Tropical Meteorology All

ESCI 344 – Tropical Meteorology
Lesson 1 – Introduction to the Tropics
References: Forecaster’s Guide to Tropical Meteorology (updated), Ramage

Tropical Climatology (2nd ed), McGregor and Nieuwolt
Tropical Meteorology, Tarakanov
Climate and Weather in the Tropics, Riehl
Tropical Meteorology in Africa, MUNITALP
Reading: McGregor and Nieuwolt, Chapter 1

Hagen et al., “Atmospheric Tides” (e-reserve)
GEOGRAPHY
For meteorological purposes, the Tropics may be roughly defined as that region
lying between the boundaries separating the easterlies and the westerlies in the
middle troposphere.
This boundary varies
seasonally
longitudinally
A knowledge of geography is important for a meteorologist. Below is a list of
places and locations that I expect you to be able to find on a map.
India Marshall Islands Midway Island
Madagascar U.S. Virgin Islands Peru
Somalia The Hawaiian Islands Pacific Ocean
Japan Guam Atlantic Ocean
Vietnam Wake Island Indian Ocean
Cuba Lesser Antilles Caribbean Sea
Taiwan Tahiti Persian Gulf
Malay Peninsula Fiji Bay of Bengal
Baja California Indonesia North Arabian Sea
Australia The Philippines South China Sea
One-half of the Earth’s surface area is contained between 30S and 30N.
MAJOR DIFFERENCES BETWEEN TROPICS AND MIDLATITUDES

The main differences between the tropical atmosphere and the midlatitudes are:
Coriolis parameter is small in Tropics. The implications of this are
The wind cannot be assumed to be geostrophic, or even quasi-
geostrophic.
The dynamic link between divergence/convergence and vorticity tendency
is much weaker than in higher latitudes.
Circulations must be much larger in order to be affected by rotation of
the Earth. One manifestation of this is that land/sea breezes penetrate
further inland in Tropics than in midlatitudes.
Weather systems in the Tropics are dominated by convection and the effects
of water vapor and latent heating.
The planetary boundary layer (PBL) is much deeper in tropics (because of
convection).
The tropopause is much higher and colder in Tropics.
Sea-level pressure exhibits pronounced diurnal and semi-diurnal cycle due to
gravitational and thermal tides in the atmosphere.
THE BREAKDOWN OF THE GEOSTROPHIC APPROXIMATION

The simplified horizontal momentum equation is

DV 1
= − ∇p − kˆ × f V

Dt ρ
A B C
Term A is the inertial term which represents the acceleration of the wind (it
includes the local derivative and advection terms).
Term B is the pressure gradient term.
Term C is the Coriolis term.
The relative strengths of the terms is indicated by the Rossby number,
R0 = U f L
where U is the characteristic wind speed, f is the Coriolis parameter, and L is the
characteristic horizontal length scale.
The Rossby number can be thought of as showing the ratio of the inertial forces
(Term A) to the Coriolis force (Term C).
2
A small Rossby number (R0 << 1) implies that the inertial terms are
unimportant, and that the pressure gradient force balances the Coriolis force
(geostrophic balance).
A large Rossby number (R0 >> 1) implies that the Coriolis term is
unimportant, and that the inertial terms and the pressure gradient term
balance (cyclostrophic balance).
For Rossby numbers on the order of unity, no term can be ignored, and
gradient balance holds.
For synoptic scale motions we typically use values of U ~ 10 m s−1 and L ~ 106 m.
The graph below shows the Rossby number as a function of latitude for these
values of U and L.
Rossby Number
4.5
4
3.5
3
2.5
2
1.5
1
0.5
0
0 20 40 60 80
latitude
Poleward of 20°° the Rossby number is small enough that the geostrophic
wind and actual wind are fairly close.
Equatorward of 20°° the Rossby number is no longer small, and the
geostrophic and actual wind can differ greatly, especially as the Equator is
approached.
The comparatively large values of Rossby number in the Tropics vs. the Mid-
latitudes means that on the synoptic scale quasi-geostrophic theory isn’t very
useful in explaining the dynamics of synoptic-scale tropical circulations.
3
On the planetary scale, where L ~ 107 m, the Rossby number does remain small
through most of the Tropics. Therefore, quasi-geostrophic theory may be
carefully applied to planetary-scale circulations such as the monsoon, the
Walker circulation, etc.
WIND-PRESSURE RELATION
In the mid-latitudes, due to the smallness of the Rossby number, Term A in the
horizontal momentum equation can often be ignored, so that there is a direct
relationship between the pressure field and the wind,
1
∇p = − kˆ × f V .

ρ
This is why in mid-latitudes we analyze isobars on the surface chart, from
which we can just look at the map and readily infer wind direction and
speed.
In the Tropics, all three terms must be retained, and gradient wind balance is
the norm.
There is no longer a simple relation between the pressure field and the wind.
Pressure gradients are quite weak in the Tropics except in tropical cyclones.
Because of the weak pressure gradients and lack of a simple wind-pressure
relationship, meteorologists prefer to analyze streamlines and isotachs rather
than isobars.
GRADIENT LEVEL WINDS

The two most useful levels for tropical wind analysis are the gradient level and
the 200 mb level.
The gradient level is defined as the lowest level at which frictional influences
from the surface are no longer important (i.e., just above the planetary
boundary layer) so that the actual wind will be very close to the gradient wind.
Though the actual gradient level varies with time and location, by convention the
gradient level is assumed to be at 850 mb.
4
So, the 850 and 200 mb charts, analyzed using streamlines, are the two most
common (but by no means the only) charts used for tropical analysis and
forecasting.
STREAMLINE ANALYSIS
Streamline analysis consists of drawing lines that are everywhere parallel to the
wind.
Some features that commonly show up on streamline analyses are discussed as
follows:
Asymptotes (diverging or converging)
These are streamlines from with other streamlines are either diffluent or
confluent.
Although in general diffluence and confluence don’t necessarily imply
divergence and convergence, it is common to refer to a diffluent asymptote as
a divergent asymptote, and a confluent asymptote as a convergent asymptote.
5
Cusps
These are intermediate between a wave and a developing closed circulation.

Cusps may be either cyclonic or anticyclonic.
Neutral points
These occur at the intersection of two asymptotes of opposite sign (divergent

and convergent).
They are sometimes referred to as cols, because they represent saddle points
on a pressure surface.
Inside a neutral point the wind is very light (theoretically it is zero at the
intersection of the asymptotes.)
It is theoretically possible to have a six-sided neutral point, involving the
intersection of three asymptotes; however, most meteorologists consider
them unlikely, and they should be avoided unless there is enough data to
truly support their existence.
6
Inflow and outflow
Pure inflow and outflow doesn’t usually appear on the synoptic scale.
Usually it is accompanied by rotation, and so appears in conjunction with
cyclones and anticyclones.
Cyclones and anticyclones
Cyclones and anticyclones represent closed circulations, which would have

closed isobars on a pressure map.
7
They are usually accompanied by either inflow or outflow, and so purely
circular or closed streamlines are not usually analyzed.
Although cyclones usually have inflow and anticyclones usually have outflow,
it is possible to have outflow cyclones and inflow anticyclones.
Some general rules and reminders:
Streamlines are everywhere parallel to the wind
Neutral points should be kept relatively small, and in areas of light wind.
Streamlines are not necessarily the same as trajectories (they are equivalent
only if the wind field is steady-state.)
The spacing of the streamlines is rather arbitrary. You should draw enough
to be useful. (Some analysts draw more streamlines in areas of strong wind,
and fewer in areas of weak wind, but this is not universal.)
Here are two rather arbitrary rules that I was taught to follow by veteran
analysts while in the Navy. I’ve never found these written down in any reliable
source, and have never been able to find a dynamical argument for these two
rules. I’ve often seen them violated, so there is nothing sacred about them. I
just present them here, because it’s the way I was taught. They are:
8
Neutral points should not be connected to each other.
Don’t diverge off of a line downstream of convergence (however,

convergence after divergence is OK.)
RELATION BETWEEN ASYMPTOTES AND CLOUDS

Since the ageostrophic wind is large in the Tropics, divergence and convergence
play an even more accentuated role than they do in the mid-latitudes.
The divergence/convergence pattern can often be inferred by the cloud patterns
on a satellite image, and used to aid the streamline analysis.
The following general rules can be very helpful:
Where the satellite picture shows a distinct, linear cloud feature, there is
likely a convergent asymptote on the low-level chart, and a divergent
asymptote on the upper-level chart.
9
In large areas that are cloud free, there is likely a divergent asymptote on the
low-level chart, and a convergent asymptote on the upper-level chart.
Sun-glint on the visible image is an indication of light winds, and can be used
to help locate ridge axes or neutral points.
ABSOLUTE VORTICITY
The synoptic scale vorticity equation is
D
(ζ + f ) = − ( ζ + f ) ∇ • V .

Dt
Because the absolute vorticity is comparatively weak in the Tropics, a given
amount of convergence or divergence will have a much smaller effect on absolute
vorticity than it would in the midlatitudes.
Areas of inflow and outflow with little associated rotation appear on
streamline charts in deep Tropics.
Strong anticyclones are not common in deep Tropics, because they would have
negative absolute vorticity.
ATMOSPHERIC TIDES
Another interesting difference in the tropics compared to the higher latitudes is
the relatively large amplitude, diurnal and semi-diurnal variation of sea-level
pressure due to atmospheric tides.
Both the diurnal and semidiurnal tides are due to solar heating and cooling, and
are not gravitationally induced.
The solar heating excites three main pressure oscillations in the atmosphere,
having periods of 24, 12, and 8 hours.
The horizontal and vertical structure of each of these modes of oscillation are
different.
The 12-hour (semidiurnal) wave has the largest amplitude at the surface, and is
the most meteorologically significant of the modes, since it has a range of 1 to
2 mb.
The amplitude of the semidiurnal variation decreases with height, and is less
than 0.5 mb at the tropopause.
10
The pressure maxima occur around 1000 and 2200 local time, with the
minima at about 0400 and 1600 local.
Due to the superposition of the 24 and 8-hr period waves with the 12-hr wave,
the morning pressure maximum is slightly larger than nighttime maximum,
while the afternoon minimum is slightly lower than the early morning minimum.
The diagram below shows a 5 day pressure trace Lime Tree Bay, VI (from the
National Data Buoy Center website). The semidiurnal variation is readily
apparent, with the maxima and minima occurring at the same time every day.
Times are GMT, and not local.
Since solar heating and cooling is the primary factor for producing the
atmospheric tides, the daily range of the pressure variation can vary with
cloudiness, water vapor content, and surface moisture.
In general:
The tidal range is greater near the Equator, and decreases as you go
poleward.
The tidal range is greater over land than over water.
The asymmetry of the maximums and minimums is also more pronounced
over land.
Though the signal from atmospheric tides can be deduced in the midlatitudes, it
is much smaller than in the tropics and isn’t really considered important in
midlatitude meteorology.
11
The tides must be accounted for when comparing the pressure tendency at a
single station in the Tropics.
Failure to account for the tides may lead a forecaster to mistakenly conclude
that a tropical cyclone is developing or approaching, or may even lead to
missing such development.
12
Lesson 2 – Tropical Cyclones: Definitions and Advisory Responsibilities
SOME ACRONYMS
CPHC – Central Pacific Hurricane Center
JMA – Japan Meteorological Agency
JTWC – Joint Typhoon Warning Center
NHC – National Hurricane Center
RSMC – Regional Specialized Meteorological Center
TCWC – Tropical Cyclone Warning Center
WMO – World Meteorological Organization (part of the United Nations)
TROPICAL CYCLONES DEFINED

Definitions relevant to tropical cyclones vary from region-to-region. For this
class, we will use definitions consistent with those of the National Hurricane
Center, Central Pacific Hurricane Center, and the Joint Typhoon Warning
Center.
Tropical Cyclone – A warm-core, non-frontal synoptic scale cyclone that forms
over the tropical or subtropical oceans, and having organized deep convection
and a closed surface circulation about a well defined center.
Tropical Wave – A trough or area of cyclonic curvature in the trade winds or
equatorial westerlies. May be associated with an upper-troposphere cold low or
a shear line.
Tropical Disturbance – A persistent (24 hours or more) non-frontal system of
organized convection forming over the tropical or subtropical oceans.
Tropical Depression - A tropical cyclone in the North Atlantic or North Pacific
that has maximum sustained surface winds (1-minute mean) of less than 34
knots.
Tropical Storm - A tropical cyclone in the North Atlantic or North Pacific that
has maximum sustained surface winds (1-minute mean) of 34 knots or greater,
but less than 64 knots.
Hurricane – A tropical cyclone in the North Atlantic or Eastern North Pacific
that has maximum sustained surface winds (1-minute mean) of 64 knots or
greater.
Typhoon - A tropical cyclone in the Northwest Pacific with maximum sustained
surface winds (1-minute mean) of 64 knots or greater, but less than 130 knots.
Super Typhoon - A tropical cyclone in the Northwest Pacific that has maximum
sustained surface winds (1-minute mean) of 130 knots or greater.
The table below shows the difference in definitions and nomenclature from
region-to-region.
Wind Atlantic/ Northwest Pacific North Southwest Southeast Pacific

speed Northeast Indian Indian Ocean
(kts) Pacific* JMA JTWC* Ocean
≤16 Tropical Tropical Tropical Tropical Tropical

17-27 Depression Depression Depression Depression Disturbance Depression
28-33 Deep Tropical
Depression Depression
34-47 Tropical Tropical Tropical Cyclonic Moderate Tropical Cyclone
Storm Storm Storm Storm Tropical Storm (Gale)
47-63 Severe Severe Severe Tropical Tropical Cyclone
Tropical Cyclonic Storm (Storm)
Storm Storm
64-89 Hurricane Typhoon Typhoon Very Tropical Tropical Cyclone
Severe Cyclone (Hurricane)
90- Cyclonic Intense Tropical
115 Storm Cyclone or
115- Very Intense
119 Tropical Severe Tropical
120- Super Cyclone Cyclone
129 Cyclonic
≥ 130 Super Storm
Typhoon
* Use 1-minute average winds, instead of 10-minute average winds.
2
SUBTROPICAL CYCLONES
Cyclones occasionally form over the tropical ocean that have a mixture of
tropical and extratropical characteristics. These are referred to as subtropical
cyclones.
The energetics are a mix of baroclinic instability as well as latent heating
from the sea.
Characteristics of subtropical cyclones are
Maximum winds not concentrated near center of storm.
Asymmetric wind distribution.
Asymmetric distribution of convection, not coincident with low center.
Two main types of subtropical cyclones are:
Those that are surface manifestations of cold-core upper-level lows.
Those that are mesoscale circulations (cold or warm core) initiated near a
frontal zone or shear line.
Subtropical cyclones can transition into tropical cyclones.
WARNINGS AND WARNING AGENCIES

