Saturday, May 31, 2008

Ghost Riders -- Part Deux

Global tropical SSTs remain well above average from the East Pacific into the Atlantic Oceans with anomalies ~1-3C and totals near 30C south of Mexico and about 29C along the equator east of South America. La-Nina associated SSTs are relatively weak with negative anomalies only ~0.5-1C along and east of the equatorial date line. Large-scale warm and cold anomaly spatial horseshoes remain well defined across the Pacific Ocean basin (negative phase of the PDO?). SST anomalies across the Eastern Hemisphere are also fairly weak (generally less than 1C) having a loose warm-cool-warm spatial pattern from the Indian into the west central Pacific Oceans. The latter connects to the warm Pacific Ocean horseshoe, with totals ~28.5-30C.


Latest 5-day averaged TAO buoy data continues to indicate expanding subsurface warmth (~plus 3C) down to ~200m west of the date line to west of South America (~plus 4C at 50m). Restating from a week ago, regardless of what tools are believed to be useful, the interannual SST component including ENSO is unclear. However, careful monitoring of upcoming subseasonal variations is extremely critical (only a handful of these issues can be addressed below given this presentation medium). These variations WILL impact the global weather including the USA, irrelevant of any “Nino or whatever” indices, classifications, and so forth. Anything to the contrary is not scientifically defensible.


The following are links to global SST and related information.


http://iri.columbia.edu/climate/ENSO/currentinfo/technical.html


http://www.pmel.noaa.gov/tao/jsdisplay/


http://www.cdc.noaa.gov/forecast1/IndoPacific.frcst.html (note the initial projection)


http://www.cpc.ncep.noaa.gov/products/precip/CWlink/MJO/index.primjo.html (link 18).


Full disk satellite imagery and other diagnostic monitoring tools show there has been a significant increase of intense-severe clusters of thunderstorms from the equatorial Arabian Sea and Indian Ocean east-northeast into Southeast Asia (3-day OLRA ~minus 50-70 W/m**2). Enhanced tropical rainfall persists across northern South America and particularly equatorial Africa. The nemesis of the west Pacific Ocean is still hanging around, with Typhoon Nakri spawned since my last writing.


Updated through 30 May, the latest WH (2004) phase space plot shows a much weaker MJO projection, barely above 1 sigma, than a week ago, in octant 1 (with and without the interannual component). Should this be interpreted as a decaying MJO? There are literally an endless number of responses I could offer to that question. However, the point is we have an important evolution of tropical convective forcing and associated circulation anomalies shifting from the Western into the Eastern Hemispheres.


Continuing on the above, strong tropical-extratropical coupling continues (recall last week’s discussion of events since mid April). As was observed earlier this year, once again there is a well defined inter-hemispheric meridional symmetry of zonally symmetric zonal mean circulation anomalies. However, there are some subtle differences (a few discussed below) involving the exchange of momentum between the solid earth and free atmosphere. The WB (2007, 08) GWO, which captures the global dynamical AAM processes including any MJO contribution, has orbited in phase space to nearly octant 3. Without further confusing the reader, another point is that the global circulation, at least for now, is returning to a La-Nina base state.


Updated through 29 May, global relative AAM has dipped to ~minus 1.5 AMUs below the R1 data climatology, with its tendency ~minus 5-10 Hadleys. While the global integral of the surface torques (and tendency) may not appear that large (~0 for the torques), that is anything but true for specific latitude bands of zonal mean frictional and mountain torques as well as the AAM transports.


This above is another example where I feel rather comfortable knowing “how we got here”, but have an extremely difficult time sharing that. It is only a simple matter for me to go back through numerous animations of many fields, daily averaged and longer periods, linking those observations to WB (2007, 07) GWO variations and be able to offer scientifically defensible null hypotheses.


Most broadly, the symmetry I am writing about involves anomalous zonal mean westerlies ~60-70N and S, as well as 30-35 N and S. Anomalous easterlies are present in the upper midlatitudes and the tropical and subtropical atmospheres. Hence the zonal mean frictional torque anomaly is ~plus 2 Hadleys near 45S and ~plus 1 Hadley around 50N. The latter is significant since this is opposite to what was seen during roughly February, and represents one of those subtle base state changes I am discussing. Obviously the overall gradual increase in global relative AAM since roughly early March in addition to the dynamical processes leading to MJO #5 are another example of slow base state changes.


Summing up, there has been a tendency for the WB (2007, 08) GWO to slowly shift away from La-Nina since mid March. Please keep in mind that while I type “La-Nina”, “El-Veijo”, or whatever, this is only a component of a complex global (ocean-land-atmosphere) interannual signal I am really talking about (to not have a “failure to communicate”). Remembering that, subseasonal events constructively and destructively interfering with this basic global circulation state (which are “unusual for La-Nina” – see past postings) have led to the barrage of anomalous western USA troughs since at least December 2007. Numerous high impact/extreme weather events of several types (tornadoes, blizzards, flooding rainfall, etc.) affecting most of the country have accompanied these troughs.


