Assessment of WRF-CHEM Simulated Dust Using Reanalysis, Satellite Data and Ground-Based Observations

Abstract

One of the significant natural pollutants in the atmosphere is the mineral dust aerosols. In the northern hemisphere, Arabian Peninsula is one of the significant sources of dust aerosols with the frequency of dust storms changing seasonally. Dust emitted from the Arabian Peninsula region transports towards the Indian region through prevailing winds, therefore a positive correlation between the extreme dust episodes over the Arabian region and air quality over the Indian region is generally observed. Therefore, there is a need to monitor and forecast such an episodic event over the Arabian Peninsula and surrounding regions so that substantial measures may be taken in Indian subcontinent to mitigate the adverse impact of low air quality on human being and several other sectors such as aviation, energy and infrastructure. In the present study, the WRF-Chem simulations for dust particles are assessed against the observational data sets (i.e. MERRA-2, MODIS-Terra, Aura-OMI, CPCB). Based on the comparison of WRF-Chem simulated data sets with the satellite/reanalysis data, it is noted that in the post-monsoon season, WRF-Chem model can capture the entire dust episode (emission, transportation and dissipation) reasonably well. From the statistical analysis (PDF, CDF, Scatter plot and Temporal evolution), it was noted that there was a consistent underestimation of the simulated dust by WRF-Chem as compared to the observational data sets. A statistically robust categorical analysis has also been carried out for assessing the performance of WRF-Chem with respect to the observations for each dust event, and it is noted that on an average the probability of detection of dust event is about 77% and false alarm ratio is about 15% with an overall accuracy of 76%. Results obtained from the present analysis are encouraging and would be useful for the assessment of WRF-Chem simulations for other seasons also.