Articles | Volume 17, issue 2
Atmos. Chem. Phys., 17, 1511–1528, 2017
Atmos. Chem. Phys., 17, 1511–1528, 2017

Research article 31 Jan 2017

Research article | 31 Jan 2017

The impact of resolution on meteorological, chemical and aerosol properties in regional simulations with WRF-Chem

Paola Crippa1, Ryan C. Sullivan2, Abhinav Thota3, and Sara C. Pryor2,3 Paola Crippa et al.
  • 1COMET, School of Civil Engineering and Geosciences, Cassie Building, Newcastle University, Newcastle upon Tyne, NE1 7RU, UK
  • 2Department of Earth and Atmospheric Sciences, Bradfield Hall, 306 Tower Road, Cornell University, Ithaca, NY 14853, USA
  • 3Pervasive Technology Institute, Indiana University, Bloomington, IN 47405, USA

Abstract. Limited area (regional) models applied at high resolution over specific regions of interest are generally expected to more accurately capture the spatiotemporal variability of key meteorological and climate parameters. However, improved performance is not inevitable, and there remains a need to optimize use of numerical resources and to quantify the impact on simulation fidelity that derives from increased resolution. The application of regional models for climate forcing assessment is currently limited by the lack of studies quantifying the sensitivity to horizontal spatial resolution and the physical–dynamical–chemical schemes driving the simulations. Here we investigate model skill in simulating meteorological, chemical and aerosol properties as a function of spatial resolution, by applying the Weather Research and Forecasting model with coupled Chemistry (WRF-Chem) over eastern North America at different resolutions. Using Brier skill scores and other statistical metrics it is shown that enhanced resolution (from 60 to 12 km) improves model performance for all of the meteorological parameters and gas-phase concentrations considered, in addition to both mean and extreme aerosol optical depth (AOD) in three wavelengths in the visible relative to satellite observations, principally via increase of potential skill. Some of the enhanced model performance for AOD appears to be attributable to improved simulation of meteorological conditions and the concentration of key aerosol precursor gases (e.g., SO2 and NH3). Among other reasons, a dry bias in the specific humidity in the boundary layer and a substantial underestimation of total monthly precipitation in the 60 km simulations are identified as causes for the better performance of WRF-Chem simulations at 12 km.

Short summary
Here we quantify WRF-CHEM sensitivity in simulating meteorological, chemical and aerosol properties as a function of spatial resolution. We demonstrate that WRF-Chem at high resolution improves model performance of meteorological and gas-phase parameters and of mean and extreme aerosol properties over North America. A dry bias in specific humidity and precipitation in the coarse simulations is identified as cause of the better performance of the high-resolution simulations.
Final-revised paper