Okay, so where do we go from here? Zonal mean flux convergence of AAM transport (greater than 12 Hadleys) has maximized ~40N with sinks to the north and south. A similar process is occurring in the Southern Hemisphere. The dynamics tied to the tilts of midlatitude eddies interacting with the surface torques and the tropical forcing are at least partly responsible. Regional-scale responses have included a robust extension of the Australian jet (austral winter) and soon the jet across North Pacific Ocean. The latter is what the models are keying on for the next western USA trough to slam the northern and central Plains (etc.) with possibly another vicious and destructive round of severe local storms next week.


Animations of upper tropospheric daily mean vector wind anomalies continue to present the expected circulation responses consistent with the WB (2007, 08) GWO. Twin upper tropospheric tropical anticyclones are becoming established (coming back) in Africa and the western Indian Ocean, while cyclones develop in the region of the date line. Rossby wave energy dispersions (RWDs) are strongly zonally oriented across the midlatitudes.


My feelings are for the tropical convective forcing to become better established from roughly the Indian Ocean into the Southeast Asia (impacting the local monsoon systems) by week-2, if not sooner. I also think this enhanced rainfall may take on more of a west-northwest to east-southeast orientation from around India and particularly Southeast Asia into the west central and northwest Pacific Ocean weeks ~2-4 (also impacting Indonesia). Remember timing is white noise and I am offering a statement of probability. Should significant tropical forcing shift east along the equator with an accompanying westerly wind event, interannual ramifications may become non-trivial. Regardless, the latter is a real possibility (and worth the speculation), and may start the process of transitioning to an equatorial Pacific Ocean warm event.


The WB (2007, 08) GWO may not only orbit to phase 5 during the next 1-3 weeks, but even drift toward the “El-Nino side”. The WH (2004) MJO phase plot may also do a similar behavior. Focusing now on the USA, above average subtropical/midlatitude zonal mean westerly wind flow suggests more troughs to penetrate inland into the western states going well into June. Of course, these will be shifted northwest, and impact mainly the Northern Plains and Upper Mississippi Valley as they progress. Whatever the case, I do think the general pattern of “cool northwest-warm southeast” with wetness in the middle is probable through at least weeks 2-3. Included are possibilities of additional rounds of severe local storms (and flooding rainfall) including tornadoes perhaps focusing on Iowa and surrounding states. Taking into account the WB (2007, 08) GWO, should tropical forcing start to shift farther east during boreal summer, cool and wet may also shift into the Plains and even portions of the eastern states (ridge from Hell returns to the west coast?).


As discussed above, portions of India/Indonesia into Southeast Asia/China may be impacted by intense to severe thunderstorms during the next few weeks. During weeks 3-4 that area of enhanced rainfall may become focused from Southeast Asia into the Philippines to the west central Pacific Ocean. Please see the latest statements from the JTWC and TPC for tropical cyclone concerns. While the risk for the latter may decrease across the East Pacific Ocean and Caribbean weeks 1-2, there may an increased tropical cyclone risk for the northwest Pacific Ocean weeks 2-4. I trust the expertise of the appropriate meteorological centers internationally to alert the public of additional weather hazards worldwide.


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


This phase plot is being re-done, as is the web site. Stay tuned. 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 to the WH (2004) “convention”. In addition, the GWO plot does not have the ENSO signal removed.


Please see the revised description of the GSDM per above link. Also, I encourage the readers to study the annotated MJO and GWO phase space plots to help relate the global variations explained by those techniques to “weather”.


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/


The following is a link to information about the stratosphere and other nice monitoring tools:


http://ds.data.jma.go.jp/tcc/tcc/products/clisys/index.html


The following is a link to NCEP model verifications (surf around for lots more).


http://www.cpc.ncep.noaa.gov/schemm/z500ac_wk2_na.html .


The following is a link discussing recent global weather and related events.


http://www.wmo.ch/pages/mediacentre/news/index_en.html


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. In addition, a two-part paper is in ACTIVE preparation by WB that will formally introduce the GWO along with subseasonal composites for variables such as surface temperatures. We want to emphasize notions such as global-zonal mean-regional scale linkages as well as forcing-response-feedback (with subsequent interactions) relationships.


Given shift work and travel, updates are extremely difficult. I should be able to post another discussion the weekend of 7-8 June.


Ed Berry

Friday, May 23, 2008

Ghost Riders Storming the USA Plains (as expected)

Global tropical SSTs are well above average from the East Pacific into the Atlantic Oceans with anomalies ~1-3C and totals near 30C south of Mexico and in the Gulf of Guinea. While La-Nina associated SSTs have weakened considerably, large-scale warm and cold anomaly spatial horseshoes remain well defined across the Pacific Ocean basin (negative phase of the PDO?). SSTs across the Eastern Hemisphere are slowly warming, with totals ~30C across the western Indian Ocean including the Arabian Sea.


