Mark Lander (To Tropical Storms Mailing List)
TC Group,
It's that
time of year again for the monsoon trough to be at a low
latitude and produce TCs
in the Indonesian Seas. Late last year, Jeff
Callaghan posted a nice
listing of historical TCs in this area. There is
now another one for the
record: TC Bonnie, which has moved west across
Timor over the past 24 hours.
The attached image montage of IR, VIS,
and
QuikScat depicts the well-organized
tropical storm in the Savu Sea just
after local sunrise this
morning. The QuikScat has been quite useful for
diagnosing the location
and intensity of this TC.
A week
ago, there was a TC in the Northern Hemisphere that formed near
Chuuk and moved from there
NNE toward Wake Island. John McBride linked
the formation of that TC
to an equatorial Rossby Wave that was clearly
there in Matt Wheeler's
Equatorial modes decomposition of OLR prior to the
TC formation. The origin
of Bonnie, and another TC -- Dianne-Jery --
further to the west is not
so clear-cut. There has been no signature of
an equatorial Rossby wave
in this area for a while (at least in the OLR).
This idea of equatorial
N=1 Rossby waves as generators of TCs has worked
well so far for the few
early season TCs in the western North Pacific, but
with the two current Southern
Hemisphere TCs, it is not evident that this
is how they came to be.
Regards, Mark Lander
Matt Wheeler (the Boy Wonder)
Mark,
Thanks for your images and
discussion. You do a lot for the "community".
Just a small reply (below).
I just checked my web-page
to see what it was showing, and I don't agree
with your assessment. I've
attached two current plots, and to me, TC
Bonnie is right in an area
being "influenced" by both a westward propagating
n=1 ER (Eq. Rossby) wave,
and the enhanced phase of the eastward propagating
MJO. Note that the last
day of data that went into my monitoring and
prediction scheme in these
plots is the 8th of April, so you must look
at the 3 day forecast for
the current situation. Both the MJO and n=1 ER
wave are centred at around
10S, 120E in the 3-day forecast plot. Of
course, it doesn't always
work this well!
Plots: a) last six-months
Hovmoller of filtered OLR
anomalies 0ver 2.5-17.5 S, with MJO and n=1 Rossby signals superimposed.
b) Today's
representation of MJO, n=1 Rossby and Kelvin waves from filtered OLR
Regards,
-Matt.
Mark Lander
Matt,
I think
that with your equatorial modes product, we are getting closer
to seeing how TCs out here
in the monsoon region get their start. The n=1
Equatorial Rossby wave is
a good candidate for the background
disturbance that is associated
with many of our TC developments.
In this
e-mail. I would like to defend my earlier position a bit, and
ask you some general questions
that may help me to understand how these
equatorial modes really
work. I have attached a file
that shows a
Hovmoller type presentation
of the OLR for the period 04-10 April for two
sections: one from
EQ to 10 N, and the other from EQ to 10 S. In the
Northern section, the cloudiness
associated with our TD 04W can be seen in
the east and that convection
seems to stay put, or even move eastward.
There is no evidence in
the Northern Hemisphere that anything moves
westward. In the Southern
section, there is clearly a westward moving
cloud disturbance that evolves
into Bonnie as time goes on.
I have a few questions:
(1) Can an ER be confined to one hemisphere only?
(2) When you do your
mapping on your OLR modes web-page it seems to
dump energy into both hemispheres,
even though in this case, there was
absolutely nothing in the
norhern Hemisphere (at least in the cloud
field). What other modes
could produce the results of the attached
Hovmoller?
(3)
Do you think that it is possible that the earlier ER manifesting
as a small equatorial westerly
wind burst (That John McB pointed out last
week) initiated two vortexes
that went their separate ways in each
hemisphere (i.e., TD 04W
moved NNE, and the distubance that became Bonnie
started near 150 E and moved
westward)?
Matt Wheeler (BW)
On Fri, 12 Apr 2002, Mark A. Lander wrote:
> Matt,
>
>
I think that with your equatorial modes product, we are getting closer
> to seeing how TCs out
here in the monsoon region get their start. The n=1
> Equatorial Rossby wave
is a good candidate for the background
> disturbance that is associated
with many of our TC developments.
>
In this e-mail. I would like to defend my earlier position a bit, and
> ask you some general questions
that may help me to understand how these
> equatorial modes really
work. I have attached a file that shows a
> Hovmoller type presentation
of the OLR for the period 04-10 April for two
> sections: one from
EQ to 10 N, and the other from EQ to 10 S. In the
> Northern section, the
cloudiness associated with our TD 04W can be seen in
> the east and that convection
seems to stay put, or even move eastward.
> There is no evidence in
the Northern Hemisphere that anything moves
> westward. In the
Southern section, there is clearly a westward moving
> cloud disturbance that
evolves into Bonnie as time goes on.
Yes, I see your point. I
also often have difficulty with the way these
large scale waves in the
convection often disappear then reappear.
