Friday, March 30, 2007

GSDM Stage 1 Perturbed

The west Pacific SST horseshoe of warmth remains; however, anomalies are weakening due to recent days of intense tropical rainfall. Magnitudes are generally 0.5-1C with SSTs ~29-30C, with the warmest waters now around 10S/160W as of March 29th. Conversely, the equatorial Indian Ocean continues to slowly warm due to suppression (with the exception of a tropical cyclone in the South Indian Ocean) with SST totals and anomalies comparable to the west Pacific warm pool. Cooling along the equatorial east Pacific cold tongue has again intensified this past week with negative anomalies ~minus 1-3C east of 140W and at least minus 4C in the subsurface. These colder than normal waters extend down to roughly 200m at the date line per latest TAO buoy data. Whether or not La-Nina is evolving is an open question (recall May 2003); however, it is not unusual to see “pulses” of cooling if this is the case.

Hard times continue trying to understand the tropical convective forcing, and there is a seasonal cycle consideration to remember. There is no coherent MJO dynamical signal. Most significant behavior during the last 10 days has been an eastward shift of intense (OLRA ~ minus 50-70 W/m**2) forcing from ~120E to 160E just south of the equator. I maintain this is a transient event linked to very complicated tropical-extratropical interactions. These can only be reasonably understood observationally from disciplined detailed daily monitoring of fields such as animations of upper and lower tropospheric wind fields utilizing the GSDM framework. Some of these interactions include last week’s extension of the North Pacific jet shifting this forcing to the east, possibly reinforced by feedbacks from the extratropics including a strong positive East Asian mountain torque.

Full disk satellite imagery already shows the west Pacific tropical forcing drifting west and northwest, and the coherent modes Hovmollers suggest a weak projection on the convectively coupled Rossby mode. In fact, a tropical cyclone may develop out of this region north of the equator during the next few days per JTWC. Weaker forcing (diurnal cycle understood) also persists across northwest South America, west central Africa and ~0/120E.

Our best weather-climate signal remains global relative AAM including the zonal mean contributions. This quantity is a good diagnostic measure of the character and evolution of the global and zonal mean circulation anomalies. Relative AAM continues at around 1.5 standard deviations below the 1968-1997 reanalysis data climatology, and its tendency has been weak for the last couple of weeks. Furthermore, zonal mean upper tropospheric easterly wind anomalies remain at around 5m/s throughout much of the tropical and subtropical atmospheres. These easterlies are contributing to a zonal mean sink of AAM ~20N while flux convergence of its transports occur ~35N. All of these observations are consistent with a GSDM Stage 1 base state (most probable during a La-Nina) that has been around since at least late February.

Given the warm Indian Ocean SSTs, feedbacks from our low AAM regime and the slow westward shift of tropical forcing back to 120E discussed on 3/27, we need to monitor for the development of intense tropical convection across the Indian Ocean into Indonesia sometime weeks 2-3. In the meantime, the transient SPCZ convection discussed above has generated enough anomalous (~20-30m/s at 150mb on 3/29) upper tropospheric equatorial westerly wind flow over the east Pacific to perturb our GSDM Stage 1 state toward Stage 2 for week 1. In fact, animations of daily mean upper tropospheric vector wind anomalies show well defined twin tropical anticyclones at 160E as of 3/29. These anticyclones are already interacting with Rossby wave energy dispersion (RWD) processes from Asia to North America, including the ridge building into Alaska and the developing downstream trough across western Canada. It is possible the models may be underestimating the impacts of this trough onto the Rockies and Plains early-mid next week given the large circulation anomalies tied to this RWD and the complex dynamics. Overall, as shown by most ensemble numerical models, a colder regime is probable for the eastern two-thirds of the USA for at least week 1.

My thoughts for weeks 2-3 remain the same. I speculate whatever “renegade” North Pacific jet that does come out may lead to a similar evolution of synoptic events that was observed during the last 7-10 days. If the tropical convective scenario discussed above does occur, a shift back to a GSDM Stage 4-1 would be most probable by sometime week 3 (April 14-20). Although models do not predict the evolution of tropical convection very well after week 1, the NCEP week 2 ensemble mean total precipitation forecast lends loose support to renewed Indian Ocean convection. The latter is likely a SST response. This would lead to a renewed active pattern for much of the western and central USA, including severe local storms across the Plains. Careful monitoring will be needed to see if Eastern Hemisphere tropical forcing eventually develops into a MJO.

