppbv discrepancy the Northern Hemisphere (NH) up 8 ppbv tropical rainforests July, and seasonally forests in Indochina December. Parameterization-specific biases based on individual type hydroclimate found be two main drivers discrepancies. find statistically significant trends in the multiannual time series July daytime all parameterizations, driven warming drying (southern Amazonia, southern African savannah, Mongolia) or greening (high latitudes). The trend in July 1 % yr −1 leads to 3 ppbv changes 1982–2011. interannual coefficient of variation (CV) NH be 5 %–15 %, spatial distribution deposition parameterization. simulations suggest this contribute between 0.5 2 ppbv (IAV) ozone, but all models tend underestimate CV when compared long-term ozone flux observations. also IAV some deposition parameterizations more sensitive LAI, while others it sensitive to climate. Comparisons other published estimates of background confirm an important part of natural variability. demonstrate importance of parameterization choice modelling and thus making strong case for further measurement, evaluation, model–data integration spatiotemporal scales.