Saturday, January 26, 2008

El-Veijo Getting Juiced

There is relatively little change in the spatial distribution of global tropical SSTs (I am not forgetting the Atlantic Ocean) since the 17 January posting. Mature basin-wide cold event SST conditions prevail across the Indo-Pacific sector, with anomaly magnitudes as low as ~minus 3C (totals ~24C) around 0/150-160W per TAO array buoy data. Tied to moist tropical forcing returning to the Indian Ocean (discussed below), a recent trade wind surge has led to renewed cooling along the equatorial central and eastern Pacific Ocean. The cold anomalies extend to roughly 150-200m deep, with magnitudes ~minus 5C (100m) at 120W.

Warmer than normal waters having similar anomalies as their La-Nina counterpart prevail generally along and south of the equator from the Indian Ocean-northwest Australia. SST totals in this region are ~29-30C. There is still the well-pronounced warm SST anomaly horseshoe extending from the far equatorial west Pacific Ocean into especially the Northern Hemisphere. Finally, responding to a recent westerly wind burst south of the equator, positive anomalies (5-day mean) in excess of 4C have developed ~200m at 165E. Together with the cold anomalies farther east, this is a strong signal of a steeper than normal equatorial oceanic thermocline, typical of La-Nina. Whether or not a coherent oceanic Kelvin wave gets (or already is) generated is unclear.


Full-disk satellite imagery and other diagnostic monitoring tools indicate renewed strong-severe tropical convective forcing from South Africa-central equatorial Indian Ocean (as expected per past discussions). Sporadic thunderstorm activity has been recently occurring across Indonesia while a remnant cluster attempts to develop into a South Pacific Ocean tropical cyclone. WH(2004) phase space plots support the notion that the strong MJO signal since early-mid December 2007 has returned to the Indian Ocean. Again, it is dynamical interactions with the extratropics such as Rossby wave energy dispersion (RWD) processes that allow this kind of behavior to occur. Model predictive schemes of the MJO utilizing the WH(2004) methodology were generally mixed about the signal returning to the Indian Ocean.


MJO tropical variability thus far has been the strongest since the boreal 2004-05 cold season, and this is unusual for a strong La-Nina such as the one currently in progress. Whether or not the start of MJO#3 for the boreal 2007-08 is in progress remains unclear. However, the west central-South Pacific Ocean tropical forcing finally weakening suggests that it is, as do some other tools.


As was the case for MJOs #1-2 constructively and destructively interfering with our La-Nina base state, the redevelopment of Indian Ocean tropical forcing is probable to enhance/amplify it. Keep in mind that dynamical processes in the atmosphere cannot be “shoe-horned” into simple "recipes". For example, it is not accurate to attribute the intense USA west coast winter storm ~3-4 January 2008 simply to the MJO. In reality the moist MJO tropical convective forcing centered on Indonesia amplified an already existing La-Nina base state allowing a cold trough to slam the west coast. There is no short cut to understanding the dynamics of physical atmospheric processes involving forcing-response-feedback-subsequent interactions impacting multiple time and space scales. The latter is important if we are to improve to making predictions of especially, for instance, high impact weather events weeks 1-3. Numerical models will struggle when there are rapid and complex changes involving tropical convective forcing with subsequent extratropical interactions, no matter how much bias correction is employed.


Since early-mid December 2007, responding to MJO#2, well defined poleward propagation of zonal mean easterly wind flow anomalies occurred from the equatorial into the subtropical/mid-latitude atmospheres of both hemispheres. Currently there are westerly wind flow anomalies shifting off the equator into the subtropics, particularly the Northern Hemisphere (~10m/s at 200mb). That is why there are several low-latitude closed lows with subtropical jets, globally.


With the added subtropical westerly wind flow from the last MJO, global relative AAM continues to hang around ~minus 1 standard deviation below the R1 data climatology through 23 January. Also helping to keep global AAM from plunging like the financial markets have done so far this year is a ~10-15 Hadley positive frictional torque. The latter is a direct response to the poleward and downward propagation of the zonal mean easterly wind flow anomalies discussed above, especially to around 20-40N (weaker ~15-30S).


