Verification of the Convection-Allowing Ensemble System over the Hindu Kush Himalaya Region during the 2018 and 2019 Premonsoon Severe Thunderstorm Seasons

Some of the most intense thunderstorms on the planet occur in the Hindu Kush Himalaya (HKH) region of South-Central Asia — where many organizations lack the capacity needed to predict, observe and/or effectively respond to the threats associated with high-impact convective weather. Among the convective hazards of the HKH region include excessive lightning, damaging straight-line winds, large hail, flash flooding, and even significant tornadoes, which typically peak in the pre-wet-monsoon months (March through May). Previous studies have documented a disproportionately large number of casualties associated with severe thunderstorms in the HKH region; therefore, our project goal is to increase situational awareness and warning decision support of these hazards through the development of a short-term, convection-allowing ensemble modeling system.

As part of the NASA/SERVIR Applied Sciences Team, this project combines numerical weather prediction (NWP) strategies, satellite-based precipitation products, and land-imagery techniques into a high-impact weather assessment toolkit (HIWAT). The HIWAT tool was developed to enhance the early warning service capacity of weather-sensitive agencies of the HKH region, focusing particularly on the countries of Bangladesh and Nepal. HIWAT’s short-term NWP component involves a regional, real-time, convection-allowing deterministic and 12-member ensemble system, the latter of which contains variations in initial/boundary conditions and physical parameterization schemes (planetary boundary layer and microphysics schemes). The ensemble system provides output products similar to the Storm-Scale Ensemble of Opportunity at the Storm Prediction Center (i.e., “paintball” maps, neighborhood probabilities, diurnal summary plots, etc., of proxy convective fields such as reflectivity, updraft helicity, 10-m wind speed, and total-column graupel) to depict the most likely areas for severe weather over a 48-hour outlook. Output fields were made available in an open-source platform called “Tethys” that enables end-users to interactively visualize and interrogate the ensemble products at specific points or polygon regions.

We will present results for select case studies and composite precipitation and lightning verification results from the ensemble system during HIWAT’s 2018 and 2019 Spring Forecasting Demonstration. Quantitative precipitation forecasts (QPF) from the individual ensemble members, ensemble mean, and probability matched mean are verified against the Global Precipitation Measurement (GPM) mission’s Integrated Multi-satellitE Retrievals for GPM (IMERG)-Final precipitation rate analyses. Model total lightning flash rates are represented by the diagnostic Lightning Forecast Algorithm (LFA), which is a weighted sum of the model graupel flux at the -15C isotherm with the vertically integrated ice. Observed lightning flashes from the Earth Networks Total Lightning Network are binned and gridded at hourly intervals for validating the hourly LFA output. The Model Evaluation Tools (MET) verification package is used to generate the QPF and lightning verification scores.