Please note: These are probabilistic statements, which we will try to quantify in future posts. I am on TDY at ESRL/PSD with the HMT project until April 3rd. I am unclear when I will be able to another discussion next week due to travel. We are also working on a weather-climate discussion for the ESRL/PSD MJO web site, which will hopefully be posted during the next couple of weeks. The WB (2007) paper on the GSDM has been published in the February issue of MWR.

Ed Berry

Tuesday, March 27, 2007

Hard Times

Above average SSTs remain centered around 0/150E and continue to have a horseshoe distribution pattern into both hemispheres. Anomalies are ~0.5-1C having total SSTs of 29-30C with the warmest centered at roughly 5-10S/170E. These SSTs have cooled a bit during the last week as a response to enhanced rainfall in that region. Weak-moderate warm anomalies continue to exist across much of the equatorial Indian and north tropical Atlantic Oceans. In fact, suppression of tropical convection has allowed SSTs across the Indian Ocean to warm by about 0.5C over the past week. The equatorial east Pacific remains moderately cool east of 120W with negative anomalies as low as -4C down to roughly 200m.

Understanding the current tropical convective forcing is about as difficult as it can get. There is no coherent MJO dynamical signal, and these are the situations where additional understanding beyond linear MJO (and linear dynamics) thinking is needed. The strongest tropical convective forcing globally is centered ~5-10S/165E, in the region of warm SSTs. In fact, three-day averaged OLRA are at least minus 50-70 W/m**2 (including Hurricane Becky) and the Wheeler multivariate MJO index even projects this convection as a MJO as of March 26th. The latter is a response to the baroclinic vertical wind structure with this convection. For example, there are weak anomalous twin upper tropospheric tropical anticylones with cross-equatorial southerly flow (with lower level opposite wind flows) around the date line.

As discussed on my last Friday posting (3/23), the enhancement of the North Pacific jet helped to shift tropical convection into the southwest Pacific. I believe that additional dynamical forcing tied to a strong East Asian mountain torque around March 20th (~20 Hadleys with a global torque of 30 Hadleys) contributed to maintaining this jet and the SPCZ convection. Other weaker tropical forcing persists over northwest South America with respectable thunderstorm activity across west central Africa.

However, another region of now relatively weak tropical convection remains centered at roughly 0/120E. This is actually a westward shift out of the South Pacific from mid-January, and continues to be the most important tropical forcing to impact the global circulation. A response is that global relative AAM remains about 1.5 standard deviations below the 1968-1997 reanalysis data climatology, reflective of loosely deep zonal mean easterly flow throughout much of the tropical and subtropical atmospheres (~5m/s anomalies). I think this is a slowly evolving component and may be part of a coupled process involving a transition to La-Nina. The role of the seasonal cycle needs to be remembered, particularly around May. GSDM Stage 1 continues to define the current weather-climate/global circulation, which is consistent with La-Nina. Summing up, our strongest weather-climate signal remains global relative AAM while the subseasonal evolution of the tropical forcing is unclear. I can only offer the following speculation.

Discussion in past postings has addressed the fairly rapid subtropical variations (~10-15 days as seen from the global AAM tendency time series) within this GSDM Stage 1 base state. There are fairly symmetric anomalous wave trains across the subtropical atmospheres of both hemispheres, including twin upper tropospheric cyclones across the east Pacific and Indian Oceans. These are actually generating anomalous equatorial westerly flow that I think will locally interact with the subtropics. I also suspect tropical convection will increase across the warm Indian Ocean SSTs during the next 2-3 weeks once the twin cyclones in that region move east. A weak eastward propagating dyamical signal from the southwest Pacific may also arrive there. By weeks 3-4 tropical forcing may become very intense around 90-120E. The gist is there may be a shift toward GSDM Stage 1-2 by week 2 and then perhaps a GSDM Stage 4-1 response afterwards.