A strong zonal mean poleward AAM transport signal has appeared ~50N while weakly equatorward ~20N, meaning midlatitude split flows that have regionally been across the oceans and roughly central Asia. Finally, also consistent is a strong positive zonal mean earth AAM signal of ~1 AMU centered ~45N suggesting a lot of surface high pressure across the Northern Hemisphere midlatitudes (weaker Southern Hemisphere). The point is that in the presence of the strong MJO variability, typical of a mature La-Nina, a loose equilibrium now exists between processes wanting to remove and add angular momentum to the atmosphere that are meridionally symmetric about the equator.


Also updated through 23 January, the GWO signal has orbited weakly to phases 4-5 in the daily mean. I suspect the orbit will shift a bit toward the left into the region of phase 3 (old GSDM Stage 1 – “La-Nina attractor”) during the next week or so. In fact, supportive of this notion, animations of upper tropospheric daily mean vector wind anomalies from ESRL/PSD already show twin tropical anticyclones developing across the Indian Ocean with cyclones shifting back toward the date line. Zonally and meridionally oriented RWDs forced by both the GWO (which considers the MJO) and MJO respectively have been leading to the expected retrogression (3-week lead time, before the numerical models caught on) of the midlatitude ridge back into the central Pacific Ocean.


Summarizing, it is still unclear whether or not MJO #3 is developing across the South Indian Ocean. However, given the warm SSTs it is probable to see the enhanced tropical forcing shift eastward into the region Indonesia-Australia (~80-140E south of the equator) by roughly week-2. Perhaps this will be another MJO with several “stops along the way (coupling with SSTs, etc.)” similar to the last one. El-Viejo is probable to get pumped once again, possibly similar to late December 2007 into earlier this month. Most numerical ensemble prediction schemes now show a PNA response typical of GWO/MJO phases 3-4 (GSDM Stage 1); that is, central Pacific Ocean ridge, trough from Alaska (good block never did develop per past posts) to the USA west coast-deep southeast states ridge. I speculate this whole pattern will drift east weeks 3-4 especially if there is another MJO (timing is always white noise). Weather ramifications of this pattern for the lower 48 states should be apparent by now (see 17 January discussion).


In the longer term, poleward shifting zonal mean easterly wind flow anomalies do increase the probability of anomalous midlatitude ridges. Where these ridges become established obviously has a lot to do where drought may occur say, for this upcoming boreal summer. Right now, that is unclear. Hence statements about possible drought centered on Iowa this upcoming warm season are very much premature. In fact, I can easily speculate an opposite scenario.


Internationally, a significant tropical cyclone risk exists from Madagascar into the South Indian Ocean week 1 likely shifting to the north coast of Australia weeks 2-3. The west central-South Pacific Ocean is going to be a bit of a “wild card” from now on (not discussed above) due to warm SSTs and other issues (stay tuned). One or two tropical cyclones may impact the paradise islands week 1 with hopefully some suppression weeks 2-3. The severe flooding rainfall events for South Africa should slowly diminish during the next couple of weeks. At least diurnal/frontal seasonal thunderstorm activity is probable week 1 for tropical South America including Brasil, with possible suppression starting weeks 2-3 should there be another MJO.


Appendix

An experimental quasi-phase space plot of the GSDM utilizing time series of normalized global relative AAM time tendency (Y-axis) and normalized global relative AAM anomaly (X-axis) can be found at

http://www.cdc.noaa.gov/map/clim/gsdm.shtml

We call the behavior of this plot the Global Wind Oscillation (GWO). While the intent of the legacy GSDM is to extend current thinking beyond the MJO, the GWO quantifies variations used to derive the original GSDM in a manner that is “user friendly” analogous t0 WH(2004) “convention”. In addition, the GWO plot does not have the ENSO signal removed.

Links to CPC and PSD ENSO discussions:

http://www.cpc.ncep.noaa.gov/products/analysis_monitoring/enso_advisory/index.shtml

http://www.cdc.noaa.gov/people/klaus.wolter/MEI/

These are probabilistic statements, and work is ongoing to quantify in future posts (for example, risk assessment maps, signal to noise ratio plots and shifts of probability). We hope that an opportunity will arise for us (soon) to have a dedicated web page effort to expedite more objectively, with rigor, thoroughness and verification. The WB (2007) paper on the GSDM has been published in the February issue of MWR. Work is on-going by Weickmann and Berry to submit a paper that will formally introduce the GWO. I will attempt another posting the weekend of 2-3 February.


