FIFTH INTERNATIONAL WORKSHOP on TROPICAL CYCLONES
Topic 4.1: Formation Definitions
Rapporteur: M.A. Lander
University of Guam/WERI
Mangilao, Guam 96923 USA
Working Group: N. Davidson, H. Rosendal, J. Knaff, and R. Edson, J. Evans, R. Hart
With the advent of the meteorological satellite, the probability that a tropical cyclone (TC) would escape detection in data-void regions of the ocean basins was greatly reduced. All Tropical Cyclone Warning Centers (TCWCs) have personnel dedicated to the function of Meteorological Watch (METWATCH) who (among other things) search for suspect regions where a tropical cyclone might develop and to assess the intensity and wind distribution of existing tropical cyclones. It thus is also a function of such METWATCH personnel to determine when a tropical cyclone has formed, and to provide this guidance to the various Tropical Cyclone Warning Centers who must decide whether to initiate advisories on any given cyclone. Many factors come into play when tropical cyclone advisories are initiated: one is the risk posed by a cyclone to mariners and coastal residents, and another is the purely technical assessment of the cyclones structure. It may well be true that operational requirements and theory of tropical cyclone formation may not mesh. If only for the sake of basin tropical-cyclone climate inter-comparison, there must be some consensus on when an area of disturbed weather in the tropics or subtropics -- or a pre-existing low-level cyclonic circulation in the tropics or sub-tropics -- becomes a tropical cyclone.
The WMO (1966) defines a tropical cyclone as a non-frontal synoptic scale cyclone originating over tropical or subtropical water with well organized convection and definite cyclonic surface wind circulation. Apart from the common use of the generic term tropical cyclone, there are regional differences in the definitions of tropical cyclone categories (e.g., tropical depression, tropical storm, and typhoon). The WMO regional committees for all the tropical cyclone basins derive their own definitions. The U.S. National Hurricane Center (NHC) is part of WMO Region IV North and Central America. The Region IV Hurricane Operations Plan defines a tropical storm as a well-organized warm-core tropical cyclone in which the maximum average surface wind (one-minute mean) is in the range of 63-117 km hr-1 (39-73 mph (34-63 knots) inclusive. Below is a sampling of how definitions vary around the world:
In all of these regions, a tropical cyclone is deemed to have formed if a low-level cyclonic vortex in the tropics or subtropics has evolved to a point such that there is a well-defined low-level cyclonic wind field (e.g., marked by obvious cyclonically curved cloud lines on visible satellite imagery) and at least some persistent deep convection near the center of the vortex. The initiation of tropical cyclone advisories in operational practice begins with the vortex intensifying to some threshold wind speed or minimum sea-level pressure. Also, there are times when pre-existing cyclones that already possess extensive areas of accelerated wind speeds and associated deep convection are said to have had a transition into a tropical cyclone. Two questions immediately arise: 1) Is the origin of a tropical cyclone from its incipient disturbance an arbitrary classification? and, 2) When do other types of pre-existing cyclones that may already have intense and extensive cyclonic wind fields (e.g., subtropical cyclones) become tropical cyclones?
4.1.2 Tropical cyclones
In all tropical ocean basins there are always areas of deep convection of many different sizes. These areas of deep convection may be increasing or decreasing in areal extent, or showing an increase or decrease in organization of the deep convection into curved bands or other patterns of organization. The first stage in the formation of a tropical cyclone (aside from the transition of other types of cyclones into tropical cyclones) is the tropical disturbance. A tropical disturbance is a discrete system of apparently organized convection, generally 200 to 600 km in diameter that originated in the tropics or subtropics, has a non-frontal, migratory character and has maintained its identity for 12-to-24 hours. The system may or may not be associated with a detectable perturbation of the low-level wind or pressure field. It is the basic generic designation that, in successive stages of development, may be classified as a tropical depression, tropical storm, or hurricane/typhoon/cyclone (taking into account regional differences of nomenclature).