The World Meteorological Organization (WMO), part of the United Nations
(UN), coordinates tropical cyclone and high seas warnings.
There are six official WMO Regional Specialized Meteorological Centers
(RSMC), and five additional Tropical Cyclone Warning Centers (TCWC).
The map below shows the locations and names of the RSMC’s and TCWC’s.
3
from World Meteorological Organization web site: http://www.wmo.ch/index-en.html
The RSMC’s and TCWC’s are operationally run by the respective weather
service of the host country. The associations of these centers with the host
country’s weather service are shown in the table below.
Center Organization
RSCM Miami U.S. National Hurricane Center (NHC)
RSMC Honolulu U.S. Central Pacific Hurricane Center (CPHC)
RSMC Tokyo Japan Meteorological Agency (JMA)
RSMC New Delhi India Meteorological Department
RSMC La Réunion Météo-France
RSMC Nadi Fiji Meteorological Service
TCWC Perth Bureau of Meteorology, Australia
TCWC Darwin Bureau of Meteorology, Australia
TCWC Brisbane Bureau of Meteorology, Australia
TCWC Port Moresby National Weather Service, Papua New Guinea
TCWC Wellington Meteorological Service of New Zealand, Ltd.
In addition to the WMO centers, the U.S. Navy and Air Force maintain the Joint
Typhoon Warning Center (JTWC) in Pearl Harbor, Hawaii, which provides
advisories and forecasts to U.S. government assets in the South Pacific, the
Northwest Pacific and Indian Oceans (both hemispheres).
JTWC is maintained so that the U.S. government, and U.S. military in
particular, do not have to rely on foreign advisories and guidance, which
may not be available in times of crisis or conflict.
Other than in the North West Pacific, the JTWC only provides warning for
tropical cyclones having sustained winds of 34 knots or greater, and refers to
these systems generically as “tropical cyclones”.
Many national weather services also do their own tropical cyclone forecasts and
issue their own warnings.
There is inconsistency in what averaging times are used for reporting sustained
winds in tropical cyclones among the various agencies.
The U.S. agencies (National Hurricane Center, Pacific Hurricane Center,
and the Joint Typhoon Warning Center) use 1-minute average winds
4
Note that standard U.S. surface observations use a 2-minute average wind
when reporting wind speed (FMH-1).
Most other agencies use 10-minute average winds, while some even use 3-
minute average winds.
For the same wind conditions, a 1-minute averaged wind will be reported
as stronger than a 10-minute average wind.
For the same reported wind speed, a 10-minute averaged wind will be
about 15% stronger than a 1-minute averaged wind (e.g., a 55 kt 10-
minute averaged wind is roughly equivalent to a 63 kt 1-minute averaged
wind).
The difference in averaging explains why sometimes JTWC will sometimes
report a stronger intensity than will RSMC Tokyo.
There is also inconsistency as to the minimum sustained winds required for
issuance of tropical cyclone advisories. This varies from region-to-region and
agency-to-agency, with some providing advisories at the depression level, while
others wait until the winds are 34 knots or greater.
NHC, CPHC, JMA, and JTWC (Northwest Pacific only) provide advisories
on tropical depressions once the sustained winds are 25 to 30 knots.
In the Indian Ocean and the Southwest Pacific, JTWC provides advisories
only for those cyclones with winds 34 knots or greater.
New Delhi provides advisories only for winds 34 knots or greater.
In the recent past, names were assigned by regional agencies, and were not
necessarily internationally recognized.
For example, up through the 1990’s, JTWC assigned names to Western
Pacific cyclones, while JMA simply referred to them by number.
However, pretty much all regions now assign names to those cyclones whose
winds are 34 knots or greater, and the name lists are coordinated by the WMO
member countries in the respective regions.
5
SAFFIR-SIMPSON SCALE
The Saffir-Simpson scale is a classification scheme for tropical cyclones based on
the maximum sustained wind (1-minute mean) and also type and extent of
damage.
It is mainly used by RSMC Miami, RSMC Honolulu, and JTWC. The wind
categories are
Saffir-Simpson Max Sustained Wind
Category knots mph
1 64 - 82 74 - 95
2 83 - 95 96 - 110
3 96 - 113 111 - 130
4 114 - 134 131 - 155
5 135 + 156 +
Other countries may use different scales (e.g., Australia’s Bureau of

Meteorology)
6
Lesson 3 – General Circulation of the Tropics

Climate Dynamics of the Tropics, Hastenrath
General Circulation of the Tropical Atmosphere, Vol II, Newell et al.
“The South Pacific Convergence Zone (SPCZ): A Review”, Vincent,
Mon.Wea. Rev., 122, 1949-1970, 1994
“The Central Pacific Near-Equatorial Convergence Zone”, Ramage,
J. Geophys. Res., 86, 6580-6598
Reading: McGregor and Nieuwolt, Chapters 5 and 6 (thru page 110 only)
Vincent, “The SPCZ: A Review” (e-reserve)
Krishnamurti, “Walker Circulation” Figure 1 (e-reserve)
Lau and Yang, “Walker Circulation” (e-reserve)
James, “Hadley Circulation” (e-reserve)
Waliser, “Intertropical Convergence Zones” (e-reserve)
Hastenrath, “Tropical Climates” (e-reserve)
Madden, “Intraseasonal Oscillation (MJO)” (e-reserve)
TERMINOLOGY
Boreal refers to the Northern Hemisphere
Austral refers to the Southern Hemisphere
LATITUDINAL HEAT IMBALANCE

Net radiation flux is defined as the difference in incoming radiation flux and
outgoing radiation flux.
A positive net radiation flux indicates a surplus of energy, while a negative
net radiation flux indicates a deficit.
When the earth-atmosphere system is considered as a whole, there is a positive
net radiation flux between about 40N and 40S, while there is a negative net
radiation flux poleward of 40 in both hemispheres.
In order for a steady-state temperature to be achieved, there must be transport
of heat from the earth’s surface to the atmosphere, and from the tropics to the
polar regions.
It is this latitudinal heat imbalance that drives the general circulation of the
atmosphere and oceans.
IMPORTANCE OF LATENT HEAT VERSUS SENSIBLE HEAT
The surface provides energy to the atmosphere in two ways:
Sensible heat, which leads directly to an increase in temperature of the
atmosphere.
Latent heating, which is essentially energy stored in water vapor. As the
water vapor condenses it releases energy to the atmosphere.
Water vapor in the atmosphere can be thought of as stored energy.
80% of the energy provided by the surface to the atmosphere is via latent
heating.
This underscores the importance of water vapor, and the oceans, on the
atmosphere, and is why no meteorologist’s education is complete without
taking a course in Oceanography. (Dr. DeCaria’s personal opinion.)
Half of the latent heat supplied by the ocean to the atmosphere comes from
the tropical oceans (between 30N and 30S), and points out why what happens
in the tropics is so important to what happens in the non-tropical
atmosphere.
The ratio of sensible heat flux to latent heat flux from the surface to the
atmosphere is called the Bowen ratio.
The lower the Bowen ratio, the greater the contribution from latent heat.
Typical values of the Bowen ratio (from the Glossary of Meteorology) are
Surface type Bowen ratio
Semi-arid 5
Grassland and forest 0.5
Irrigated orchard or grass 0.2
Ocean 0.1
The Bowen ratio can actually be negative, which would occur when the air is
warmer than the surface, but evaporation is occurring.
ROLE OF CONVECTION
The overwhelming majority of heat transferred to the atmosphere from the
surface is in the form of latent heat stored in water vapor.
2
The heat is released to the atmosphere when the water vapor condenses in
convective clouds.
Convection is therefore an extremely important process for the general
circulation of the atmosphere, especially in the tropics.
In two papers, Malkus and Riehl estimated that the heat balance of the tropics
can be maintained by around 30 synoptic-scale disturbances consisting of a total
of several thousand giant cumulonimbus clouds.
THE HADLEY CIRCULATION

The Hadley circulation is a meridional circulation with an ascending branch in
the extreme low-latitudes and a sinking branch in the subtropics.
If the earth were not rotating the Hadley circulation would be expected to
reach all the way to the poles.
The ascending branch is associated with the zone of maximum solar heating, and
migrates with the seasons.
If the earth’s surface was uniform, the mean position of the ascending branch
would be at the Equator.
Due to the asymmetric distribution of land between the Northern and Southern
Hemispheres, and the very different thermal properties of land versus water, the
mean position of the ascending branch of the Hadley cell is at about 5N.
5N is often referred to as the Meteorological Equator.
The ascending branch varies from about 5S to 15N over the course of the year.
In the ascending branch, heat (primarily latent) is transported from the surface
to the upper troposphere, where it is then transported poleward.
In each hemisphere, the Hadley cell is strongest in winter, and weakest in
summer.
The Hadley cell in the winter hemisphere is stronger than its counterpart in the
summer hemisphere.
3
ZONAL CIRCULATIONS
In addition to the meridional circulation of the Hadley cell, numerous attempts
have been made to establish the existence of large-scale zonal circulations in the
Tropics.
There have been several, often contradictory theoretical and observational
studies which have attempted to describe and document the structure of these
circulations.
These studies differ mainly in the number and locations of the cells. However,
the existence of three large cells centered over the Pacific, Atlantic, and Indian
Oceans is generally agreed upon.
Krishanmurti’s 1971 map of velocity potential and divergence field shows:
Large upper-level divergence maximum in summer over Bay of Bengal.
Large upper-level convergence maximum in summer off tip of Baja
California.
Pacific Ocean – The largest cell if located over the Pacific, and involves rising
motion in the Western Pacific in the vicinity of Indonesia (Indonesia is often
referred to as the Maritime Continent), with sinking motion in the Eastern
Pacific off of South and Central America.
This circulation is known as the Walker Circulation.
Atlantic Ocean – Over the Atlantic there is rising motion in the West, with
sinking motion in the East.
Indian Ocean – In the Indian Ocean there is rising motion over Indonesia, with
sinking motion off of the east coast of Africa.
THE SUBTROPICAL HIGHS

Found over the oceans.
In both hemispheres they are more intense in July than in January.
In both hemispheres the anticyclones are more westward in July and more
eastward in January.
The exception to the above is the South Pacific high.
Centers are closer to Equator in respective winters, and closer to Poles in
respective summers.
4
Image provided by the NOAA-CIRES Climate Diagnostics Center, Boulder, Colorado,
USA, from their Web site at http://www.cdc.noaa.gov/.
January
July
The upper-level high is usually located to the west of the surface high.
Because of the asymmetry in the upper-level versus surface high, as well as the
fact that ocean temperatures are generally warmer in the western ocean basins,
the weather characteristics associated with the subtropical highs are distributed
asymmetrically as follows:
5
Characteristic Western portion Eastern portion
Vertical motion Weak subsidence or rising Strong subsidence
motion
Surface wind From Equator From Pole
Advection Warm Cold
Temperature Warmer Colder
Dew point Higher Lower
Stability Weakly stable Strongly stable
Clouds Open-cell Clear or closed-cell
The longitudinal axis of the high can often be inferred from satellite imagery
based on the transition from open-cell to closed-cell convection.
Formation and maintenance of the subtropical highs is complex, and is likely the
result of a combination of dynamic and thermal processes. Possible important
processes are:
Poleward moving air in the upper-branch of the Hadley cell experiences
upper-level convergence in the subtropics, resulting in higher surface
pressure.
This process by itself could not produce the subtropical highs because the
resulting subsidence would cause adiabatic warming, which would push
the lower-level pressure surfaces downward and work against creating of
a ridge (recall the hypsometric equation).
If diabatic cooling of the air were added to the above process, it could
offset the adiabatic warming during descent. Two sources of diabatic
cooling are:
Long-wave radiational cooling of the upper atmosphere.
Cooling of lower atmosphere from cold ocean currents or wintertime
continents.
Another possible dynamic process is the non-linear Beta effect. This effect
results in Polar anticyclones moving equatorward, providing mass to
maintain the subtropical highs.
6
THE TRADE WINDS
The trade winds result from the flow on the equatorial side of the subtropical
highs.
The trades blow generally from ENE in the NH and from ESE in the SH.
The trades extend over about 20°° of latitude in the summer hemisphere, and
about 30°° of latitude in the winter hemisphere.
They blow more toward the equator in winter than in summer.
Mean velocities are 3.6 – 7.2 m/s, and are stronger in winter than in summer.
The trades are very steady, though there is large inter-annual variability.
The trades have a 3-layer vertical structure:
Sub-cloud layer – The layer below the cloud bases.
Cloud layer
Inversion layer – Characterized by negative lapse rate, and therefore, the
tops of the convective clouds.
The trade-wind inversion is a subsidence inversion.
The inversion height and strength vary spatially as follows:
Zonally Meridionally
Height Increases toward the west Increases toward the Equator
Strength Decreases toward the west Decreases toward the Equator
As the inversion height and strength vary, so do the type and height of the
clouds, with small cumulus or stratocumulus prevailing in the east, while taller
cumulus clouds become more prevalent toward the west, or toward the Equator.
The spatial variation of the inversion is explained by the fact that subsidence is
strongest, and convection weakest, to the east of the subtropical high, resulting
in a low, strong inversion in that region.
As the air embedded in the trades moves westward and equatorward it
encounters less subsidence, as well as enhanced convection (since it is picking up
latent and sensible heat from the ocean surface.) This results in a weaker, higher
inversion toward the Equator, or toward the west.
7
The warming of the air as it moves westward helps to maintain the trade winds
by resulting in a lowering of surface pressure to the west (recall again the
hypsometric equation).
The maximum wind in the trades is usually found near the bottom of the cloud
layer.
Friction causes the wind to increase with height.
Thermal wind is westerly, so causes trades to decrease with height.
Result of these two effects is a wind max near the bottom of the cloud layer.
This vertical shear causes the trade-wind cumulus clouds to have their
characteristic appearance of leaning upstream.
THE INTERTROPICAL CONVERGENCE ZONE

