Re the current MJO, twin sets of twin vortices thread, I think there may be a third pair forming at about 150 E  (N Hem) and 160 E (Southern Hem).  I am in the office for only about 15 minutes today; so won't dwell on them.  I have put the latest GMS image, 850 hPa analysis and 900 vorticity analysis on my web page

cheers

JMcB

Mark Lander

Hi John, TC group,

    This current MJO and its spawn of twin TCs (now up to quads or maybe
even sextuplets -- Gary Padget suggested this nomenclature), is a weather
event worth recording for posterity.

    I have enclosed a pic at a time when the twin cyclone pattern was at
its most impressive (09 UTC 09 May).  The other image involves a little
trickery:  the left panel shows an image in which the southern hemishpere
is just a mirror image of the northern hemisphere, and it is hard at first
glance to see the differences between this manipulated image and the real
thing in the right panel.

    Now we wait for the western Pacific to get into the act.

    Regards,  Mark Lander

George Kiladis

Date: Fri, 10 May 2002 16:59:20 -0600 (MDT)
From: George N. Kiladis <gkiladis@al.noaa.gov>
To: tropical-storms@tstorms.org
Subject: [Tropical-storms] Eq Rossby dispersion

John and TC'ers,

just a quick word on the apparent vs. theoretical dispersion of
equatorial Rossby modes:

We've always been puzzled by the fact that in certain observations
(such as Kiladis and Wheeler, 1995 JGR pg. 22,981) we see a mean
westward phase vel. of 7 m/s and eastward group of 2.5 m/s in
Pacific ER modes (zonal wave 6'ish), but that the implied dispersion
relationship for the "ER region" of OLR space-time implies eastward
dispersion out to at least wavenumber -10 or so (see the spectrum at the
top right of Matt's monitoring page at:

http://www.bom.gov.au/bmrc/clfor/cfstaff/matw/maproom/OLR_modes/index.html

By the way, eastward group of this past March's ER packet in the w. Pac.
seems quite evident to me in the time-longitude plots for the last 6 months
on Matt's page, pointers are just below the spectra plots.

One possibility is that ER modes morph themselves quite regularly into
smaller off-equatorial Rossby gyres, as we've just seen in the Indian Ocean,
TDs that are of the right scale to have eastward group, but that still doesn't
explain the space-time OLR signal in the spectra.  However, the spectral
ER peak does grade into a region starting at around wave -6 into something
that does have power tilting the other way in space-time, i.e. with eastward
group.

Done for the day, I'll leave it to you guys to comment further next week!

George Kiladis

Paul Roundy
From: Paul Roundy <proun@essc.psu.edu>
To: George N. Kiladis <gkiladis@al.noaa.gov>
Cc: tropical-storms@tstorms.org
Hi all. . .  I thought this would be an appropraite time to jump into this
discussion.  It was good to see several of you at the tropical conference.
. .

See notes below.

        Paul E. Roundy                proun@essc.psu.edu
        Department of Meteorology     Http://barks.met.psu.edu/~proun/
        The Pennsylvania State University

I have been pondering this dispersion question for many months, and like George,  I have made the
obvious conclusion that the apparent dispersion that we see in convection is not the same as the
dynamical dispersion that we see in ideal shallow-water equatorial wave models.  My take on the
processes is that convection feedbacks to wave dynamics favor those portions of wave groups that
are located in regions where convection is enhanced by other large-scale processes.  Eastward and
westward-propagating disturbances interact to produce the observed 'dispersion' behavior.  I attached a
poor quality (appologies) longitude-time image of OLR for 10 degrees north that illustrates my
point. Blue regions represent active convection.  The contours represent my MJOw filtered OLR (wavenumbers 1-6 westward and periods of 15-104 days), and shading represents my MJOe filtered OLR (wavenumbers 0-6
eastward and periods of 30-104 days).  The MJOw filter captures most of the Wheeler and Kiladis ER n=1 filter region (except on the high frequency end), but includes a broader portion of the spectrum at lower frequencies.
The MJOe filter captures the eastward-propagating intraseasonal oscillations.  The contoured groups of westward-propagating waves are usually associated with an eastward-propagating MJO event, and the
apparent group velocities of the westward-propagating waves are usually similar to the phase speed of the intraseasonal oscillations.

 
 
 

The Rossby waves amplify or weaken in regions of amplified or supressed convection associated with the MJO and similar eastward propagating intraseasonal disturbances.  It follows that though the Rossby waves may have a dynamical westward group velocity, the anomalies that propagate westward into regions of supressed convection fail to amplify,
while those anomalies to the east that would have decayed in a 'dry' wave group actually amplify because of the large-scale state of enhanced convection that the waves experience there.

The group propagation patterns in the Wheeler and Kiladis ER n=1 filter do not match these MJO patterns as clearly as the MJOw filter result does, probably because their ER n=1 filter is not broad enough.

This process is probably also involved in producing an apparent westward group velocity associated with the MJO, but this should be the topic of another note. . .

Thanks. . .

Paul