A hybrid model to improve WRF-Chem performance for crop burning emissions of PM2.5 and secondary aerosols in North India

Abstract

In north India (about 30,000 km2 area), October–November months witness crop residue burning (CRB). This study aims to assess the impact of CRB emissions on PM2.5 and secondary aerosol (SA) at distances far from the sources. Chemical transport model, WRF-Chem (with chemical mechanisms, RADM2 and MADE/SORGAM), was applied to a domain of 840 × 840 km2 (centered around Delhi) with inputs of emissions of anthropogenic sources of EDGAR-HTAP and fire inventory from the National Centre for Atmospheric Research, USA. In parallel, measurements of PM2.5 and its constituents, CO, and NO2 were undertaken in Delhi and Kanpur. The model underestimated PM2.5 and showed poor predictions (r < 0.15) with insignificant formation of secondary organic aerosol (SOA) (< 1% in PM2.5). To improve the model, a linear regression-based hybrid model (HM) was developed that utilized the mechanistic processes of WRF-Chem and observed levels of PM2.5, CO and NO2. The HM performed adequately at multiple sites in Delhi (r = 0.19–0.47) and Kanpur (r = 0.38–0.43). The contribution of CRB to PM2.5 was at 31 ± 16% (in 232 ± 36 μg/m3) in Delhi and 21 ± 15% (in 233 ± 38 μg/m3) in Kanpur. The contribution of SOA in CRB-contributed PM2.5 was 18 ± 9% (Delhi) and 29 ± 21% (Kanpur).