Edward Berry, NWS and Klaus Weickmann, CDC

1. Weather-Climate Overview

The purpose of this document is to update the real-time discussion issued on 30 September 2004.  A more complete discussion will be posted in about 2 to 3 weeks.  Ideally, short updates such as this one will allow the readers to maintain a time continuity of weather-climate events.  Reference will be made to the 30 September discussion available here .

On 30 September 2004 the convection anomaly was north of the equator near 100E.  Since then the convectively active phase of the MJO has progressed to about 0/160E.  This represents an eastward shift of ~60 degrees and a southward displacement of ~10 degrees.  The latter behavior may be a response to the relatively warm SSTs along the equator west of the dateline (~30C; 1.5C anomaly).  A westerly wind event is accompanying this MJO with strongest anomalies now north of the equator, as seen in Figure 1.

Figure 2 is a Hovmoller of outgoing longwave radiation (OLR) anomalies (shading) and space-time filtered OLR (contours).   The space-time filter utilized in Fig. 2 does not depict an MJO, partly because the eastward shift was very rapid in association with a Kelvin wave (green contour).  An alternate index (shown  here ) has a weak to moderate MJO signal currently moving into the western Pacific.  The main point is that convective forcing exists at around 160E and has persisted for about a week.

Figure 3 shows 3-day average maps of OLR and OLRA for the period of 9-11 October 2004.  The bottom panel illustrates the coherent negative OLRA centered at the equator, 160E.  A comparison of Fig. 3 with statistical predictions shown in the last discussion (Figs. 15 and 16) indicates convection is more consolidated and further east than predicted.  Also, there are positive OLRA over most of the Indian Ocean and Indonesia instead of the predicted lingering negative anomalies.

2. Predictive Insights

As stated in the last discussion,

"Should the convection develop more rapidly across the western Pacific (~ 150E), the atmosphere may go to stage 2 of the Synoptic Dynamic Model (SDM) even before the end of week 2".

This evolution appears to be underway.  The convection with the "MJO" is at ~160E, and a trough has developed downstream in the northern extratropics just east of the dateline.  A ridge has been amplifying just off the USA west coast and a downstream trough is starting to dig around 85-90W.  This is about 10-15 degrees farther east than predicted in the previous discussion.  The result is a positive phase of the Pacific North American teleconnection pattern (PNA) and a first step toward stage 2 of the Synoptic Dynamic Model (SDM).

Satellite imagery already shows evidence of the MJO convection "splitting", with a portion shifting northwest toward southeast Asia (with 1 or 2 embedded tropical cyclones) and the other moving along the South Pacific Convergence Zone.  As this process continues, it is expected (based on linear modeling studies) that the above evolving positive PNA's trough-ridge-trough structure will retrograde by roughly 20 degrees, leading to the positive phase of the western Pacific wavetrain by around 19 October (day 7).  The wavetrain is depicted here  in Stage 2 of the SDM.

Operational models such as the NCEP and CDC ensembles are predicting this behavior.  For the USA, this will mean a significant precipitation event for at least the eastern half of the country, as well as colder than normal temperatures by this weekend (15-17 October).  Afterwards, particularly for week 2, the northern and western portion of the USA should cool while the east warms.  The best opportunities for precipitation should also shift back into the Pacific Northwest, Rockies and Plains.

There are several uncertainties concerning the upcoming evolution of atmospheric circulation anomalies over the hemisphere.  From the broad perspective of the seasonal cycle and the weak-moderate El Nino, the persistent circulation anomalies associated with El Nino should develop abruptly some time during the seasonal transition (mid-December to mid-January 2005) in association with an MJO moving into the west-central Pacific Ocean region.  Until then, there are several possible scenarios to consider.  An immediate concern is the duration of the positive PNA currently in place.  Typical decay time scales for the PNA are in the range 7-10 days.  It thus seems reasonable to expect Stage 2 of the SDM to develop at the end of this time frame as the PNA weakens.  The next question is whether convection will establish itself over the El Nino SST anomalies and help excite Stage 3 of the SDM (e.g., strong west coast storms).  The seasonal cycle (i.e., time of year) may argue against this since warm Pacific SSTs (e.g., >29C) do not yet excite persistent convection anomalies.  Alternatively, the atmosphere may quickly go into Stage 4 (e.g., strong subtropical jets in the western hemisphere) where convection of moderate amplitude occurs in three regions, the dateline, South America and the west Indian Ocean.  At this point, there is little insight about when the next Indian Ocean convection may develop.  The last two flare-ups there occurred only 25 days apart (e.g., 8/30/04 and 9/22/04).