Latest 5-day averaged TAO buoy data indicates expanding subsurface warmth (~plus 3C) down to ~200m west of the date line and even 50m west of South America. However, regardless of what tools are believed to be useful, the interannual SST component including ENSO is unclear.


The following are links to global SST and related information.


http://iri.columbia.edu/climate/ENSO/currentinfo/technical.html


http://www.pmel.noaa.gov/tao/jsdisplay/


http://www.cdc.noaa.gov/forecast1/IndoPacific.frcst.html (note the initial projection)


http://www.cpc.ncep.noaa.gov/products/precip/CWlink/MJO/index.primjo.html (link 18).


Full disk satellite imagery and other aids do not indicate any concentrated region of enhanced tropical convective forcing. Instead, the most robust signal is the strong suppression covering much of the Indian Ocean into Australia. Three-day averaged OLRA have been ~plus 50W/m**2. There has been enhancement across the Western Hemisphere particularly in the regions of Northern South America and Africa. The Northwest Pacific remains sporadic.


Keying on the suppression as well as the 850mb and 200mb winds, the WH (2004) phase space plot indicates ~3 sigma projection (through 22 May) in octant 8. Even with the interannual component left in, the projection is still well over 2 standard deviations. There is other evidence to support what this equatorially confined empirically derived index is telling us, including the WB (2007, 08) GWO. The latter is a dynamical measure of the global circulation that links global-zonal mean-regional scale variations.


The point is that there is strong tropical-extratropical coupling once again occurring, including a very robust MJO signal in the Western Hemisphere and exchanges of momentum between the solid earth and atmosphere. Since about mid-April, dynamical processes captured by the WB (2007, 08) GWO have worked to add global westerly wind flow to the atmosphere. Part of these processes included the extratropics rejuvenating the current MJO. This has included not only a weakening of zonal mean subtropical easterly wind flow anomalies, but also westerly wind flow anomalies being added to the equatorial atmosphere. Hence global relative AAM recently increased too slightly above the R1 data climatology, the highest since ~ 1 December 2007.


During this month, these anomalous equatorial westerlies (~5m/s zonal mean anomalies at 200mb) have propagated poleward and downward particularly into the Southern Hemisphere. Interactions with the midlatitude storm tracks including Rossby wave energy dispersions (RWDs) tied to Northern Hemisphere north-south mountain massifs led to the recent remarkable extension of the East Asian jet and ridge from Hell along the USA west coast. In fact, the global mountain torque was ~plus 20 Hadleys ~ 11 May with much of that contribution from East Asia.


As was expected probabilistically 2 weeks ago, there is currently frictional dissipation of the added westerly wind flow. The global frictional and mountain torques are strongly negative leading to a computed AAM tendency of ~minus 25 Hadleys. Much of that negative global AAM tendency has ~1-2 Hadley zonal mean contributions from the equatorial and subtropical atmospheres. Hence the GWO has orbited in phase space to octant 1 and a regional-scale response has been the retrogression of the “PNA ridge” into the eastern Pacific Ocean. The anomalous North Pacific Ocean jet has collapsed into the unseasonably very strong western USA trough leading to another severe local storms outbreak on Plains as I write.


As stated in previous postings, the WB (2007, 08) GWO, which considers the MJO and other tropical variability, offers a global dynamical weather-climate linkage framework to intelligently evaluate the numerical models. A couple of weeks ago the feeling was that the positive PNA USA ridge pattern was unlikely to persist. What adds confidence to offering a prediction based on the WB (2007, 08) GWO is when the numerical models start to support what scientific thinking suggests. The Day-10 deterministic ECMWF solution as well as the ESRL/PSD ensemble were the first models to capture the “PNA retrogression”. I applaud particularly ECMWF for such as wonderful Day-10 forecast (for the USA). Nonetheless, other models such as the GEFS were slow to catch on, and because of model disagreement so were the official forecasts for particularly week-2. My feeling is that this was a blown forecast of opportunity for several reasons.


As tropical convective forcing returns to the Eastern Hemisphere during the next 1-3 weeks (with the MJO), the WB (2007,08) GWO is probable to circuit into phase 3 as zonal mean easterly wind flow anomalies increase in the tropical/subtropical atmospheres. Supportive are animations of fields such as 150mb daily mean vector wind anomalies. They indicate twin tropical anticyclones ~30-60W, with some “hints” of returning to the Indian Ocean. Thus there may be June rendition of a circumglobal teleconnection of midlatitude ridges as our EL-Viejo base state may get some “new life”. A serious and critical monitoring issue from the interannual perspective will be to see if the tropical MJO remains strong and shifts into the west central Pacific Ocean during the next several weeks (monsoon systems understood).


