Interpolating station quantile biases for tropospheric ozone MDA8 bias correction

Abstract

Chemistry transport models (CTMs) consistently exhibit systematic errors in ozone concentrations, which can be partly compensated by bias correction. There are several bias correction strategies suitable for using station data, but they are likely to introduce statistical artifacts when applied at high resolution. We propose a new bias correction strategy based on parametric interpolation of quantile biases (PIQB) suitable for high resolution simulations, which is designed to avoid such artifacts. In this study, we evaluate and compare the performance of our strategy with other older strategies with a focus on ambient maximum daily 8 h average ozone concentrations (MDA8). Our experimental setup consisted of two simulations from the CTMs WRF-Chem and CAMx at horizontal resolution of 9 km within the time period of 2007-2016 and 165 ground-based stations in central Europe. Our results show that each strategy brings the simulated MDA8 closer to observations, but PIQB performs the best in terms of mitigating systematic errors while retaining the modeled fine resolution structure of spatial variability. We conclude that out of the considered strategies, PIQB is the most suitable one for bias correction at high resolution, suggesting its possible applications for correcting climate projections of ozone MDA8.