Ed Berry

Thursday, January 17, 2008

Another Injection due for La-Nina; Update

The spatial pattern of global tropical SSTs is generally unchanged from at least a week ago. Above normal waters remain across the Indian Ocean while negative anomalies, including those with La-Nina, extend from north of Australia into the equatorial Pacific to the coast of South America. Anomaly magnitudes are around 1-3C, with the warmest across the South Indian Ocean (totals ~30C) and coolest in the region of the equatorial date line to 120W. Latest TAO array buoy data shows that the colder than normal equatorial Pacific SSTs are deep, extending to around 200m. A pronounced SST horseshoe of warmth extends into the extratropics from the west Pacific, with ~2C magnitudes around 30N/180. The bottom line is that El-Viejo ocean conditions are well pronounced and mature. Several of the latest dynamical and statistical models suggest La-Nina may continue most of this year.

Full disk satellite imagery and other monitoring tools indicate the strongest moist tropical convective forcing is in the region of the date line/South Pacific Ocean, with 3-day averaged OLRA ~minus 70-90 W/m**2. The latter is part of a loosely organized region of tropical rainfall that extends from the Philippines east-southeast crossing the equator near 160E into the South Pacific Ocean along the South Pacific Convergence Zone (SPCZ). I would not have expected this tropical forcing to have propagated into the west central Pacific Ocean along the equator a couple of weeks ago. An equally important region of convective suppression is across the Indian Ocean, leading to the warming SSTs there. Other regions of enhanced diurnal convection are over tropical South America and South Africa.

The MJO signal remains strong per WH(2004) methodology, having ~2.7 standard deviation projection in phase 7 on 16 January. Circulation wind flow anomalies (discussed below) along with the Indian Ocean suppression are contributing. Unlike the late boreal fall 2007 MJO occurrence, during the last week the current MJO event has stalled. Several forecast tools indicate the MJO signal will redevelop across the Indian Ocean during ~week 2-3. The very warm waters of the South Indian Ocean support this prediction, as does daily monitoring of other diagnostics (more said below).

There is, however, some concern the west central Pacific Ocean tropical forcing may persist longer than was earlier thought, possibly another ~1-2 weeks. The latter is a monitoring issue, and does add uncertainty to making predictive statements about probable circulation anomalies to evolve during the next few weeks. Stay tuned.

Updated through 12 January (there are once again data issues) global relative AAM was slightly below the R1 data climatology while its tendency was ~minus 20 Hadleys. Much of that negative tendency was a response to the removal of westerly wind flow by the mountains, particularly East Asia. The GWO signal was weakly in phase 2 (5-day average), after its circuit into phase 5 roughly two weeks ago. Even though the global AAM signal is relatively “benign” (near climatology), zonal mean anomalies remain robust. Strong negative anomalies (~1-2 AMUs), due to intense easterlies, dominate the subtropical atmospheres flanked by positive midlatitude AAM anomalies. Positive zonal mean AAM anomalies are also present in the equatorial band with a contribution coming from above average upper tropospheric westerly wind flow (~5 m/s at 200mb). This type of inter-hemispheric meridional symmetry of zonal mean AAM anomalies is a signature of the on-going strong tropical convective forcing and their distribution is generally consistent with the La-Nina composite.

Animations of upper tropospheric daily mean vector wind anomalies for about the last week loosely show stationary twin subtropical anticyclones (cyclones) around 140E (east of the date line). The stationarity of these wind flow anomalies at least partly explains the recent MJO behavior indicated by the WH(2004) phase space plot. The expected amplification across the PNA sector discussed in our 10 January posting can be seen. There are both meridionally and zonally oriented Rossby wave energy dispersions (RWDs) associated with dynamical processes explained by both the MJO and GWO. However, unlike around 1 December 2007 and also different than earlier thought, the circulation amplification has occurred in MJO phase 7, not phase 8.

