WAVES AT LOW LATITUDES

1
Tropical cyclone intensification
Roger Smith Ludwig-Maximilians University of
Munich Collaborators Michael Montgomery, Naval
Postgraduate School, Monterey, California Ph. D.
students Sang Nguyen, Seoleun Shin
(LMU) Postdoc Hai Hoang, Vietnam National
University
2
Topics

• How do tropical cyclones intensify?
• The basic thought experiment for intensification
• Important physical principles
• Paradigms for intensification
• Recent discoveries using idealized model
  simulations with simple physics
• Dynamics of vortex spin up
• Is WISHE relevant?
• Axisymmetric view of spin up comparison with
  the other paradigms
• New frontiers

3
The basic thought experiment for intensification
Initial condition
Mean sounding
Axisymmetric vortex
T(z) q(z)
V(r,z)
r
p
28oC
sea
4
The primary circulation
Pressure gradient force
LO
r
v
sea
Centrifugal force and Coriolis force
5
Frictionally-induced secondary circulation
primary circulation
Secondary circulation
Pressure gradient force
Friction layer
r
v
v
Centrifugal force and Coriolis force are reduced
by friction
6
Tea cup Experiment
7
Hurricane intensification

• Basic principle

- Conservation of absolute angular
momentum M rv r2f/2
f Coriolis parameter 2Wsin(latitude)
r
v
v M/r - rf/2
If r decreases, v increases!
Spin up requires radial convergence
8
Paradigms for intensification

• Conventional view articulated by Ooyama (1969,
  1982), Willoughby (AMM 1998, WMO, 1995) involving
  convectively-induced convergence together with
  absolute angular momentum conservation above the
  boundary layer.
• Thermodynamic view (E-theory) articulated by
  Emanuel (1989, 1994, 1995, 1997) involving the
  WISHE mechanism.
• Asymmetric view (M-theory) invoking Vortical Hot
  Towers or VHTs (Hendricks et al. 2004,
  Montgomery et al. 2006, Nguyen et al. 2008, Shin
  and Smith 2008, Montgomery et al. 2009, Smith et
  al. 2009, Hoang et al. 2009).

9
Conventional view
15
z km
10
5
M conserved
0
50
100
r km
M not conserved,
10
Thermodynamic view A steady hurricane model
1986
Journal of the Atmospheric Sciences
11
Emanuels 1986 steady hurricane model
z
M M(qe) ? qe qe(p)
Boundary layer controls dqe/dM
h
Region I Region II
Region III (RH 80)
rmax
re
r
E86 ignores BL dynamics here!
12
2008
Thermodynamic view has problems!
Available on my website
13
Revised
Cannot ignore unbalanced BL dynamics!
14
Accepted subject to minor revision
WISHE Wind induced surface heat exchange
Basic air-sea interaction feedback loop Increase
in surface wind speed gt Increase in surface
moisture transfer from the sea surface
gt Increase in fuel supply to the storm
gt Increasing wind speed
15
Asymmetric view
Available http//www.meteo.physik.uni-muenchen.de
/roger

• Idealized numerical model simulations with
  simple physics (MM5)
• 5 km (1.67 km) resolution in the finest nest,
  24 s-levels

16
Evolution of Intensity
17
Vertical velocity\vorticity pattern at 24 h
850 mb 1.5 km 850 mb
Vertical velocity
Relative vorticity
18
Vertical vorticity evolution at 850 mb
10 h
12 h
19
Vertical vorticity pattern at 850 mb
18 h
24 h
300 km
300 km
300 km
20
Vertical vorticity pattern at 850 mb
36 h
48 h
21
Interim conclusions

• The flow evolution is intrinsically asymmetric.
• The asymmetries are associated with rotating
  convective structures that are essentially
  stochastic in nature.
• We call these structures vortical hot towers
  (VHTs).
• Their convective nature suggests that the
  structures may be sensitive to the low-level
  moisture distribution, which is known to possess
  significant variability on small space scales.
• Suggests a need for ensemble experiments with
  random moisture perturbations.

22
Evolution of local intensity 10 ensembles
control
VTmax
From Nguyen et al. 2008
23
Vertical velocity pattern at 850 mb at 24 h
control
Ensemble 1
Ensemble 3
Ensemble 2
From Nguyen et al. 2008
24
Is WISHE relevant?
Capped flux experiments
25
In press
26
Azimuthal average of the Nguyen et al. control
calculation
Tangential wind speed
27
Radial wind speed
vertical velocity
28
Movie
Time-height sequence of Absolute Angular Momentum
29
Revised view of intensification two mechanisms
15
10
z km
5
M conserved
0
50
100
r km
M reduced by friction, but strong convergence ?
small r
30
Exciting times!

• There is much work to do to pursue all the
  consequences of our recent findings and the new
  paradigm for intensification.
• We need to determine the limits of predictability
  for intensity and especially for rapid
  intensification.
• We need to much better understand the flow in the
  inner core, beneath and inside the eyewall, and
  to determine the utility of conventional
  (boundary layer) representations of this region
  in models.
• We need to develop a new theory for the potential
  intensity of tropical cyclones for climate
  assessments.

31
Thank you