Needless to say, forecast confidence after week 1 is very low. I like the increasing number of solutions being displayed by most ensembles through around days 5-7. Blocking developing across Alaska is highly probable since I can already see this process, linked to the 120E tropical forcing, from animations of daily mean 150mb and 250mb vector wind anomalies. This will allow wave breaking troughs to develop across the central Pacific hence supporting a ridge off the USA west coast and a downstream central North American trough. This suggests a colder and wetter regime for much of the northern and central USA in the wake of the present western states trough. Daily monitoring will be needed in regard to the amplitude of these features, particularly for upwelling concerns off the California coast and southward penetration of Arctic air into the Northern Rockies and Plains. Afterwards, my thought would be for this regime to shift west and north, going back into an active (above climatology) spring pattern for especially the Plains by week 3. Careful monitoring will be needed to see if the Eastern Hemisphere tropical forcing eventually develops into a MJO.

Please note: These are probabilistic statements, which we will try to quantify in future posts. I am on TDY at ESRL/PSD with the HMT project until April 3rd. I will try to post another (hopefully shorter) discussion this Friday. We are also working on a weather-climate discussion for the ESRL/PSD MJO web site, which will hopefully be posted during the next couple of weeks. The WB (2007) paper on the GSDM has been published in the February issue of MWR.

Ed Berry

Friday, March 23, 2007

Renegade North Pacific Jet

Above average SSTs still persist across the tropical west central Pacific with the warmest centered ~ 5S/165E having magnitudes of 1-2C and actual SSTs ~30C. Weak-moderate warm anomalies are also present across much of the equatorial Indian and north tropical Atlantic Oceans. The equatorial east Pacific remains moderately cool east of 120W with negative anomalies as low as -4C down to roughly 200m.

The tropical forcing discussed on my March 20th posting has loosely separated into 2 regions. One area remains quasi-stationary ~0/120E while another region has moved into the warm waters of the southwest Pacific centered at roughly 5-10S/170E. The former region continues to have the strongest interactions with the extratropics while the latter may be a response to the transient strong Pacific jet stream embedded within the low relative AAM, GSDM Stage 1.

We speculate that a Rossby wave energy dispersion from west Pacific forcing last week contributed to the snow storm for the USA northeast. Animations of 150mb and 250mb daily mean vector wind anomalies suggest that both tropical forcing present across Indonesia and a strong east Asian mountain torque of roughly 20 Hadleys (30 Hadleys globally) combined to add more westerly flow to the Pacific subtropics. Now there is an intense jet across the North Pacific having anomalies of at least 30-40 m/s. This feature is “out running” the tropical forcing back at 120E, and perhaps shifting the convection east across the southwest Pacific. In fact, according to the coherent modes Hovmollers, the convection across the southwest Pacific projects onto a Kelvin wave.

The future evolution of the tropical forcing still remains unclear. However, for at least the next 1-3 weeks, the most important convection to impact the extratropics should remain quasi-stationary centered in the region of 120E (consistent with a low AAM regime) while the southwest Pacific has episodic flare-ups. The strong North Pacific jet is transient. As most models show, this jet should amplify a strong trough across the western USA by early next week. For weeks 2-3, given persistence of GSDM Stage 1, ridge amplification across the central and east Pacific into Alaska with more troughs impacting the western and central states is probable. This is an active pattern for the Rockies and Plains favorable for late season snowstorms and outbreaks of severe thunderstorms with individual synoptic baroclinic cyclones. Careful monitoring will be needed to see if the Eastern Hemisphere tropical forcing develops into a MJO, and if the current cooling of the tropical east Pacific develops into a La-Nina.

Please note: These are probabilistic statements, which we will try to quantify in future posts. I am on TDY at ESRL/PSD with the HMT project until April 3rd. I will try to post another discussion next Tuesday. The WB (2007) paper on the GSDM has been published in the February issue of MWR.

Ed Berry and Klaus Weickmann

Tuesday, March 20, 2007

GSDM Stage 1

This discussion will be “short and sweet”. The distribution of global tropical SSTs remains similar to that discussed on my posting last Friday (3/16/07). The horseshoe pattern of warmth persists across the west central Pacific with anomalies of ~1-2C and 29-31C actual SSTs while similar cool SST anomalies exist along the equator from 150W to South America. The latter have actually cooled a bit during the last few days while colder anomalies ~2-4C persist in the subsurface down to roughly 200m.

