Most probable mixing state of aerosols in Delhi NCR, northern India

Unknown mixing state is one of the major sources of uncertainty in estimating aerosol direct radiative forcing (DRF). Aerosol DRF in India is usually reported for external mixing and any deviation from this would lead to high bias and error. Limited information on aerosol composition hinders in resolving this issue in India. Here we use two years of aerosol chemical composition data measured at megacity Delhi to examine the most probable aerosol mixing state by comparing the simulated clear-sky downward surface flux with the measured flux. We consider external, internal, and four combinations of core-shell (black carbon, BC over dust; water-soluble, WS over dust; WS over water-insoluble, WINS and BC over WINS) mixing. Our analysis reveals that choice of external mixing (usually considered in satellite retrievals and climate models) seems reasonable in Delhi only in the pre-monsoon (Mar-Jun) season. During the winter (Dec-Feb) and monsoon (Jul-Sep) seasons, 'WS coating over dust' externally mixed with BC and WINS appears to be the most probable mixing state; while 'WS coating over WINS' externally mixed with BC and dust seems to be the most probable mixing state in the post-monsoon (Oct-Nov) season. Mean seasonal TOA (surface) aerosol DRF for the most probable mixing states are 4.4 +/- 3.9 (- 25.9 +/- 3.9), - 16.3 +/- 5.7 (- 42.4 +/- 10.5), 13.6 +/- 11.4 (-76.6 +/- 16.6) and -5.4 +/- 7.7 (- 80.0 +/- 7.2) W m(-2) respectively in the pre-monsoon, monsoon, post-monsoon and winter seasons. Our results highlight the importance of realistic mixing state treatment in estimating aerosol DRF to aid in policy making to combat climate change