The region where the trades from each hemisphere converge is commonly
known as the inter-tropical convergence zone (ITCZ).
Other names sometimes used for all or some of the ITCZ are equatorial trough,
monsoon trough, or meteorological equator.
The trough associated with the ITCZ is a thermal trough.
The ITCZ has a complex structure.
The zone of lowest pressure, highest temperature, and maximum wind
confluence is separated by 300 - 1000 km from the zone of maximum
cloudiness, rainfall, and convergence.
The convergence maximum is equatorward of the confluent zone in the wind
field.
A somewhat simplistic explanation for the equatorward position of the region of
maximum convergence is as follows:
As cross-equatorial flow from the winter hemisphere enters the summer
hemisphere and moves away from the Equator, Coriolis acceleration begins
to curve it anticyclonically.
In the region of maximum anticyclonic curvature the flow will be faster than
it is downstream, when it becomes more straight-line.
8
The deceleration of the flow downstream from the region of maximum
anticyclonic curvature (but before reaching the equatorial trough) results in
convergence.
The separation of the regions of maximum cloudiness and minimum pressure is
necessary to maintain the thermal trough.
If the max cloudiness were directly over the thermal trough, the solar energy
at the surface would be decreased, which would be a negative feedback for
maintaining the thermal trough and maximum in surface temperature.
The role of the sea-surface temperature (SST) maximum is not completely clear.
Some argue that SST maximum directly contributes to the formation of the
equatorial trough.
Others (e.g., Ramage) argue that the SST maximum is caused by the
convergence of the ocean surface waters from the converging trades, and is
therefore an artifact, rather than a cause, of the trough.
There is a positive feedback mechanism between the upper ocean and the
atmosphere that also aids in the formation of the equatorial trough.
Where the surface winds are strong there is more mixing of the upper ocean,
resulting in cooler surface temperatures.
In the region of the trough the surface winds are lighter, resulting in less
ocean mixing and warmer ocean temperatures, which result in warmer
atmospheric temperatures and lower surface pressure.
If the earth’s surface were uniform the position of the ITCZ would be oriented
along the lines of latitude, and its annual migration would be symmetric with
respect to the Equator.
Because of the land-water contrasts, the ITCZ is not oriented exactly zonally,
but meanders north and south.
The ITCZ takes its largest poleward excursions in the summer hemisphere over
large land masses.
Because the NH has much more land than the SH, the ITCZ is closer to the poles
in boreal summer than in austral summer.
Mean position of ITCZ
15 N in boreal summer
9
5 S in austral summer
Maps of highly reflective clouds gives an idea of the location of the ITCZ. Below
are monthly means (Image provided by the NOAA-CIRES Climate Diagnostics
Center, Boulder, Colorado, USA, from their Web site at http://www.cdc.noaa.gov/.)
10
There are two general types of circulations associated with the equatorial
trough.
A trade-wind trough occurs when there is a confluence of the trade winds
from each hemisphere, with no directional shear across the trough.
The trade-wind trough occurs over the open ocean areas of the North
Atlantic, and over the Northeast and North Central Pacific.
A monsoon trough occurs when the trades from the winter hemisphere
recurve as they cross the Equator, and so there is a westerly flow on the
equatorial side of the trough.
The monsoon trough occurs near large continental areas, and is prevalent
in the Western Pacific and Indian Oceans.
A region where the cross-equatorial winds recurve from easterly to westerly is
referred to as a near-equatorial buffer zone.
Since the buffer zone is so close to the Equator, and Coriolis is negligible in
this region, terms such as cyclonic or anticyclonic have rather nebulous
meaning when applied to circulations in the buffer zone.
Trade-wind trough Monsoon Trough and Buffer Zone
The cloudiness associated with the ITCZ varies.

Generally narrow band of clouds over Atlantic and Eastern Pacific
associated with a trade-wind trough.
Broader over Western Pacific and Indian Ocean associated with the monsoon
trough.
11
THE SOUTH PACIFIC CONVERGENCE ZONE
Region of persistent convection in South Pacific.
Generally oriented east-southeastward.
More zonally oriented in the western portion, and diagonally oriented in the
eastern portion.
Present year round, but more active in austral summer.
Several theories for origin and maintenance.
SST gradients – Gradients in SST lead to moisture convergence in the low-
levels, with enhanced convection. Most applicable to zonal portion.
Land-sea distribution.
Equatorial/midlatitude wave dynamics, interaction, and energy convergence.
Tropical/mid-latitude interaction – Most applicable to diagonal portion.
The strength and location of the diagonal portion appears related to the
subtropical jet.
EL NINO, LA NINA, AND THE SOUTHERN OSCILLATION

Under “normal” conditions the easterly winds in the Tropics result in an
elevation of the thermocline (and cold surface waters) in the eastern tropical
Pacific ocean, and a lowering of the thermocline (with warm surface waters) in
the western tropical Pacific.
During an El Nino event the trade winds weaken or even reverse direction.
We’ve already seen that the tropical oceans adjust relatively quickly to changes
in equilibrium. The response during El Nino is a progressive lowering of the
thermocline from west to east.
The lowering of the thermocline is believed to be in part the result of an
equatorially trapped Kelvin wave traveling along the thermocline from west
to east.
El Nino occurs in conjunction with the Southern Oscillation, a shift in the
pressure patterns between the eastern and western tropical Pacific.
El Nino and the Southern Oscillation are closely linked. This is why the
phenomenon is often abbreviated and referred to as ENSO.
12
ENSO really isn’t an abnormal phenomenon. It can be thought of as just one of
several stable global climate modes. For reasons not completely understood, the
general circulation of the ocean/atmosphere system switches periodically
between modes.
The question then becomes, “Why aren’t the shifts regular and predictable”?
The reason is because “weather” is superimposed upon these climate modes.
The entire system is non-linear (perhaps chaotic?), so the response to
“weather” cannot be predicted.
MADDEN-JULIAN OSCILLATION (MJO)

This is a shorter-period (30-50 day) oscillation.
The oscillation is manifest as enhanced convection in the Indian Ocean which
then moves eastward into the Pacific.
The theory behind the MJO is not fully developed, but it is believed to be
partially explained by equatorial Kelvin-Rossby waves.
13
Lesson 4 – Seasonal-mean Wind Fields

General Circulation of the Tropical Atmosphere, Vol. II, Newell et al.
Reading: McGregor and Nieuwolt, Chapter 5 (e-reserve)

“Climatological Streamlines” (e-reserve)
“Climatological Streamlines (detailed)” (e-reserve)
“Tropical Cyclone Forecaster’s Reference Guide”, Chapter 2, Section 1.2 thru
1.5, (web resource) http://www.nrlmry.navy.mil/~chu
LOW-LEVEL WIND FIELD

January
Oceanic subtropical ridges merged with continental anticyclones.
Max easterly trade flow at about 10N.
Heat lows over Southern Africa.
In Atlantic and Eastern Pacific, trade-wind trough lies between 0 and 5N.
In Western Pacific and Indian Ocean, monsoon trough in SH, with northeast
monsoon firmly in place over Asia
April
Subtropical ridges close to January location.
African monsoon trough near 10N
In Atlantic and Eastern Pacific, trade-wind trough still lies between 0 and
5N.
Heat lows forming over China, Southern India, and Sahara Desert.
In Western Pacific and Indian Ocean, SH monsoon trough has moved
toward Equator.
A nearly symmetric trough forms in North Indian Ocean, and begins the first
of two tropical cyclone seasons in the Bay of Bengal and North Arabian Sea
July
African monsoon trough near 20N
Heat lows over Saudi Arabia and Mexico
Eastern Pacific now has a monsoon trough, instead of a trade-wind trough.
Monsoon trough sets in over South China sea and Western Pacific, and is
oriented northwest to southeast.
South Asia firmly entrenched in southwest monsoon flow, with monsoon
trough well inland.
October
Monsoon trough over North Africa is weak and lies near 10N.
Trade wind trough in Eastern Pacific.
Heat lows over Asia are gone.
Monsoon trough over South China Sea moves southward.
Monsoon trough draped across North Arabian Sea and Bay of Bengal,
leading to second tropical cyclone season in North Indian Ocean.
UPPER-LEVEL WIND FIELD

January
West Pacific and Indian Ocean under long ridges in both hemispheres.
Eastern Pacific and Atlantic have westerly flow, broken by anticyclone over
South America.
NH subtropical jet stream well developed, with strong maximum south of
Japan.
April
NH subtropical jet still present, but weaker.
SH subtropical jet moves toward Equator.
Ridges in both hemispheres have moved closer to equator.
Westerlies still dominate Eastern Pacific and Atlantic, but have pronounced
dip into SH.
July
Ridge over Asia has moved drastically northward.
NH subtropical jet has disappeared.
SH subtropical jet very strong near Australia.
Tropical easterly jet (TEJ) has formed over Asia.
Continuous ridge present in SH.
2
Upper-level troughs, termed tropical upper-tropospheric troughs (TUTT) have
formed in North Pacific and North Atlantic.
Anticyclone established over Mexico.
October
Anticyclone in Asia-Africa region moved back toward Equator.
NH subtropical jet reestablished.
Anticyclone that was over Mexico has moved south.
Variable westerlies reestablish over Atlantic and North Eastern Pacific.
SH subtropical jet weakens somewhat.
SUBTROPICAL JET STREAM

The subtropical jet stream located in the upper troposphere above where the
Hadley cell descends.
In the SH the subtropical jet is present year-round.
In the NH the subtropical jet disappears in summer.
There are three speed maxima, over the Middle East, Asia, and America.
The subtropical jet is often very close to the polar jet.
Forecasters often refer to the subtropical jet as the “southern branch” of the
jet stream, though this is misleading as it implies that the subtropical jet is
just an offshoot of the polar jet.
The dynamics of the subtropical jet are very different than those of the polar jet.
The origin and maintenance of the subtropical jet is complex, and not completely
understood. Some significant factors are:
conservation of angular momentum as the air in the upper branch of the
Hadley cell moves poleward.
meridional gradients in the surface heat budget
Central Asian topography
Though there is no surface front or weather characteristics associated with the
subtropical jet, there is an upper-tropospheric front associated with it. This is
consistent with thermal wind balance.
3
The gradients across the upper-tropospheric front are due in part to the
upper air convergence in this region as the upper branch of the Hadley cell
converges with the mid-latitude westerlies.
Though there is no surface weather associated with the subtropical jet, it does
interact with the midlatitudes and does play a role in midlatitude dynamics and
influences midlatitude weather.
TROPICAL EASTERLY JET

The tropical easterly jet (or TEJ) forms over Southern Asia and East Africa
during the boreal summer.
It exists because the mid-tropospheric thickness gradient is reversed due to the
intense heating over the Tibetan Plateau.
WEST AFRICAN MID-TROPOSPHERIC JET

The West African mid-tropospheric jet is found at about 600 mb, and extends
from the Red Sea to the Atlantic Coast.
Mean position is about 15N.
Strongest in boreal summer (~10 m/s).
Exists due to easterly thermal wind caused by heating over Sahara Dessert
contrasted with cooler, maritime air to South.
EAST AFRICAN LOW-LEVEL JET (SOMALI JET)

Found near 850 mb
Velocities 12-15 m/s.
African highlands are important geographic feature for creation and
maintenance of jet.
Maximum speeds in early morning, with minimum in early afternoon.
Likely due to changes in boundary layer stability between morning and
afternoon.
Important for mass transport and water vapor transport across Equator during
southwest monsoon.
4
Lesson 5 – Tropical Cyclones: Climatology
References: A Global View of Tropical Cyclones, Elsberry (ed.)

The Hurricane, Pielke
Tropical Cyclones: Their evolution, structure, and effects, Anthes
Forecasters’ Guide to Tropical Meteorology, Atkinsson
Forecasters Guide to Tropical Meteorology (updated), Ramage
Global Guide to Tropical Cyclone Forecasting, Holland (ed.)
Reading: A Global View of Tropical Cyclones, Chapter 3 (e-reserve)

Tropical Climatology, pp 151 - 160
Global Guide to Tropical Cyclone Forecasting, Chapter 1 (online at
http://www.bom.gov.au/bmrc/pubs/tcguide/globa_guide_intro.htm)
Tropical Cyclone Forecasters Reference Guide, Chapter 3 (online at
http://www.nrlmry.navy.mil/~chu/)
REQUIREMENTS FOR FORMATION

In order for a tropical cyclone to form, the following general conditions must be
present:
Deep, warm ocean mixed layer.
Sea-surface temperature at least 26.5°°C.
Mixed layer depth of 45 meters or more.
Relative maxima in absolute vorticity in the lower troposphere
Need a preexisting cyclonic disturbance.
Must be more than a few degrees of latitude from the Equator.
Small values of vertical wind shear.
Disturbance must be in deep easterly flow, or in a region of light upper-
level winds.
Mean upward vertical motion with humid mid-levels.
GLOBAL CLIMATOLOGY
Note: Most of the statistics given in this section are from Gray, W.M., 1985:
Tropical Cyclone Global Climatology, WMO Technical Document WMO/TD-72, Vol.
I, 1985.
About 80 tropical cyclones per year world-wide reach tropical storm strength
(≥
≥ 34 kts).
About 50 – 55 each year world-wide reach hurricane/typhoon strength
(≥
≥ 64 kts).
The rate of occurrence globally is very steady.
Global average annual variation is small (about 7%).
Extreme variations are in the range of 16 to 22%.
Variability within a particular region is much larger than global variability.
Most (87%) form within 20°° of the Equator.
Those that form farther than 20°° from the Equator are usually in the
Northern Hemisphere.
Vast majority form in or near monsoon troughs.
Most of the remainder form in tropical waves
A few form along old frontal zones or shear lines.
Cyclogenesis tends to cluster in 2-3 week active periods that occur after 2-3 week
inactive periods.
NORTH ATLANTIC
11.6% of global total tropical storm strength and higher.
12% of global total hurricane strength and higher.
97% occur between June 1 and November 30
There is no month without at least one tropical cyclone
Peak of season is September 10
Position and strength of subtropical jet is key factor in seasonality, along with
variations in SST.
Genesis regions migrate throughout season in response to upper-level winds
and SST.
Per year there are on average
10.1 named cyclones
5.9 hurricanes
2.5 category III or greater
Variability
Have been as few as 1, to over 20 cyclones
Modes of genesis different from global averages
2
More than half form in tropical waves coming off of Saharan Africa.
Less than half form along ITCZ, with a few forming in baroclinic zones
associated with old fronts or shear lines.
Possible 25 to 40 year cycle in numbers (not enough data to conclusively
determine).
SOUTH ATLANTIC
Tropical cyclones are very rare in the South Atlantic.
No ITCZ
Strong vertical shear
In satellite era there have been two tropical cyclones in the South Atlantic.
One, in March, 2004, becoming a Category I hurricane.
Formed from upper-level low.
In April, 1991 there was a nominal tropical storm off of Africa.
NORTHEAST AND NORTHCENTRAL PACIFIC

19.9% of global total hurricane strength and higher.
Numbers undercounted in pre-satellite era (not many ships in this area
compared to Atlantic).
Season similar to Atlantic.
Peak in late-August
16.4 named cyclones
9.2 hurricanes
4.0 category III or greater
Region off of Central America has highest density of genesis points on the globe.
Little genesis south of 10°° due to cold upwelling of ocean waters.
Modes of genesis not well documented.
Controversy concerning whether “African” waves propagate across Central
America and spawn Pacific cyclones.
3
WESTERN NORTH PACIFIC
Granddaddy of all basins
Noted for high occurrence of very large and very intense storms.
Lowest SLP on record was 870 mb in Super Typhoon Tip (October, 1979).
Only basin that is active throughout the year.
Highly seasonal, with peak in late-July/early-August
Can occur in any month
Per year the average is
26 named storms
16 typhoons
Genesis regions migrate seasonally with position of monsoon trough.
Vast majority of cyclones form in monsoon trough, with a small minority
forming in tropical waves.
Tropical upper-tropospheric trough (TUTT) can play significant role in genesis
and development.
NORTH INDIAN OCEAN

