Uncertainties Caused by Major Meteorological Analysis Data Sets in Simulating Air Quality Over India

Many places in India suffer from severe air pollution. Regional air quality simulations are essential to develop effective strategies for improving air quality, considering the nonlinear relationships between ambient pollutants and their precursor emissions. Meteorological fields used in simulations are derived from regional meteorological models with analysis data sets as inputs. This study reveals that two major analysis data sets provided by the National Centers for Environmental Prediction and the European Centre for Medium-Range Weather Forecasts (ECMWF) cause significant differences in simulated meteorological fields over India. Especially, relative humidity values simulated using the ECMWF data set are much higher and closer to the observed values than those simulated using the National Centers for Environmental Prediction data set. Results simulated using the ECMWF data set show better model performance for most meteorological parameters over India. Differences in relative humidity originate in the data contained in the analysis data sets through grid nudging. It is not possible to avoid this underestimation by simply turning off grid nudging. The meteorological fields simulated with two major analysis data sets also lead to differences in pollutant concentrations simulated by regional chemical transport models through various physical and chemical processes. Differences originating in the two analysis data sets could be comparable with the uncertainties originating in various emission inventories in some regions and seasons in India. However, discrepancies between observed and simulated pollutant concentrations cannot be explained only by differences of the meteorological fields. Other aspects need to be explored for better performance required to develop effective strategies for India, based on accurate regional air quality simulations.