Development of NWS GSI-Based EnVar System for convective scale radar data assimilation: Methodology, Impact of sub-kilometer resolution and Experiment with the 8 May 2003 Oklahoma City Tornadic Supercell

A GSI-based EnVar data assimilation system is extended to directly assimilate radar reflectivity to initialize convective-scale forecasts. When hydrometeor mixing ratios are used as state variables (option-mixing-ratio), the large difference of the EnVar cost function gradients with respect to the small hydrometeor mixing ratios and wind prevents efficient convergence. Using logarithmic mixing ratios as state variables (option-logarithm) fixes this issue, but generates spuriously large hydrometeor increments. The tangent linear of the reflectivity operators further contributes to the spuriously small and large hydrometeor increments in option-mixing-ratio and option-logarithm respectively. A new method is proposed and developed to avoid the use of the tangent linear and adjoint of the nonlinear operator and therefore overcomes the aforementioned issues.

The newly proposed method is examined on the analysis and prediction of the 8 May 2003, Oklahoma City, tornadic supercell storm at 2-km model grid. Both the probabilistic forecast of a strong low-level vorticity and the maintenance of the strong updraft and vorticity in the new method are more consistent with the reality than in option-logarithm and option-mixing-ratio. Detailed diagnostics suggest that a more realistic cold pool due to the better analyzed hydrometeor mixing ratios in the new method than in other methods leads to the constructive interaction between the surface gust front and the updraft aloft associated with the mid-level mesocyclone. Similar low-level vorticity forecast and maintenance of the storm are produced by WSM6 and Thompson microphysics schemes in the new method. The Thompson scheme better forecasts the reflectivity distribution but led to southward track bias. Details can be found in Wang and Wang (2016).

Because the 2-km grid spacing is not able to explicitly depict a tornado circulation, the GSI-based EnVar system is further developed with the dual resolution (DR) capability where the analysis is produced a sub-kilometer resolution grid and the ensembles ingested by EnVar are still at relatively coarse resolution.  The newly developed dual-resolution EnVar system is also examined on the 8 May tornadic supercell with a 2-km grid ensemble and a 500-m grid analysis. Our results show the DR system alleviates the aforementioned southeast displacement of the vorticity swaths in the coarser 2km single resolution (SR) system. The initial diagnostics indicate the better vorticity swaths in the DR system than the SR system result from more realistic rear flank gust front.  Such more realistic rear flank gust front is due to the better analyzed size and location of the hydrometeors aloft.