Deadliest in world.
Low-lying, flood prone areas of Bangladesh with large population density.
1970 cyclone killed over 300,000 people.
Several cyclones have killed 100,000 to 200,000 people each.
Most recently, Tropical Cyclone Nargis (May 2008) killed up to 150,000 in
Myanmar (Burma).
Highly seasonal, with two peaks in May and November.
The bimodal distribution is associated with the transition seasons of the
monsoon, as the monsoon trough is over water during these times.
4
During height of summer monsoon, the monsoon trough is well inland, and
there is also strong vertical shear due to the Tropical Easterly Jet, which
suppresses cyclogenesis.
November peak is more pronounced than May peak.
Majority occur in the Bay of Bengal, with far fewer in the Arabian Sea.
5 – 6 tropical storm strength cyclones per year
Range is 1 – 10
2 to 3 hurricane strength cyclones per year
SOUTHWEST INDIAN OCEAN (West of 100°°E)

Seasonal (October through May)
Peaks in January and February/March
Can get off-season genesis.
10.4 tropical storm strength cyclones
4.4 hurricane strength cyclones
Monsoon trough is important genesis feature.
SOUTHEAST INDIAN OCEAN (100°°E to 142°°E)

Seasonal (October through May)
Peaks in January and February/March
Monsoon trough is important genesis feature.
5
SOUTHWEST PACIFIC OCEAN (East of 142°°E)
Seasonal (October through April)
Peak in February/March
Monsoon trough and SPCZ are important genesis features.
Twin cyclones (one on each side of the Equator) sometimes form in early and
late season, often in conjunction with westerly wind burst along the Equator.
6
Lesson 6 – Local and Diurnal Circulations

“The interaction of trade wind and sea breeze, Hawaii,” Leopold,
J. Meteor., 6, 312-320, 1949
“The effects of a large island upon the trade-wind air stream,” Malkus,
Quart. J. Roy. Meteor. Soc., 81, 538-550, 1955
Reading: “The interaction of trade wind and sea breeze, Hawaii” (e-reserve)
“The effects of a large island upon the trade-wind air stream” (e-reserve)
GENERAL
Diurnal variation of solar radiation and temperature in Tropics is much larger
than annual variation.
In midlatitudes, annual variation is larger than diurnal variation.
Diurnal and mesoscale process are very dominant in the Tropics.
LAND-SEA BREEZE
Land-sea breezes caused by the differential heating across land-water
boundaries occur both in the midlatitudes and the tropics. However, the
influence of land-sea breezes is generally more pronounced in the tropics.
Inland or offshore extent of land-sea breeze is greater in the tropics compared
with the midlatitudes (~100 km, vs. ~10 km).
The difference in extent is explained by the weaker Coriolis acceleration in the
tropics.
Rossby radius of deformation is the ratio of the gravity wave speed divided by
the Coriolis parameter,
λR = c f .
The radius of deformation is essentially a measure of how large a circulation
must be in order for the effects of the earth’s rotation to be important.
Circulation smaller than the radius of deformation are not affected by the
earths rotation.
Coriolis helps to limit the horizontal effects of the land-sea breeze by turning
the wind and therefore limiting its inland-offshore influence.
In the tropics the radius of deformation is nearly an order of magnitude
larger than in the midlatitudes, so a circulation such as the sea breeze can
have a much larger horizontal extent without being turned by the Coriolis
acceleration.
The type of cloudiness and/or precipitation associated with the land-sea breeze
depends on the orientation of the coast with respect to the prevailing wind, as
well as the inland topography.
In a 1949 study, Leopold categorized four main types of land-sea breeze
interactions with the trade winds over islands. His four categories are:
Lanai type – Appropriate for small, low island in which the trade winds blow
up and over the peaks, and do not split.
Sea-breeze front is narrow and perpendicular to trades.
Maui type – Appropriate when peaks are tall enough to split trade wind flow.
Flanks of sea-breeze front exhibit shear lines trailing downwind.
Mauna Kea type – Appropriate for windward side of island with very tall
peaks.
Cloudiness forming on windward side during daytime as trades are
reinforced by sea breeze.
At night, land breeze front causes convergence offshore, with offshore
cloudiness and showers.
Kona type – Appropriate for leeward side of islands with very tall, broad
peaks.
During daytime, sea breeze moves in unimpeded by trade flow, which has
been blocked by tall, broad peaks.
Convective showers form inland as sea breeze brings in moist air that is
heated from below.
At night, down-slope land breeze suppresses clouds.
Leopold’s categories likely apply to other islands besides Hawaii.
2
Land-breeze interaction with trades has been documented as producing
enhanced rainfall at night and early morning on the east coasts of Brazil and the
Ivory Coast.
Large, mountainous islands (such as Puerto Rico) often have a clear ring of
suppressed cloudiness surrounding them in the afternoon.
This is due to the compensating subsidence from the vigorous convection
over the interior.
Land-lake breeze circulations occur over and around large lakes, such as Lake
Victoria in Africa.
MOUNTAIN CIRCULATIONS
Mountain-valley breezes are common in the tropics.
Valley breezes occur in afternoon, leading to cloudiness over mountains, with
clear skies in subsiding air over valleys.
Mountain breezes occur at night, leading to clearing over the mountains, and
cloudiness in valleys.
Mountain rainfall patterns in tropics often show a band of maximum rainfall
somewhere at around 1000 meters above the valley floor, with decreasing
rainfall at the foot or toward the top of the mountain.
This is mainly associated with areas where convective clouds dominate the
rainfall.
Lower slopes don’t experience as much orographic lift, and also experience
evaporation of rain below cloud base.
3
Upper slopes suffer from lower moisture availability.
Where stratiform clouds dominate, the rainfall maximum tends to be near the
top of the mountain.
STRESS-DIFFERENTIAL INDUCED DIVERGENCE

An abrupt change in friction (surface stress) can result in convergence or
divergence and impact rainfall and cloudiness patterns.
The force balance for steady flow with friction is shown in the diagram
In order to balance the friction and Coriolis terms, the following relation must
hold,
tan ϕ = Fr f V
where Fr is the deceleration from friction. If friction is considered to be linear in

V, then
tan ϕ = CD f
where CD is the drag coefficient.

Where the drag coefficient is larger, the cross-isobar angle is larger.
For the same drag coefficient, the cross-isobar angle is larger at low latitudes
than at higher latitudes.
An air stream paralleling a coast-line will result in
Divergence and drier conditions over the coastline if the lower pressures are
inland.
4
Convergence and moister conditions over the coastline if higher pressures
are inland.
This may partially explain the dry conditions over certain coastal regions such as
the North coast of Venezuela, and other locations in the Tropics.
5
Lesson 8 – Tropical Weather Systems
References: Tropical Climatology (2nd Ed.), McGregor and Nieuwolt

Forecaster’s Guide to Tropical Meteorology (AWS TR 240 Updated),
Ramage
“Conceptual Models of Tropical Waves,” Burton and Burton (online MetEd
module), https://www.meted.ucar.edu/meteoforum/tropwaves/
African Easter Waves, (online MetEd module),
https://www.meted.ucar.edu/tropical/synoptic/Afr_E_Waves
“The origin and structure of easterly waves in the lower troposphere of
North Africa,” Burpee, J. Atmos. Sci., 29, 77-90, 1972
“Three dimensional structure and dynamics of the African easterly waves,
Part II: Dynamical models,” Hall et al., J. Atmos. Sci., 63, 2231-2245,
2006
“African easterly wave variability and its relationship to Atlantic tropical
cyclone activity,” Thorncroft and Hodges, J. of Climate, 14, 1166-1179,
2001

Burton and Burton, online module
Thorncroft and Hodges, 2001 (available online via library periodical
collections)
“Conceptual Models of Tropical Waves,” Burton and Burton (online MetEd
module), https://www.meted.ucar.edu/meteoforum/tropwaves/
African Easter Waves, (online MetEd module),
https://www.meted.ucar.edu/tropical/synoptic/Afr_E_Waves
GENERAL
 Deep convection in Tropics tied generally to where there is upper-level
divergence and outflow, not so much on static stability.
 Tropical weather systems are typically divided into three categories:
o Waves
o Vortexes
o Linear disturbances
 Observational studies of tropical weather systems find that often the lowest
observed relative humidities are actually located in the region of highest
precipitation!
o This is due to the fact that even in the areas of greatest precipitation and
convection, compensating subsidence occupies most of the area, and so
statistically the observations will be taken in subsiding (dry) regions.
WAVES IN THE EASTERLIES

 There is lots of different nomenclature with waves. Common terms you will see
in the literature are:
o Easterly waves
o Waves in the easterlies
o African waves
o Equatorial waves
o Tropical waves
 Not all waves are formed the same way, or have similar dynamics.
o Some are surface manifestations of upper-level lows, while others are
low-level phenomenon.
 The National Hurricane Center uses the generic term tropical wave to refer to
wave-like features, regardless of their dynamics.
o They define a tropical wave as “a trough or area of cyclonic curvature in
the trade winds or equatorial westerlies.”
 The remainder of this section deals only with those waves that are not
manifestations of upper-level lows or associated with shear lines, but instead are
embedded in the trade-wind easterlies. We will refer to these as waves in the
easterlies.
 There are several conceptual models for waves in the easterlies. These are:
o Riehl’s classical model – Applies mainly to Caribbean waves
o Frank’s inverted “V” model – Applies to eastern and mid-Atlantic waves
o African wave model – Applies to waves over Equatorial Africa and the
western coast of Africa.
 These are only conceptual models, meant to help explain and categorize different
waves. This doesn’t mean the every wave fits into one of these categories, or that
a wave cannot transition between categories as it propagates.
2
 Waves in the Easterlies are primarily a Northern Hemisphere phenomena
o This is because they need deep, easterly current to form an propagate,
and these are more prevalent in the Northern Hemisphere versus the
Southern Hemisphere.
WAVES OVER THE NORTH ATLANTIC AND CARIBBEAN

 Riehl and Frank models generally apply
 Persist for 1 – 2 weeks
 Vorticity maximum is near 700 mb.
 Cold-core in the low levels, warm-core aloft
 Associated trough tilts upstream with height.
o This is due to hypsometric considerations, since air is cooler behind the
wave.
 The relative magnitudes of the mean wind velocity, U, versus the propagation
speed of the wave, C is important in explaining the divergence/convergence
pattern associated with waves.
o For any case, as parcels approach the wave axis from whichever relative
direction they are coming from, they must acquire positive absolute
vorticity, which can only occur through convergence as shown by the
vorticity equation
D
    V
Dt
o If the wave is moving slower than the mean wind speed (U – C > 0), then
the air parcels approach the wave axis from the East, and the
convergence will lie behind (to the East) of the wave axis.
o If the wave is moving faster than the mean wind speed (U – C < 0), then
the air parcels approach the wave axis from the West, and the
convergence will lie ahead of the wave axis.
 In most cases the winds in the lower troposphere are faster than the wave speed,
while winds in the mid and upper levels are slower than the wave speed. The
simplified divergence/convergence pattern would look like the figure below.
3
o This pattern of convergence/divergence is associated with upward motion
and cloudiness behind the wave axis, with subsidence and clearing ahead.
 The wave needs anticyclonic outflow aloft for maintenance of the convection.
 Convection often amplifies as the waves approach the eastern Caribbean Sea.
o This is due to the presence of the TUTT, which can enhance the upper-
level outflow from the convection.
 Waves in the easterlies are usually thought to be a North Atlantic phenomenon.
There is considerable differences of opinions among researchers and operational
meteorologists as to whether or not there are true waves in the easterlies in the
Western Pacific, or to whether or not Atlantic waves actually traverse Central
and South America and maintain their form into the Eastern Pacific.
BAROTROPIC INSTABILITY
One form of hydrodynamic instability that can occur in the atmosphere is
barotropic instability. The derivation of the condition for barotropic instability is beyond
the scope of this course. But, the condition for barotropic instability involves the
horizontal shear of the mean wind. The necessary condition for barotropic instability to
occur is that, somewhere within the flow, the following condition must be true:
d 2u
   0. (1)
dy 2
This means that for barotropic instability to occur that the second derivative of the mean
zonal wind must be equal to  somewhere in the flow. Condition (1) can also be written
as
d  du 
  f 0 (2)
dydy 
or
4
d
 0. (3)
dy
We can interpret this to mean that the absolute vorticity must have a minimum or
maximum value somewhere in the flow in order for barotropic instability to occur.
BAROTROPIC INSTABILITY IN AN EASTERLY JET STREAM

Barotropic instability is dependent upon horizontal shear of the mean flow. To
examine if barotropic instability is possible, the horizontal profile of the absolute
vorticity must be examined. The plot below shows the zonal velocity, absolute vorticity,
and the second derivative of the velocity for an idealized easterly jet stream on the beta
plane. The dashed line on the third diagram is the value of beta.
There are absolute vorticity minima and maxima on both flanks of the jet, near the
locations of the inflection points in the velocity profile. Thus, the condition for
barotropic instability is met in these two regions.
However, the presence of an inflection point does not automatically mean that there is a
minimum or maximum in the absolute vorticity. If beta is large compared to the second
derivative of the velocity, such as for a broad, weak jet at low latitudes, as shown below,
then there will not be any maxima or minima in vorticity, even though there are inflection
points in the velocity profile. Thus, beta acts as stabilizing influence against barotropic
instability.
5
ENERGETICS OF BAROTROPIC DISTURBANCES
Barotropic disturbances derive their energy from the mean flow. Energy
considerations show that for a barotropic disturbance to grow it must tilt opposite to
du dy .1 Since midlatitude disturbances tend to tilt in the same direction as du dy , they
actually lose energy back to the mean flow due to barotropic instability. Thus, barotropic
instability is not a viable way for midlatitude disturbances to form and grow. However,
interestingly enough, since midlatitude disturbance decay due to barotropic instability,
they give up energy to the mean flow and help maintain the mean flow against friction.
Thus, barotropic instability is somewhat important for the maintenance of the mean flow
in the midlatitudes.
AFRICAN WAVES
 Western Africa is major source of waves in the easterlies.
 Their dynamics appears to be a combination of both barotropic and baroclinic
instability associated with the West African mid-level (easterly) jet.
o Convection may play a role acting as a forcing mechanism to create a
perturbation in the flow from which the disturbance can grow into a
wave.
 Wave axis generally tilted from southwest to northeast, and is opposite to
horizontal shear south of the jet.
1
See Haltiner and Williams, pp.74-75.
6
o When tilt and shear are opposite, barotropic instability removes energy
from the jet and puts it into the perturbation (wave).
o Thus, growth due to barotropic instability will mainly be south of the jet
core.
 Often have multiple vorticity centers, with southern one at around 600 mb and
south of the jet axis, and the northern one nearer to 850 mb and north of the jet
axis.
o Thorncroft and Hodges found that those 850 mb vorticity centers that are
found over the ocean are not continuations of the 850 mb over the land.
Instead, it appears that as the 600 mb vorticity center passes the coast
that development occurs downward, possibly influenced by latent
heating, so that a new 850 mb vorticity center is formed while the
previous, northerly 850 mb vorticity center dissipates.
 These waves are preferentially formed in the half-year associated with the
Boreal summer.
 Waves increase in amplitude as they approach the west coast of Africa, and
weaken as they head out to sea.
 Divergence/convergence pattern opposite of waves over the North Atlantic.
o Low-level convergence, clouds and rain ahead of wave axis, rather than
behind.
 Waves in the Easterlies account for the majority of all North Atlantic tropical
cyclones.
 An increased number of waves means an increased number of tropical cyclones.
 Meteorological factors in the wave formation region over Africa impact the
Atlantic tropical cyclone season.
UPPER-LEVEL CYCLONES
 Most intense in the 200 – 300 mb level.
 Many don’t even show up on the 700 mb chart.
 Associated with upper-level convergence and suppressed convection.
 Often surrounded by rings of thin cirrus.
7
 Occasionally convection fires in center of low since cool upper-levels are
associated with static instability.
o Convection usually short-lived due to:
 upper-level convergence
 entrainment of dry air
 Upper-level lows rarely transform directly into warm-core surface lows, though
they can interact with surface features to form new surface lows to the east of
the upper-level cyclone.
TROPICAL UPPER-TROPOSPHERIC TROUGHS

