The Effect of Latent Heat on the Extratropical Transition of Tropical Cyclone Sinlaku By: Michael Kevin Hernandez Key JTWC ET onset JTWC Post ET Fig. 1: JTWC best track data on TC Sinlaku (2008). ECMWF analysis ET completion ECMWF analysis ET onset
Jan 13, 2016
The Effect of Latent Heat on the Extratropical Transition of Tropical Cyclone Sinlaku
By: Michael Kevin Hernandez
Key JTWC ET onset JTWC Post ET
Fig. 1: JTWC best track data on TC Sinlaku (2008).
ECMWF analysis ET completion
ECMWF analysis ET onset
Two primary reasons for studying Extratropical Transitions (ET)
Societal RelevancePost Hurricane Noel caused floods and wind
damage to Maryland and Eastern Quebec & Labrador
Scientific RelevanceET can modulate the baroclinic zone,
enhancing midlatitude cyclogenesis and associated high-impact weather (Danielsen 2004; Harr et al. 2008)
Literature reviewVeiga et al. (2008) used kinetic energy
to calculate energetics of Hurricane Catarina (2004)Explored Kinetic Energy (KE) – use KE as a
diagnostic hereKuo et al. (1990) used the NCAR/PSU
MM5 model to explore the impact of diabatic processes on the development of a midlatitude cycloneTwo different runs (regular and fake dry
physics) – will also use this approach here
High resolution initial conditions
Infrared satellite imagery
325K Ertel PV500 hPa KE and geopotential height
Fig. 2: Observed storm (left), 25 x 25km ECMWF resolution diagnostics (below) which will be fed into the model.
Model formulation for the simulations analyzed here
Resolution: 9km (740 x 370), 3km (331 x 331) vortex following
LW radiation: RRTM schemeSW radiation: Goddard schemePBL physics: YSU schemeSurface layer: MM5 similarityMicrophysics: WSM 5-class scheme
Full physicsSurface fluxes includedCumulus physics: Kain–Fritsch scheme (9km only)
Fake drySurface fluxes includedno convection or grid scale precipitation
Diagnostics Kinetic Energy at 500mb
Other Parameters: MSLP, accumulated precip, ω, vertical wind shear, Tv, θe, θe anomaly, RH, vorticity, Ertel PV etc.
Difference between the two model forecasts
Fig. 3: Observed (black), full physics (pink) and fake dry (blue) simulated tracks for Typhoon Sinlaku.
Numerical Difference
Table 1: Error metrics of distance [km] for each simulation of TC Sinlaku compared to the JTWC best track.
Forecast time
12 24 36 48 60 72 84 96
full physics
17 95 326 247 241 223 583 764
fake dry 303 350 696 843 n/a n/a n/a n/a
HR 12 24 36 48 60 72 84 96full
physics12 3 5 10 17 13 15 8
fake dry 27 24 29 34 n/a n/a n/a n/aTable 2: Error metrics of minimum central pressure [hPa] for each simulation of TC Sinlaku compared to the JTWC best track.
Fig. 4: KE analyses from the ECMWF Analysis, Full Physics and Fake Dry runs at the 500 mb. The red dots show the storm and the black line shows the trough axis.
FakeDry
FullPhysics
IR imagery ECMWF .
Reanalysis .
500 hPa KEFore
cast
lead
tim
e48 h
r 2
4 h
r 00
hr
Fig. 5: KE analyses from the ECMWF Analysis, Full Physics and Fake Dry runs at the 500 mb. The red dots show the storm and the black line shows the trough axis.
IR imagery ECMWF Full Fake .
Reanalysis Physics Dry
500 hPa KEFore
cast
lead
tim
e96 h
r 7
2 h
r 48
hr
In the full physics run, the TC underwent ETET completion time consistent with ECMWF analysesTrough to north of TC at 24h (simulation and analyses)Simulated trough moved eastwards much faster in
subsequent 24h (transition phase) than analyzed (model fast)
Model TC moved NNE (rather than NE) in this period, leading to significantly different wind structure
Model TC moved due east from 48-72 h (rather than ENE) giving smaller track errors at 72h than 48h
Analyzed trough stronger and further east than full physics simulation at 72h (model slower)
Very different evolution of trough in simulation than analyses
Take home message from full physics simulation
In the fake dry run TC Sinlaku could no longer be tracked by 72 hours into the simulationWithout latent heat continuously feeding the
TC, it became shallower and diedAdvected northwards by the low-level steering
current
TC west of trough axis by 24h and there was no interaction of the trough with the storm
Subtropical high became much stronger than in the full physics
Take home message from fake dry simulation
Cloud permitting simulations of the ET of Typhoon Sinlaku initializing with high resolution ECMWF reanalysis focused on the 500 hPa trough with and without latent heating with the storm filtered out
QG diagnostics to evaluate the impact of diabatic heating on ET
Future work
THANKYOU
QUESTIONS