Tropical convective forcing is getting better organized across the Eastern Hemisphere centered at about 0/120E. Latest 3-day averaged OLRA are ~minus 50-70W/m**2. Other pockets of enhancement still remain across the South Pacific, Brasil and equatorial Africa. The dynamical MJO signal discussed in my last posting has lost coherence. The future evolution of the tropical convection is unclear. Based on its current position around 120-130E and the existing low AAM circulation anomalies, we speculate that convection will remain quasi-stationary or shift slightly west over the next one to two weeks. Kelvin waves excited by the convection should also help excite transient convective episodes over the west Pacific. This complicated picture may all be part of a slow process of the atmosphere transitioning to a more pronounced La-Nina base state.

GSDM Stage 1 (most probable during La-Nina) still best describes the current weather-climate situation. It is probable this situation will persist for at least the next 2-3 weeks. Relative atmospheric angular momentum (AAM) remains well below the 1968-1997 reanalysis data climatology at ~1.5 standard deviations. For the Pacific-North American sector, more strong troughs are expected for the western USA with a southwest flow storm track across the Plains probable at least into early April. This pattern is favorable for late season snowstorms across portions of the Rockies and northern/central High Plains with a greater than climatology risk for severe local storms across the Plains and Mississippi Valley.

Please note: These are probabilistic statements, which we will try to quantify in future posts. I am on TDY at ESRL/PSD with the HMT project until April 3rd. I will try to post another discussion this Friday. The WB (2007) paper on the GSDM has been published in the February issue of MWR.

Ed Berry and Klaus Weickmann

Friday, March 16, 2007

Recap on the Complexity

Tropical west Pacific SSTs still remain above average in a horseshoe pattern centered on 155E with anomalies ~0.5-1.5C and actual temperatures from 29-31C. There has also been some warming across portions of the equatorial Indian Ocean with positive anomalies ~0.5-1C and SSTs near 30C. Below normal SSTs remain from the north coast of Australia to the equator, with anomalies ~1.0C. With the exception of the equatorial Pacific ~120W, eastern hemisphere tropical SSTs are weakly positive. While the negative anomalies along the central equatorial Pacific cold tongue are currently minimal, colder subsurface anomalies remain down to roughly 200m. The latter are important for possible La-Nina development.

Largely forced by the warm SSTs, favorable upper tropospheric winds and Rossby wave energy propagation from the extratropics, tropical convective forcing remains loosely organized centered ~0/140E. This area extends from the central Indian Ocean into the South Pacific. However, the convection in this region is not a large departure from climatology. There is still evidence from various fields such as animations of daily mean 150mb vector wind anomalies and velocity potential Hovmollers that a weak dynamical signal is moving through the Western Hemisphere. I think it is centered at roughly 0/60W and moving east at about 10 deg long/day. We expect this signal to come back into the Eastern Hemisphere sometime during week 2, and allow the tropical convective forcing to increase from Africa to the warming SSTs of the Indian Ocean. We may see the third MJO since late last fall evolve in that region by late week 2 or 3.

GSDM Stage 1 best describes the current weather-climate situation. Relative atmospheric angular momentum (AAM) remains well below the 1968-1997 reanalysis data climatology at ~1.5 standard deviations. The global tropical and subtropical atmospheres are dominated by zonal mean easterly wind anomalies with magnitudes of 5-10m/s. The fast dynamical signal discussed above should reinforce what is already a low AAM regime during the week 2-3 time frame. Synoptically, the most probable response across the Pacific North American sector would be for a discontinuous retrogression of existing circulation anomalous by at least 20-30 degrees of longitude. This means the ridge currently over the western North America should redevelop ~150W.

Even with the above reasoning, forecast uncertainty for weeks 2-3 remains very high. If a coherent MJO develops across the Indian Ocean by week 3, forecast confidence in subseasonal outlooks may improve. Most models forecast a fairly strong trough to impact the west coast early next week and then shift into the Rockies and Plains afterwards. It is possible this first trough may “split” with northern and southern stream developments during the course of its evolution. In fact, recent signals of AAM transports would support this possibility. Afterwards, a more “classic” GSDM Stage 1 springtime active pattern for the western and central USA may emerge for weeks 2-3. This would argue for late season snowstorms across portions of the Rockies and northern/central High Plains with a greater than climatology risk for severe local storms across the Plains and Mississippi Valley.