Animations of additional fields including mean sea level pressures and surface temperatures show a significant cold air surge coming off of East Asia into the northwest Pacific at the time of this writing (17 January). I think, when updated, there will also be a strong positive mountain torque as a response to a ~1045 mb anticyclone east of the Tibetan Plateau associated with this cold outbreak. In fact, the R1 data AAM plots updated through 14 January “moments ago” support the assertion of a positive East Asian mountain torque. One consequence of this synoptic evolution will be an intensification of the East Asian jet (250mb anomalies ~40m/s 17 January) that will lead to additional amplification of the ridge near 140W. I also speculate (which can be verified observationally) that feedback processes from this synoptic evolution will lead to tropical convective forcing returning to the Indian Ocean weeks 2-3.

Summarizing, the eastward shift of the MJO tropical convective forcing into the west central-South Pacific Ocean has destructively interfered with the La-Nina atmospheric circulation by adding global westerly wind flow. The onset of the lower 48 states cold regime is a direct response. Once the tropical forcing returns to the Indian Ocean-Indonesia region (~100-140E), constructive interference with La-Nina is probable as easterly wind flow anomalies are added. The GWO should then orbit back to phase 3 (GSDM Stage 1) and current circulation anomalies across the PNA sector are probable to retrograde. In fact, latest week-2 ensemble means from various operational weather centers are now generally predicting this response as their initial conditions become more representative of the on-going circulation. However, timing and details of synoptic evolution are always unclear, especially now given the west Pacific forcing discussed above. One can only offer probabilistic statements based on signals like those given by the MJO and GWO (which considers the MJO), as well as the numerical models.

In the face of increasing uncertainty, we are still staying the course in regard to the North American outlooks discussed 5 and 10 January. Amplification across the PNA sector including the east Pacific Ocean ridge has already occurred (this possibility was first discussed on the 15 December 2007 posting, well before the models “caught on”). Additional amplification of this pattern is likely during the next several days, and most models currently predict this. We think these anomalies are probable (synoptic details are unclear) to shift west by ~late week 2/week 3 possibly leading to Alaska blocking tied to an anomalously strong north central Pacific anticyclone and cold troughs that would impact the USA west coast, before coming inland.

Weather ramifications of this pattern for the lower 48 states have already been detailed in past discussions. The Aleutian Islands of Alaska may get hammered by the strong ridge building cyclones mainly week-1 while Kona lows may become a concern for the Hawaiian Islands after week-1. In general, a west-southwest flow storm track across the central part of the country downstream from western USA troughs and upstream from a deep southeast states anticyclone, consistent with GWO phase 3 “amplifying” La-Nina, is probable for the next several weeks (~3-6 weeks). Most of the country may feel significant impacts (severe local storms warm sectors, blizzards cold sectors, etc.) from what could be a severe winter weather regime at times.

Internationally, the current westerly wind burst across the South Pacific Ocean will maintain a tropical cyclone hazard (possibly severe) for the paradise islands most likely week-1. That same but much lesser concern cannot be ruled out for locations around the Philippines. Portions of Brasil and South Africa may also experience enhanced frontal thunderstorm activity weeks 1-2. As moist tropical convective forcing returns to the South Indian Ocean-Indonesia regions weeks 2-3, the risk of tropical cyclones is probable to increase around the region of Madagascar. This may be start of MJO #3 for the 2007-08 boreal cold season.

Appendix

An experimental quasi-phase space plot of the GSDM utilizing time series of normalized global relative AAM time tendency (Y-axis) and normalized global relative AAM anomaly (X-axis) can be found at

http://www.cdc.noaa.gov/map/clim/gsdm.shtml

We call the behavior of this plot the Global Wind Oscillation (GWO). While the intent of the legacy GSDM is to extend current thinking beyond the MJO, the GWO quantifies variations used to derive the original GSDM in a manner that is “user friendly” analogous t0 WH(2004) “convention”. In addition, the GWO plot does not have the ENSO signal removed.

