Aerosol levels over Southeastern Brazil retrieved by Remote Sensing and simulated by WRF-Chem Transport Model

<p>In this work, we combine Aerosol Optical Depth (AOD) products derived from MODIS and AERONET instruments, and WRF-Chem simulations in order to analyze the aerosol levels over the Metropolitan Area of São Paulo (MASP), in southeastern Brazil. Five MODIS AOD products were initially considered for validation against AOD data obtained from four AERONET stations during a period of six years (2014-2019). For the analysis, a spatiotemporal window methodology was used, assuming that the aerosol plume is homogeneous within a certain time-space limit. This methodology showed that the most adequate space-time limit for the validation of the MODIS AOD data is less than 15 km, and 15 minutes before and after the overpass of the Terra and Aqua satellites. A time series and statistical analysis were necessary to find the best product that represents the aerosols on the MASP. Satellite-derived AOD products reach a good accuracy when more than 66% of retrievals fall within Expected Error (EE) envelope (within<sub>EE</sub> > 66%). Using this approach, Dark Target of 3km spatial resolution (DT-3K) has shown good accuracy (within<sub>EE</sub> > 86%) compared to the other satellite products. With this information, both MODIS and AERONET data were then compared with AOD fields derived from WRF-Chem simulation for June 2017. On cloudy days, aerosol products do not provide AOD data information, hence in-situ PM<sub>2.5</sub> data from CETESB air quality stations over the MASP were analyzed to complement the WRF-Chem model performance. Analysis between simulated and in-situ PM<sub>2.5</sub> surface concentrations showed similarities on some days. Both the model and the air quality stations reached maximum peaks in some days of June 2017, even though the model did not reach the high values as the air quality stations. Discrepancies between model results and observations at site-specific locations at both surface and total-column are related to a misrepresentation of local conditions, not only in terms of emissions but also in terms of land-use and atmospheric stability. This work represents a first effort that combines different remote sensing and modeling tools to improve understanding of how aerosol emissions impact the air quality in the MASP and surrounding urban areas. </p>