Please note: These are probabilistic statements, which we will try to quantify in future posts. I am on TDY at ESRL/PSD with the HMT project until April 3rd. I will try to post another discussion next Tuesday. The WB (2007) paper on the GSDM has been published in the February issue of MWR.

Ed Berry and Klaus Weickmann

Tuesday, March 13, 2007

Regrouping or What?

SSTs remain warm in a horseshoe pattern centered on 0/150E and extend to just southwest of Hawaii as well as into the South Pacific. Anomalies are ~plus 0.5-1.0C with actual SSTs from 29-30C. Other weak-moderate positive tropical SST anomalies are also present across portions of the Indian Ocean, far eastern Pacific and the North Tropical Atlantic. Negative anomalies of roughly 0.5-1.0C are present from Indonesia into Northern Australia as the result of recent tropical cyclone and heavy precipitation activity. Cool anomalies of ~0.5C also remain on the equator from about 120-150W with colder subsurface values down to 200m at the date line per latest TAO data. The latter may be attributable to a developing La-Nina. If one does develop, we expect the cooling to occur in “steps” related to forcing from subseasonal events.

Resolving the situation with the tropical convective forcing is difficult and about as tough as it gets in terms of uncertainty. A very weak dynamical signal with the recent weak-moderate MJO remains around 90W and appears to be stalling just west of South America. Tropical convection has also tried to get better organized ~0/140-150E for the past several days, largely forced by the warm SSTs in that region and favorable upper level winds. A Rossby wave energy dispersion from the Southern Hemisphere extratropics has led to an increase in upper tropospheric easterly wind anomalies across the west central Pacific. The locally strengthened upper level divergence is providing a source of lift for additional thunderstorm activity.

Animations of 150mb daily mean vector wind anomalies suggest twin subtropical anticyclones redeveloping ~150E. However, whether or not this convection will then shift east toward the warm SSTs as a MJO-like signal is unclear. Currently, a component actually shifting northwest from the west Pacific toward the Philippines is only complicating matters. Furthermore, should a strong eastward shift occur, the bulk of the forcing will stay south of the equator given the cool east Pacific SSTs and the time of year.

The uncertainty in the near term tropical convective forcing is not present in all circulation fields. Global relative atmospheric angular momentum (AAM) remains low at about -1.5 sigma anomaly. A large contribution is coming from the northern subtropical atmosphere centered ~25N. Zonal mean easterly wind anomalies at 200mb are from 5-10m/s in that region, having propagated off the equator starting back in mid February tied to the recent MJO. There have been a couple of episodes of weak positive global AAM tendency since then (~10 Hadleys), partly linked to variations involving mountain and frictional torques in addition to the MJO dynamical signal moving into the Western Hemisphere. For the latter, weak upper tropospheric westerly wind anomalies developed along the Western Hemisphere equator and are interacting with a subtropical wave train currently moving through South Asia.

A very strong storm system is currently pounding Tibet and northern India, part of the wave energy dispersion along the South Asian wave train. As this system interacts with the west central Pacific convection, an extension of the East Asia Jet (EAJ) is probable. Synoptically, this translates to the current trough just east of Japan moving east toward western North America while the upstream system replaces it. This was a possibility discussed in my March 9th posting, and most models predict this situation starting around next Tuesday. However, details remain very unclear. Nevertheless, there is hope for at least one IOP for the American River Basin HMT project with this system. This trough should then move into the Rockies and Plains during week 2 increasing the probability of high impact weather.

Afterwards, for weeks 2-3, uncertainty remains very high. I would think a GSDM Stage 1 base state (consistent with La-Nina) is most probable to persist, with subseasonal variations. Considering the seasonal cycle more strong troughs for particularly the Rockies and Plains while ridge amplification occurs across the central/east Pacific would be expected. However, if the tropical forcing shifts east along the equator to the date line (and/or a strong central Pacific flare-up occurs), the circulation could quickly transition to Stage 3.

Please note: These are probabilistic statements, which we will try to quantify in future posts. I am on TDY at ESRL/PSD with the HMT project until April 3rd. I will try to post another discussion this Friday. The WB (2007) paper on the GSDM has been published in the February issue of MWR.