Links to CPC and PSD ENSO discussions:

http://www.cpc.ncep.noaa.gov/products/analysis_monitoring/enso_advisory/index.shtml

http://www.cdc.noaa.gov/people/klaus.wolter/MEI/

These are probabilistic statements, and work is ongoing to quantify in future posts (for example, risk assessment maps, signal to noise ratio plots and shifts of probability). We hope that an opportunity will arise for us (soon) to have a dedicated web page effort to expedite more objectively, with rigor, thoroughness and verification. The WB (2007) paper on the GSDM has been published in the February issue of MWR. Given travel, I am unclear when my next posting will be. Hopefully I will be able to post a discussion the weekend of 26-27 January. Please stay tuned.

Ed Berry

Thursday, January 10, 2008

La-Nina on Steroids; Severe USA Winter Weather Regime Coming???

Per latest 5-day averaged TAO buoy data, SST anomalies along the equatorial Pacific Ocean from 160E-South America remain strongly negative with magnitudes ~1-3C extending to ~100-150m depth. While significant cooling of SSTs has occurred during the last 1-2 weeks north of Australia due to recent intense convective rainfall, comparable warming has been in progress for the suppressed South Indian Ocean. Regardless of recent subseasonal weather events, well-pronounced mature La-Nina SSTs continue, as part of the global tropical SST system.

Although weakened, moist tropical convective forcing remains loosely consolidated across the Eastern Hemisphere centered ~0/140E. Per WH(2004) phase space plots; there is still a significant MJO projection, ~2 standard deviations located in phase 7. However, the eastward movement has again slowed. In fact, OLR/A Hovmoller plots suggest a portion of this tropical forcing has started to drift back to the west. Whether or not a weak dynamical signal will propagate into the Western Hemisphere is unclear. Recent full disk satellite imagery does show some enhancement of diurnal/frontal thunderstorm activity across tropical South America and perhaps South Africa. Although it is probable the MJO signal will re-emerge across the South Indian Ocean weeks 2-3, its exact path through phase space per WH(2004) methodology is unknown.

The MJO variability observed since boreal autumn 2007 is the strongest in at least 2 years per time series plots of the WH(2004) RMMs. What is so remarkable is to see this kind of tropical convective variability during a moderate-strong La-Nina, as is currently the case. Specifically, there have been 2 moderate-strong MJOs, including the on-going event. Depending on the MJO’s phase, these variations have alternately weakened or strengthened the La-Nina tropical convective forcing. As part of a shift by tropical convection to the west Pacific Ocean in late November, away from the typical La Nina location over Indonesia, a large amplification of the circulation occurred across the PNA sector. The response was an intense ridge from the east Pacific Ocean into Alaska and a deep western USA trough. These anomalies then retrograded as zonal mean tropical easterly wind flow anomalies developed, linked to a negative mountain torque and tropical convection returning to Africa and the western Indian Ocean.

Most recently, strong poleward fluxes of relative AAM ~35N during the last week of December dramatically increased the zonal mean westerly wind flow across the Northern Hemisphere midlatitudes. In fact, wind speed anomalies were ~70m/s at 250mb on 31 December 2007 north of a central Pacific Ocean anticyclone, the latter a response to the phasing of the MJO and the preferred region of La Nina convection over Indonesia. The point is through cooperative dynamical interactions involving the La-Nina base state, MJO variability and Rossby wave energy dispersions, the global circulation during the first week of January 2008 can be thought of as an “amplified La-Nina response”. Thus we attribute the west coast storm of ~3-5 January 2008 to “El-Viejo on steroids”!

Updated through 7 January, global relative AAM is near normal. In addition to the MJO, GWO processes such as possibly the strongest North American positive mountain torque of (~30-40 Hadleys) on record which occurred around 3 January also led to the AAM increase. Synoptically, the large and intense ~1045 hPa anticyclone that plunged across the Southern Plains and Mexico contributed to this torque. The rapid moisture return in the Plains in its wake was a needed ingredient to the severe local storm outbreak in that region a few days ago. Through eddy exchange processes linked to a negative global frictional torque (~10 Hadleys) and what will soon be a negative global mountain torque, AAM tendency is probable to become strongly negative again. In fact, the negative AAM tendency may be comparable to that observed in early December. We do expect global AAM to again drop substantially during the next few weeks, especially as subtropical zonal mean easterly wind flow anomalies remain strong. After a brief circuit through GWO phases 5-8, phase 3 is most probable by ~week 2. Again, timing is always an issue.