 The tropical upper-tropospheric trough (TUTT) is a persistent feature in the
North Pacific and North Atlantic during the summer months.
 A mid-latitude short-wave trough passing poleward of the TUTT can sometimes
result in a surge of cold air wrapping around the TUTT and forming a cut-off
low at the Equatorward end of the TUTT (called a TUTT cell).
 The region to the east of the TUTT cell is one of enhance upper-level divergence,
and Sadler has documented cases where this has led to the generation a tropical
cyclone to the southeast of the TUTT cell.
SQUALL LINES
 Non-frontal lines of active thunderstorms.
 Can be hundreds of miles in length
 Persists much longer than lifetime of individual cells that make up line.
o Usually last for 3 – 15 hours
o May last for several days in West Africa
 Occur in northern Australia, northwestern India, Bangladesh, malysia,
Indonesia, South America/Caribbean, and West Africa
 Most frequently occur in mid to late afternoon.
 Require moist low levels with relatively dry mid levels.
 Tropical squall lines have trailing anvil, while mid-latitude squall lines have
preceding anvil.
8
 Leading edge of gust front sparks new development and propagates line.
SUMATRA
 Form during SW monsoon over Straits of Malacca
 Usually form at night
 Last 1 to 2 hours
 Three main factors
o Daytime heating over Sumatra followed by nighttime radiative cooling
over the Straits of Malacca
o Orographic lifting over Malaysia
o Converging land breezes from Malaysia and Sumatra
SHEAR LINES
 Equatorward extension of a midlatitude cold front
 No large temperature or humidity shift
 Zone of convergence
 Characterized by stratiform clouds with embedded convection
 Can stall out and cause low ceilings and rain for extended periods.
COLD SURGES
 Surge of cold air from midlatitudes into Tropics.
 Surge from winter hemisphere can enhance Equatorial westerlies in summer
hemisphere, and increase cyclonic vorticity.
o Can result in development of tropical cyclones in monsoon trough.
9
Lesson 9 – The Tropical Oceans
References: Climate Dynamics of the Tropics, Hastenrath

Descriptive Physical Oceanography, Pickard and Emery
Ocean Circulation, Brown and Colling
“On the wind-driven ocean circulation”, W.H. Munk, J. Meteor., 7, 79-93,
1950
Reading: Brown and Colling Section 5.1 (e-reserve)
BASIC OCEAN STRUCTURE

The ocean can be divided into three layers
ο Mixed (or surface) layer
ο Thermocline
ο Deep layer
The mixed layer is akin to the atmosphere’s planetary boundary layer.
The mixed layer gets its name from the fact that it tends to be well mixed, with
the temperature being nearly isothermal with depth.
The depth of the mixed layer varies with location and season. Typical ranges are
from 25 to 500 meters.
The depth is determined primarily by how rough the seas are. The rougher the
seas, the deeper the mixing.
ο Since seas are generally rougher in winter, the mixed layer depth is usually
deeper in winter than in summer.
At the bottom of the mixed layer is the beginning of the thermocline.
The thermocline is characterized by a decrease in temperature with depth.
The thermocline is a very stable layer. Because of this, vertical mixing in the
ocean at depths below the mixed layer is very slow.
Because the ocean typically has a strong thermocline that inhibits mixing
between the mixed layer and the deep layer, it is sometimes conceptually and
mathematically convenient to model the ocean as a two-layer fluid.
SEA-SURFACE TEMPERATURE
The images below show the monthly mean sea-surface temperature (SST) for
January and July. (Images provided by the NOAA-CIRES Climate Diagnostics
Center, Boulder, Colorado, USA, from their Web site at http://www.cdc.noaa.gov/.
January
July
Things to note about the SST distribution

Tongue of cold SST along Equator in Atlantic and Pacific Oceans.
Relatively colder SST along west coasts of continents (except Australia) as
compared to the east coasts.
2
WIND STRESS AND THE EKMAN SPIRAL
The horizontal momentum equations for the ocean are
Du 1 ∂p 1 ∂τ x
=− + fv +
Dt ρ ∂x ρ ∂z
Dv 1 ∂p 1 ∂τ y
=− − fu +
Dt ρ ∂y ρ ∂z
where the stress terms (those involving τ ) are due to vertical turbulent
momentum fluxes (Reynolds stress).
For steady flow the equations are
1 ∂ p 1 ∂τ x
− fv = − +
ρ ∂x ρ ∂z
1 ∂ p 1 ∂τ y
fu =− +
ρ ∂ y ρ ∂z
The flow can be broken up into a part driven by the pressure gradient and that
driven by the stresses. The part driven by the stresses is governed by the
equations
1 ∂τ x
v=−
f ρ ∂z
1 ∂τ y
u=
f ρ ∂z
In the early 1900’s an oceanography students named Ekman solved the above
two equations to find the vertical structure of flow driven by the wind stress. His
result is
u ( z ) = exp(γ z )[U 0 cos(γ z ) − V0 sin (γ z )]
v( z ) = exp(γ z )[V0 cos(γ z ) + U 0 sin (γ z )] .
where U0 and V0 are the zonal and meridional components of the surface
current, and
γ ≡ f 2K
where K is the eddy viscosity.
If the solution above is plotted on a hodograph it traces a decaying clockwise
spiral with depth (in the Northern Hemisphere). This is known as the Ekman
spiral.
3
The depth of the Ekman layer is taken to be that point at which the spiral has
made one complete rotation (or the e-folding scale of the decay). Therefore, the
depth of the Ekman layer is
1 2K
DE = = .
γ f
Two interesting and important facts about the Ekman spiral:
ο #1 – The surface current flows at exactly 45° to the right of the surface wind (in
the NH).
Objects such as icebergs, life rafts, etc. will actually drift to the right of
the prevailing wind.
ο #2 - The vertically-integrated mass transport in the Ekman layer is directed at
90° to the right of the surface wind.
EKMAN TRANSPORT AND UPWELLING/DOWNWELLING

The fact that the vertically-integrated mass transport in the Ekman layer is
directed at 90° to the right of the surface wind is important in the distribution of
ocean surface temperature.
Off of the east coast of North America, the winds from the Pacific High in spring
and summer blow nearly parallel down the coast.
The resultant Ekman transport is to the west, away from the coast.
The surface divergence along the coast results in the upwelling of cooler
water from below, explaining the cooler SST observed off of the west coast of
North America.
This cool water helps keep the summertime climate of the coast cool and
often foggy.
The Ekman transport forced by the South Pacific high also explains cool
water off of west coast of South America.
Along the Equator, the trade winds result in Ekman transport away from the
Equator in both hemispheres.
The resultant surface divergence along the Equator results in upwelling of
cooler water from below, and explains the tongue of cold water observed in
the SST distribution in the Atlantic and Pacific Oceans.
4
A persistent anticyclone over the ocean will result in a net transport of ocean
water toward the center of the anticyclone, with down-welling pushing the cooler
deep water even deeper.
A persistent cyclone over the ocean will result in a net transport of ocean water
away from the center of the cyclone, with up-welling pulling the cooler deep
water toward the surface.
WIND-DRIVEN SURFACE CURRENTS IN A RECTANGULAR OCEAN

In 1950, Walter Munk used a realistic latitudinal wind profile to derive what the
steady-state surface circulation would be in a rectangular ocean.
His results are recreated in the figure below.
The major features of the circulation are:

A cyclonically rotating subpolar gyre
An anticyclonically rotating subtropical gyre
Two westward flowing equatorial currents symmetric with the ITCZ (not the
Equator).
5
An eastward flowing equatorial counter current between the equatorial
currents.
Strong, western boundary currents in the subtropical and subpolar gyres,
contrasted with weaker return flows east of the gyre centers.
One striking feature of this circulation pattern is the strong, well-defined western
boundary current, as opposed to the weaker, broader flow to the east of the gyre.
The western boundary currents are due to the fact that Coriolis changes with
latitude. If Coriolis were constant, the western boundary currents wouldn’t
exist.
EQUATORIAL CURRENTS
The surface currents in the Pacific and Atlantic have a similar structure, and
can be explained at least in part by convergence and divergence associated with
Ekman transport.
6
The diagram above shows the air flow (open arrows) and the resultant Ekman
transport (dark arrows).
The “DIV” and “CONV” denote regions of divergence and convergence in the
Ekman transport.
Regions of divergence will result in a lowering of the sea-surface, while
regions of convergence will raise the sea surface.
Other than within a degree or so of the Equator, the ocean flow will be parallel
to the sea-surface contours with low heights to the left in the Northern
Hemisphere, and to the right in the Southern Hemisphere.
The ocean currents and their directions are indicated by the “W” and “E”
annotations, with “W” indicating a westward current, while “E” indicates an
eastward current.
The resultant surface currents are the
North Equatorial Current – Westward flowing
Equatorial Counter Current – Eastward flowing, more-or-less aligned with
the ITCZ.
South Equatorial Current – Westward flowing, and in both hemispheres.
There is also an Eastward flowing Equatorial Undercurrent that more-or-less
flows along the Equator at depth.
The equatorial currents in the tropical Indian Ocean differ in that since the
atmospheric flow switches directions seasonally due to the monsoon, so do the
currents.
EAST-TO-WEST STRUCTURE OF THE THERMOCLINE

In the tropical Pacific Ocean, the trade winds result in a net transport of surface
waters toward the western part of the basin, toward Indonesia and Australia.
This net transport of surface waters results in a suppressing of the thermocline
toward the west, and a deep pool of warm surface water in the western part of
the basin.
7
Lesson 10 – Tropical Cyclones: Structure

Global Perspectives on Tropical Cylones: From Science to Mitigation,
Chan and Kepert (ed.)
Global Guide to Tropical Cyclone Forecasting, Holland (ed.), online at
http://www.bom.gov.au/bmrc/pubs/tcguide/globa_guide_intro.htm
Reading: A Global View of Tropical Cyclones, Chapter 3, Frank (e-reserve)

Tropical Cyclones: Their Evolution, Structure, and Effects, Chapter 2
(e-reserve)
Hurricane, Chapter 2, Pielke (e-reserve)
Global Guide, Chapter 2 (online)
GENERAL
Tropical cyclones are primarily driven by latent heating.
As the air spirals in toward the center, it picks up latent heat and sensible heat
through evaporation from the ocean.
Once the air is saturated, it can still pick up some sensible heat, but latent
heating is the dominant mechanism.
As the air approaches the center of the vortex, it rises in convection either in the
eye wall, or in the spiral convective bands. As it rises, it cools and the water
vapor condenses, giving off its latent heat.
The warming of the air due to the latent heat release results in low-level height
falls, and upper-level height rises, which helps maintain the low-level
convergence of the warm-moist air.
The air is exhausted out and away at the upper levels.
STRENGTH, SIZE, AND INTENSITY

We will use the following definitions:
Core intensity, based on maximum winds or minimum sea-level pressure.
Size, based on the mean radius of the outermost closed isobar.
Strength, based on the shape of the outer core wind profile.
There is great variability in the size, intensity, and strength of tropical cyclones.
Storms can be large and intense, small and intense, large and weak, etc.
You can’t infer intensity based on size or strength.
DISTRIBUTION OF WIND
The diagram below shows the typical tangential wind structure in a tropical
cyclone.
The wind structure is often represented by a modified Rankin vortex,

where C, D, and a are empirically determined constants.
vr a = C r > rmax wind
(1)
vr −1 = D r ≤ rmax wind
a is usually 0.4 to 0.6.

a = 1 would be a pure Rankin vortex, in which relative angular momentum is
conserved.
In cylindrical coordinates, the radial and tangential components of the
momentum equation are
Du v 2 1 ∂p
− =− + fv + Fr (2)
Dt r ρ ∂r
Dv uv 1 ∂p
+ =− − fu + Fθ (3)
Dt r ρ r ∂θ
where u is the radial velocity, v is the tangential velocity, r is the distance from
the center of the storm, ࣂ is the angular measure, and Fr and Fθ represent
turbulent friction.
Equations (2) and (3) are valid for both cyclonic and anticyclonic flow, as
long as we use the following convention:
u is positive outward from the vortex
2
r is positive outward from the vortex
v is positive for a cyclonic vortex, and negative for an anticyclonic vortex.
If the vortex is steady, axisymmetric, and friction is ignored, then the tangential
wind is
v2 1 ∂p
+ fv − = 0, (4)
r ρ ∂r
and the vortex is in gradient balance.
Dividing (4) by fv, we get
v 1 ∂p
+1− = 0. (5)
fr f vρ ∂r
The first term is the Rossby number, RO.
In the core of the vortex, the Rossby number is large, so the Coriolis effects
can be ignored. Therefore, the core of the storm is in cyclostrophic balance.
Outside the core, the Rossby number is of the order of unity, so gradient
balance holds.
In either case, the wind speed depends on the pressure gradient.
Broadly speaking, the lower the central pressure, the faster the maximum wind
will be.
The relation between maximum winds and the central pressure is closely
approximated by
vmax = A p∞ − pc (6)
where p∞ is the ambient sea-level pressure outside of the circulation, and pc is

the minimum central pressure.
The constant A is empirically determined. A value of 6.3 is often used for
Atlantic hurricanes.
ABSOLUTE ANGULAR MOMENTUM

If no energy were added or subtracted from an air parcel as it spiraled toward
the center of the vortex, then its absolute angular momentum would have to be
conserved.
The absolute angular momentum is the relative angular momentum plus the
angular momentum due to the rotation of the Earth.
3
An air parcel at distance r from the center of the vortex, and stationary with
respect to the Earth, will have a specific absolute angular momentum
(absolute angular momentum per unit mass) of
M f = ω r 2 = f0 r 2 2 (7)
just due to the rotation of the Earth.