Shifting northwest with the seasonal cycle, more periodic and anomalous western USA troughs are probable for the next 2-3 weeks. At some point the “Desert Southwest” monsoon ridge will appear; however, that may not be until July (which is climatology). The storm track is probable to remain active from the northern and central Rockies into the Upper Mississippi Valley leading to more rounds of heavy rain and severe local storms. Summertime heat should become established across much of the southern USA. Finally, please see statements from the Tropical Prediction Center for possible upcoming USA tropical cyclone hazards. During weeks 1-2 both the East Pacific Ocean and Caribbean may have the latter.


Intense to severe thunderstorm activity is probable to spread east from equatorial Africa into the Indian Ocean weeks 1-2. In fact, most recent satellite imagery do show clusters of thunderstorms increasing in those regions. The warming SSTs discussed above are often a precursor to the active phase of the MJO. During weeks 2-4, intense rainfall may become established from the Arabian Sea into Southeast Asia, eventually focusing on the Indian and Asian monsoon systems. The tropical cyclone hazard still cannot be ruled out for both the Arabian Sea and Bay of Bengal. The west Pacific Ocean “should” remain relatively suppressed at least week-1. I trust the expertise of the appropriate weather centers internationally to alert the public of additional weather hazards worldwide.


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


This phase plot is being re-done, as is the web site. Stay tuned. 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 to the WH (2004) “convention”. In addition, the GWO plot does not have the ENSO signal removed.


Please see the revised description of the GSDM per above link. Also, I encourage the readers to study the annotated MJO and GWO phase space plots to help relate the global variations explained by those techniques to “weather”.


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/


The following is a link to information about the stratosphere and other nice monitoring tools:


http://ds.data.jma.go.jp/tcc/tcc/products/clisys/index.html


The following is a link to NCEP model verifications (surf around for lots more).


http://www.cpc.ncep.noaa.gov/schemm/z500ac_wk2_na.html .


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. In addition, a two-part paper is in ACTIVE preparation by WB that will formally introduce the GWO along with subseasonal composites for variables such as surface temperatures. We want to emphasize notions such as global-zonal mean-regional scale linkages as well as forcing-response-feedback (with subsequent interactions) relationships.


Given shift work and travel, updates are extremely difficult. I should be able to post another discussion the weekend of 31 May – 1 June.


Ed Berry

Thursday, May 15, 2008

Very Short Update -- Some Just do not Get It!

The following are links to global SST and related information. The interannual component (including ENSO) is unclear and only monitoring will tell. Stay tuned.

http://iri.columbia.edu/climate/ENSO/currentinfo/technical.html


http://www.pmel.noaa.gov/tao/jsdisplay/


http://www.cdc.noaa.gov/forecast1/IndoPacific.frcst.html (note the initial projection)


http://www.cpc.ncep.noaa.gov/products/precip/CWlink/MJO/index.primjo.html (link 18).


Full disk satellite imagery shows tropical convective forcing literally breaking up into tropical cyclones from the Bay of Bengal into the Northwest Pacific Ocean. Thunderstorms are increasing from the East Pacific ITCZ all the way to western Africa, while the Indian Ocean remains suppressed. As demonstrated by the WH (2004) phase space plot, a robust (not weak, per official sources) MJO (can be viewed as a “component” of the WB (2007, 08) GWO) signal is propagating into the Western Hemisphere and the tropical forcing is responding. The MJO, working with dynamical processes explained by the WB (2007, 08) GWO, has resulted in an increase of global relative AAM (updated through 13 May) too slightly above the R1 data climatology. A zonal mean response has included a fundamental shift of tropical/subtropical circulation anomalies leading to twin anticyclones just west of date line. Regionally, an anomalous extension of the East Asian jet has occurred (wind speed anomalies ~50m/s at 250mb) supporting a ridge (from Hell) along the North American west coast (strong positive phase of the PNA for this time of year).


The first point to make is that MJO variability is currently strongly impacting North America, and will continue to do so for ~weeks 1-3. Any statement to the contrary is scientifically not defensible. Secondly, I am quite confident that tropical convective forcing will return to equatorial Africa and the Indian Ocean (west Pacific Ocean nemesis understood) during weeks 2-3. Sparing the “dynamics lesson”, the GWO is probable to collapse to octant 1 then orbit in phase space to octant 3 during the next few weeks. The physical processes responsible (ex., surface torques and AAM transports) will remove zonal mean westerly wind flow (anomalies ~5-10m/s at 200mb) from the tropical and subtropical atmospheres. In fact, the El-Viejo base state may get rejuvenated during the next few weeks. The latter suggests a return to anomalous twin tropical upper tropospheric anticyclones (cyclones) in the region of the Indian Ocean (date line) and perhaps another Branstator (2002) circumglobal teleconnection of midlatitude ridges.


For the USA the above means the anomalous North Pacific Ocean jet is probable to collapse into an energetic western states trough by week-2 leading to a southwest flow storm track on the Plains. This situation may persist through weeks 3-4, shifting northwest with the seasonal cycle. The notion of this retrogression has already been discussed in past postings.


