Inverse modelling approach to assess air pollutant emission trends, and source contributions in highly polluted cities

Abstract

Pollutant concentrations are regularly measured in many cities, but emission loads often remain unknown and are irregularly estimated. This paper presents a methodology to estimate total emissions in a highly polluted megacity using an inverse modeling approach. An inverted Fixed Box Model (FBM) has been used with pollutant concentrations and meteorological variables to estimate emissions of pollutants by discounting the meteorological variability from the measured concentrations. Model-predicted emissions are validated with existing inventories and then used to assess annual, monthly, and diurnal trends in emissions. The applicability of the model is demonstrated for Delhi. Trend analysis shows that pollutants are emitted in a unique pattern that also differs over time periods of consideration. Yearly trends in Delhi show that PM10, NOX, and SO2 emissions have increased by 33%, 4%, and 16%, respectively, during 2008–2018. However, due to controls, the emission intensities have been found to stabilize for PM10 and decrease for NOX and SO2 pollutants in Delhi. Monthly trends indicate that summer months have higher emission rates owing to intensive energy demands (in automobiles, power plants), dust storms, and increased re-suspension of dust due to higher wind speeds. However, due to dispersive meteorology, ambient concentrations are lower in summers. The inverse modeling shows that PM10 emissions are dominated by dust sources (73%), while PM2.5 has the largest contribution from transport sector emissions (34%). NOX and SO2 emissions in the city are contributed heavily by the transport (84%) and industrial (92%) sectors, respectively. The model helps in understanding temporal emission trends and source contributions, which can be useful to implement effective control measures and track the progress of control.