Lightning Assimilation Techniques Nonthinned Lightning Experiment NLDNLR lightning

Lightning Assimilation Techniques • Non-thinned Lightning Experiment – NLDN/LR lightning strike is detected – Lightning strike is converted into 30 minute lightning rate from nearby LTNG observations – Lightning rate converted into “observation” of convective rainfall rate using Pessi/Businger convective rain rate/lightning rate relationship – Convective rainfall is scaled up to 6 hour cumulative total to match model background forecasted 6 hour convective rainfall – Convective rainfall (mm) is assimilated into WRF-En. KF

Lightning Assimilation Techniques Pessi/Businger Lightning rate/Convective rainfall rate relationship

Lightning Assimilation Techniques • Thinned Lightning Experiment – NLDN/LR lightning strike is detected – Lightning strike is converted into 30 minute lightning rate from nearby LTNG observations – Any lightning strikes used in the density calculation are no longer allowed to be an assimilation point, resulting in a thinning out of the lightning “observations” (although strikes will be used to calculate nearby densities) – Lightning rate converted into “observation” of convective rainfall rate using Pessi/Businger convective rain rate/lightning rate relationship – Convective rainfall is scaled up to 6 hour cumulative total to match model background forecasted 6 hour convective rainfall – Convective rainfall (mm) is assimilated into WRF-En. KF

Lightning Assimilation Techniques • Thinned Lightning Experiment – 1 hr rain – NLDN/LR lightning strike is detected – Lightning strike is converted into 30 minute lightning rate from nearby LTNG observations – Any lightning strikes used in the density calculation are no longer allowed to be an assimilation point, resulting in a thinning out of the lightning “observations” (although strikes will be used to calculate nearby densities) – Lightning rate converted into “observation” of convective rainfall rate using Pessi/Businger convective rain rate/lightning rate relationship – Convective rainfall is NO LONGER scaled up as model background forecasted convective rainfall is a 1 hour cumulative total – Convective rainfall (mm) is assimilated into WRF-En. KF

Experiment Design • Pacific Ocean – Low observation density; location of important storm tracks; errors propagate downstream to mainland United States • North America – High observation area; potential of forecast improvement; • Similar studies with similar domain – Pessi/Businger previously studied domain for lightning assimilation

Experiment Design • Observations – Control case • • Radiosondes Surface stations (ASOS, ship, buoy) ACARS Cloud drift winds (no sat. radiances) – Experimental cases • Control observations • Lightning

Experiment Design • WRF Setup – – – – – WRF 2. 1. 2 (Jan 27, 2006) 100 by 86 grid 45 -km horizontal resolution 33 vertical levels 270 second timestep Shortwave: Dudhia Longwave: Rrtm Surface: Noah land-sfc PBL: MYJ TKE scheme Cumulus: Kain-Fritsch (new Eta)

Experiment Design • En. KF Setup – – – – – 90 ensemble members 6 -hr Analyses 24 -hr Forecasts (starting every 12 hours) 8 assimilations of “spin-up” before lightning assimilations Square root filter (Whitaker and Hamill, 2002) Horizontal localization – Gaspari and Cohn 5 th order piecewise Fixed covariance perturbations to lateral boundaries Zhang covariance inflation method Localization radius – 2000 km

Test Cases • Test Case #1 – December 16 -21, 2002 • Test Case #2 – October 4 -8, 2004 • Test Case #3 – November 8 -12, 2006

Experiment observations example ACARS observations spatial distribution

Experiment observations example Cloud track wind observations spatial distribution

Experiment observations example Radiosonde, surface station and buoy observations • Radiosonde Obs • Surface Stations • Buoys

Experiment observations example Lightning Observations