If the parcel has a tangential velocity v then the specific relative angular
momentum is
M r = vr , (8)
so the specific absolute angular momentum is

M a = M r + M f = vr + f0 r 2 2 . (9)
If the parcel started out at rest at a distance of 500 km, and spiraled in to a
radius of 15 km, it would have attained a tangential velocity of over 600 m/s
if its angular momentum were conserved.
Obviously, parcels don’t conserve angular momentum as they spiral into the
center of a tropical cyclone.
In fact, the absolute angular momentum decreases toward the center of tropical
cyclones, which means that air parcels must be losing angular momentum as
they spiral inward.
The parcels lose angular momentum through turbulent dissipation.
The absolute vorticity of an axisymmetric vortex is
∂v v
η = f0 + + , (10)
∂r r
where the second and third terms are just the shear and curvature terms for
relative vorticity. The absolute vorticity and the specific absolute angular
momentum are related via
1 ∂M a
η= . (11)
r ∂r
INERTIAL STABILITY OF A VORTEX

A fundamental parameter for assessing how a vortex interacts with its
environment is the inertial stability, which is developed mathematically below.
4
In pressure coordinates the momentum equations for an axisymmetric vortex
without friction are
Du ∂Φ v 2
=− + + f 0v (12)
Dt ∂r r
Dv uv
= − − f 0u . (13)
Dt r
If an air parcel starts out in gradient balance, but is impelled outward at a
velocity u, the change in radial acceleration with time is given by taking the time
derivative of (12),
D  Du  D  ∂Φ  D  v 2  D
 =−  +  + ( f 0v ) . (14)
Dt  Dt  Dt  ∂r  Dt  r  Dt
The terms on the right-hand side of (14) are evaluated as follows:

D  ∂Φ  ∂  ∂Φ  ∂  ∂Φ  v ∂  ∂Φ  ∂ 2Φ
−  =−  −u  −   = −u 2 ; (15)
Dt  ∂r  ∂t  ∂r  ∂ r  ∂r  r ∂θ  ∂r  ∂r
D  v 2  1 Dv 2 v 2 Dr 2v Dv uv 2 2v  uv  uv
2
3uv 2 2 f uv
 = − 2 = − 2 =  − − f 0u  − 2 = − 2 − 0 (16)
Dt  r  r Dt r Dt r Dt r r  r  r r r
D Dv uv
( f 0v ) = f 0 = − f 0 − f 02u . (17)
Dt Dt r
Substituting (15) thru (17) into (14) results in
D  Du  ∂ 2 Φ 3uv 2 3 f 0uv
  = − u 2
− 2 − − f 02u
Dt  Dt  ∂r r r
or
D 2u  ∂ 2Φ 3v 2 3 f0v 
2
+ 2 + 2 + + f 02  u = 0 . (18)
Dt  ∂r r r 
Defining
2∂ 2Φ 3v 2 3 f 0v
ω = 2 + 2 + + f 02 (19)
∂r r r
we can write (18) as
D 2u
+ ω 2u = 0 . (20)
Dt 2
Equation (20) has solution of

5
u = Aeiωt + Be −iωt (21)
if ω 2 > 0 , in which case a parcel perturbed outward will stable oscillations
around its initial radius with angular frequency ω .
If ω 2 < 0 the solutions to (20) are
u = Aeωt + Be −ωt , (22)
in which case a parcel perturbed outward will accelerate a way and never
return.
We thus have established the criteria for inertial stability of a vortex:
An inertially stable vortex will have
∂ 2 Φ 3v 2 3 f 0 v
2
+ 2 + + f 02 > 0 , (23)
∂r r r
while an inertially unstable vortex will have
∂ 2 Φ 3v 2 3 f 0 v
+ 2 + + f 02 < 0 . (24)
∂r 2 r r
The inertial stability of a vortex is an indication of how readily the vortex can
interact with its environment.
A highly inertially stable vortex will resist any perturbations and outside
influences, and will remain symmetric. A low stability vortex, on the other hand,
is more readily influenced by outside disturbances, and will be less symmetric.
INERTIAL STABILITY AND ABSOLUTE VORTICITY

The inertial stability condition, (23), can be written in terms of absolute vorticity
as follows:
For a parcel in gradient balance, the radial geopotential gradient (pressure
gradient) is
∂Φ v 2
= + f0v . (25)
∂r r
Substituting (25) into (23) results in
∂  v2  3v 2 3 f 0 v
 + f 0 +
v 2
+ + f 02 > 0
∂r  r  r r
which expands to
6
2v ∂ v v 2 ∂ v 3v 2 3 f 0 v
− 2 + f0 + 2 + + f 02 > 0
r ∂r r ∂r r r
and through the following steps
2v ∂ v v 2 ∂ v 2v 2 v 2 2 f 0 v f 0 v
− 2 + f0 + 2 + 2+ + + f 02 > 0
r ∂r r ∂r r r r r
2v ∂ v ∂ v 2v 2 2 f 0 v f 0 v
+ f0 + 2 + + + f 02 > 0
r ∂r ∂r r r r
 2v  ∂ v  2v   v   2v 
 + f 0  +  + f0    +  + f 0  f 0 > 0
 r  ∂r  r  r   r 
results in
 2v   ∂v v 
 + f0   + + f0  > 0 . (26)
 r   ∂r r 
The second term in brackets is the absolute vorticity, and so we finally have the
criteria for inertial stability as
 2v 
 + f 0 η > 0 . (27)
 r 
Equation (27) shows that negative absolute vorticity is not a sufficient condition
to have inertial instability. A vortex with negative absolute vorticity can still be
inertially stable as long as
 2v 
 + f0  < 0 . (28)
 r 
In straight-line flow, or for weak vortexes where the curvature term in (27) can
be ignored, the stability condition can be written as
η f0 > 0 . (29)
This is why you often hear it said that anytime absolute vorticity is negative
that the flow is inertially unstable. However, keep in mind that for stronger
vortexes, particularly at low latitudes, negative absolute vorticity doesn’t
automatically imply inertial instability.
7
And finally, a reminder that the oscillation frequency of a stable vortex is given
by the stability condition [recall Error! Reference source not found.], so that for a
stable vortex
 2v 
ω 2 =  + f 0 η . (30)
 r 
Sometimes it is helpful to write the stability condition or oscillation frequency in
terms of the specific absolute angular momentum,
1 ∂M a2
ω2 = . (31)
r 3 ∂r
ROSSBY RADIUS OF DEFORMATION
A fundamental horizontal length scale for a disturbance in a rotating fluid is the
Rossby radius of deformation.
The Rossby radius of deformation is the distance that a gravity wave (which is
the means by which the fluid adjusts to equilibrium) will travel in one inertial
period (ω−1),
λR = cgτ f = cg ω . (32)
where cg is the speed of an inertial gravity wave.

The Rossby radius of deformation is therefore
cg
λR = . (33)
η ( f0 + 2 v r )
Note: For flows whose absolute vorticity is primarily due to planetary

vorticity (i.e., flows where ζ << f), (33) becomes
λR = cg f 0 , (34)
which is the form of the Rossby radius of deformation most commonly used
in dynamic meteorology textbooks. However, (34) is just a less general form
of (33).
The gravity wave group velocity, cg, depends on the stratification of the fluid.
For a barotropic fluid, it is given by c g = gH where H is the depth of the
fluid.
8
For a baroclinic fluid, there are multiple baroclinic modes of oscillation, each
with its own group velocity. For waves that travel primarily horizontally
(long wavelengths) the group velocity can be approximated as
cg ≅ NH nπ . (35)
where N is the Brunt-Vaisala frequency, H is the scale height of the

atmosphere, and n refers to the different baroclinic modes of oscillation
(n = 1, 2, 3, …).
The baroclinic modes have a significantly slower group velocity than does the
barotropic mode, and so the Rossby radius of deformation is smaller for the
baroclinic modes.
The response of the fluid to a disturbance can be assessed by comparing the
horizontal length scale of the disturbance, L, to the Rossby radius of
deformation.
For disturbances whose horizontal length scale, L, is much less than 2πλR ,
the mass field will adjust to the velocity field.
For disturbances whose horizontal length scale, L, is much greater than 2πλR,
the velocity field will adjust to the mass field.
For disturbances in the intermediate range, mutual adjustment occurs.
The size of the disturbance compared to the Rossby radius of deformation can
also give us an idea of how persistent a circulation can be.
For disturbances whose size is comparable or larger than the Rossby radius
of deformation, we know that the wind will adjust to the mass field, while for
small disturbances the mass will adjust to the wind field.
Diabatic heating will cause the mass field to be perturbed.
If the heating is confined to a region much smaller than the Rossby radius of
deformation, the disturbance in the mass field will not influence the wind.
All that will happen is that gravity waves will propagate away from the
disturbance, and eventually nothing will be left of it.
Only if the heating is over a region comparable to the Rossby radius of
deformation will a circulation in the wind field develop in response.
Disturbances whose sizes are much less than the Rossby radius of deformation
tend to die out quickly, and not develop a persistent circulation.
9
STRUCTURE AND DYNAMICS OF A TROPICAL CYCLONE
The diagrams below show the wind speed, angular momen
momentum,
tum, inertial
frequency,, and Rossby radius of deformation (cg = 10 m/s) for a modified
Rankin vortex at a latitude of 15°, with a radius of maximum winds of 25
2 km, an
intensity of 48 m/s and a = 0.5.
In particular, note that the Rossby radius of deformat

deformation
ion is very large in the
outer region of the cyclone, whereas it is much smaller in the core.
Also notice the difference in the inertial frequency, with
ith large values
indicating high inertial stability in the core region, as compared with the
outer region.
STRUCTURE OF THE CORE REGION

The core region is 3 to 6 times the radius of maximum winds.
The Rossby number (not shown in graphs above) is large in the core region, so
the balance is cyclostrophic.
The wind structure of the core has the following genera
generall characteristics:
10
Radius of maximum winds is generally in the eye wall.
There is a sharp increase in speed near the radius of maximum winds.
Winds decrease aloft.
Maximum winds tilt outward with height.
Maximum winds are found in the front-right quadrant of the cyclone.
This is true, even if the forward speed of the storm is factored out.
The reason isn’t completely understood.
Strongest horizontal temperature gradients are near the radius of maximum
winds.
Warm core is narrow in the lower troposphere, and bulges outward aloft.
Some features of the core region are the eye and rainbands.
Some definitions:
Rainband – Also called spiral bands, or feeder bands, they are areas of cloud
and of precipitation spiraling in toward the center of the circulation.
Convective ring – Rainbands that near completely circle the center.
Moat – Clear region between two convective rings.
Eyewall – Innermost convective ring.
Eye – Clear region at center of circulation, surrounded by the eyewall.
DYNAMICS OF THE CORE REGION

The inner core region has very high inertial stability due to the large absolute
vorticity and small radius.
This means it doesn’t readily interact with its environment, and is why the
inner core tends to be very symmetric.
The inner core regions also has a small Rossby radius of deformation.
In the core, the wind tends to adjust to changes in the mass field. Any
changes in the mass field due to heating/cooling or convergence/divergence
will result in changes to the wind.
The intensity of the cyclone can therefore react rapidly to fluctuations in
diabatic and latent heating, as well as fluctuations in vertically integrated
divergence.
11
The dynamics and fluctuations in the core are dominated by primarily by
convection and heating.
THE EYE
An eye generally doesn’t form until the cyclone reaches hurricane or typhoon
intensity.
The mechanisms by which an eye is formed and maintained are not completely
known. One conceptual model is presented below.
As air spirals inward toward the center of the cyclone, its speed will increase
due to conservation of angular momentum.
However, if angular momentum were truly conserved, wind speeds of 600
m/s or more would be reached.
Therefore, angular momentum must be dissipated via convection,
turbulence, and other processes that retard the flow.
The angular momentum balance at a point is given by
∂M a ∂M a ∂M a
= −u −w − Fr . (36)
∂t ∂r ∂z
The three terms on the right-hand-side are radial advection, vertical
advection, and dissipation of angular momentum.
At some distance near the center of the vortex, the dissipation of angular
momentum can no longer keep up with the horizontal advection. This leads
to a horizontal convergence of angular momentum.
This horizontal convergence of angular momentum leads to an increase in
the winds above the value that the pressure gradient can support (i.e., they
are super-gradient).
The super-gradient winds develop a radially outward component, since the
pressure gradient cannot supply the required centripetal acceleration. The
consequences of this are (see figure):
There is radial convergence at distances outside of the radius of
maximum winds.
There is radial divergence at distances inside the radius of maximum
winds.
12
The convergence leads to upward motion, and results in the convection
within the eye wall.
The resultant convection within the eye wall serves to balance the
horizontal convergence of angular momentum through vertical advection.
The upward motion in the eye wall results in outflow aloft, and high
perturbation pressure over the center of the storm.
The upper-level pressure perturbation, combined with the low-level
divergence within the eye itself, results in compensating subsidence in the
eye.
Due to the strong inertial stability in the eye wall, the strongest compensating
subsidence is found on the inside of the radius of maximum winds.
This results in an abrupt clearing just inside the eye wall.
As the air subsides, the resultant compressional warming actually works against
subsidence (through buoyancy).
All that is needed in order to keep the eye relatively clear is enough subsidence
to balance the buoyancy.
Thus, in the steady state, there doesn’t have to be strong subsidence in the
eye in order to maintain the eye.
The maintenance of the eyewall is a balance between the horizontal and vertical
advection of angular momentum. The radius of the eye can expand or contract
depending on this balance.
EYE WALL REPLACEMENT CYCLE

Sometimes an outer convective ring will form and will often cause the inner ring
(the eye wall) to dissipate for two reasons.
The subsidence from the outer ring suppresses convection in the inner ring.
The outer ring robs the inner ring of inflow.
13
Eye wall replacement usually results in a lowering of the intensity, since the new
eye wall is at a larger radius, and from angular momentum arguments, would
have a lower wind speed.
Intensity may build again one the new eye wall contracts.
The causes and dynamics of eye wall replacement cycles are not well known.
Eye wall replacement tends to occur in very intense cyclones.
RAINBANDS
Rain bands are bands of convection that occur in the core region, and also can
extend into the outer region of the cyclone.
Rain bands can be tens of kilometers wide, and hundreds of kilometers in length.
Convection often forms on the inside of the rain bands and moves up and
forward through them.
The region between the rain bands is characterized by subsidence.
Rain bands may be of three types:
Moving spiral bands – These bands appear to rotate with the circulation.
Convective rings – A rain band that completely encloses the cyclone.
A convective ring may transition into an eye wall
Principle spiral bands – A non-moving rain band that wraps into the cyclone
on the east side of the cyclone, and feeds lower latitude air into the vortex.
In a rough sense, principle spiral bands separate the inner region of the
cyclone from the outer regions.
In the inner region, air makes several circuits as it spirals into the core,
where in the outer region the air doesn’t make it to the core.
STRUCTURE OF OUTER REGION