a) Physical characteristics
Prior to the advent of the meteorological satellite, features that could be observed from the surface defined tropical cyclones. The universally recognized signature of a tropical cyclone is the characteristic rapid decrease in surface pressure and increase in surface winds towards the center, with the heavy rain area and storm-lashed seas, and the relatively calm eye. Early knowledge of tropical cyclone structure included the increase of wind as one proceeded into the region of rapid pressure fall, the sense of rotation of the winds around the center, the presence of a calm region at the center, and the typical asymmetry of wind and heavy weather to the right of the direction of motion in the Northern Hemisphere and to the left in the Southern Hemisphere. The hydrostatic requirement of a warm core extending throughout the troposphere was confirmed with the first upper-air observations in tropical cyclones in the 1940s and 1950s. Other well-known characteristics of tropical cyclones observable from ground-based observations include the thermal and moisture fields, and the characteristic banded distribution of cloud and rainfall. By the late 1960s, meteorologists had observed many tropical cyclones during their entire life cycles using good quality satellite pictures. One of the most important observations made during these years was that the cloud patterns of tropical cyclones evolve through recognizable states as the intensity of the cyclone changes. The cloud patterns in general showed that the dense clouds (or coldest clouds on Infrared images) formed around the center in the shape of one or more curved bands in the early stage of development. It was further determined that the cloud patterns normally showed indication of tropical cyclogenesis 36 hours before the TC reached tropical storm intensity. Dvorak (1975, 1984) made revolutionary advances in using satellite imagery to detect tropical cyclones and to estimate their intensity. Dvorak observed that it was the pattern formed by the clouds that is related to the cyclones intensity and not the amount of clouds in the pattern. He made the further observation that most tropical cyclones exhibit cloud patterns of the curved-band pattern type through much of their life times.
b) Thermodynamic considerations
Kleinschmidt (1951) first recognized that the energy source of hurricane resides in the thermodynamic disequilibrium between the tropical atmosphere and oceans. This may not be reflected in an actual temperature difference between air and sea, which in the tropics in usually less than 1ºC, but rather in the under-saturation of near-surface air. The evaporation of water transfers heat from the ocean, whose effective heat capacity is enormous in comparison with the overlying atmosphere. The rate of transfer of heat from the ocean to the atmosphere is a function of the surface wind speed. The actual rate of heat transfer is a subject of much controversy and research. The dependence of the transfer rate on wind speed is the principal feedback mechanism that allows hurricanes to develop. In its essence, the mature tropical cyclone (hurricane or typhoon) may be thought of as a wind-induced surface heat exchange instability, in which increasing surface winds lead to increased heat transfer from the sea, which lead to intensification of the winds, and so on. The energy cycle of the mature tropical cyclone has been idealized by Emanuel (1986, 1991) as a as a Carnot engine that converts heat energy extracted from the ocean to mechanical energy. We now come to one of the most important milestones in the evolution of a tropical cyclone, and one that has bearing on the occurrence of tropical cyclogensis: the incipient tropical vortex must reach a point such that it is self-sustaining and self-amplifying by virtue of enthalpy extraction from the ocean (R. Zehr and K. Emanuel, personal communication).
c) Other types of cyclones in the tropics and subtropics
Between the great divide of ordinary mid-latitude cyclones and the typical mature tropical cyclone, there exist other types of cyclones. Hebert and Poteat (1975) were among the first to identify and codify the characteristics of subtropical cyclones. Both the tropical cyclone and the subtropical cyclone have deep convection organized into bands around the center. The primary difference between the tropical and subtropical cyclone was the existence, in the tropical cyclone, of dense overcast of nearby deep convective over the center. A subtropical cyclone that develops persistent dense (or cold) cloud of convective origin over its low-level circulation center is said to have completed its transition into a tropical cyclone. As to the subtropical cyclone and other cyclones possessing some of the physical characteristics of tropical cyclones (e.g., arctic hurricanes, Mediterranean hurricanes, warm-seclusions and hybrids (Kuo et al. 1992), and post-tropical (Keith 1988)), the evidence is that there is a continuous spectrum of cyclones from the pure air-sea enthalpy flux limit of the classical tropical cyclone to the pure baroclinic limit of the ordinary mid-latitude cyclones. For events in the middle of this spectrum (e.g., subtropical cyclones) any definition of when it has become a tropical cyclone will tend to be arbitrary.
