In the second installment of the much-delayed learning series for the off-season, we'll take a look at the idea of cyclone phase and the cyclone phase diagrams, as suggested some time ago by Margie. As a brief reminder or introduction, the phase diagrams are available in real-time at http://moe.met.fsu.edu/cyclonephase/.
It has been understood for some time now that, in a general sense, there are two types of cyclones: warm-core cyclones such as tropical storms and cold-core cyclones such as most mid-latitude systems. While both are areas of lower pressure, they have several key differences structurally: warm-core cyclones are strongest in low levels, while cold-core cyclones are often strongest at upper levels (this point becomes critical later on); warm-core systems often have their peak winds near the center of circulation, decaying rapidly with increasing distance, while cold-core systems often have their peak winds located further from the center and not decaying as rapidly with distance; and warm-core tropical cyclones draw their energy from heat and moisture drawn from the oceans, while cold-core cyclones largely draw their energy from shear and gradients of vorticity.
These differences (and others not listed) lead to quite different structures for warm-core tropical cyclones and cold-core cyclones. The former are generally symmetric and upright in nature (i.e. the wind field is rather circular, as is the temperature field), while the latter are generally tilted with height and highly asymmetric in nature (somewhat owing to frontal structures, conveyor belts of moisture and temperature, and a lot of other factors beyond the scope of this piece). These differences are noticeable both at lower and upper levels.
Inherent to their definitions, warm-core cyclones are denoted by warmer temperatures at their core than in the surrounding environment, while cold-core cyclones are denoted by colder temperatures at their core. This has impacts on something called the thickness field, or the distance between two height surfaces (e.g. the 1000-850mb thickness is often used when forecasting snow and is just the height of the 850mb surface subtracted by the height of the 1000mb surface). We have equations to define the thickness that I won't go into; however, the key thing to note is that as the mean temperature increases, so does the thickness. Symmetric storms -- whether cold-core or warm-core -- will have relatively symmetric thickness fields (as the temperature fields should, by definition, be relatively symmetric); the opposite is true for asymmetric (often cold-core) storms.
This leads us into the discussion of categorizing storms by their structures. The idea of a quick classification scheme first arose in the 1990s from Jack Beven at the , a noted purveyor of "oddball" storms. Later, Bob Hart, then a grad student at Penn St., came along and found that the basic fields that define the differences between warm-core and cold-core cyclones -- their symmetry and intensity with height -- can be used to classify any given cyclone based upon their observed structure. This is not to say that the other features differing between the two types of cyclones aren't important; in fact, the reverse is true, as these can be used to further distinguish between specific types of storms and are often the factors we are most concerned with here on the ground.
Thus, the cyclone phase space came about, featuring two diagrams: a lower level diagram comparing the asymmetry in the thickness fields (called B) and the change in intensity of the cyclone with height over the low levels (900-600mb; here called VtL). Essentially, the lower-level diagrams try to show how symmetric a storm is and the inner-core structure (warm vs. cold) of the cyclone. An example can be found here for Hurricane . The vertical axis denotes B, while the horizontal axis denotes VtL. A tropical cyclone is going to be relatively symmetric (no matter the intensity) and be stronger at lower levels than it is higher up (even moreso the stronger the storm is). Thus, the lower right hand corner of the diagram is where you will find tropical cyclones in most cases. A typical cyclone will be asymmetric and have an intensity that increases with height, both moreso with increasing intensity; thus, the upper left hand corner of the diagram is where you will find these storms in most cases. Storms undergoing transition often weaken from the top-down, leading to them going through the upper right portion of the diagram as they become . A good example of this is Hurricane Karl from 2004.
The upper level diagram features the VtL parameter along with a measure of how the storm is structured in intensity at upper levels (here, 600-300mb and called VtU). These diagrams attempt to show how vertically deep a cyclone is; a deep warm-core system will show an continued decrease in intensity with height, a shallow warm-core system will show an increase in intensity in the upper levels (but a decrease with one in lower levels), and a deep cold-core system will show an increase in intensity at all levels. An example is given here again with Hurricane Karl. Note how it initially starts out as a deep warm-core system and gradually transitions into a deep cold-core system.
The real utility in these diagrams is with a) forecasting the onset of transition, b) forecasting the potential for tropical transition, and c) analyzing warm-core structures in midlatitude cyclones (the so-called warm-seclusions). With regards to point (a), the diagrams above with Karl give a good example of ET. When a system in the lower diagram crosses the barrier between symmetric and asymmetric warm-core and crosses the barrier between deep and shallow warm-core in the upper diagram, ET is said to have begun. When it crosses from asymmetric warm-core to asymmetric cold-core and from shallow warm-core to deep cold-core, ET is said to have completed. After ET, any number of evolutions are possible; a storm could decay, in which case all of the parameters will move back toward zero, it could continue to develop as a cold core system (a good example is given with this storm in 1978), or it could become a warm-seclusion, point (c) above. These storms often combine the worst of the features of both warm- and cold-core cyclones, with strong winds found both near the center and outward to larger distances, and thus pose dangers to property and shipping interests. These storms are defined by asymmetric and often shallow warm-core structures -- the netherworld between tropical and -- as well as very deep pressures (note the color guide on the right side of each diagram) and large gale-force wind radii (note the circle size diagram on the right side of each diagram). A good example of a storm that undergoes warm-seclusion development is given by Irene of 1999, one of the most intense summer/fall cyclones on record in the midlatitudes in the N. Hemisphere. Finally, point (b) details the case where a cold-core system will transition into a warm-core system, the process known as tropical transition. This often occurs through the processes noted in the last installment, Trough Interactions, so I refer you back to that article for more. Owing to where these systems usually form -- well out into the Atlantic away from our land-based observing systems -- and their relatively small nature, they are not always well-represented by the phase analyses. Nevertheless, when a cold-core system is forecast to start becoming asymmetric warm-core in nature and trending toward the symmetric warm-core region, that sets off a bell that something might happen -- subtropical if it stays in the asymmetric warm-core region (these systems often are similar in structure to warm-seclusions, albeit weaker) or tropical if it keeps going. There aren't many examples of this phenomenon available in the analysis archive, but Delta from 2005, Olga from 2001, and Karen from 2001 are three decent examples of this process.
As noted above, all of these products are available in real-time. Using model data, it is possible to forecast the evolution of a storm structurally with a great degree of success. This is done by the over the course of the tropical season both with forecasting transition as well as the potential for tropical transition -- something we saw several times this past season. Further information about these diagrams is available at http://moe.met.fsu.edu/cyclonephase/help.html and the citation of the paper initially detailing this work is:
Hart, R., 2003: A cyclone phase space derived from thermal wind and thermal asymmetry. Mon. Wea. Rev., 131, 585-616.
I hope this helps to better understand these tools to some degree. It's late, so there may well be some things that need clarification. Please let me know if so and if you have any questions. Next time: tropical cyclone track forecasting & steering influences.
Current Tropical Model Output Plots
(or view them on the main page for any active Atlantic storms!)