Official forecasts have not yet picked up on this (at the time of this writing), being fooled by model disagreement, etc.. This again demonstrates the need of an atmospheric dynamics based weather-climate linkage framework to intelligently evaluate numerical model guidance (insert my angry Rottweiler)! One purpose of the WB (2007, 08) GWO is to facilitate the latter. There is no cookbook to making subseasonal forecasts, and that includes blending numerical model predictions!!! Ensemble prediction schemes such as from ESRL/PSD as well as the deterministic ECMWF model have provided the most realistic solutions for the USA. Link below is verification for the GEFS.


http://www.cpc.ncep.noaa.gov/schemm/z500ac_wk2_na.html .


The final point is that a potentially explosive pattern for severe local storms on the Plains exists for at least week-2 (exact timing is white noise). Locations to get hit the hardest may be farther north than observed during the last few weeks (climatology understood). The latter may extend from portions of Oklahoma into Iowa then into the Ohio Valley. Late season heavy snow (with intense thunderstorms) may also pound portions of the Rockies. While the western states cool down, the southeast should warm to above normal. Other weather ramifications should be well known by now. The latest official week-2 forecast for the USA does not indicate the latter, and I strongly disagree based on scientific reasons.


Locations from portions of the Bay of Bengal into Indonesia/Southeast Asia and the Philippines are still likely to get hammered with intense to severe thunderstorm activity week-1 including tropical cyclones (please see JTWC). This activity is probable to weaken weeks 2-3 as tropical thunderstorm activity increases from Africa into the Indian Ocean. By weeks 3-4 intense-severe thunderstorm activity may extend from the Arabian Sea into Bay of Bengal and Southeast Asia.


The best shot for enhanced Western Hemisphere convection, impacting northern South America into the Atlantic ITCZ appears to be weeks 1-2. I trust the expertise of the appropriate weather centers internationally to alert the public of additional weather hazards worldwide.


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


This phase plot is being re-done, as is the web site. Stay tuned. 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 to the WH (2004) “convention”. In addition, the GWO plot does not have the ENSO signal removed.


Please see the revised description of the GSDM per above link. Also, I encourage the readers to study the annotated MJO and GWO phase space plots to help relate the global variations explained by those techniques to “weather”.


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/


The following is a link to information about the stratosphere and other nice monitoring tools:


http://ds.data.jma.go.jp/tcc/tcc/products/clisys/index.html


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. In addition, a two-part paper is in ACTIVE preparation by WB that will formally introduce the GWO along with subseasonal composites for variables such as surface temperatures. We want to emphasize notions such as global-zonal mean-regional scale linkages as well as forcing-response-feedback (with subsequent interactions) relationships.


Given shift work and travel, updates are extremely difficult. I should be able to post another discussion the weekend of 24-25 May.


Ed Berry

Saturday, May 10, 2008

Hell Ride Update

Numerous issues dictate that this writing must be shorter. In fact, given matters briefly mentioned in the Appendix particularly scientific concerns, brevity may become the norm (perhaps with more updates). Stay tuned. I will never be close to satisfied with these discussions until we have a properly supported dedicated web page effort (with figures, composites, verification, etc…..). The latter is the right and only way to provide useful information about the dynamics of subseasonal atmospheric variability, having ~week1-4 predictions expressed probabilistically for users to make important decisions.


There is little change to the global SSTs as discussed in my 3 May posting. Perhaps most notable is the westward shift of equatorial subsurface warmth, ~4-5C at 150-200m depth from 140-160E per 5-day averaged TAO buoy data. That is “not good” for those cheering for El-Nino. However, May is a critical month for ENSO variability given Indian Ocean-Asian monsoon issues, annual cycle of solar input to the tropical oceans, etc. Also, as more folks are starting to recognize, all wide tropical ocean basins (and outside) have global-regional scale circulation impacts. Please see links below.


http://iri.columbia.edu/climate/ENSO/currentinfo/technical.html


http://www.pmel.noaa.gov/tao/jsdisplay/


http://www.cdc.noaa.gov/forecast1/IndoPacific.frcst.html (note the initial projection)


http://www.cpc.ncep.noaa.gov/products/precip/CWlink/MJO/index.primjo.html (link 18).


The story with the tropical convective forcing is complicated, and it should be. Arguably a weak “rejuvenation” of MJO #5 (for the past boreal cold season) occurred during the last few weeks. The WH (2004) MJO phase space plot suggests a projection in octant 6 (barely so leaving ENSO in) updated through 9 May. Dynamical processes explained by the WB (2007, 08) GWO contributed to the latter, and led to severe Tropical Cyclone Nargis and currently Super Typhoon Rammasun. A westerly wind burst on the equator accompanied the tropical forcing as it propagated into the west central Pacific Ocean. Intense-severe thunderstorm activity has now weakened across the west central Pacific Ocean as has the SPCZ. The lower tropospheric westerly wind anomalies are also shifting off the equator.


