Scholarlink

Abstract <p>The sensitivity of the Weather Research and Forecasting (WRF) model to cloud microphysics parameterizations plays a critical role in simulating the evolution of severe convective systems. This study examines the relative performance of three widely used bulk microphysics schemes (Thompson, Morrison, and WSM6) for a high-impact pre-monsoon thunderstorm over Hyderabad, India, on 7 April 2020, using a convection-permitting model configuration with 3 km horizontal grid spacing. Model outputs are evaluated against satellite-based rainfall estimates from GPM-IMERG, lightning observations from the IITM Lightning Location Network, and thermodynamic fields from ERA5 reanalysis. The results reveal notable contrasts among the schemes. The Thompson scheme shows comparatively better skill in quantitative precipitation forecasting (QPF), reproducing the large-scale spatial organization and temporal evolution of rainfall with the highest correlation (CC = 0.86) and the lowest root mean square error (RMSE = 0.66). In contrast, the Morrison scheme provides a more realistic representation of lightning activity, capturing the broad spatial distribution and relative intensity of observed flashes, and more consistently reproducing key pre-convective thermodynamic features, including CAPE, KI, TTI, and vertical moisture structure. The WSM6 scheme generally underperforms, producing fragmented rainfall patterns and weaker correspondence between atmospheric instability and convective development. A key outcome of this study is that microphysics scheme performance is application-dependent. While the Thompson scheme appears more suitable for rainfall-oriented diagnostics, the Morrison scheme better represents storm electrification and associated thermodynamic structures. This contrast is likely linked to Morrison’s double-moment formulation, which allows a more flexible representation of mixed-phase microphysical processes relevant to charge separation. Overall, the findings emphasize the need for purpose-driven microphysics selection in convection-permitting simulations of severe thunderstorms over the Indian region.</p> Research&#xa0;highlights <p><UnorderedList Mark="Bullet"> <ItemContent> <p>Identifies a performance dichotomy in microphysics schemes. The Thompson scheme excels for rainfall prediction, while the Morrison scheme better captures lightning activity and pre-convective thermodynamics revealing that optimal scheme choice is application-dependent.</p> </ItemContent> <ItemContent> <p>First multi-faceted evaluation for Hyderabad thunderstorm: Provides a high-resolution (3 km) integrated assessment of microphysics schemes for a severe pre-monsoon thunderstorm over complex Indian terrain, validating against satellite rainfall, lightning network, and reanalysis data.</p> </ItemContent> <ItemContent> <p>Links microphysical processes to forecast performance: Through vertical hydrometeor analysis, demonstrates that Morrison's double-moment formulation enhances ice/graupel in the mixed-phase layer (favoring lightning), while Thompson's efficient warm-rain processes improve rainfall simulation</p> </ItemContent> </UnorderedList></p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Sensitivity of simulated rainfall, lightning, and thermodynamic structure of a pre-monsoon thunderstorm over Hyderabad to cloud microphysics schemes in WRF

  • 期刊论文

  • 发表日期：

    2026年6月18日

查看全文
  • Nandivada Umakanth,
  • Rupraj Biswasharma,
  • Rajesh Gogineni,
  • Kondaveeti Sivakrishna,
  • Unashish Mondal,
  • Prafull P Yadav,
  • Kambala Anand Raju,
  • Yarlagadda Ramakrishna

摘要

Abstract

The sensitivity of the Weather Research and Forecasting (WRF) model to cloud microphysics parameterizations plays a critical role in simulating the evolution of severe convective systems. This study examines the relative performance of three widely used bulk microphysics schemes (Thompson, Morrison, and WSM6) for a high-impact pre-monsoon thunderstorm over Hyderabad, India, on 7 April 2020, using a convection-permitting model configuration with 3 km horizontal grid spacing. Model outputs are evaluated against satellite-based rainfall estimates from GPM-IMERG, lightning observations from the IITM Lightning Location Network, and thermodynamic fields from ERA5 reanalysis. The results reveal notable contrasts among the schemes. The Thompson scheme shows comparatively better skill in quantitative precipitation forecasting (QPF), reproducing the large-scale spatial organization and temporal evolution of rainfall with the highest correlation (CC = 0.86) and the lowest root mean square error (RMSE = 0.66). In contrast, the Morrison scheme provides a more realistic representation of lightning activity, capturing the broad spatial distribution and relative intensity of observed flashes, and more consistently reproducing key pre-convective thermodynamic features, including CAPE, KI, TTI, and vertical moisture structure. The WSM6 scheme generally underperforms, producing fragmented rainfall patterns and weaker correspondence between atmospheric instability and convective development. A key outcome of this study is that microphysics scheme performance is application-dependent. While the Thompson scheme appears more suitable for rainfall-oriented diagnostics, the Morrison scheme better represents storm electrification and associated thermodynamic structures. This contrast is likely linked to Morrison’s double-moment formulation, which allows a more flexible representation of mixed-phase microphysical processes relevant to charge separation. Overall, the findings emphasize the need for purpose-driven microphysics selection in convection-permitting simulations of severe thunderstorms over the Indian region.

Research highlights

Identifies a performance dichotomy in microphysics schemes. The Thompson scheme excels for rainfall prediction, while the Morrison scheme better captures lightning activity and pre-convective thermodynamics revealing that optimal scheme choice is application-dependent.

First multi-faceted evaluation for Hyderabad thunderstorm: Provides a high-resolution (3 km) integrated assessment of microphysics schemes for a severe pre-monsoon thunderstorm over complex Indian terrain, validating against satellite rainfall, lightning network, and reanalysis data.

Links microphysical processes to forecast performance: Through vertical hydrometeor analysis, demonstrates that Morrison's double-moment formulation enhances ice/graupel in the mixed-phase layer (favoring lightning), while Thompson's efficient warm-rain processes improve rainfall simulation