The Structure and Dynamics of  Coherent Vortices in the Eyewall Boundary Layer of Tropical Cyclones Daniel P. Stern

The Structure and Dynamics of Coherent Vortices in the Eyewall
Boundary Layer of Tropical Cyclones
Daniel P. Stern
George H. Bryan
National Center for Atmospheric Research
NCAR is sponsored by the National Science Foundation
Motivation
• Dropsondes occasionally sample 10‐25 m/s updrafts within (and near) the TC boundary layer (Stern and Aberson, 2006).
Vertical Velocity vs. Height
• Such extreme updrafts are almost exclusively a phenomenon of the eyewall
region of Category 4 and 5 hurricanes.
22 m/s
Motivation
• The extreme updrafts appear to be associated with extreme horizontal windspeeds (> 90 m/s).
Wind Speed vs. Height
height (m)
Windspeed (m/s), Isabel 030913
4000
3800
3600
3400
3200
3000
2800
2600
2400
2200
2000
1800
1600
1400
1200
1000
800
600
400
200
0
0
175248
175422
175427
175436
5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 105
windspeed (m/s)
• Hypothesis: Extreme updrafts and near‐surface wind speeds are associated with small‐scale (~1 km) coherent vortices.
100 m/s
Goal: Use LES simulations to investigate the structure and dynamics of these features.
• What is the relationship between extreme updrafts and horizontal wind speed?
• Where do near‐surface wind maxima originate?
• Is buoyancy important for driving strong updrafts in and near the boundary layer?
• How should we interpret dropsonde “vertical” profiles?
Numerical Model
• CM1 (Bryan and Rotunno 2009, MWR)
• A single domain, with grid stretching
•
Fine‐mesh part of domain:
Fine‐mesh part of domain: – 70 km × 70 km × 3 km – Δ = Δx = Δy = constant; Δz = Δx/2
– Modified Deardorff subgrid model (TKE scheme, e.g., “1.5 order”)
•
Rest of domain:
– Δx, Δy, Δz increase gradually
– Parameterized turbulence (i.e., PBL scheme)
Model Setup
• 125 m horizontal grid spacing
• 62 m vertical grid spacing below 4 km height.
• 28 C SST
• Homogeneous environment (Dunion
“Moist Tropical” mean sounding).
• No mean flow or shear
• f‐plane (20 N)
+20 m/s W
+90 m/s Wind Speed
+0.18 Vorticity
Where Do Near‐Surface Extreme Windspeeds
Originate?
Where Do Near‐Surface Extreme Windspeeds
Originate?
• A number of studies (Wurman and Winslow 1998, Franklin et al. 2003, Aberson et al. 2006) have hypothesized that extreme surface winds result from downdrafts.
Where Do Near‐Surface Extreme Windspeeds
Originate?
• A number of studies (Wurman and Winslow 1998, Franklin et al. 2003, Aberson et al. 2006) have hypothesized that extreme surface winds result from downdrafts.
• We can test this idea in our simulations, through parcel trajectory analysis.
Where Do Near‐Surface Extreme Windspeeds
Originate?
• A number of studies (Wurman and Winslow 1998, Franklin et al. 2003, Aberson et al. 2006) have hypothesized that extreme surface winds result from downdrafts.
• We can test this idea in our simulations, through parcel trajectory analysis.
• Next, we examine a trajectory which enters a near‐
surface wind maximum.
What is the relationship between the updraft, horizontal winds, and vorticity?
• We next examine composite fields from many updrafts
1. At a given level, find all points where w >= 12 m/s.
1. Interpolate to cylindrical coordinates.
1. Take a 1x1 km box (in radius/azimuth) around each point.
1. Average all such boxes.
z=100 m
Vertical Velocity
Vertical Vorticity
Perturbation Vt
Perturbation Vr
Downstream/Cyclo
nic
Towards Center
z=100 m
Vertical Velocity
Perturbation v
Buoyant Acceleration
Dynamic Acceleration
Downstream/Cyclo
nic
Towards Center
How Should We Interpret Dropsondes?
• Dropsondes are advected by the horizontal and vertical wind, but also fall relative to the air at 10‐12 m/s.
• Profiles are not actually vertical, nor are dropsondes
necessarily following along with dynamical features.
• We can use “simulated” dropsondes to try to understand what real dropsondes are actually sampling.
Wind Speed vs. Height
Isabel
Wind Speed vs. Height
2400
2200
2000
Isabel Dropsonde
W vs. Height
Isabel
1800
height (km)
1600
1400
1200
1000
800
600
400
200
0
55
60
65
70
75
80
85
90
95
100
105
windspeed (m/s)
Wind Speed vs. Height
2400
Simulated Dropsonde
Simulated 2200
2000
1800
height (km)
1600
1400
1200
1000
800
600
400
200
0
55
60
65
70
75
80
85
windspeed (m/s)
90
95
100
105
Simulated z=560 m
z=10 m
z=560 m
z=10 m
Wind Speed=75 m/s
Wind Speed>90 m/s
z=560 m
z=10 m
z=560 m
z=10 m
z=800 m
z=560 m
z=10 m
Summary and Conclusions
• Strong updrafts in the boundary layer are closely associated with intense small‐scale vortices, as well as extreme near‐surface horizontal wind speeds.
• Near‐surface extreme winds can result from surface‐
layer inflow, without the need for a downdraft to bring high momentum air to the surface. • We quantitatively showed that buoyancy is not a significant contributor to extreme low‐level updrafts.
• Apparent sharp vertical gradients in dropsonde profiles are likely horizontal.
Bonus Slides
How does the distribution of simulated updrafts compare to observations?
Observed Updrafts
Jorgensen et al. (1985)
Simulated Updrafts
How does the distribution of simulated updrafts compare to observations?
Observed Updrafts
Jorgensen et al. (1985)
Simulated Updrafts
Storm Center
Increasing Radius
Marks et al. (2008)
Storm Center
Increasing Radius
Wind Speed
Marks et al. (2008)
Storm Center
Increasing Radius
Vertical Velocityy
Marks et al. (2008)
Examine updraft in a 1x1x1 km cube
w =+12 m/s
z (km)
x (km)
y (km)
Add Vorticity Isosurface
w =+12 m/s
z (km)
x (km)
+0.15 s‐1
y (km)
Total Acceleration (ms‐2; +/‐ 0.5 black)
w =+12 m/s
z (km)
x (km)
+0.15 s‐1
y (km)
Acceleration from “dynamics” (ms‐2; +/‐ 0.5 black)
w =+12 m/s
z (km)
x (km)
+0.15 s‐1
y (km)
Acceleration from buoyancy (ms‐2; 0 black)
w =+12 m/s
z (km)
x (km)
+0.15 s‐1
y (km)
`