The study investigates the characteristics of localized rainfall and associated convective processes using observations and model data sets for three different monsoon low-pressure systems that occurred over the Bay of Bengal and moved inland. The performance evaluation of five microphysics parameterization schemes (WSM6, WDM6, THOM, MORR and AA_THOM) are carried out for a location-specific rainfall event up to a lead time of 72 h. The model forecast results are validated against available observations over the study location (i.e. Argul, Odisha, 85.7°E 20.17°N). The results suggested that both MORR and THOM are relatively better performers among all schemes in terms of timing, evolution and intensity of hyperlocal rainfall events. The WDM6 has better performance in terms of predicting 2-m temperature. In general, the results suggest a robust and coherent relationship among moisture convergence, diabatic heating and rainfall peaks with better estimates in MORR. Further, it is noted that upper-level heating and lower level cooling pattern is distinct in all the cases and AA_THOM has the lowest heating rate as compared to others. It is also found that snow and ice hydrometeors are playing a key role in better rainfall estimates in MORR. It is demonstrated that factors (i.e., moisture convergence, vertical motion, hydrometeors and diabatic heating) highly modulated by choice of cloud microphysical parameterizations are dominantly influencing the localized rainfall and associated convection. In addition, the robust impact of cloud microphysical parameterization schemes in modulating the evolution, amount and distribution of hydrometeors and intensity of diabatic heating is evident. The findings of this study have also provided valuable clues about adopting a particular cloud microphysical parameterization for the hyperlocal forecast with special reference to moderate to heavy rainfall events. The results of the study have significant implications for operational forecasting agencies to optimize their model configuration for predicting location-specific heavy rainfall events, which are on the rise due to climate change scenarios over the Indian region.

该研究使用对发生在孟加拉湾上空并移动到内陆的三个不同季风低压系统的观测和模型数据集，研究了局部降雨和相关对流过程的特征。对五个微物理参数化方案（WSM6、WDM6、THOM、MORR 和 AA_THOM）的性能评估针对特定位置的降雨事件进行了长达 72 小时的提前期。模型预测结果根据研究地点（即阿尔古尔，奥里萨邦，85.7°E 20.17°N）的可用观测结果进行验证。结果表明，就超局地降雨事件的时间、演变和强度而言，MORR 和 THOM 在所有方案中都表现相对较好。WDM6 在预测 2 米温度方面具有更好的性能。一般来说，结果表明水分收敛、非绝热加热和降雨峰值之间存在稳健和一致的关系，并在 MORR 中得到更好的估计。此外，注意到上层加热和下层冷却模式在所有情况下都是不同的，与其他情况相比，AA_THOM 具有最低的加热速率。还发现冰雪水凝物在 MORR 更好的降雨估计中发挥着关键作用。结果表明，通过选择云微物理参数化高度调制的因素（即水分辐合、垂直运动、水凝物和非绝热加热）主要影响局部降雨和相关对流。此外，云微物理参数化方案在调节演化方面的强大影响，水凝物的数量和分布以及非绝热加热强度是明显的。这项研究的结果还为采用特定云微物理参数化进行超局域预报提供了宝贵的线索，特别是针对中到强降雨事件。该研究的结果对业务预测机构优化其模型配置以预测特定地点的强降雨事件具有重要意义，这些事件由于印度地区的气候变化情景而呈上升趋势。