Full-disk satellite imagery has shown a nice consolidation of enhanced rainfall centered ~120E from 5-15N during the last few days. The region of intense-severe tropical convection extends from the Bay of Bengal into the northwest Pacific Ocean. This is farther north than climatology, and may signal an early onset to the monsoon systems in those regions, particularly Southeast Asia.


As demonstrated by the WB (2007, 08) GWO, since mid April global westerly wind flow has been increasing. That has produced anomalous zonal mean westerlies in the equatorial and subtropical atmospheres while removing anomalous midlatitude easterlies. Hence the global relative AAM has increased to near the R1 data climatology putting a “dent” in our El-Viejo base state. One circulation response has been the extension of the East Asian Jet (EAJ; for this time of year having 250mb wind speed anomalies ~30-40m/s at ~40N) and a negative phase of the NAO (latter not by some cookbook process “thinking”).


Tropical forcing returning to the Eastern Hemisphere and physical processes tied to the earth-atmosphere AAM budget (mainly the frictional and mountain torques) are now removing global westerly wind flow. That does include the subtropical atmospheres, where zonal mean AAM tendencies are ~minus 4 Hadleys around 10-20N through 8 May. Hence, the WB (2007, 08) GWO is orbiting in phase space toward octant 1, meaning global relative AAM is probable to decrease during the next few weeks.


After weeks of struggling, the numerical models are performing better (link below). Their prediction of the “wicked ridge of the USA west coast (ridge from Hell per others)”, a large pattern change from the past several weeks, for the next 7-10 days is reasonable (along with the downstream USA storm discussed 3 May). In fact, Whitaker and others (ex., MWR 1998) have suggested, from an ensemble numerical modeling perspective, the positive phase of the PNA to be a relatively predictable pattern for roughly week 1-2 time scales. My thoughts are to understand why this appears to be the case, and believe the large-scale character to extended EAJs is a contributor (keeping in mind the dynamical processes that cause that response).


http://www.cpc.ncep.noaa.gov/schemm/z500ac_wk2_na.html .


While having underestimated the above mentioned “wicked ridge (arguably a systematic bias of mine!)”, there is no change to my outlooks from 3 May. By ~ weeks 2-3, I do think tropical convective forcing will become more robust across the Indian Ocean while possibly remaining intense from northern Indonesia/Southeast Asia-west central Pacific Ocean. The GWO may circuit through phases 1-2 into phase 3 by that time. A probable regional scale response will be for the +PNA pattern to slowly retrograde weeks 2-3. Thus anomalous cold/wet likely to dominate much of the eastern two-thirds of the USA should shift northwest to what has been a familiar regime since at least last January. Weather ramifications should be understood. Whether or not there are additional episodes of adding global westerly wind flow and perhaps a base state change linked to interannual variations (including ENSO) is unclear.


Locations such as western Kansas had significant welcome rainfall last week. The anomalous zonal mean westerly wind flow (~5m/s at 200mb) leading to a regional-scale response of a strong STJ into the USA Desert Southwest was a direct contributor. However, this upcoming week looks generally dry (white noise "surprises" can "always" happen), and what may become a northwest shifted storm track by week-3 means the dryness concern is still there.


Locations from portions of the Bay of Bengal into Indonesia/Southeast Asia and the Philippines are likely to get hammered with intense to severe thunderstorm activity week-1, and may signal an early onset to the wet monsoon phase. This region of severe tropical rainfall may expand into the west central and South Pacific Ocean weeks 2-3, while persisting across especially Southeast Asia and the Philippines. Locations from east Pacific Ocean ITCZ into South America are likely to be sporadic weeks 1-2. Per above, what may be a poor excuse of a dynamical MJO signal linked to the GWO appears probable to increase convection across the Indian Ocean (and perhaps portions of equatorial Africa) weeks 2-3. As mentioned last week, our nemesis of 2 dominate regions of tropical forcing may return going into boreal summer.


Please see the latest official JTWC statements about Super Typhoon Rammasun. While an on-going oncern for the northwest Pacific Ocean including the South China Sea the next few weeks, the risk of tropical cyclone development may again increase by week-2 across the Bay of Bengal (perhaps sooner). That would include coastal sections of Myanmar. I trust the expertise of the appropriate weather centers internationally to alert the public of additional weather hazards worldwide.


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


This phase plot is being re-done, stay tuned. 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 to the WH (2004) “convention”. In addition, the GWO plot does not have the ENSO signal removed.


Please see the revised description of the GSDM per above link. Also, I encourage the readers to study the annotated MJO and GWO phase space plots to help relate the global variations explained by those techniques to “weather”.


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/


The following is a link to information about the stratosphere and other nice monitoring tools:


http://ds.data.jma.go.jp/tcc/tcc/products/clisys/index.html


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. In addition, a two-part paper is in ACTIVE preparation by WB that will formally introduce the GWO along with subseasonal composites for variables such as surface temperatures. We want to emphasize notions such as global-zonal mean-regional scale linkages as well as forcing-response-feedback (with subsequent interactions) relationships.