Ed Berry and Klaus Weickmann

Friday, March 09, 2007

Details are Important

Not much change has occurred with the tropical sea surface temperatures (SSTs) since my posting on Tuesday (3/6). While equatorial Pacific SSTs remain cool east of 150W (~minus 1-2C), warm anomalies are still present across the west central into the South Pacific. The latter are roughly 0.5-1C extending to depths of about 100m with actual SSTs varying from 29-31C. Only weak anomalies exist to the west.

There are at least 2 important regions of tropical convective forcing. One is nearly stationary centered around 10S/120E, and may be part of a slower process linked to a transition from El-Nino to La-Nina. The other is the dynamical signal with the MJO ~15-20S/160W. About 10 days ago these regions consolidated, and there were at least 2-3 tropical cyclones in the area of Australia left in the wake.

An important point is that a MJO signal remains. In fact, this signal can be traced back to late January 2007 in the near equatorial velocity potential Hovmollers courtesy of CPC. The signal is weakening especially in the convective field while it continues to evolve as expected in the upper level wind field (see below). The convective field now has multiple regions of tropical forcing with well defined anomalous twin upper tropospheric subtropical anticyclones near 120E and also near 160W (but mainly south of the equator).

At this time upper tropospheric zonal mean anomalous easterly flow continues to propagate poleward through the northern subtropical atmosphere, with magnitudes of ~10m/s around 25N (as a response to the MJO). This has contributed to a -2 sigma global relative angular momentum anomaly. Locally, there are weak upper tropospheric equatorial westerly wind anomalies starting to interact with the subtropics.

The MJO dynamical signal is moving east at about 10 deg long/day, and it may re-emerge into the Eastern Hemisphere by late week 2 or week 3. During this time zonal mean westerly flow should increase throughout the northern subtropics possibly allowing the East Asian Jet to undercut the recent blocking across the North Pacific. Convective flare-ups from the west central-South Pacific are also probable while a stationary convective signal may remain near 120E. Forecast uncertainty remains tremendously high after week 1 but I think we will see persistence of a GSDM Stage 1 anomaly pattern.

I think the models have a reasonable handle on week1 across the lower 48 states with ridge amplification just off the west coast and a central and eventually eastern states trough. This will keep the western states relatively warm and dry with cooler and wetter weather for the central and east USA. However, while I did underestimate in recent posts the impacts from the subtropical easterlies discussed above, I do think there will be more strong troughs to impact the western USA by late week 2 into week 3. The latter is most probable during GSDM Stage 1, and also during seasonal transition into spring. In fact, should the tropical convective forcing become quite intense across the Eastern Hemisphere during week 3 (leading to MJO #3 since December 2006?) with central Pacific flare-ups, an active regime for the USA may persist well into April.

Please note: These are probabilistic statements, which we will try to quantify in future posts. I am on TDY at ESRL/PSD with the HMT project until April 3rd. I will try to post another discussion next Tuesday. The WB (2007) paper on the GSDM has been published in the February issue of MWR.

Ed Berry and Klaus Weickmann

Tuesday, March 06, 2007

Not What I Expected

Tropical sea surface temperatures continue to cool across the equatorial Pacific Ocean while weak-moderate positive anomalies remain from the west central to the South Pacific. In fact, negative temperature anomalies as low as -2C have recently been observed at around 120W per TAO buoy data, extending to depths of at least 200m. An evolution to La-Nina is looking more probable.

The dynamical signal with the MJO continues to propagate east at roughly 5m/s (4 deg long/day) toward the west central and South Pacific per diagnostic monitoring tools and satellite imagery. For instance, as of March 5th the Wheeler index suggests this signal to be at ~160E. We prefer to describe the recent eastward shift as part of a weather-climate or subseasonal event. These typically consist of multiple time scales, one of which is the MJO but others of which are less oscillatory.

Tropical convective forcing has also been getting better organized from the central Indian Ocean to Indonesia during the last few days, centered at about 10S/120E. Hovmoller plots of both total and anomalous OLR suggest this forcing may become a stationary possibly linked to a developing cold event. The global circulation has responded accordingly with anomalous zonal mean easterly flow propagating off the equator into the subtropical atmosphere particularly for the Northern Hemisphere (magnitudes ~minus 10m/s for the latter). Global relative AAM remains very low at about 2 standard deviations below the 1968-1997 reanalysis climatology with its tendency at roughly minus 10 Hadleys. The recent peak of AAM tendency to plus 10 Hadleys was the result of the subseasonal signal discussed above.

