Storm-Scale Ensemble Hail Analysis and Prediction using EnKF for the Supercell Thunderstorms of 20 May 2013

- Indicates paper has been withdrawn from meeting
- Indicates an Award Winner
Wednesday, 7 January 2015
Nathan Snook, CAPS/Univ. of Oklahoma, Norman, OK; and J. Brotzge, Y. Jung, and M. Xue

During the afternoon and evening of 20 May 2013, a line of supercell thunderstorms developed and moved over southern and central Oklahoma, producing numerous hail reports, as well as multiple tornadoes (including the EF-5 Newcastle-Moore tornado). For this case, ensembles of 40 storm-scale forecasts, produced using the Advanced Regional Prediction System (ARPS) NWP model with a horizontal grid-spacing of 500m, are initialized shortly prior to convective initiation at 1800 UTC. Radar and conventional observations are assimilated using an ensemble Kalman filter (EnKF) between 1830 and 1930 UTC, and 0-3 hour ensemble forecasts are produced from the resulting analyses.

At sub-kilometer grid-spacing, explicit prediction of hail is possible for storm-scale ensembles using multiple-moment microphysics schemes. Storm-scale analyses and forecasts are verified against observed hail reports and radar-observed maximum estimated size of hail (MESH) obtained from the WSR-88D radar network. The EnKF produces analyses of the supercell storms which match well with observations in terms of radar reflectivity and predicted MESH. The ability of the model to correctly represent and predict hail is assessed, and factors influencing the skill of the hail forecast are investigated, including varying the length of the EnKF assimilation period and altering the dual-moment microphysical scheme used.