Assessment of Aerosol indirect effects (AIE) in regional and global climate models is critical in climate projection considering AIE is of the largest uncertainty in climate forcing. Bulk microphysics employed in regional and global models needs to be extensively evaluated and improved. Previous study shows the striking differences in cloud droplet (Nc) and raindrop concentrations (Nr) and the opposite aerosol effects on convection and heavy rain between spectrual-bin (SBM) and bulk microphysical scheme. In this study we target on improving the Morrison two-moment bulk microphysical scheme using SBM simulations as benchmarks in the Weather Research and Forecasting (WRF) model framework. Different from the fixed aerosol approach in the original Morrison scheme, we implement a prognostic aerosol representation with the prediction of both aerosol number concentration and mass mixing ratio. Two typical cloud regimes are investigated in our study, including maritime stratocumuli (SC) over Southeast Pacific Ocean from VAMOS Ocean-Cloud-Atmosphere-Land Study (VOCALS) and the continental deep convective cloud (DCC) over southeast China from the DOE AMF-China field campaign. Results with the prognostic CCN approach exhibit better agreement with SBM in terms of Nc, effective cloud radius and precipitation rate in SC compared with original Morrison scheme. More importantly, the 2-moment aerosol scheme is able to reproduce the larger heavy precipitation and enhanced convection strength under high aerosol loading in the DCC case, which is simulated by SBM but failed by original Morrison scheme. Since there are still large differences in raindrop number and mass concentrations from the SBM simulations in both cases, sensitivity tests on four different types of autoconversion schemes are performed to examine contribution from autoconversion parameterization. In addition, different droplet diffusion growth approaches: the saturation adjustment and explicit prediction of saturation, are examined to address the discrepancy on liquid water content between SBM and bulk microphysics, particularly in SC. Therefore, an improved bulk scheme with the optimal configurations will be presented and further evaluated with cloud cases from DOE Atmospheric Radiation Measurement (ARM) filed campaigns.