Given shift work and travel, updates are extremely difficult. I may not be able to post another discussion the weekend of 24-25 May. Stay tuned.


Ed Berry

Saturday, May 03, 2008

Exorcism Falling Short???

SSTs from the tropical Indian into the Pacific Ocean basins loosely exhibit a spatial cool-warm-cool anomaly pattern. Anomaly magnitudes are approximately 1-2C for the cool and roughly 0.5-1C for the above average waters, latter centered in the region of Indonesia. SST totals are generally in the 28-30C range west of the date line. An interesting observation is that this spatial pattern of Indo-Pacific SSTAs is essentially opposite to that during boreal fall-winter 2006-07.


The La-Nina component of these SSTs has weakened considerably during the past 1-2 months. Latest 5-day averaged TAO buoy data shows that these cool anomalies extend only to a depth of ~100m, while positive anomalies ~4-5C persist at approximately 200m deep west of the date line. The warm horseshoe spatial SSTA pattern is still well defined from Indonesia into the extratropics of the North and South Pacific Ocean basins, while local El-Nino SST conditions are present west of South America. The reader can refer to the links below for additional SST details.


The point I want to make is there still exists no observational evidence of a coherent evolution toward an equatorial Pacific Ocean warm event. A few of these matters will be discussed below. In general, there are an endless number of scenarios anyone can propose for global SSTs numerically, statistically, both, etc. These can include the IOD, ENSO, PDO, AMO, and so forth. Whatever the case, confidence in any predictive scheme is currently low no matter what is offered. Careful detailed rigorous daily monitoring within a dynamical framework of weather-climate linkage is a must to gain some understanding on the future of El-Viejo, etc. The current global circulation base state is still La-Nina, but recently weakened. The following are links to additional SST information.


http://iri.columbia.edu/climate/ENSO/currentinfo/technical.html


http://www.pmel.noaa.gov/tao/jsdisplay/


http://www.cdc.noaa.gov/forecast1/IndoPacific.frcst.html (note the initial projection)


http://www.cpc.ncep.noaa.gov/products/precip/CWlink/MJO/index.primjo.html (link 18).


Full disk satellite imagery and other diagnostic monitoring tools show that the Eastern Hemisphere tropical convective forcing has been getting better organized during the last 5-7 days in the region of Indonesia. In general, this region of intense to severe tropical thunderstorm activity, responsible for recent high-impact weather including what was category 3-4 cyclone Nargis, extends from the eastern Bay of Bengal into the southwest Pacific Ocean. There has been some east-northeast drift of this forcing; however, it is generally stationary. Sporadic but at times intense thunderstorm activity continues across portions of tropical South America and especially Africa. No matter what anyone wants to “use”, there is little projection onto a MJO.


The WB (2007, 08) Global Wind Oscillation (GWO), a much better dynamical measure of the atmospheric circulation than, for example, an empirically derived equatorially confined MJO index designed to isolate ~10-20% of tropical rainfall and zonal wind variations (on average), is more appropriate to monitor the current situation. Recall that the GWO considers all tropical convective forcing including the MJO, in addition to red noise extratropical processes.


A well defined orbit in phase space to nearly octant 5 occurred ~23 April (3-day averaged), before collapsing approaching phase 1, updated through 1 May. This trajectory through GWO phase space was expected based on a friction-mountain torque index cycle discussed in my 19 April posting. Recall a week-2 low confidence probabilistic prediction of literally a ferocious USA Great Plains baroclinic storm leading to multiple high-impact weather was offered. The weather events during the past few days speak for themselves.


The forecast discussed above was well before ANY of the models caught on. In fact, the week-2 NCEP GFS ensemble mean forecast had a 500mb height ACC of nearly minus 0.3 for the North American sector this past week (see link below). Also, this prediction was offered well before anything “official”. My point is not to “pat myself on the back”. Past blog postings also speak for themselves of the many poor assessments I have made. However, there is, on average, additional skillful information for making subseasonal predictions based on weather-climate linkage and some understanding of atmospheric dynamics. The models are not stand alone! They are only a component of a forecast process. WB will demonstrate the latter scientifically and objectively once needed resources are given.


http://www.cpc.ncep.noaa.gov/schemm/z500ac_wk2_na.html .


A lot of complex dynamics involving the earth-atmosphere angular momentum budget including the surface torques and transports, as well as Rossby wave energy dispersions (RWDs) “spreading information around”, continue to occur as I type. I want to be as brief as possible in the following.


