A 1-D model to retrieve the vertical profiles of minor atmospheric constituents for cloud microphysical modeling: II. Simulation of diurnal cycle.

The accurate forecast of the diurnal cycle of the number concentration of trace gases is vital due to their influence on precipitation processes by controlling the number concentration of cloud condensation nuclei (CCN). 1-D hybrid Monte Carlo-Gear solver developed to retrieve vertical profiles of the number concentration of CCNs for microphysics modeling has been tested for representation of the diurnal cycle in the present paper. The retrieved profiles of CH and SO have been tested with the Copernicus Atmosphere Monitoring Service (CAMS) model at 3-hour time intervals for four megacities: Delhi, Kolkata, Chennai, and Mumbai for rainy and non-rainy days. The retrieved profiles have shown diurnal variation up to 18 UTC at all pressure levels with lead or lag with that of the CAMS model. After 18 UTC there was a furious increase in the number concentrations. During non-rainy days, the 1-D model slightly overestimated (underestimated) the maximum (minimum) number concentrations of CH over Delhi whereas concentrations are overestimated over Kolkata, Chennai, and Mumbai. Forecasted CH has a good (weak) correlation over Chennai (Mumbai) respectively. The 1-D model overestimated (overestimated) the maximum (minimum) number concentrations of SO over Delhi but the maximum (minimum) concentrations are underestimated (overestimated) over Kolkata, Chennai, and Mumbai. The number concentrations of SO have shown a good correlation for all megacities except Delhi. CH number concentration is overestimated during rainy days. Delhi and Kolkata show a good correlation of CH during rainy days. SO during rainy days is underestimated except over Chennai and both models show a good correlation except over Mumbai. Overall, it can be stated that the 1-D hybrid solver is successful in simulating the monthly mean diurnal variation of vertical profiles of CH and SO and its implementation in the global model may estimate the number concentrations with better accuracies.