Fig. 4.1.1: An infrared view of the western North Atlantic Ocean on 12 September 2000 illustrates the spectrum of cyclone types. Small Hurricane Florence is northeast of Florida, developing Tropical Storm Gordon (that was not officially numbered at this time) is south of Cuba, and a subtropical cyclone is east-northeast of Florence and west-southwest of a mature extratropical cyclone.
4.1.3 Formation definitions
a) Conventional tropical cyclones
Because the incipient stage of most tropical cyclones is over the data-sparse tropical oceans, it is likely to be first viewed by meteorological satellite. Thus, satellite-observed properties of the deep convection and low-level cloud lines will likely be used operationally to decide when a tropical cyclone has formed. Many of the physical and thermodynamic properties of tropical cyclones (e.g., their middle- and upper-tropospheric warm anomaly, and their self-sustaining and self-amplifying energy extraction from the sea are not variables that can be assessed from visible and infrared satellite imagery. New satellite microwave imagers allow a better view of the structure of the deep convection in a tropical cyclone, an estimate of the vertical temperature profile in the core region of a tropical cyclone, and (for the active microwave sensors) a direct measurement of the wind field in the core and environment of a tropical cyclone. The four tropical cyclone pattern types defined by Dvorak the curved band pattern, the shear pattern, the central dense overcast (CDO) pattern, and the eye pattern are suitable for classifying most tropical cyclones (therefore, these will be the conventional tropical cyclones). In operational practice, almost any area of deep convection that has persisted for at least 12 hours and that has some degree of organization of deep convection into cyclonically curved bands, or accompanies a field of cyclonically curved low-level cloud lines, may be classified in the T 1.0 stage of Dvoraks scale. A tropical cyclone classified in the T 0.5 through T 1.5 categories indicates a weak tropical cyclone possessing 1-minute average winds of 25 kt (13 m s-1). At the T 2.0 stage, the tropical cyclone has 30 kt (15 m s-1) winds, and it is at this stage that many Tropical Cyclone Warning Centers initiate advisories. A T 2.5 classification is given to tropical cyclones that have acquired winds of 35 kt (18 m s-1).
Other cyclonic circulations in the tropics do not fall into Dvoraks pattern types, including monsoon depressions (MD) (JTWC 1996) and monsoon gyres (MG) (Lander 1994). Some have argued that these types of cyclones are not tropical cyclones in the conventional sense as defined by Dvorak, and should not be numbered or named by Tropical Cyclone Warning Centers. The primary structural difference between these types of cyclones and conventional tropical cyclones is the larger displacement of the band of maximum surface winds from the center in the MD and MG, an incomplete ring of high winds in the MD and MG, and a much broader light-wind core in the MD and MG cyclones. In the WMO 2001 Tropical Cyclone Operational Plan for the South Pacific and Southeast Indian Ocean, the requirement for gales to encircle the center was dropped. This was refined at a WMO meeting in 2002 to require that the gales be close to the center. There is no single definition that suits everyone, because not everyone has the same goals. By not including MD cyclones and MG cyclones as tropical cyclones, there is a possibility that a community could not be warned of damaging wind. On the other hand, including (or not including) the MD and MG cyclones as tropical cyclones may lead to a corruption of the historical tropical cyclone databases.
