TROPICAL CYCLONES 
likely at some altitude high above the storm center, 
has summarized how considerable outflow may take 
place otherwise. Following Durst and Sutcliffe [6] he 
has considered the effect of angular momentum of 
rising surface air. Because of the decrease of pressure- 
gradient force with height, the rising air seeks to draw 
away from the center as soon as the centrifugal force 
exceeds the pressure-gradient force. Values indicating 
the variations of the pressure-gradient force with height 
are tabulated below. The values in the right-hand 
column are the ratios (expressed in per cent) of the 
pressure-gradient force at different elevations to that 
at the surface. 
H(km) Per cent Al(km) Per cent 
SiCMEE eee. 100 
6 0b tg ORE ee 98 
Ch o.6.0 Soe ee 92 
OMe RE WeY eetersnmtesere Bee 90 
Gh sg ad ACO oe 80 
Dac amines ee eo eee 70 
Os coum Seca meee 63 
Coo bees Sa eee 54 
G)occ)a:p.0 6 6 Rr oe 28 
9. ee 25 (After Richl) 
If it is assumed that the ascending current retains 
its angular momentum, withdrawal from the central 
area would start approximately at the 3-km level, 
according to Riehl. At this elevation the magnitude 
of the pressure-gradient force directed toward the hur- 
ricane center begins to drop significantly with height. 
In reality, outflow will begin at even lower heights, 
owing to the decrease of the frictional force above 
the surface. 
Durst and Sutcliffe also postulate an imcrease of the 
vertical velocity with height and show that such an 
imcrease may contribute to the radial outflow of the 
air aloft. In view of the probability that convergence 
is restricted to the vicinity of the surface, Riehl con- 
cludes it is unlikely that the upward vertical motion 
increases with height except near the surface. 
Height of Tropical Cyclones. Tropical cyclones are 
associated with stronger pressure gradients than are 
observed in any other weather phenomenon where we 
must deal with hydrostatic equilibrium. The equaliza- 
tion of this pressure difference between the center 
and the periphery of the storm at some point in the 
upper air is dependent primarily upon a thermal gra- 
dient directed so that the warmest air is in the core 
of the storm. Haurwitz [10] has shown impressively 
how high this temperature must be if the circulations 
of very small storms are to disappear at levels at or 
below twenty thousand feet. For example, in a rela- 
tively moderate tropical cyclone with a pressure differ- 
ence of only 40 mb from the center to the outer edge of 
the surface vortex, and assuming a reasonable lapse 
rate in the outer vortex, he found that the temperature 
in the eye must increase with elevation at the rate 
of 18C km if the circulation of the storm is to disap- 
pear at ten thousand feet. The lapse rate must be 
zero, or isothermal, if the circulation is to disappear 
at twenty thousand feet. For the circulation to dis- 
appear at thirty thousand feet the lapse rate must be 
only slightly more than 4.5C km. The two storms for 
893 
which soundings in the eye are available had average 
lapse rates of not less than 4C km=. Since the more 
severe hurricanes are not infrequently associated with 
pressure differences of 70-90 mb between the center 
and the periphery, it can readily be seen that such 
storms must extend high into the upper troposphere. 
Figure 3 shows a sounding in the eye made at 
Tampa, Florida, on October 18-19, 1944, and two 
wo 
a 
<t OCTOBER 19, 1944 
= 7 0800 EST 
a ~ 
= 
500 S 
=90° =80° -70° -60° -50° =40° =30° =20° -10° 0° 
TEMPERATURE (°C) 
400 
500 
OCTOBER 19, 1944 
ce 0600 EST 
o OCTOBER 18, 1944 X\. \ (in THE EYE) 
& 600 1700 EST WS 
5 (EYE MINUS I3H) \\i 
=) EN 
= 700 Ms 
OCTOBER 18,1944 Ay * 
00 EST \} 
B00) (EYE MINUS 19H) \\ 
X. 
900 
1000 
1050 
=50° -40* =30° =20° =|0° 0° 10° 50° 
2OSmRSOS 
TEMPERATURE (°C) 
40° 
Fie. 3.—Soundings at Tampa, Florida, during the approach 
and arrival of a hurricane center, October 18-19, 1944. 
soundings made earlier. Figure 4 presents a similar 
series for October 7-8, 1946. Both series represent 
conditions during the approach of a hurricane, and 
the warmth of the eye is clearly indicated. It should be 
noted, however, that both storms were in a state of 
recurvature and it is not certain that the radiosonde 
did not pass out of the eye. 
Thermal Structure. There is as yet little actual data 
to verify theoretical models of temperature distribu- 
tion in tropical cyclones above the surface. A number 
of pressure-anomaly cross-section charts by Simpson 
have indicated that these storms extend to very high 
levels and have a double cellular structure with the 
lower warm-core cell capped by an upper cold-core 
cell. 
Tropopause Heights in Tropical Cyclones. Until re- 
cently it was generally thought that low tropopauses 
accompanied tropical cyclones. The reasons most com- 
monly advanced for the low tropopause were the sup- 
posed absence of convection in the eye and the stronger 
