Hello,

In trying to track down the double warm line on either side of the equator in SST's (pointed out by Mark Lander), I browsed  a few web sites and was looking at zonal winds as measured by theTAO-TRITON  array (http://www.pmel.noaa.gov/tao/jsdisplay/).  Have a look at the plot on my web-page for zonal wind and zonal wind anomalies along the equator over the past 5 months.  There is a terrific eastward moving zonal wind anomaly that has occurred over the past couple of weeks.  It did not stand out on the day-to-day charts as the actual winds were weak, around zero... but on the time longitude plot of anomalies, it stands out like the proverbial cats' ears.

I have put up another plot of it over the region from the dateline to 140 West, using the NOAA CDC site (http://www.cdc.noaa.gov/map/clim/glbcir.shtml); and beside that I have put the last six months Boy-Wonder OLR diagnostics for the 7.5 N to 7.5 South equatorial belt.  First look at the u-wind anomaly plot.  The eastward travelling anomaly  (almost certainly an atmospheric equatorially-trapped kelvin wave) is located at 60E on the 30 April, and has shot across to the dateline on 15 May.  Now look at Matt the Boy Wonder's OLR anomaly plot.  His Kelvin wave filter is in green.  See the Kelvin wave at 60E on 30 April and at 180 E on 15 May.

I was asked three days ago to give a talk on conceptual models for the tropics to a WMO sponsored course titled APSATS 2000 (Asia Pacific Satellite Applications and Training Seminar).   I spent a lot of time explaining the importance of upper-tropospheric warm cored structure for tropical cyclones, monsoon troughs and MJO's and explained that anything worth a damn in the tropics has a warm core in the upper troposphere.  Consistent with that, lets look at our Kelvin wave:  Also on my webpage is the Hovmoller of 200 hPa temperature for the past 2 months.  See the nice warm area once again at 60E on 30 April and at 180 E on 5 May.

To make sure it is a Kelvin wave, I also looked at the meridional wind anomaly; and there is no clear feature (which fits).  I also looked at the Hovmollers of sea surface temperature and also there is no equivalent feature.  Presumably this is because the Kelvin wave moved across so fast  the upper layers of the ocean did not have time to respond.   The phase speed is 120 degrees longitude in 15 days, which corresponds to about 10 m/sec; this is way too slow for a dry Kelvin-wave; so there must be some moist Kelvin wave dynamics involved.

Looking at Matt's OLR diagnostics, the Kelvin wave was clearly tied in with the recent MJO episode.  How it related to the other phenomena we have been discussing, e.g. the series of twin vortices, I don't know yet.  Its Friday afternoon here and I have an important appointment at the pub.

Cheers

John McB