Electrified Simulations of Hurricane Rita (2005) with Comparisons to LASA Data

Electrified Simulations of Hurricane Rita (2005) with Comparisons to

LASA DataSteve Guimond1,2 , Jon Reisner2,

Chris Jeffery2 and Xuan-Min Shao2

1 Florida State University2 Los Alamos National Laboratory

Motivation• Improve understanding and

forecasting of TC intensification• Finish PhD and get a “real” job

Latent Heat

Updraft

Background Vortex

Microphysics

Hurricane Intensification

Roadmap

Eddy Heat and

Momentum Fluxes

Balanced response

Adjustment

Symmetric heating

Asymmetricheating

Adjustment

Balanced response

Adjustment

Intensity and

Structure Change

Nolan and Grasso (2003)

Motivation• Convective obs hard to come by over

ocean• Microwave satellite overpasses intermittent and

coarse• Doppler radar coverage very sparse• Lightning fills gaps in convective monitoring ?

Latent Heat

Updraft

Background Vortex

Microphysics

Hurricane Intensification

Roadmap

Eddy Heat and

Momentum Fluxes

Balanced response

Adjustment

Symmetric heating

Asymmetricheating

Adjustment

Balanced response

Adjustment

Intensity and

Structure Change

Nolan and Grasso (2003)

Lightning

Collisions &

Charging

Motivation• Understand relationship between

latent heating and lightning– Heating dynamics– Add energy to system

» When» Where» Magnitude» Structure

New Research Tools– Observational component

• Los Alamos Sferic Array (LASA; Shao et al. 2000)– Existing VLF/LF array

» Records full EMP (allows detection of intracloud and cloud-to-ground strokes)

» Lat/Lon, time• New Dual VLF-VHF 4-D lightning mapping array

– Deployed along banks of Gulf of Mexico– VLF (~2000 km range)– VHF (~500 km range)

» Provides precise height retrieval

New Research Tools– Theoretical component

• Advanced atmospheric model HIGRAD (Reisner et al. 2005)

– Compressible Navier-Stokes, non-hydrostatic, bulk or explicit microphysics

– Differentiable (smooth) numerics with greatly reduced time errors (option)

• Coupled to electrification model (Mansell et al. 2005)

– Non-inductive collisional charge separation (Saunders)– Lightning discharge model requires significant tuning

» Flash initiated when EF exceeds “floor”» What is a good “floor” for hurricanes?» Limit “floor” to ~50 kV/m for reasonable results

Hurricane Rita Simulations• Current configuration

– Grid• 1,980 km on a side; 4 km inner mesh, stretch to 20 km• 35 m stretching to 15 km

– Relaxation boundary conditions– Weak, top gravity wave absorber– F plane

• Initialization procedure– Barotropic vortex, max wind of 40 m/s– Initialize mass from Key West 88D reflectivity

• Storm-centered, gridded, native 1 km• Below melting rainwater saturate lower levels• Above melting graupel or snow hydrometeor drag, phase changes

– Gaussian water vapor function from eyewall to ~200 km radius– ECMWF operational analyses for large scale– Satellite SSTs, High-res topography

3 Hours Into Simulation

HIGRAD vs. LASA

Model

Observations

Initializing with LASA data

Rainwater mixing ratio

Potentially relevant work• Understand the non-linear response of observed

vortices to retrieved heating– Airborne Dual-Doppler Radar: Hurricane Guillermo (1997)– Latent heat retrieval (Guimond 2008)

• What spatial/temporal scales of heating does the hurricane “feel” ?– Resolution dependence (i.e. 100 m vs. 2 km)

• Impact on azimuthal mean• Balanced adjustment

– Are small scale details of lightning necessary to capture intensification?

• Governed by model grid cells• Is bulk heating sufficient?

Idealized Calculation

Realistic CalculationLatent Heating Slice @ z=5

km

• New area of research with physics not well understood– Not all deep convection is created equal

• Hurricane Initialization– Dual-doppler vortices– Dual-doppler latent heat retrieval– LASA data

• How is lightning tied to latent heating (4D)?• What scales matter for the hurricane?

– Azimuthal mean sensitive to resolution of heat• ≥ factor of ~4

– Balanced adjustment process

Summary and Science Questions

Acknowledgments• LANL Hurricane Lightning Team

References• Reisner et al. (2005)• Mansell et al. (2005)• Guimond (2008)

P-3

EDOP

Do Eyewall Hot Towers Produce Lightning?

• Next slides…– ER-2 Doppler Radar observations of Hot Towers

• Linear Depolarization Ratio (LDR)– particle canting angle or asymmetry – dielectric constant (i.e. wet or dry)

• Retrieved vertical velocities (nadir beam)

– Lightning Instrument Package (LIP)• Aircraft (20 km) electric field mills (x,y,z components)• ~1 s sampling, ~200 m horizontal resolution

10log vh

vv

ZLDR

Z

Hot Tower #1: CAT 2 Dennis (2005)-8 to -15 dB large, wet, asymmetric ice to large, wet snow aggregates

-13 to -17 dB medium, wet graupel or small hail

-18 to -26 dB small, dry ice particles to dry, low density snow

Hot Tower #2: CAT 4 Emily (2005)-8 to -15 dB large, wet, asymmetric ice to large, wet snow aggregates

-13 to -17 dB medium, wet graupel or small hail

-18 to -26 dB small, dry ice particles to dry, low density snow

Some Model Results

4 hours into simulation

Cloud Liquid Water (g/kg)

Vertical Velocity (m/s)

Graupel (g/kg)

Ice (g/kg)