TY - JOUR
T1 - Evaluation of a CMAQ simulation at high resolution over the UK for the calendar year 2003
AU - Chemel, C.
AU - Sokhi, R.S.
AU - Yu, Y.
AU - Hayman, G.D.
AU - Vincent, K.J.
AU - Dore, A.J.
AU - Tang, Y.S.
AU - Prain, H.D.
AU - Fisher, B.
PY - 2010
Y1 - 2010
N2 - A comprehensive ‘operational’ evaluation of the performance of the Community Multiscale Air Quality (CMAQ) modelling system version 4.6 was conducted in support of pollution assessment in the UK for the calendar year 2003. The model was run on multiple grids using one-way nests down to a horizontal resolution as fine as 5 km over the whole of the UK. The model performance was evaluated for pollutants with standards and limit values (e.g. O3, PM10) and species contributing to acidic and nitrogenous deposition (e.g. NH3, SO42–, NO3–, NH4+) against data from operational national monitoring networks. The key performance characteristics of the modelling system were found to be variable according to acceptance criteria and to depend on the type (e.g. urban, rural) and location of the sites, as well as on the time of the year. As regards the techniques that were used for ‘operational’ evaluation, performance generally complied with expected levels and ranged from good (e.g. O3, SO42–) to moderate (e.g. PM10, NO3–). At a few sites low correlations and large standard deviations for some species (e.g. SO2) suggest that these sites are subject to local factors (e.g. topography, emission sources) that are not well described in the model. Overall, the model tends to over predict O3 and under predict aerosol species (except SO42–). Discrepancies between predicted and observed concentrations may be due to a variety of intertwined factors, which include inaccuracies in meteorological predictions, chemical boundary conditions, temporal variability in emissions, and uncertainties in the treatment of gas and aerosol chemistry. Further work is thus required to investigate the respective contributions of such factors on the predicted concentrations.
AB - A comprehensive ‘operational’ evaluation of the performance of the Community Multiscale Air Quality (CMAQ) modelling system version 4.6 was conducted in support of pollution assessment in the UK for the calendar year 2003. The model was run on multiple grids using one-way nests down to a horizontal resolution as fine as 5 km over the whole of the UK. The model performance was evaluated for pollutants with standards and limit values (e.g. O3, PM10) and species contributing to acidic and nitrogenous deposition (e.g. NH3, SO42–, NO3–, NH4+) against data from operational national monitoring networks. The key performance characteristics of the modelling system were found to be variable according to acceptance criteria and to depend on the type (e.g. urban, rural) and location of the sites, as well as on the time of the year. As regards the techniques that were used for ‘operational’ evaluation, performance generally complied with expected levels and ranged from good (e.g. O3, SO42–) to moderate (e.g. PM10, NO3–). At a few sites low correlations and large standard deviations for some species (e.g. SO2) suggest that these sites are subject to local factors (e.g. topography, emission sources) that are not well described in the model. Overall, the model tends to over predict O3 and under predict aerosol species (except SO42–). Discrepancies between predicted and observed concentrations may be due to a variety of intertwined factors, which include inaccuracies in meteorological predictions, chemical boundary conditions, temporal variability in emissions, and uncertainties in the treatment of gas and aerosol chemistry. Further work is thus required to investigate the respective contributions of such factors on the predicted concentrations.
U2 - 10.1016/j.atmosenv.2010.03.029
DO - 10.1016/j.atmosenv.2010.03.029
M3 - Article
SN - 1352-2310
VL - 44
SP - 2927
EP - 2939
JO - Atmospheric Environment
JF - Atmospheric Environment
IS - 24
ER -