Evaluation of the MERRA-2 PM2.5 mass and its constituent chemical species concentrations over a COALESCE network site in Bhopal, India 

<p>Modern-Era Retrospective analysis for Research and Applications (MERRA-2) generated  PM<sub>2.5</sub>  concentrations are widely used to understand the spatio-temporal variability of PM<sub>2.5</sub> across the globe. Only PM<sub>2.5</sub> data from black carbon, organic carbon, sulphate, sea-salt, and dust are provided by MERRA-2. However, previous studies validated MERRA-2 PM<sub>2.5</sub> concentrations obtained by combining all five species data against in-situ ​total PM<sub>2.5</sub> concentrations. To the best of our knowledge, this is the first study over India to validate MERRA-2 species wise PM2.5 concentrations utilizing in-situ surface measurements made over a site at Bhopal (23.285° N, 22.277° E). Bhopal is one of the eleven COALESCE (Carbonaceous Aerosol Emissions, Source Apportionment and Climate Impacts) network regionally representative sites in India. 24 hour integrated filter-based samples (N = 165) collected during 2019, using the MetOne SASS® speciation sampler were used to measure mass and aerosol species concentrations by a variety of analyses. Our results show that the MERRA-2 well captures the aerosol species data at Bhopal. However, MERRA-2 underestimated the annual mean in-situ concentration of organic carbon, black carbon, and sulphate by 1.9 µg m<sup> -3</sup> (~22 %), 1.3 µg m<sup> -3</sup> (~47 %) and 0.9 µg m<sup> -3</sup> (~11 %), respectively and overestimated the sea salt and dust components by 0.75 µg m<sup> -3 </sup>(~95 %) and 8.5 µg m <sup>-3</sup> (~153 %), respectively.  It is pertinent to note that dust from surface aerosol chemical species measurements was re-constructed using elemental aluminium, silicon, potassium, calcium, titanium, manganese concentrations. Further, the annual mean MERRA-2 PM<sub>2.5</sub> mass (reconstructed from its constituent species) underestimated the average in-situ PM<sub>2.5</sub> mass by 13.55 µg m <sup>-3</sup> (~26.16 %). This underestimation is likely due to aerosol nitrate not being included in the MERRA2 PM<sub>2.5</sub> mass and uncertainties in aerosol species concentrations resulting from limitations in the chemical transport model set-up and emissions inventories. This study discusses the possible causes of disagreements between in-situ measurements and MERRA2 products, in addition to estimating the effect of including nitrate in the MERRA2 PM<sub>2.5 </sub>mass reconstruction.</p>