In the outer region, the absolute vorticity is weaker and the radius is larger, so
this region has much lower inertial stability than does the core region.
This means the outer region is less symmetric, and also more readily
influenced by the environmental flow.
The outer region also has a larger Rossby radius of deformation than does the
core.
14
In the outer region the mass field adjusts to the wind field.
The tangential circulation can extend 1000 – 2000 km from the center of the
storm.
The radial circulation is much smaller in scale (~600 km).
There is usually subsidence in the outer region, with areas of weak upward
motion.
A typical ring-like area of subsidence is found around 5° latitude or so from the
center, and is sometimes referred to as the moat, since it results in clearing.
OUTFLOW REGION
The intense updrafts in the eye wall carry cyclonic vorticity upwards. This
results in the upper-level outflow near the cyclone center having cyclonic
rotation.
As it spreads outward, it loses positive vorticity and gains negative vorticity, so
that it becomes anticyclonic as it moves away.
The outflow region has weak inertial stability, and so doesn’t resist horizontal
flow.
The outflow region interacts readily with the environmental flow.
The outflow mainly takes place in one or two outflow jets or channels.
These outflow channels are shallow, and are in the upper troposphere.
Cyclones with two outflow channels tend to be more intense than those with a
single outflow channel.
DIURNAL CYCLES
Tropical cyclone convection shows a pronounced diurnal cycle.
Convection is enhanced in the early morning hours (0300 to 0600 local), likely
due to the same reasons that tropical convection over the open oceans has an
early morning maximum.
Areal coverage in cirrus clouds is maximum in the late afternoon (around 1800
local time).
15
Lesson 11 – Tropical Cyclones: Formation, Maintenance, and Intensification

Global Perspectives on Tropical Cylones: From Science to Mitigation,
Chan and Kepert (ed.)
‘Tropical cyclogenesis in a tropical wave critical layer: easterly waves’,
Dunkerton, Montgomery, and Wang Atmos. Chem. and Phys. 2009.
Global Guide to Tropical Cyclone Forecasting, Holland (ed.), online at
http://www.bom.gov.au/bmrc/pubs/tcguide/globa_guide_intro.htm
Reading: A Global View of Tropical Cyclones, Chapter 3, Frank (e-reserve)

Tropical Cyclones: Their Evolution, Structure, and Effects, Chapter 2
(e-reserve)
Hurricane, Chapter 2, Pielke (e-reserve)
Global Guide, Chapter 2 (online)
GENERAL CONSIDERATIONS
Tropical convection acts as a heat engine, taking warm moist air from the
surface and converting the latent heat into kinetic energy in the updraft, which is
then exhausted into the upper troposphere.
If the circulation can overcome the dissipating effects of friction it can become
self-sustaining.
In order for a convective cloud cluster to result in pressure falls at the surface,
there must be a net removal of mass from the air column (net vertically
integrated divergence).
Since there is compensating subsidence nearby, outside of a typical
convective cloud, there really isn’t much integrated mass divergence.
Pressure really won’t fall unless there is a mechanism to remove the mass
that is exhausted well away from the convection.
Compensating subsidence near the convection also serves to decrease the
buoyancy within the clouds, because the subsiding air will also warm. This
reduces the temperature difference between the in-cloud and outside air.
This illustrates the importance of upper-level outflow in tropical cyclones.
Even if there are pressure falls at the surface, the atmosphere could just
generate gravity waves and readjust back to the original pressure, unless the
Rossby radius of deformation is small enough to allow a residual circulation
to form.
Frictional effects will work against formation as well.
The mechanisms by which a tropical cloud cluster can overcome the above
effects and develop a sustained circulation are complex, and much is still
unknown.
Some literature refers to these mechanisms as providing a ‘crank-start’ for
the tropical cyclone, in analogy to starting internal combustion engine.
CONDITIONAL INSTABILITY OF THE SECOND KIND (CISK)

An older theory for how tropical convection develops into a sustained tropical
circulation is that of conditional instability of the second kind or CISK.
In its simplest form, CISK can be explained as follows:
Latent heating of the atmosphere leads (through the hypsometric
relationship) to a lowering of surface pressure.
The lowering of the surface pressure leads to enhanced radial inflow and
convergence, which enhances the convection and latent heat release, which
further decreases the surface pressures.
CISK is then a positive feedback loop.
There are several difficulties with using CISK to explain tropical cyclone
formation.
One problem is that in order for a circulation to form from the latent heating,
the scale of the heating must be of the order of the Rossby radius of deformation
(L ~ 2πλR).
In the Tropics, λR is large even for the baroclinic modes, so the mass field
disturbances caused by latent heating from cloud clusters aren’t large
enough to form circulation.
In order to excite modes with a small enough Rossby radius of deformation, a
heating profile that is concentrated in the lower levels is ideal.
However, the latent heating profile in most tropical cloud clusters is
concentrated in the middle and upper levels, so it doesn’t excite the modes
with low Rossby radius of deformation.
2
Most studies of CISK have involved a linear approximation in which the
different wavelengths, or modes, do not interact and exchange energy.
It is also possible that the non-linear interaction of the wave modes is
important, which would allow energy that is input into one mode to be
transferred into other modes that do have a horizontal scale closer to their
Rossby radii of deformation.
Although CISK has trouble explaining how cloud clusters form into a tropical
cyclone, it can be used to explain how a tropical cyclone, once formed,
intensifies, since as the vortex becomes stronger the Rossby radius of
deformation in the core can become small enough to allow a mass-field
disturbance to force a velocity field adjustment.
MESOSCALE CONVECTIVE VORTICES AND SYSTEMS

Three relatively recent (1990’s) theories for the development of tropical cyclones
from tropical cloud clusters focus on the dynamics of mesoscale convective
vortices which are mid-level vortices found in many tropical cloud clusters.
Before discussing these theories a very brief review of mesoscale convective
vortices is presented.
Mesoscale convective vortices (MCV) are formed through the adjustment of the
troposphere to convective heating. The occur both in mid-latitude and tropical
mesoscale convective systems.
The figure below shows an idealized vertical cross-section through an MCV.
The solid lines are isobars and the dashed lines are lines of constant potential
temperature.
3
In the
he diagram the maximum in vorticity is in the middle, with decreasing
intensity above and below.
Since the MCV is in thermal wind balance there is a cold anomaly below, and
a warm anomaly
aly above the vorticity maximum.
Mesoscale convective vortices often form in and are associated with mesoscale
mes
convective systems (MCS)
MCS).
An idealized mesoscale convective system contains two regions:
A convective region where new convective cells are formed and progress to
maturity
In the convective region there is generally low to mid-level
level convergence,
with divergence aloft.
A stratiform region formed from the remnants of the old cells from the
convective region.
In the stratiform region there is generally mid
mid-level
level convergence, with
low-level
level and upper
upper-level divergence.
The three recent theories all attempt to explain how an MCV associated with an
MCS can result in the
he formation and/or amplification of a vortex at the surface,
which can then further develop into a tropical cyclone.
The three theories are broadly classified as: Top-down Merger; Top-down
Top
Showerhead; and Bottom
ottom-up Development
4
TOP-DOWN MERGER THEORY FOR DEVELOPMENT
In the Top-down Merger theory, two or more smaller mid-level vortices merge
into a larger mid-level vortex. The influence of this larger vortex is then felt
through a deeper depth of the atmosphere, influencing development at the
surface.
The depth or vertical thickness to which a vortex penetrates is given by the
Rossby penetration depth,
DR = ω L N (1)
where ω is the inertial frequency, L is the horizontal scale of the vortex, and N is
the Brunt-Vaisala frequency.
The larger in horizontal extent, or the more intense a vortex is (more intense
vortices have higher inertial frequencies), the deeper the Rossby penetration
depth.
If two or more smaller, mid-level vortices merge then they will form a larger
vortex that has a deeper Rossby penetration depth, and can therefore build
down to lower levels.
TOP-DOWN SHOWERHEAD THEORY FOR DEVELOPMENT

In the Top-down Showerhead theory we start with an existing mid-level vortex in
the stratiform region of an MCS.
Rain falling from mid-level stratiform clouds causes evaporative cooling in the
low levels, and results in subsidence, which advects positive vorticity downward.
The evaporative cooling also cools the lower levels, resulting in a cold anomaly
that bows the isentropes upwards. Through thermal wind balance, this requires
the pressure surfaces to bulge downward, extending the mid-level circulation
downward.
The lower levels have become humid and cool due to the evaporation of the rain
falling through them. Once the downdrafts abate, convection can now fire in the
low levels, resulting in low-level convergence and spin-up of low-level cyclonic
vorticity.
5
BOTTOM-UP DEVELOPMENT THEORY FOR DEVELOPMENT
In the theory of Bottom-up Development a low-level potential vorticity (PV)
anomaly produced from a separate convective updraft moves underneath the
MCV.
The interaction between the PV anomaly and the MCV shifts the profile of
divergence and convergence such that there is enhanced low-level convergence
and spin-up of a low-level vortex, at the expense of the original MCV.
NEED TO ELIMINATE CONVECTIVE DOWNDRAFTS

Another general consideration for development is the need to eliminate
convective downdrafts.
Regions with convective downdrafts are not favorable for development.
This is because convective downdrafts lead to low-level divergence, which
will reduce, rather than enhance, cyclonic spin up.
Having moisture in the mid and low levels will reduce convective downdrafts by
reducing the evaporative cooling and entrainment which lead to downdrafts.
THE MARSUPIAL PARADIGM

Regardless of which theories, Top-down Merger, Top-down Showerhead, or
Bottom-up Development, explain development, it is important to have the region
of development remain in a favorable environment long enough for the
circulation to become self-sustaining.
One very recent theory attempting to explain why some cloud-clusters develop
into tropical cyclones while others don’t is the Marsupial Paradigm.
In the marsupial paradigm the favorable environment is located where the mean
wind speed is equal to the phase speed of the tropical wave.
In this location, air parcel trajectories in a reference frame moving with the
wave are nearly closed curves and are isolated from entrainment from the
outside.
A vortex residing in this location can develop through the continual
moistening and latent heat release of the air parcels ‘trapped’ within
favorable location.
6
Once the vortex is strong enough to maintain itself on its own it breaks away
from the parent wave.
The marsupial hypothesis gets its name from an analogy to how a marsupial
such as a kangaroo rears its young in a protective pouch until the young are
ready to fend for themselves.
MAINTAINING INTENSITY
The coupling between the strong winds in the boundary layer and the transfer of
latent and sensible heat into the air as it spirals into the cyclone.
Frictional convergence of the tangential winds plays a key role in
transporting latent heat into the core of the vortex.
The stronger the winds, the greater the frictional convergence into the core.
Also, the stronger the winds, the greater the transfer of latent and sensible
heat from the sea surface into the inflowing air.
The formation of an anticyclone aloft by the heating aids in the upper-level mass
divergence which is necessary to sustain and intensify the cyclone.
Tropical cyclones can maintain themselves as long as there is sufficient inflow of
warm, moist air into the cyclone, and there is adequate outflow aloft. Factors
that can cause a tropical cyclone to fluctuate in intensity (either up or down) are:
Variations in SST.
Interaction with land, which can result in less evaporation and latent heat
inflow.
Tropical cyclones can momentarily increase in intensity as they make
landfall due to enhanced low-level convergence due to the increased
friction over land.
Enhanced or suppressed outflow.
Increased vertical shear.
Vertical shear can cause the upper-level anticyclone and outflow to
decouple from the low-level inflow. If the outflow is weakened, then the
mass that is ejected into the upper troposphere by the convection can
subside in the vicinity of the tropical cyclone and weaken the convection,
as well as limit the surface pressure falls.
7
SYNOPTIC SCALE INFLUENCES
Since tropical cyclone formation requires a pre-existing cyclonic disturbance,
they normally form in
Monsoon troughs
Tropical waves
Old frontal zones or shear lines
Any enhancement of vorticity is favorable for formation.
A cold surge in the wintertime hemisphere often will enhance the Equatorial
westerlies, and is favorable for formation.
Existing storms can influence the formation and development of new storms.
For storms that have not yet recurved, the path ahead of the storm is
unfavorable for formation of another tropical cyclone, while the wake area
behind it is favorable.
This is due to the large scale vertical motion pattern forced by the
tropical cyclone.
The outflow from an existing storm can also sometimes provide too much
shear over a region and suppress formation and development.
Formation of upper-tropospheric outflow jets is key for development, since the
upper-level mass divergence must be larger than the lower-level mass
convergence.
A north-east quadrant outflow jet is enhanced by linking with
the subtropical jet
a tropical upper-tropospheric trough (TUTT)
a deep mid-latitude trough
an upper-level cold low
The position of the cyclone in relation to the above features is crucial for an
enhanced north-east outflow jet. Although they enhance outflow, which aids
deepening of the surface low, if they are too close then the increased westerly
shear can inhibit growth.
A south-west quadrant outflow jet is enhanced by an intense upper-level
anticyclone in the opposite hemisphere.
8
Lesson 12 – Tropical Cyclones: Motion and Analysis

The Use of Satellite Imagery in Tropical Cyclone Analysis, WMO
Reading: The Hurricane, Pielke, Chapter 3 (e-reserve)

The Use of Satellite Imagery in Tropical Cyclone Analysis, WMO, Chapter 5
(e-reserve)
GENERAL
Tropical cyclone motion can be thought of as the cumulative effect of the
following three influences:
Environmental steering
The beta effect
Asymmetrical convection
Of the three, environmental steering is usually of primary importance, followed
by the beta effect.
Asymmetrical convection is mainly responsible for short term “eye wobble”,
but is of little importance for long term motion.
ENVIRONMENTAL STEERING
To a first approximation, environmental steering can be thought of as “cork-in-
a-stream”, with the tropical cyclone merely advected by the mean environmental
flow.
The environmental flow can be represented by various means, either through a
single level, or through a mass-weighted mean flow.
A deep-layer mean is best for intense, mature cyclones.
A medium-layer mean or even shallow-layer mean is more suited to weak
systems, especially if they are highly sheared.
In any case, the cyclone circulation needs to be removed from the wind field, in
order to determine the environmental flow.
Synoptic-scale influences are very important for the environmental steering
currents.
An approaching trough can alter the steering current such that the cyclone
recurves, or at least tracks more northerly.
A strong subtropical ridge will keep a cyclone entrenched in the trade winds,
and tracking westward.
An adjacent tropical cyclone can influence the steering flow.
This is called the Fujiwhara effect, and can result in some unusual tracks as
the cyclones spiral around each other, occasionally even merging.
LINEAR BETA EFFECT

