This is not a self-contained discussion.  Rather it is a number of small contributions to a much wider discuussion held on the tropical stroms mailing list:  http://tstorms.org/tropical-storms/

Date: Tue, 30 Mar 2004 12:32:37 +1000
From: Jeff Callaghan
 

Thanks to members of the group for the illuminating discussion re this  strange system

We in the Coral Sea (lately the shear capital of globe) experience the  whole continuum between severe extra-tropical cyclones and severe tropical cyclones and consequently we have, from necessity, studied many of these
systems in attempt to better make the decision when to declare a system a  named TC or not.

We have examined as best we can the large scale environment in which this
system evolved using the UK model analyses.

The UK had the low initially developing downstream (upper levels) of a  tropopause undulation (200hPa warm anomally) at 1200UTC 22 March 2004 . As an example see Figure 25 Paul A. Hirschberg and J. Michael Fritsch 1999 Tropopause Undulations and the Development of Extratropical Cyclones. Part II: Diagnostic Analysis and  Conceptual Model., MWR 119, pages 518-550

The undulation then weakened and by 1200UTC 24 March 2004 the low was  nearly vertical with a weak cyclonic circulation developing at 200hPa and it was located in a 700hPa to 500hP cold thermal trough.

 From 1200UTC 25 March 2004 to 1200UTC 26 March 2004 a new tropopause  undulation approached from the southwest and came close to the low which was intensifying. This new undulation brought warmer air over the system at 200hPa but without the strong 200hPa warm air advection associated with the  earlier intense extra-tropical development. This is what we find in our region - that the Hirschberg and Fritch model describes all our intense  extra-tropical development but as we move across the spectrum towards the tropical systems end,  the upper warm air advection is not evident and  presumably convection begins to provide much of the warming at upper levels necessary for MSL pressure falls.

By 0000UTC 28 March 2004 the low straddled an increasing  low to middle  thermal gradient between a warm thermal high overland to its southwest and a cold 700hhPa to 500hPa cold low near and northeast of the centre. The southeasterly low to middle southeasterly vertical wind shear associated with this thermal gradient can be verified from the Porto Alegre winds
(provided by Jiann-Gwo Jiing ) after 1200UTC 26th when the winds there mostly turned  more southeasterly with height which indicated colder air to the northeast of the station (in the vicinity of the low) and warmer air to the southwest.

We see all our systems (TCs and others) straddling such  thermal gradient  (from real observations) but with the TCs the shear and vertical tilt is much weaker. The 850/500hPa shear at Porto Alegre  at 0000UTC 28th is  around 20 knots which we have observed from observations around 100km from intense TCs.
 

In summary this system when compared with our examples in eastern OZ  appears to be similar to the hybrid systems located towards the tropical cyclone end of the continuum, operationally I think we would call it a TC.

Jeff

Date: Tue, 30 Mar 2004 07:07:03 +0000 (GMT)
From: John McBride
 

I too am finding the discussion interesting.  For us to be able to understand why  tropical cyclones develop in the first place and how their frequency,  structure etc will change with anthropogenically induced climate change, it is certainly relevant to spend time on the question of what constitutes a tropical cyclone. The "looks like a duck, quacks like a duck" school have a strong point as  the formation of an eye and spiral bands are certainly characteristic of a  set of dynamics we would refer to
as tropical cyclone eye-dynamics.

In that context, I would like to clarify my statement on the  importance of the cyclone's location relative to the sub-tropical jet.  My conceptual model of cyclone development is that there are two families of system representing the extremes of what we find in nature.  The first, caused by baroclinic instability, has its dynamics governed by baroclinic energy conversion processes. The characteristic structure of these is cold core through the depth of the troposphere (and warm above); and in the upper troposphere these systems have the jet stream as part of their circulation on the equatorward side. (As a corollary, once a system has the jet stream "as part of its circulation" on the equatorward side, it is almost trivially cold-cored through the depth of the troposphere)

The second family is tropical cyclones.  The dynamics are governed by latent heat release through the middle troposhere; and so the resulting potential vorticity couplet (negative/anticylonic above; positive/cyclonic below) is consistent with or requires a warm core structure through the upper half of the troposphere.   Genesis/development (and "tropical-cyclone eye dynamics") depend on some form of air-sea interaction in that it requires a pre-existing low-level vortex overlying warm seas. Examples are the ITCZ vortex  roll-up mechanism in the East Pacific and the n=1 Rossby wave twin vortex mechanisms we have discussed in recent years in this forum.

>From the cases I have looked at (mainly Australian cases) the hybrid systems are not entirely inconsistent with this two family model.  What seems to happen is that the tropical-cyclone type dynamics (air-sea interaction, mid-troposphere latent heat release, formation of an eye) occurs on a small scale in an area of very weak horizontal  pressure and temperature gradient all embedded within a larger-scale  mid-latitude baroclinic system.   Thus you have two systems, a  tropical-type one embedded with the envelope of a mid-latitude one. Presumably this can happen as convective dynamics and tropical cylone dynamics do not know about spatial scale.  Rather the scale can be non-dimensionalised relative to the Rossby deformation radius.  Thus if this becomes small (low static stability, smallish vertical depth) presumably the relevant dynamics can occur over the scale of a meso-vortex.

It would help if we could see the relevant analysed charts.  I just called Jeff Callaghan to see if he could send down the UK model analyses he was looking at, so I could put them on my web-page.  He had gone home for the day, however.  If no-one beats me to it, I'll put them up tomorrow sometime.

Thanks to Roger Edson for putting out a "working best track".  It will be very useful for us to have a look at this system.

Cheers

John McBride