Figure 1 (recent OLRA plots: link for equator ;  link for north of eq. ;  most recent AAM plot)
 

The colored horizontal lines on the left panel match the times of the vertical lines on the right panel.  These times are the last week in February 2005, denoted by the brown lines, around 21 April 2005, orange lines, and at roughly 7 May 2005, light blue lines.  A general relationship to observe from Fig. 1 is that as enhanced tropical convection moves east from the regions of the Indian Ocean (IO) and Indonesia into the area of the equatorial date line (vertical dashed brown line on the left panel), there is an increase of AAM meaning more westerly flow in the atmosphere (latitudes affected shown by "W" on the top right panel).  During the CP coupling, there was a large increase of AAM starting with strong zonal mean westerly flow across the tropics, which then propagated to about 30 deg N and S.  The global AAM anomaly with this coupling was 3-4 standard deviations above the 1968-97 climatology.  As MJO #1 developed, the westerly anomalies were replaced by subtropical easterlies ("E") with a return to weak subtropical westerly anomalies afterwards (red line).  Westerly anomalies returned to the equatorial latitudes by about mid-May more linked to KW activity than to MJO#2. These have rapidly weakened recently.

Some points to take from Fig. 1: 1) AAM averaged from around 1 February to late May 2005 has been above climatology, 2) since late January 2005, the easterly (westerly) anomalies that were at ~40N (25N) during fall 2004 shifted to 55N (30N), thereby initiating a wintertime El Nino circulation state, 3) since the CP coupling the largest OLR anomalies have apparently been shifting westward toward the Indian Ocean. This is partially a seasonal trend.

On 10 June 2005 MJO #3 appears to be getting more defined at about 10N/100E, a location typical for an MJO in early June.  For the latest satellite imagery see: Latest Indian Ocean Satellite Picture here ;  Latest Western Pacific Satellite Picture  here  ; Latest GOES West Satellite Picture  here ; Lastest GOES East Satellite here .

Figure 2 (below) shows a four-panel set of 150mb vector wind anomalies with the top left for the period from February 1-May 27, 2005, when AAM was above average.  The bottom left and top right are for two 7-day averages 17-23 May and 24-30 May 2005, when enhanced tropical convective forcing moved from the western to the eastern hemisphere.  The bottom right panel is for the 7-day mean centered on 6 June, 2005.

Figure 2

The pattern shown in the top left is intermediate between Stages 3 and 4 of the subseasonal Synoptic-Dynamic Model (SDM; see Fig. 3 where the same plot is shown next to the SDM).  A large contribution to the anomaly comes from February 2005 when the atmosphere was in Stage 3.  There has been general persistence of the atmosphere in Stage 3-4 since February 2005.  This follows northern fall 2004 when a Stage 4-1 pattern was preferred.  The resumption of MJO activity after February did allow for periodic AAM fluxes out of the tropics, leading to subtropical jets (STJ) in both hemispheres and supporting a "Stage 3-4" component to the circulation.  Notice also the anomalous northerly flow across the USA during this averaging period.  This reflects numerous outbreaks of cold air that penetrated well into the Gulf of Mexico and deep tropics, which suppressed the severe local storm season on the USA Plains during this time.

The circulation during 17-23 May 2005 (bottom left) has similarities to what is shown in the top left.  During this time tropical convection was most intense across the western hemisphere as twin subtropical anticyclones were quite strong around 60W.  Strong westerly flow continued over the North Pacific although it was further north compared to the 4 month mean.

By 24-30 May 2005 (Fig. 2 top right) a major change in the circulation had occurred.  Related to the development of MJO #3 enhanced tropical convective forcing shifted from the western into the eastern hemisphere (see Fig. 1).  Westerly anomalies, particularly over Africa and the Indian Ocean collapsed, and AAM reached a low point in late May 2005.  Twin anticyclones (instead of cyclones) were now observed around 30E and 100E, tied to the convective outflow across Africa and the Indian Ocean.  One response to this change was for the North Pacific jet was to contract (seasonal cycle also contributed) and allow the trough previously across the eastern USA to retrograde to the Rockies.  More recently, 3-9 June (Fig. 2 bottom right), twin anticyclones directly forced by MJO #3 have become more prominent around 120E with meridional outflow between them especially into the southern hemisphere.  Across the southern hemisphere, Rossby wave dispersion from the tropical convection was leading to symmetry between the hemispheres, including a deep trough just southwest of Chile and a trough across the western USA.  This pattern represents a transition from Stage 4 to 1 (weakeninsg subtropical jet, La Nina-like circulation anomalies), and will be an important consideration for the predictive insights given in Section 2.

2.   Predictive Insights

Our previous forecast for the period of 19 April-2 May suggested SDM Stage 3 would be most probable, followed by Stage 4 during week three, 3-9 May.  This was too slow since the atmosphere was already in Stage 1 of the next MJO by 9 May.  The forecast argued for continued split flow across much of North America, with a southerly displaced storm track.  Additionally, we felt there would be an anomalous ridge from western Canada back into Alaska and trough along the east coast of the USA.  The forecast was for below normal temperatures to shift out of the western states into the eastern half of country, with above normal rainfall most likely across the southeast half, and perhaps the states of Oregon and California.  Finally, possible hazards included severe local storms across much of the Deep South and late season snowfall across the Rockies (including the Front Range) and far western High Plains.  Please see the previous discussion for details.