There is no change to the North American weeks 1-3 outlooks discussed 5 January. Similar to December 2007, a large amplification across the PNA sector is probable within the next 10 days. As more models are showing, ridge amplification from the east Pacific Ocean into Alaska with a downstream central/eastern USA trough is a likely outcome. We think these anomalies will shift west leading to Alaska blocking and a western states trough weeks 3-4. This will allow the delivery of Arctic air into the lower 48 states, along with increasing southwest flow storm track activity across the Plains. Later weeks 2-3 the west coast may again be hammered by cold troughs as circulation anomalies retrograde. Again, timing and details about the upcoming circulation evolution are still unclear. Most of the country may feel significant impacts (severe local storms warm sectors, blizzards cold sectors, etc.) from what could be a severe winter weather regime.

Internationally, at least enhanced possibly severe diurnal thunderstorm activity is probable weeks 1-2 for locations such as tropical South America and west central-South Africa. Tropical cyclone activity (including Elisa) should continue across the South Pacific Ocean mainly week-1. We think consolidation of tropical forcing should occur by about week-3 from the South Indian Ocean into western Indonesia. This may be start of MJO #3 for the 2007-08 boreal cold season.

Appendix

An experimental quasi-phase space plot of the GSDM utilizing time series of normalized global relative AAM time tendency (Y-axis) and normalized global relative AAM anomaly (X-axis) can be found at

http://www.cdc.noaa.gov/map/clim/gsdm.shtml

We call the behavior of this plot the Global Wind Oscillation (GWO). While the intent of the legacy GSDM is to extend current thinking beyond the MJO, the GWO quantifies variations used to derive the original GSDM in a manner that is “user friendly” analogous t0 WH(2004) “convention”. In addition, the GWO plot does not have the ENSO signal removed. Links to CPC and PSD ENSO discussions:

http://www.cpc.ncep.noaa.gov/products/analysis_monitoring/enso_advisory/index.shtml

http://www.cdc.noaa.gov/people/klaus.wolter/MEI/

These are probabilistic statements, and work is ongoing to quantify in future posts (for example, risk assessment maps, signal to noise ratio plots and shifts of probability). We hope that an opportunity will arise for us (soon) to have a dedicated web page effort to expedite more objectively, with rigor, thoroughness and verification. The WB (2007) paper on the GSDM has been published in the February issue of MWR. Given travel, I am unclear when my next posting will be. Please stay tuned.

Ed Berry and Klaus Weickmann

Saturday, January 05, 2008

El-Viejo Circulation Phase Space Rules

The spatial pattern of global tropical and even extratropical SST anomalies remains similar to a week ago. Some changes include cooling of the ocean waters generally north of Australia having anomalies ~minus 1-2C as a response to persistent enhanced rainfall, and corresponding warming across the suppressed Indian Ocean. Nevertheless, SST totals in excess of 28C dominate most of the Eastern Hemisphere equatorial and southern tropics extending well into the South Pacific Ocean.

Significant cold anomalies of ~minus 1-3C remain across the equatorial Pacific Ocean from ~155E-west coast of South America. In fact, recent anomalously strong trade winds have expanded and intensified these El-Viejo SST anomalies during the last couple of weeks. These below normal waters extend to at least 100m deep per latest 5-day averaged TAO buoy data (see http://www.pmel.noaa.gov/tao/jsdisplay and other relevant links for SST details). Subsurface warmth is still present ~150m depth in the region of the equatorial date line. The weak oceanic Kelvin wave discussed in the last few postings has lost coherence, likely a response to the enhanced trades.

I again want to make the point the La-Nina SSTs are only a component of the global tropical SSTs forcing (and responding with feedbacks) the atmospheric circulation. In fact, speculation suggests to me that the very warm Eastern Hemisphere tropical SSTs observed for the past several years may have recently tilted the odds toward La-Nina rather than El-Nino. The latter goes back to subseasonal weather events during boreal autumn 2006 which led to a La-Nina global circulation state (as seen from an understanding of the earth-atmosphere AAM budget). An abrupt and unpredicted end to the El-Nino SSTs subsequently occurred.