A way to address this problem could be to state, a priori, what the purpose of the definition is, then pose the definition, and then adjudge accordingly. So, for example:
Circumstance 1. The primary purpose is to protect life and property from gale-force (or stronger) winds associated with tropical low-pressure systems. Then a relatively non-restrictive definition would be appropriate: A tropical cyclone is a low-pressure system of tropical origin with associated winds of at least gale force, or some other pertinent wind-speed threshold such as the 30 kt of the Dvorak T 2.0 classification. On this basis, many MD cyclones, and even an occasional MG cyclone would qualify as a tropical cyclone.
Circumstance 2. The primary purpose is to monitor tropical cyclone climate change and an internally consistent database that contains only those systems that possess the stringent scientific/meteorological attributes that have been traditionally ascribed to tropical cyclones. Then, a restrictive definition is required, such as: the cyclone must be of tropical origin, possess a warm core in the middle and upper troposphere, have persistent deep convection near the center, and have wind speeds near the center of 30 kt(15 m s-1) or more, etc.
In light of the typical formation of conventional tropical cyclones in the western North Pacific (and other basins to a lesser extent) from MD type cyclones, it is recommended that any cyclone in the tropics with gales in any part of its circulation be considered to be a tropical storm (or whatever the regional nomenclature is for tropical cyclones that have acquired gales). Discretion is urged for the MG type cyclones that may have a long swath of nearly straight-line gales that are displaced very far from the center. Those Tropical Cyclone Warning Centers that issue advisories for cyclones with winds less than gale force the so-called tropical depression -- should also consider the MD type cyclones, if only for their propensity to intensify and become conventional tropical cyclones with large outer wind distributions.
b) Transition of subtropical cyclones to tropical cyclones
Beginning in 2002, the U.S. National Hurricane Center (NHC) began to number and name subtropical cyclones. In addition, the numbering and naming of the subtropical cyclones follow the natural sequence of the numbering and naming system for the tropical cyclones. Thus, should a subtropical cyclone have a transition into a tropical cyclone, the number and name carry over. Hebert and Poteat (1975) provided a description of the characteristics of the subtropical cyclone that still applies today. The subtropical cyclone typically has only a weak lower tropospheric warm-core structure resulting from the lack of sustained convection near the cyclone center. A large radius of maximum winds is associated with these cyclones. To obtain the satellite classification of a subtropical cyclone is a simple process of looking at a series of satellite pictures of subtropical cyclones of various intensities and choosing the one that best applies to the case in question. They recommended that a subtropical cyclone is considered to have a transition to a tropical cyclone if the persistent deep convection becomes located near the cyclone center so as to cover the low-level center with dense (cold on IR imagery) overcast. This conversion can be diagnosed using satellite imagery, but is often quite difficult to forecast since the evolution within numerical models is so subtle and poorly indicated using conventional analysis. So far in 2002, the NHC strategy has worked well, and all of the named subtropical cyclones have made the transition to tropical (though this need not be the case).
Recent work by Hart (2002) gives a quantitative way to track a cyclone through a 3-D phase space based on the thermal wind in the core of the cyclone in the lower-troposphere, the thermal wind in the core of the cyclone in the upper troposphere, and the horizontal temperature gradient in a section drawn across the core of the cyclone (See Fig 4.1.2a,b). The previously sharp boundary distinguishing tropical cyclones from extratropical cyclones in the first half of the 20th century has been substantially weakened. As a result of the numerous examples of unconventional cyclone structure and development, the meteorological community is now attempting to use a more flexible approach toward classification of the three-dimensional nature of cyclone structure. However, an objective and consistent approach has not yet been agreed upon. The objective scheme proposed by Hart (2002) can be applied to any cyclone that has a well-resolved three-dimensional height field (this may limit its use where data is sparse, but new satellite techniques for deriving soundings and dense three-dimensional wind information may widen the applicability of the proposed scheme). After reviewing many parameters to delimit cyclone types, it was found that two similar, yet fundamental, measures of cyclone structure were the most robust: thermal wind and thermal asymmetry. The three parameters used to describe the general structure of cyclones are: the lower tropospheric thermal asymmetry, the lower tropospheric thermal wind (cold core versus warm core parameter VtLOWER), and the upper tropospheric thermal wind (cold core versus warm core parameter VtUPPER). These three parameters are successful at describing and differentiating the structure of tropical, subtropical, and extratropical cyclones. Using the work of Hebert and Poteat (1975) and the phase diagrams of Hart (2002), it should be relatively easy to identify subtropical cyclones and their transition into tropical cyclones.