The beta effect refers to the tendency of a circulation to move, even in the
absence of a mean flow, due to the conservation of absolute vorticity
The quasi-geostrophic barotropic vorticity equation is
∂ζ g
+ V g • ∇ζ g + β v g = f ∇ • V z (1)
∂t
where the terms are the local vorticity tendency, advection of relative and
planetary vorticity, and divergence.
When equation (1) is linearized the phase speed for the waves it supports (called
Rossby waves) is
β
c=u − (2)
k + l2
2
where u is the speed of mean zonal flow, and k and l are the zonal and
meridional wave numbers.
The dispersion relation (equation 2) shows two important characteristics of
Rossby waves:
They will move westward in the absence of a mean flow.
They are dispersive, with the shorter waves moving more slowly than the
longer waves.
If a tropical cyclone is thought of as the superposition of several different linear
Rossby waves, then in the absence of a mean flow the cyclone would be expected
to move westward.
This tendency for the cyclone to move westward in the absence of a mean flow is
known as the linear beta effect.
2
NONLINEAR BETA EFFECT
If the Rossby waves comprising the cyclone were truly linear, there would be no
interaction between the different wavelengths, and the waves would just pass
through each other, with the longer waves moving faster.
However, if nonlinear interactions are allowed, the resultant motion of the
cyclone is not only westward, but also poleward.
This is known as the nonlinear beta effect.
The nonlinear beta effect can be explained somewhat qualitatively by imagining
that the cyclone is comprised of two Rossby waves.
The outer cyclone is represented by a long wavelength (small wave number)
wave while the inner cyclone is represented by a short wavelength (large
wave number) wave.
We will imagine that initially the cyclone is symmetric, and the vorticity
isopleths will be concentric with the streamfunction contours (see left side of
diagram below).
There will initially be no advection of relative vorticity.
The cyclone will move westward due to the advection of planetary vorticity.
Since Rossby waves are dispersive, at some time later the outer Rossby wave will
have traveled farther westward than the inner Rossby wave.
The cyclone is no longer symmetric, and the vorticity maximum will no
longer be in the center of the cyclone (see right side of diagram).
The result of the dispersion is that there is now positive advection of relative
vorticity poleward of the cyclone, which will cause the cyclone to move poleward.
3
The net result of the nonlinear beta effect is a westward and poleward cyclone
track in the absence of a mean flow.
Some qualities of the nonlinear beta effect are:
It is the size and strength of the outer region that are most important for
determining the beta effect.
Larger cyclones will have a stronger beta effect.
The intensity of the inner core has little influence on the beta effect.
The beta effect acts in addition to the mean flow steering.
The interaction between the beta effect and the mean flow steering may not
be linear.
ASYMMETRICAL CONVECTION
Though the inner core of a tropical cyclone has large inertial stability, there are
asymmetries in convection.
The asymmetrical convection can be due to several factors, including
SST gradients
Differential stress (different surface roughness, especially near land).
The asymmetrical convection can lead to local pressure falls in the eyewall,
which essentially can displace the eye in the direction of the pressure falls.
This leads to short-term “eye wobble” often seen in tropical cyclone tracks.
This eye-wobble is one reason why the longer term tropical cyclone motion
should not be estimated from two eye positions, but should be based on a longer
term average of several positions or fixes.
INTERACTION WITH MOUNTAINS

Topography can influence tropical cyclone motion. The prime example is the
Island of Taiwan, which is very mountainous.
As tropical cyclone cross Taiwan, the eye is sometimes seen to “jump” to the
other side, presumably due to pressure changes induced by the lee of the
topography.
4
ANALYSIS
One important key for a successful prediction of the cyclone track and intensity
is an accurate analysis of the current position and intensity of the cyclone.
Location is determines primarily by satellite imagery, except near land, when
radar and aircraft can be brought to bear.
Synoptic fixes (positioning based on adjacent synoptic observations) was the
staple of location techniques in the pre-satellite era, but is now a “quaint”
pastime for bored analysts.
Satellite imagery can give a very accurate fix when there is an eye, or an easily
visible, exposed low-level circulation center.
Positions based on curved cloud features or when the eye is obscured by a
central dense overcast (CDO) are less accurate.
The best intensity estimate is a direct observation from an aircraft penetration.
The aircraft can not only measure flight-level winds, which can be
extrapolated to surface winds, but pressure measurements can also be made
using drop sondes.
If the cyclone is close enough to land, Doppler radar observations can be used to
estimate intensity.
In the absence of aircraft observations or Doppler radar observations, the
Dvorak technique is the staple for determining intensity, and is described below.
DVORAK ANALYSIS
The Dvorak technique is based on appearance of the storm from IR and visible
satellite imagery.
The Dvorak technique assigns a T-number based on the appearance from the
satellite imagery.
The T-number is itself a combination of a CF (central feature) number and a
BF (banding feature) number.
The “T-number”, along with some rules which are formulated to avoid wildly
differing intensity estimates from one observation to the next, are used to assign
the current intensity, or CI.
5
The CI is usually close to the T-number for developing storms, but is higher for
weakening storms.
The assumption is the cloud features dissipate faster than the momentum of
the circulation.
The relation between CI and intensity is shown in the table below.
CI 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0
Intensity 25 25 30 35 45 55 65 77 90 102 115 127 140 155 170
(knots)
A CI of 2.5 implies a tropical storm, while a CI of 4.0 implies a
hurricane/typhoon.
TRACK PREDICTION
There are several different methods for predicting tropical cyclone tracks.
The historical progression of forecasting techniques began with persistence,
which is useful in the first few hours of the forecast.
Later, methods based on climatology were developed, such as finding analogs, or
past storms that had similar characteristics and synoptic environments.
Statistical methods use regression techniques to correlate information about the
storm and its environment with the likely track.
The earliest successful prediction technique was a statistical blend of
persistence and climatology, called CLIPER.
CLIPER had no dynamical input, and yet was very successful. It is often
used as the reference by which new methods are judged. If they can’t do
better than CLIPER, then they aren’t worth the effort.
Those statistical techniques that use dynamical properties as independent
variables are known as statistical-dynamical methods.
As understanding of the physical processes affecting motion was gained, pure
dynamical methods were developed.
Examples of the dynamical methods are
Numerical models
BAM (Beta plus Advection Models) which used output from numerical
models to define the steering flow, to which calculations of the beta effect
were added.
6
FBAM, MBAM, and SBAM stood for full BAM, medium BAM and
shallow BAM, and denoted whether a deep-layer, medium-layer, or
shallow-layer mean was used for the steering flow.
Output tracks from numerical models are also ensembled, in a sense creating a
type of statistical dynamical method.
INTENSITY AND SIZE PREDICTION

Methods for intensity and size prediction have lagged those for track prediction.
While track prediction has progressed into the dynamical methods stage,
intensity and size prediction are still in the climatology and statistical stages,
owing to an incomplete understanding of the processes that result in intensity or
size changes.
7
Lesson 13 – Monsoons
References: Tropical Climatology, McGregor and Nieuwolt

“The Elementary Monsoon,” Webster, in Monsoons, Fein (Ed.)
“Monsoon Overview,” Slingo, in Encyclopedia of Atmospheric Science
“Monsoon Dynamical Theory,” Webster and Fusillo, in Encyclopedia of
Atmospheric Science

“Monsoon Overview,” Slingo (e-reserve)
“Monsoon Dynamical Theory,” Webster and Fusillo (e-reserve)
GENERAL
• Monsoon comes from “mausim,” which is Arabic for season.
• There are a number of different definitions.
• Ramage’s definition
o Prevailing wind directions shifts by at least 120°° between January and July.
o Prevailing wind direction persists at least 40% of the time in January and
July.
o Mean wind speed exceeds 3 m/s in either January or July.
o Fewer than one cyclone-anticyclone alternation every 2 years in either
January or July.
• According to Ramage’s definition, only Asia, Australia, and Africa have distinct
monsoons.
• A more relaxed definition commonly used:
o Wind must reverse in direction between summer and winter.
o Summer season must be very wet, and winter season very dry.
• Using this definition, North America also experiences a monsoon (in the
Southwest U. S. and Northern Mexico.)
• Using the relaxed definition, over ½ of the tropics and ¼ of entire globe
experience monsoon-type climates.
• Three main driving mechanisms for the monsoon:
o Differential heating of land and ocean
o Moist processes
o Rotation of the Earth
DIFFERENTIAL HEATING OF LAND AND OCEAN

• The specific heat of water is much larger than that of dry soil.
• Effective heat capacity difference is even larger, because mass of ocean is much
larger.
o Only the upper most few centimeters of land are heated, due to slow
molecular transfer of heat vertically.
o In oceans, heat is effectively mixed downward tens of meters via turbulent
mixing.
• The difference in heat capacities, rather than specific heats themselves, is most
important.
• Moist soil has higher specific heat than dry soil.
o Saturated soil behaves more like “ocean” than land.
• Differential heating sets up a horizontal pressure gradient (similar to land/sea
breeze only on much larger scale.)
ROLE OF MOIST PROCESSES

• Differential heating is not enough to explain the strength and extent of monsoon
circulations.
• Moisture acts as “stored energy” through latent heat release.
• Evaporation occurs over the oceans, and then moisture is transported over the
land, where it is released through condensation.
o This essentially “focuses” the effects of the solar heating collected over the
ocean onto the land areas.
o This process is often referred to as the “solar collector.”
• Latent heating results in a more intense monsoon flow, and also a vertically
deeper monsoon flow.
o A moist monsoon has a depth on the order of the troposphere.
o A dry monsoon is much shallower, extending only to the mid-troposphere.
• Moisture also changes the character of the heating of the land.
2
o Moist land acts more like ocean.
o If land is dry, rising motion will occur closer to the coast, since the land will
be very warm.
o As land becomes wet from precipitation, the rising motion will move inland
over drier land.
o Precipitation will progress inland, allowing coastal area to dry out. Cycle
will then repeat itself.
o This is one factor in monsoon variability and monsoon “breaks.”
ROTATIONAL AND FRICTIONAL EFFECTS

• The Coriolis effect causes the air to “swirl” into the monsoon rather then flow
directly in. It results in
o cyclonic inflow at the surface
o anticyclonic outflow aloft
• The longitudinal extent of the low-level, versus upper-level circulations are
influenced by friction.
o There is more cross-isobaric flow at the surface than aloft, so surface
circulation has less of a longitudinal extent than does the upper-level
outflow.
LAND/OCEAN GEOMETRY
• Latitude and orientation of land masses effect monsoon circulation, as shown in
figure below, in which grey areas are land.
• Areas on or near Equator do not experience much rotation.
• Uniform surface (either all land or all water) result more in trade-wind trough
ITCZ, and not monsoon (e.g., Central Pacific Ocean).
• Western boundary results in atmospheric jet (e.g., East African Low-level Jet).
3
After “Physics of Monsoons: The Current View,” J.L. Young, in Monsoons (Fein, ed)

Tropical Meteorology All

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Tropical Meteorology All

Uploaded by

Copyright:

Available Formats

ESCI 344 – Tropical Meteorology

Lesson 1 – Introduction to the Tropics

References: Forecaster’s Guide to Tropical Meteorology (updated), Ramage

Reading: McGregor and Nieuwolt, Chapter 1

MAJOR DIFFERENCES BETWEEN TROPICS AND MIDLATITUDES

THE BREAKDOWN OF THE GEOSTROPHIC APPROXIMATION

GRADIENT LEVEL WINDS

These are intermediate between a wave and a developing closed circulation.

These occur at the intersection of two asymptotes of opposite sign (divergent

Cyclones and anticyclones represent closed circulations, which would have

Don’t diverge off of a line downstream of convergence (however,

RELATION BETWEEN ASYMPTOTES AND CLOUDS

TROPICAL CYCLONES DEFINED

Wind Atlantic/ Northwest Pacific North Southwest Southeast Pacific

≤16 Tropical Tropical Tropical Tropical Tropical

WARNINGS AND WARNING AGENCIES

Other countries may use different scales (e.g., Australia’s Bureau of

References: Forecaster’s Guide to Tropical Meteorology (updated), Ramage

LATITUDINAL HEAT IMBALANCE

THE HADLEY CIRCULATION

THE SUBTROPICAL HIGHS

THE INTERTROPICAL CONVERGENCE ZONE

Trade-wind trough Monsoon Trough and Buffer Zone

The cloudiness associated with the ITCZ varies.

EL NINO, LA NINA, AND THE SOUTHERN OSCILLATION

MADDEN-JULIAN OSCILLATION (MJO)

References: Forecaster’s Guide to Tropical Meteorology (updated), Ramage

Reading: McGregor and Nieuwolt, Chapter 5 (e-reserve)

LOW-LEVEL WIND FIELD

UPPER-LEVEL WIND FIELD

SUBTROPICAL JET STREAM

TROPICAL EASTERLY JET

WEST AFRICAN MID-TROPOSPHERIC JET

EAST AFRICAN LOW-LEVEL JET (SOMALI JET)

References: A Global View of Tropical Cyclones, Elsberry (ed.)

Reading: A Global View of Tropical Cyclones, Chapter 3 (e-reserve)

REQUIREMENTS FOR FORMATION

NORTHEAST AND NORTHCENTRAL PACIFIC

NORTH INDIAN OCEAN

SOUTHWEST INDIAN OCEAN (West of 100°°E)

SOUTHEAST INDIAN OCEAN (100°°E to 142°°E)

References: Forecaster’s Guide to Tropical Meteorology (updated), Ramage

STRESS-DIFFERENTIAL INDUCED DIVERGENCE

where Fr is the deceleration from friction. If friction is considered to be linear in

where CD is the drag coefficient.

References: Tropical Climatology (2nd Ed.), McGregor and Nieuwolt

Reading: McGregor and Nieuwolt, Chapter 8 (e-reserve)

WAVES IN THE EASTERLIES

WAVES OVER THE NORTH ATLANTIC AND CARIBBEAN

BAROTROPIC INSTABILITY IN AN EASTERLY JET STREAM

TROPICAL UPPER-TROPOSPHERIC TROUGHS

References: Climate Dynamics of the Tropics, Hastenrath

Reading: Brown and Colling Section 5.1 (e-reserve)

BASIC OCEAN STRUCTURE

Things to note about the SST distribution

EKMAN TRANSPORT AND UPWELLING/DOWNWELLING

WIND-DRIVEN SURFACE CURRENTS IN A RECTANGULAR OCEAN

The major features of the circulation are:

EAST-TO-WEST STRUCTURE OF THE THERMOCLINE

References: A Global View of Tropical Cyclones, Elsberry (ed.)

Reading: A Global View of Tropical Cyclones, Chapter 3, Frank (e-reserve)

STRENGTH, SIZE, AND INTENSITY

The wind structure is often represented by a modified Rankin vortex,