Yet when you take an even
more large-scale view than your plots, and look
at the OLR from as far back
in time as the 20th of March in these two
separate north-of-the-equator
and south-of-the-equator bands, there does
appear to be something consistent
about the convection on the very large
scale. For my latest OLR
time-longitude plots in these two bands, see
http://www.bom.gov.au/bmrc/clfor/cfstaff/matw/maproom/OLR_modes/h.6.ALL.N.html
and
http://www.bom.gov.au/bmrc/clfor/cfstaff/matw/maproom/OLR_modes/h.6.ALL.S.html
and on these plots the black
sloping contours representing the n=1 ER waves
are by definition the same
in each hemisphere.
Can't you see, on the large
scale, the westward propagation in both
hemispheres that started
at around 160W on the 20th of March?
Of course, this large-scale
westward propagation is modulated by things
(like you TD 04W in the
NH) that move eastward and elsewhere, just in the
same way as on an even larger
scale the eastward propagating envelope
of convection of the MJO
contains a lot of smaller and shorter time scale
variability. And amazingly,
a lot of this appears to be linear! That is,
you sum up the MJO, n=1
ER waves, TD-disturbances, and the TCs, and you
get most of the total field!
Maybe this linear superposition of different
disturbances is wrong, but
it often appear to work! Now to answer your
questions....
>
>
I have a few questions:
>
(1) Can an ER be confined to one hemisphere only?
No, by definition (and theory),
an n=1 ER wave is symmetric about the equator,
and should be in both hemispheres.
If, however, you superimpose another
disturbance that is in one
hemisphere only, then the total field may give
the impression of an ER
wave in one hemisphere only. Don't you think this
is feasible?
>
(2) When you do your mapping on your OLR modes web-page it seems to
> dump energy into both
hemispheres, even though in this case, there was
> absolutely nothing in
the norhern Hemisphere (at least in the cloud
> field). What other modes
could produce the results of the attached
> Hovmoller?
Yes, my "mapping" of the
n=1 ER waves is defined to be symmetric about the
equator in the OLR field,
and that is why you see it in both hemispheres.
As for what mode could be
added to the n=1 ER wave to give the current
appearance in the total
field, I can only speculate. A n=0 MRG wave
would be a good candidate,
as it is antisymmetric (in OLR and u), as
would an eastward n=0 inertio-gravity
wave. Of course, with complex
basic states, these can
be modified a lot from their usual structure from
the linear theory.
>
(3) Do you think that it is possible that the earlier ER manifesting
> as a small equatorial
westerly wind burst (That John McB pointed out last
> week) initiated two vortexes
that went their separate ways in each
> hemisphere (i.e., TD 04W
moved NNE, and the distubance that became Bonnie
> started near 150 E and
moved westward)?
Yes, I think that is possible.
Of course, the TD and TC are operating on
smaller scales, so have
their own dynamics. The ER wave is on a larger scale.
I think of the ER wave as
a large-scale "modulating" factor on the smaller
scale disturbances, just
in the same way that other people have found the
MJO to be a "modulating"
factor on TCs elsewhere.
....and John McBride tells me that he is brewing up a reply himself.....
-Matt.
Paul
Roundy (Pennsylvania State University)
From: Paul Roundy <proun@essc.psu.edu>
To: m.wheeler@bom.gov.au
Cc: tropical-storms@tstorms.org,
Oz discussion group started by JMB. <synoptic_discussion@bom.gov.au>
Subject: Re: [Tropical-storms]
TC in the Indonesian Seas
I have been following the
equatorial modes discussion with great interest,
and am pleased that knowledge
of this topic is finally spilling over into
forecast applications!
See notes below.
>
> No, by definition (and
theory), an n=1 ER wave is symmetric about the equator,
> and should be in both
hemispheres. If, however, you superimpose another
> disturbance that is in
one hemisphere only, then the total field may give
> the impression of an ER
wave in one hemisphere only. Don't you think this
> is feasible?
>
This is true, but in practice
one must be careful applying the theoretical
forms to OLR signatures
we see in the atmosphere. Matt discussed my
concerns somewhat in his
reply, but I though it would be useful to
comment on them a little
more. The theoretical wave
forms are obtained from
equatorial beta-plane shallow-water theory. The
theoretical cross-equatorial
symmetry patterns apply to divergence
patterns associated with
the wave itself. The convection is modulated by
the divergence patterns
associated with the wave and by the base state
thermodynamics. This
means that if an n=1 ER wave propagates into a
region with relatively high
SSTs north of the equator and low SSTs south
of the equator, the convection
anomaly north of the equator will be more
significant than the anomaly
that appears south of the equator. Sometimes,
convection associated with
such a wave may be entirely suppressed in one
hemisphere. This raises
interesting questions about the how heating from
the convection in the active
hemisphere modifies the wave structure.
Several studies have already
discussed how convection associated with
mixed-Rossby gravity waves
may occasionally cause the circulations that
the waves induce to refract
off the equator to behave more like TD-type
disturbances. As another
example, Kelvin waves that propagate into the
Pacific basin often induce
convection anomalies that are centered near the
ITCZ, instead of on the
equator where maximum divergence anomalies are
expected to be found.
