Ian Shepherd

John and wx group,

At 09:56 25/01/2002 +1100, John McBride wrote:
>And another question is "Why Broome?"; why not Learmonth or Wyndham?"

Or Darwin ...

Conditions suitable for the formation of long-lived squall lines may be
found anywhere over northern Australia at various times during the wet
season - often in association with middle-level easterly winds north of the
subtropical ridge or south of a tropical low/trough.

Squall lines of this type are a significant factor in the wet season
climate of the western 'Top End' and the Darwin region. The detection of
conditions conducive to the formation of squall lines and severe
thunderstorms is a daily task for Darwin forecasters - the 'convective
analysis' - in which the potential contribution of CAPE and vertical wind
shear to convective organisation are assessed.

A summary of the techniques used is contained in the attached paper:
'Assessing Severe Thunderstorm Potential Days and Storm Types in the
Tropics' by Lori Chappel, NTRO Severe Weather Section; published in the
Proceedings of the International Workshop on the Dynamics and Forecasting
of Tropical Weather Systems, Darwin January 2001.
 

The convective analysis procedures followed by Darwin forecasters are also
described in the NT Severe Thunderstorm directive:
http://www.nt.bom.gov.au/ntregion/sevwx/storms/stsdir_contents.html

Interestingly, situations similar to the Broome one described by Baz are
occasionally observed in Darwin i.e. when the shear profile ahead of a
long-lived squall line is weak; strong middle-level rear inflow is observed
after the line which then weakens rapidly back to environmental values.
Sometimes large squall lines over the Top End are also observed at the
leading edge of a surge in the low to middle-level easterlies.

Forecasting squall lines (and associated severe wind squalls) for the
Darwin area is rather difficult as CAPE and shear profiles must be forecast
in the data-sparse upstream region where long-lived squall lines originate
and also in the coastal zone where sea breezes and local convection
complicate the environment.

 Ian Shepherd,
Senior Meteorologist, Severe Weather Section
Northern Territory Regional Office
Bureau of Meteorology
 

John McBride

Gedday all,

Here is the response by Ian Shepherd to my query on "Why Broome" for the severe thunbderstorms.

I have put the links to the papers by Lori Chappel and the NT Thuderstorm forecasting directive on my web page.

I have a few immediate  responses:

a) valid as Ian's point is, I had the impression from Barry Hanstrum's  note that Broome was actually a preferred location.  Perhaps Bazza could expand on this?

B) The Lori Chappel paper is extremely interesting  I have to give it some thought, but thanks very much for posting this.

c) I was surprised by the comment in the paper that all the CAPE program available give CAPE only for surface parcels.   I am surprised by this.  I wrote a CAPE program myself many years ago.
There are a lot of tricks:
*    you have to incoporate some iterative procedure in the equation for the moist adiabat;
*    to get the moist adiabat to agree with our F160's you have to make some of the coefficients (e.g L, cp) temperature dependent, following the Smithsonian table;
*    it is useful to incoporate virtual temperature, and options for switching to ice values of L at -10, and/or -20;
*    you need to muck around a bit to be able to feed in significant levels, and to take account of the fact that the LCL can fall anywhere (e.g in the current layer, the next layer, two or three layers up)
*    and so on...

Its one of these things that is  is conceptually straightforward ; but it probably took me several weeks to code.
However, my point is that once you go to the trouble of coding it up, it is essentially trivial to generalise it to be able to start the calculation from any level.

 Harald Richter is on this list.  I am sure with his severe storm connections, he has access to a suite of CAPE programs.

cheers

JMcB

Harald Richter

 A lonely voice in the convective wilderness typed:

 > c) I was surprised by the comment in the paper that all the CAPE
 > programn available give CAPE only for surface parcels.   I am
 > surprised by this.  I wrote a CAPE program myself many years ago.
 [snip]
 >  Harald Richter is on this list.  I am sure with his severe storm
 > connections, he has access to a suite of CAPE programs.
 

 Bait taken.  There are some widely used programs that compute
 different types of CAPE (RAOB, BUFKIT, NSHARP),  none of which I have worked
with
 (as I wrote my own toy program, just like JMB).
 I have (almost) no idea what these programs can/can't do in terms of CAPE,
 but I could find out if prodded suffiently.

 Cheers,  Harald

Matt Wheeler

To add to this, quite general Fortran code, with some documentation,
is available for download from Kerry Emanuel's web-site at
http://www-paoc.mit.edu/~emanuel/home.html
and go to "Fortran codes for Atmospheric Convection". This code
is designed to go with his text book called "Atmospheric Convection".

His code, called calcsound.f, calculates CAPE for parcels lifting
from any level (as well as many other things).

I downloaded it once, but I found that it was more
complicated/sophisticated than I needed, so wrote my own simpler
code instead - like JMB and HR.

Kerry is a smart cookie, so I'm sure his code works well.

-Matt.