Inter-comparison of different NOX emission inventories and associated variation in simulated surface ozone in Indian region

Abstract

In this work, we compare for the first time different anthropogenic NOX emission inventories and examine the associate variation in simulated surface ozone (O3) in India. Six anthropogenic NOX emission inventories namely Emission Database for Global Atmospheric Research (EDGAR), Intercontinental Chemical Transport Experiment-Phase B (INTEX-B), Regional Emission Inventory in Asia (REAS), MACCity, Indian National Emission Inventory (India_NOx), and Top-Down NOX emission inventory for India (Top-Down) are included in the comparison. We include these emission inventories in regional chemical transport model WRF-Chem to simulate tropospheric column NO2 and surface O3 mixing ratios for the month of summer (15-March to 15-April) and winter (December) in 2005. Predicted tropospheric column NO2 using different NOX emission inventory are evaluated with the OMI satellite observations. All emission inventories show similar spatial features, however uncertainty in NOX emissions distribution is about 20–50% over rural regions and about 60–160% over the major point sources. Compared to OMI, the largest bias in simulated tropospheric NO2 columns is seen in the REAS (−243.0 ± 338.8 × 1013 molecules cm−2) emission inventory, followed by EDGAR (−199.1 ± 272.2 × 1013 molecules cm−2), MACCity (−150.5 ± 236.3 × 1013 molecules cm−2), INTEX-B (−96.8 ± 199.5 × 1013 molecules cm−2), India_NOx (−87.7 ± 159.9 × 1013 molecules cm−2) and Top-Down (−30.8 ± 69.6 × 1013 molecules cm−2) inventories during winter. Simulations using different NOX emission inventories produces maximum deviation in daytime 8-h averaged O3 of the order of 9–17 ppb (15–40%) in summer and 3–12 ppb (5–25%) in winter over most of the land area. The simulation suggests that choice of NOX emission inventories have significant effect on surface O3 concentration for air quality studies over India.