As predicted, GSDM Stage 1 best describes the current global circulation. However, regionally from the Asian to North American sector the distribution of circulation anomalies has become very complicated. Animations of daily mean 150mb and 250mb vector wind anomalies show twin anticyclones centered around 115E with downstream troughs near the date line. However, there are also residual anomalies linked to past forcing from the tropical west Pacific including those across the Arctic. From a synoptic view point the ridge that is most probable to be centered ~140-150W in this situation has been split in two; one portion is at the date line with the other over western North America. Most ensemble means (not details) suggested this possibility 7-10 days ago. There is some hint from recent model runs for current Gulf of Alaska trough to slowly move east and southeast into western North America during the next couple of weeks. My own thought a week ago was for an enhanced probability for this process to happen sooner and farther south than the models were showing, allowing a resumption of an active regime for much of the country. This appears less likely in the immediate future.

Whatever the case over the next 7-10 days, I think forecast uncertainty for a week 2-3 prediction for the USA is very high, despite what appears to be good model agreement of a low amplitude trough across western North America and a northward displaced storm track. In fact, I think it probable that a strong synoptic scale trough will impact the western and central USA with the extreme weather typical of GSDM Stage 1 during weeks 2-3. This would mean at least one prominent “southwest flow cyclonic baroclinic event” across the Plains similar to the 2 recent events. Until then, a rather benign weather pattern with warm and dry across the southern states and a bit unsettled across the north can be expected.

Please note: These are probabilistic statements, which we will try to quantify in future posts. I am on TDY at ESRL/PSD with the HMT project until April 3rd. I will try to post another discussion this Friday. The WB (2007) paper on the GSDM has been published in the February issue of MWR.

Ed Berry and Klaus Weickmann

Friday, March 02, 2007

Brief Update -- Global models may perform like the recent global markets

The dynamical signal with the MJO is centered at ~10S/140E per several monitoring tools including the multivariate EOF Wheeler index and animations of circulation anomalies. Rough calculations of its phase speed give me anywhere from 5-8m/s depending on what variable you use; for example, velocity potential or OLRA. In any case, this recent MJO event has been a very difficult one to monitor and understand. In fact, I think what has happened during the past week is a consolidation between the MJO dynamical signal and westward propagating convection associated with west Pacific SST boundary forcing. The latter is a slower process that started about mid-January 2007 and may be part of the evolution of a transition from El-Nino to La-Nina.

My point in this short writing is I do not think the global models have “much of a clue” of the responses to the above. Global relative AAM tendency is ~plus 10-15 Hadleys per reanalysis data with much of that signal coming from the equatorial and subtropical atmospheres of the Western Hemisphere. Contributions are from the tropical forcing crossing Indonesia and an increase in the global frictional torque (while global relative AAM remains ~minus 2 sigma). I think the global circulation is trending toward GSDM Stage 2. This argues for ridge amplification across the eastern Pacific (~140W given seasonal cycle) during the next few weeks, most likely week 2. The global models (ensembles) do not depict this very well. They generally want to persist a trough across the Gulf of Alaska downstream of the central Pacific blocking.

I would not want to be a person investing money banking on “good” global model performance during the next few weeks. A decline in, for example, the skill of week-2 ensemble model means going into the middle of this month, appears probable (similar to the recent global markets?). Sources would include not only the excitation of a west Pacific wavetrain forced by Rossby wave energy dispersion tied to the east Indonesia/west Pacific tropical forcing, but also seasonal transition. My thought would be for the Gulf of Alaska trough to dig southeast along the USA west coast (impacting ARB) during week 2, then shift into the Rockies leading to a resumption of a stormy regime for much of the USA. This would be similar to recently observed, but with an Arctic cold air source. If this MJO signal stalls, say ~150E, then shifts back to west, this regime may persist for a period longer than 1 synoptic event.

Please note: These are probabilistic statements, which we will try to quantify in future posts. My next 1 month period at ESRL/PSD with the HMT project will be from 3/4-4/3. I will try to post a more complete discussion ~Tuesday next week while at ESRL/PSD. The WB (2007) paper on the GSDM has been published in the February issue of MWR.

Ed Berry