In general, this prediction did reasonably well.  Toward the end of April there were at least one or two occurrences of severe storms across portions of the southeast and during this period there were several episodes of significant snowfall over much of the central and southern Rockies.  The latter included at least one event along the Front Range of Colorado.  Finally, much of the West Coast from northern California to Washington did start to get much needed precipitation.  The temperature forecast did very well.  For example, during the first few days of May a large portion of the Plains and Great Lakes states had record low minimum temperatures, including late season freezes.

For about the past 2 weeks the circulation of the atmosphere has behaved as would be expected from SDM Stage 4, and is believed to have transitioned from Stage 4 to Stage 1 as of 14 June 2005 (link to latest 30-day animation).  The top left panel of Fig. 3 depicts the most recent daily mean of 150mb vector wind anomalies.  Twin anticyclones linked to the divergent outflow are located at about 120E and are embedded in a complicated wave field.  An apparent Rossby wavetrain can be seen linked to the cyclonic wind anomaly across the western USA.  In addition, a well defined STJ can be seen interacting with the western states trough.  Several rounds of severe local storms have already occurred across the Plains due to this transition.

Figure 3  (Most recent 150mb daily mean vector wind anomaly)

Currently satellite imagery shows MJO #3 centered around 10N/100E.  This MJO has been strengthening during the past several days while interacting with a convectively coupled Rossby mode, with OLRA ~ -50 to -70 W/M**2.  The MJO have been nearly stationary for about the past week, and its future movement is in doubt adding uncertainty to the outlook for weeks 1-3.  Independent flare-ups of tropical convection over the relatively warm SSTs near the date line (anomalies ~0.5-1.0 deg. C; actual SSTs ~29-31 deg. C) and the Indian Ocean also add uncertainty.

Since the MJO signal is weak, statistical and numerical models of the MJO are inconclusive (see latest MJO forecasts ).  However, the coherent modes Hovmollers do suggest the MJO envelope will shift across the northern tropics into the western Pacific (~160E) by the end of week 2 (link to Coherent Modes Hovmollers).  Various global forecast models are also predicting MJO #3 to move into the western Pacific by the end of week 2 or sometime week 3.  Additional MJO forecasts and tools can be found  here.  Close weather-climate monitoring suggests the MJO is now nearly stationary, while separate convective flare-ups continue to occur west of the date line.  Daily monitoring will be needed to improve predictions, especially of high impact weather.

Week 1 (15-21 June 2005):  The circulation has transitioned into Stage 1, as was expected from weather-climate monitoring.  Numerous rounds of severe local storms (including tornadoes) occurred across the central USA ( link to SPC), with rain and higher elevation snowfall occurs across the Rockies, during the past week.  Meanwhile, tropical cyclone Arlene moved ashore from the Gulf of Mexico last weekend (11-12 June) and continued northward into the Ohio Valley (please see latest NCEP/TPC and HPC advisories and storm summaries here ).

Toward the middle-end of this period an anomalously deep trough is expected to develop along the USA west coast as part of Rossby wave dispersion across the North Pacific.  That system should bring more needed precipitation for the Pacific Northwest, and perhaps a concern for severe local storms across the northern Rockies west of the Continental Divide.  As shown by most ensembles and seems reasonable, a large amplitude ridge is expected develop across the central USA with a down stream trough across the east.  Temperatures are likely to turn cooler to below climatology once again across much of the east, especially the Great Lakes and Northeast States.  The western High Plains and Rockies should warm to above normal.  By the end of this period excessive heat and humidity may surge northeast across the Northern Plains into at least the Upper Mississippi Valley.  The Pacific Northwest is expected to stay cooler than normal.  Finally, nocturnal clusters of thunderstorms are expected across portions of the Front Range and High Plains (with the possibility of severe storms), and locally heavy rainfall may occur across the Gulf coast (please monitor TPC for tropical cyclone development that could affect the Gulf coast).

Week 2 (22-28 June 2005):  The atmosphere is expected to remain in Stage 1.  Hence the trough that develops along the west coast week 1 should progress slowly into the Northern Rockies by the end of this period, while perhaps extending southwestward  to just off the California coast.  Meanwhile, what could be a large amplitude ridge initially across the Plains would be expected to move into the eastern half of country, while a residual trough lingers across the southeast and south central states (latter feature may retrograde).  Whatever the details, the strongest westerlies are expected be across the northern USA (as climatology would also suggest).  Particularly during the mid-latter half of this period, severe local storms and mesoscale convective complexes/clusters may once again be a hazard for especially the central and northern Plains.  While temperatures return to below normal across the northern Rockies and northwest states, excessive heat and humidity is a possibility for much of the Plains and eastern USA.  Temperatures are expected to return to seasonable along the east coast by the end of this outlook period.

Week 3 (29 June- 5 July 2005):  Should the MJO coherently propagate into the western Pacific, then a transition to a summer time version of Stage 2, similar to what was observed at times during summer 2004 (see August 16, 2004 discussion), would be most probable.  This would mean an anomalously deep trough setting up across the Rockies and western High Plains, while a ridge amplifies off the west coast into Alaska.  The southeastern part of the country would also be under the influence of a ridge.  The westerlies would then come anomalously southward across the country.  If this indeed does happen, that would suggest cooler than normal temperatures and quite stormy (including several rounds of severe local storms and meso-scale convective systems with heavy rainfall) for the Northern Rockies and much of the central USA into the Great Lakes and Ohio Valley.  The mid-Atlantic and southeastern states would be warmer and dryer than normal, perhaps with excessive heat, while locations along the west coast also warm.  If the MJO remains stationary, the weather described above for week 2 may be more likely.
 

 Latest CDC Ensemble Forecast

 Latest NCEP Ensemble Forecast

Additional NCEP Ensemble output

Latest Canadian Ensemble Output