Evaluation of black carbon emission inventories using a Lagrangian dispersion model – a case study over southern India

Abstract. We evaluated three emission inventories of black carbon (BC) using Lagrangian particle dispersion model simulations and BC observations from a rural site in southern India (Gadanki; 13.48° N, 79.18° E) from 2008 to 2012. We found that 93 to 95% of the BC load at the observation site originated from emissions in India and the rest from the neighbouring countries and shipping. A substantial fraction (33 to 43%) of the BC was transported from northern India. Wet deposition is found to play a minor role in reducing BC mass at the site because of its proximity to BC sources during rainy season and relatively short rainy season over western and northern parts of India. Seasonally, the highest BC concentration (approx. 3.3 μg m⁻³) is observed during winter, followed by spring (approx. 2.8 μg m⁻³). While the model reproduced well the seasonal cycle, the modelled BC concentrations are significantly lower than observed values, especially in spring. The model bias is correlated to fire radiative power – a proxy of open biomass burning activity. Using potential emission sensitivity maps derived using the model, we suggest that underestimation of BC mass in the model during spring is due to the underestimation of BC fluxes over southern India (possibly from open-biomass-burning/forest-fires). The overall performance of the model simulations using three different emission inventories (SAFAR-India, ECLIPSE and RETRO) is similar, with ECLIPSE and SAFAR-India performing marginally better as both have about 30% higher emissions for India than RETRO. The ratio of observed to modelled annual mean BC concentration was estimated as 1.5 for SAFAR, 1.7 for ECLIPSE and 2.4 for RETRO.