WRF high resolution simulation of an extreme rainfall event over Nepal: a case study of August 11-12,2017 event.

Floods and landslides due to extreme rainfall are major hazards to both people and infrastructure in the Himalayan regions of Nepal. Here steep Orography enhances precipitation and the complex terrain channels the runoff into narrow valleys and small rivers. A better understanding of the dynamics causing extreme rainfall events is crucial in order to improve their prediction and to minimmize the associated impacts.In this study we investigate extreme precipitation events in the Central Nepal on 11-12 August 2017 using Weather Research and Forecasting(WRF-ARW) model. The purpose of this study is to determine how available Cumulus Parameterization (CU) and Microphysics(MP) schemes in the WRF(version 3.9.1) simulate extreme precipitation with multiple Domains (25,5,1km).Possible combination among two MP (WSM6, Kessler) and three CU (New Tiedtke, New-GFS, Modified Kain-Fritsch )schemes were tested. The model output has been compared with point observation data where maximum precipitation amount of 192mm/24hr was observed at Simara station. Our result shows that the essential features of the extreme events are well simulated by WRF, although the exact location and intensity varies from scheme to scheme with Domains. Configuration with WSM6 with Kain-Fritsch Produce better Quantity of Precipitation Forecast (QPF) than other schemes. Hourly precipitation maps showed the south-westward motion and intensification of development rain area over Central Nepal after 12:00 GMT 12 August 2017.The Upper air sounding data also shows that there was persistent instability in Nepal-India Border during the period. Additionally the satellite image from infrared channel shows that there was the cloud with strong vertical development.Further the simulated precipitation also compares favorably to the GPM in terms of spatial and total accumulated amount. Depending on the severity level of different MP and CU demonstrate that there is not a unique configuration that solves exactly the studied event. Further research with multiple parameterizations is extremely important.