I can further speculate the warm SSTs discussed above may be a response to the global warming signal “everyone talks about”. Make no mistake; the recent global warming trend is real (started ~1960). So, can we suggest this La-Nina is a response to global warming as a feedback??? Whatever the case, there is lots of uncertainty crossing multiple time and space scales out there. As may be case investing in the financial markets these days, the same may be said about making weather forecasts.

There has not been any substantial change in the character of the tropical convective forcing discussed a week ago. Full disk satellite imagery and other monitoring tools show strong-severe clusters of tropical thunderstorm activity extending from Indonesia-South Pacific Ocean. Consistent with La-Nina, the latter is along a southwestward shifted South Pacific Convergence Zone (SPCZ). However, I think there is a component propagating southeast toward the region of ~20S/180. Recent 3-day averages of OLRA have been less than minus 70W/m**2 in that part of the world (with suppression along the equatorial date line). My own off the cuff phase speed calculation suggests an eastward movement of ~7 m/s within the convective envelope discussed above. I think the latter represents a dynamical signal that will quickly move through the Western Hemisphere during the next 1-3 weeks. Again, who knows about the timing?

The WH(2004) phase space plots show a significant phase 6 MJO signal through 4 January 2008. The MJO partition along with the rest of the Eastern Hemisphere tropical forcing slowed its eastward propagation considerably during the last few weeks due to the atmosphere-ocean coupling discussed in my 30 December 2007 edition. Continuing on my path, I would expect "some" tropical forcing to persist in the region of 120-140E with attendant circulation anomalies while perhaps a MJO dynamical signal moves through the Western Hemisphere south of the equator given the La-Nina SSTs. How the WH(2004) real-time multivariate MJO (RMM) phase space plots maintained by several weather centers show this speculative behavior is unclear.

Animations of upper tropospheric daily mean vector wind anomalies show a strong signal of what would be expected given the spatial distribution of the tropical forcing discussed above. Twin subtropical anticyclones (cyclones) have shifted east to ~120-140E (east of the date line) with 150mb anomaly magnitudes in excess of 40m/s for the equatorial Pacific westerlies. In fact, this Western Hemisphere equatorial anomalous westerly wind flow is now coming back around into the Eastern Hemisphere (more said below). Finally, there is strong cross-equatorial wind flow around the date line from the Southern Hemisphere, tied to the convection passing south of the equator.

Worth repeating from a week ago, there have been several episodes of meridionally oriented Rossby wave energy dispersions (RWDs) into the extratropics tied to circulation features such as these for the last several weeks. For the PNA sector these RWDs have led to robust cyclonic circulation anomalies around Alaska poleward of a central Pacific Ocean anticyclone. Several strong troughs have impacted the USA west coast on into the Plains as a result. The recent (~4 January 2008) severe cyclonic winter weather event that impacted the USA west coast was an extreme manifestation of these RWDs. In fact, daily mean 250mb wind speed anomalies associated with the central Pacific anticyclone was in excess of 70m/s with this event.

Having good initial condition information largely from the extratropics (ex., blocking around Scandinavia), most models did a superb job advertising this extreme weather event starting around Christmas. Of course, the magnitude of the anomalies was underestimated particularly at longer leads given the role of the tropical forcing. Additionally, as shown in the WB ((200?); publication is planned) composites for both the GWO and MJO, extreme USA west coast precipitation events are most probable to occur in GWO phases 7-8 and to a lesser degree MJO phases 7-8. However, the recent storm occurred with GWO phase 3 and MJO phase 5. The latter again emphasizes the roles played by the dynamics of non-linear feedback processes that I can only wish the resources existed for us to demonstrate in real-time. I should also add that discussion was presented in the 18 November 2007 posting of suspicion that a higher probability existed for these kinds of non-linear feedbacks within our 2007-08 La-Nina base state.

Global relative AAM remains low ~minus 1.5 standard deviations below the R1 data climatology updated through 1 January 2008. Since mid-December 2007, in the wake of the earlier MJO, there has been substantial poleward propagation of anomalous zonal mean easterly wind flow anomalies into the subtropical atmospheres. At 200mb zonal mean anomaly magnitudes are ~10m/s poleward of similar anomalies for the equatorial westerlies. These subtropical easterly wind flow anomalies have been contributing to intense midlatitude ridges, particularly across the North Pacific Ocean.