During the past couple of weeks, the same processes responsible for the recent upward GWO orbit have added anomalous zonal mean westerly wind flow (by definition) to the upper tropospheric tropical and subtropical atmospheres (~3m/s weekly average at 250mb). Much of the latter has come from twin cyclones near the date line and the Western Hemisphere Pacific Ocean. In fact, global relative AAM updated through 1 May per ESRL/PSD R1 data plots show the global integral just barely below climatology. In a sense, the persistent El-Viejo base state has been “dented”. There is evidence from the animations and other tools that this westerly wind flow is propagating poleward (and downward) into both hemispheres. That includes anomalous westerlies across central Asia. As discussed below, I do expect the North Pacific Ocean jet to extend (as a response) during the next couple of weeks.


The GWO is collapsing into phases 1-2, with a global mountain torque of ~minus 10 Hadleys including East Asia contributing. The AAM transport signal showing a strong zonal mean sink around 60N and source near 40N is consistent (in a complicated way) with this. In fact, the higher latitude sink is tied to recent blocking structures including a negative phase of the NAO. The latter has been forced from RWDs tied to Eastern Hemisphere tropical forcing and the recent upward circuit of the GWO discussed above. My point is to show a quick example of the NAO being tied to red noise processes. In other words, this is a response to decay time scale dynamical forcing, and not a cookbook for making subseasonal forecasts (my rottweiler dog from Hell is growling)!


Latest animations of various daily mean vector wind anomaly fields give weak evidence of a zonal wave number 2 pattern of tropical circulation anomalies. However, the most dominate features are the Eastern Hemispheric baroclinic mode consisting of twin tropical upper tropospheric anticyclones near 100E with cyclones near the date line. Zonally oriented chains of anomalies generally dominate the global midlatitudes consistent with the notion of a low AAM base state circumglobal teleconnection (Branstator 2002; there is also a high AAM base state rendition).


Having gone through a lot of details, I do see some evidence that another circuit of the GWO is probable during the next 2-3 weeks. I think it will involve a mountain-frictional torque index cycle and that the GWO orbit will be shifted toward phases 2-4. The tropical convective forcing will be tied to this GWO evolution. It is only a matter of time before the Indian Ocean becomes active. My feeling is during the next few weeks, while the Indian Ocean convection increases, the west central into the South Pacific Ocean will remain active. However, I do not think there will be an abrupt eastward shift of the current Eastern Hemisphere tropical convective forcing toward the date line having a robust WWB, etc. The latter is one behavior needed to “exorcise El-Viejo”.


One or two relatively weak western USA troughs are probable during the next 1-2 weeks. Understanding the seasonal cycle, a much more energetic system, linked to an extended North Pacific Ocean jet, may first slam the USA west coast then the Plains by week-3. Feedbacks from higher latitude blocking (and any west Pacific Ocean convection) will tend to keep the storm track depressed south along with much of the country being cooler than normal at times. Other weather ramifications are much more than obvious by now. However, I will emphasize that another Rockies-Plains high impact weather situation appears probable (above May climatology) during ~ weeks 2-3. At some point perhaps WB can show online GWO composites for forecasters to judge for themselves.


My concerns of prolonged dryness remain for the central and southern High Plains. The latest drought monitor from CPC shows increasingly large areas of severe drought. Unfortunately, another situation of a dry intrusion with high wind and dust greeted portions of western Kansas this past week. I continue to hold out some “hope” for rainfall for these areas. The added subtropical westerly wind flow will help. For example, to reduce the likelihood of USA cold fronts penetrating the deep tropics to as far south as at least the Panama Canal disrupting moisture transport. There were at least 2 occurrences of that during April.


Locations centered on Indonesia including Southeast Asia and the Philippines are likely to get hammered with intense to severe thunderstorm activity week-1, shifting eastward weeks 2-3 into the west central and South Pacific Ocean. While the tropical cyclone hazard for the Bay of Bengal appears diminished, that may not be the case for particularly the northwest Pacific Ocean including the Philippines by weeks 2-3 (could be sooner). Portions of equatorial Africa including the Gulf of Guinea are also getting intense precipitation. As discussed above, that activity may shift east into the Indian Ocean during weeks 2-3. I continue to leave it to the expertise of the appropriate weather centers internationally to alert the public of additional weather hazards worldwide.


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


This phase plot is being re-done, stay tuned. 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 to the WH (2004) “convention”. In addition, the GWO plot does not have the ENSO signal removed.


Please see the revised description of the GSDM per above link. Also, I encourage the readers to study the annotated MJO and GWO phase space plots to help relate the global variations explained by those techniques to “weather”.


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/


The following is a link to information about the stratosphere and other nice monitoring tools:


http://ds.data.jma.go.jp/tcc/tcc/products/clisys/index.html


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. In addition, a two-part paper is in ACTIVE preparation by WB that will formally introduce the GWO along with subseasonal composites for variables such as surface temperatures. We want to emphasize notions such as global-zonal mean-regional scale linkages as well as forcing-response-feedback (with subsequent interactions) relationships.


Given shift work and travel, updates are extremely difficult. I will try to post another discussion next weekend, 10-11 May.


Ed Berry