Jim Marquis, Yvette Richardson, Paul Markowski , David Dowell,

1. Our experiences with EnKFassimilation of high-res radar observations collected in the 5 June 2009 Vortex 2 tornadicsupercell. Jim Marquis, Yvette Richardson, Paul Markowski, David Dowell, Josh Wurman, Karen Kosiba, Paul Robinson Photo by Sean Waugh

2. Goals of EnKF analysis • Increase availability of 3-D kinematic and • thermodynamic data (dual-Doppler and in situ obs • are spatially/temporally limited). Courtesy of Paul Markowski

3. Goals of EnKF analysis • We want a set of smoothly evolving EnKF analyses (i.e., We are not using EnKF to initialize a forecast). • Analyze roles that mesocyclone-scale processes play in tornadogenesis, maintenance, and decay: • - trajectory analysis, • - circulation/vorticity budgets, • - mid-upper level features.

4. Model Specifics • WRF-ARW 3.2.1: -Δx,y = 500 m, 80m < Δz < 2km, [120 x 80 x 20] km3 -LFO microphysics, -open lateral BCs, -no surface fluxes, no radiation, flat terrain. Homogeneous environment:

5. DA Specifics • DART: • Ensemble adjustment filter, • 50 members, • Localization: • Gaspari-Cohn, W = 0 @ r = 6 km • Ensemble initiation: • 10 randomly placed warm bubbles at model t0 for each member • Ensemble spread maintained with: • Additive noise added to T, Td, U, V every 5 min where radar reflectivity is > 25 dBZ, (Dowell and Wicker 2009) • Perturbations smoothed to 4 km (horiz), 2 km (vert) scales

6. Experiment timeline “synthetic Data” Courtesy www.vortex2.org • Radar velocities assimilated every 2 minutes • OBAN: Cressman weighting • 500 m horizontal grid spacing (for 500m-model grid experiment) • data along conical slices σ2obs= (2 m/s)2

7. Thinning the amount of observations assimilated: 50% assim’ed Unthinned 25% assim’ed Y X Radar vel. ob. Model grid No thinning done in vertical direction shown today; though, experiments show poor storm structure when low-level elevation angles are neglected.

8. unthinned Z = 0.5 km AGL θρ (K) ζ dBZ W > 0 Y 50% assimilated Y 25% assimilated X (Note: these experiments conducted with 2-km model grid)

9. Dual-Doppler – EnKF (ensemble mean) kinematics comparison (DOW6 & DOW7) W m/s EnKF Y (km) Dual-Doppler ζ radars X (km) Z = 400 m AGL

10. Storm structure with/without radar assimilation: Top row: Series of EnKF (Ens. Mean) analyses. Z = 150 m Y (km) ζ X (km) Bottom row: Single member forecasted forward from 2157 (no DA). Model errors/idealized conditions require DA for a good storm.

11. Ensemble mean analyses – temperature fields Z = 50 m ζ (Note: these experiments conducted with 1-km model grid)

12. Z = 50 m ζ Mob. Mesonets

13. Z = 50 m ζ Mob. Mesonets

14. MM obs – EnKF outflow comparison • Sequence of EnKF analyses of theta’ and • MM obs(valid +/- 1 min from EnKF analysis) overlayed: Z = 50 m θ’(K) MM’s only just getting into storm, but so far - not good Not great, particularly in far left FF Decent, but still disagreement in FF Better overall, but still disagreement in FF Warmth (lasts 1-2analysis times) - Seems to disagree With few available MM obs Pretty good ζ ζ decent ok Z = 50 m My < 1 sentence impression of the agreement of MM obs and each EnKF analysis dBZ MM obs: Low-level wmax trace:

15. Mesocyclone buoyancy (function of height & time) Positively buoyant Negatively buoyant Circulation and radial motion (function of radius & time) Inbound 0 m/s Outbound Pre-tornadic T-genesis Intensification Maturity Time (UTC) Periods of tornado lifecycle Weakening

16. Trajectories (storm-rel.) calculated from ens. mean analyses Ring (radius = 1 km) of 20 parcels centered on peak ζ at z = 200 m; integrated forward in time from 4 times ‘ θρ (K) Most parcels rising into updraft Some parcels rising into updraft No parcels rising into updraft Few/no parcels rising into updraft W (z = 200m)

17. Comparison across model grid resolutions θρ ‘ (K) Maturity T-genesis Weakening Pre-tornadic Intensification

18. RMSI Total Spread Consistency Ratio

19. θ’(K) ζ MM obs: Low-level wmax trace:

20. Summary • EnKF Kinematic Analyses: • -Compare well with dual-Doppler fields of similar scale. • -Storm-scale structure robust with different model scale. • More details come with finer model grid res. • EnKF Thermo Analyses: • Mixed success with comparisons to MM in situ obs (where available). • 500-m model grid seems verify with MM obs best in mesocyclone area (partially due to more details captured?) • Seemingly random, possibly unphysical(?) anomalies appear more prominently with finer model resolution.

21. Questions • Different way to maintain ensemble spread (adaptive methods)? • How else can we determine plausibility of analyses (kinematic and thermodynamic)? • Forcing terms along parcel trajectories? • More rigorous methods for optimizing scales/amounts of observations (compared to model resolution)?

22. Acknowledgements • The EnKF experiments were performed using NCAR CISL supercomputing facilities with the Data Assimilation Research Testbed (DART) and WRF-ARW software. • Thank you to Glen Romine, Lou Wicker, Dan Dawson, Chris Snyder, and Nancy Collins for advice/help. • Thanks to all VORTEX2 crew for their dedication while collecting data on 5 June 2009. • This research is funded by NSF grants: NSF-AGS-0801035, NSF-AGS-0801041. The DOW radars are NSF Lower Atmospheric Observing Facilities supported by NSF-AGS-0734001.