Any definition of tropical cyclone formation is arbitrary, whether the tropical cyclone forms from a weak and highly disorganized tropical disturbance, or is the result of the transition of other types of cyclones (e.g., the subtropical cyclone) into tropical cyclones. In the former case, it is recommended that any cyclone in the tropics with gales in any part of its circulation be considered to be a tropical storm (or whatever the regional nomenclature is for tropical cyclones that have acquired gales). Discretion is urged for the MG type cyclones that may have a long swath of nearly straight-line gales that are displaced very far from the center. Those TCWCs that issue advisories for cyclones with winds less than gale force the so-called tropical depression -- should also consider the MD type cyclones, if only for their propensity to intensify and become conventional tropical cyclones with large outer wind distributions. In the latter case of transition, it is proposed that the work of Hebert and Poteat (1975) and the phase diagrams of Hart (2002) be used. It should be relatively easy to identify subtropical cyclones and their transition into tropical cyclones.
Acknowledgments The input of working group members was valuable in drafting this report. Also, comments solicited from Kerry Emanuel were used extensively. Other commentary recently posted to the email@example.com bulletin board were necessary to push this work to completion. Bulletin board commentators used in this report include Gary Padget, David Roth, Andrew Burton, Kenneth Schaudt, Jon Gill, and Chris Fogarty.
Figure 4.1.2 Summary of the general locations of various cyclone types within the proposed phase space: a) VTLOWER (lower tropospheric thermal wind) vs. VTUPPER (upper tropospheric thermal wind) and b) 900-600 hPa across-storm temperature gradient vs. VTLOWER (lower tropospheric thermal wind). Cyclones may move throughout the phase space during their evolution. Figure adapted from Hart (2002).
Dvorak, V.F., 1975: Tropical cyclone intensity analysis and forecasting from satellite imagery. Mon. Wea. Rev., 103, 420-430.
_______, 1984: Tropical cyclone intensity analysis using satellite data. NOAA Tech. Rep. NESDIS 11. U.S. Dept. of Commerce.
Emanuel, K.A., 1986:An air-sea interaction theory for tropical cyclones. Part I. J. Atmos. Sci., 40, 2368-2376.
_______, K.A., 1991: The theory of hurricanes. Annu. Rev. Fluid Mech., 23, 179-196.
Hart, R.E., 2002: A cyclone phase space derived from thermal wind and thermal asymmetry. Mon. Wea. Rev., In Press
Hebert, P.H., and K.O. Poteat, 1975: A satellite classification technique for subtropical cyclones. NOAA Tech. Memo. NWS SR-83, 25 pp.
Joint Typhoon Warning Center (JTWC) 1996: Definitions. 1996 Annual Tropical Cyclone Report. Appendix A. pp 320-322. JTWC, Naval Pacific Meteorology and Oceanography Center, Pearl Harbor Hawaii.
Kleinschmidt, E, Jr., 1951: Gundlagen einer theorie des tropischen zyklonen. Arch. Meteorol., Geophys. Bioklimatol., Ser. A, 4, 53-72.
Kuo, Y-H., R.J. Reed, and Low-Nam, 1992: Thermal structure and airflow in a model simulation of an occluded marine cyclone. Mon. Wea. Rev., 120, 2280-2297.
Lander, M.A., 1994: Description of a monsoon gyre and its effects on the tropical cyclones in the western North Pacific during August 1991. Weather and Forecasting, 9, 640-654.
WMO, 1966: International Meteorological Vocabulary. WMO-No. 182, TP. 91, World Meteorological Organization, Geneva, Switzerland. 276 pp.