Although slightly positive ~20 December 2007, a strong East Asian Mountain torque did not occur as speculated a week ago. It was, instead, a perturbation in a low AAM base state. However, largely from the strong subtropical Northern Hemisphere trades, the global friction torque has peaked to plus 20 Hadleys. The latter is ~50-60 days after the last big frictional torque event, consistent with the GWO.

Since late December 2007 an intense sink-source transport signal has appeared ~10-40N with flux convergence of this transport ~40N. The dynamics responsible for this poleward transport signal were another contributor to the recent severe USA west coast storm. AAM tendency has been slightly positive since early December largely from processes tied to the friction and Coriolis torques. The point is we have a mature La-Nina base state where secondary processes in the atmosphere are trying to generate westerly flow to offset it. The GWO updated through 31 December shows the atmosphere “drifting” in phase 3 (legacy GSDM Stage 1). I again emphasize that measured in terms of the AAM budget, the global circulation has been strongly La-Nina, nothing moderate about it.

Uncertainty remains HUGE (above whatever climatology anyone chooses) for any subseasonal, etc., prediction. I do think there will be a weak GWO circuit possibly through phases 5-8 during the next 2-3 weeks. Per above, a weak MJO dynamical signal appears probable to propagate through the Western Hemisphere, only to come back around before the end of this month. Most models are suggesting some form of east Pacific Ocean ridge amplification by around week-2. Given the cross equatorial wind flow per above that notion is reasonable. However, systematic bias issues with the NCEP GFS ensemble tell me to favor the slightly farther west solutions of other models such as the ESRL/PSD week-2 ensemble mean. I also think that during ~weeks 2-3 more strong cold storms/troughs are once again probable to slam the USA west coast.

Hence I remain a broken record for any week 1-3 USA outlooks. After relatively warm and weak systems week-1, a colder and wetter regime appears probable from the Desert Southwest (bucking the composite La-Nina signal) into the Plains and east week-2. I would expect a westward shift of this situation weeks 3-4. I also think it is only a matter of time before blocking develops in the region of the Gulf of Alaska allowing Arctic airmasses to interact with a southwest flow storm track across the Plains. With very low confidence, I offer this synoptic evolution is possible around week 3 (~19-26 January). In addition to the winter weather hazards, concerns also must be expressed for an above average occurrence of severe local storms for locations such as the lower Mississippi into perhaps the Ohio Valleys not only weeks 1-3, but perhaps the rest of this cold season.

Internationally, heavy-severe rainfall/thunderstorms (including tropical cyclones) will be a concern for Eastern Hemisphere-South Pacific Ocean areas already discussed above for weeks 1-2. Heavy rainfall and severe thunderstorms may impact much of Brasil and South Africa by week-3, if not much sooner. Bitterly cold Arctic air will continue to dominate much of Siberia into Alaska. Alaska will see significant warming once the blocking develops.

Appendix

An experimental quasi-phase space plot of the GSDM utilizing time series of normalized global relative AAM time tendency (Y-axis) and normalized global relative AAM anomaly (X-axis) can be found at

http://www.cdc.noaa.gov/map/clim/gsdm.shtml

We call the behavior of this plot the Global Wind Oscillation (GWO). While the intent of the legacy GSDM is to extend current thinking beyond the MJO, the GWO quantifies variations used to derive the original GSDM in a manner that is “user friendly” analogous the WH(2004) “convention”. In addition, the GWO plot does not have the ENSO signal removed.

Links to CPC and PSD ENSO discussions:

http://www.cpc.ncep.noaa.gov/products/analysis_monitoring/enso_advisory/index.shtml

http://www.cdc.noaa.gov/people/klaus.wolter/MEI/

These are probabilistic statements, and work is ongoing to quantify in future posts (for example, risk assessment maps, signal to noise ratio plots and shifts of probability). We hope that an opportunity will arise for us (soon) to have a dedicated web page effort to expedite more objectively, with rigor, thoroughness and verification. The WB (2007) paper on the GSDM has been published in the February issue of MWR. Given travel, I am unclear when my next posting will be. Please stay tuned.

Ed Berry