Saturday, November 29, 2008

La-Nina Roper-Doping the Models???

Please keep in mind the ESRL/PSD GSDM web link, below, while reading this discussion.


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


The 91-day signal to noise ratio (snr) anomaly composites are currently centered on 23 November due to technical issues. There are also numerous other delayed real-time products such as the AAM plots and vector wind animations. We hope these problems will be resolved soon. WB (2009; likely not to be published until next year), part-1 of a 2 part paper where the GWO is formally introduced, is in press for MWR publication. There is a link in the Appendix to download the manuscript.


Please remember that these discussions are part of an experimental effort involving linking weather and climate. Until this work is formalized at the national level, many important scientific issues we are easily well aware of cannot be addressed. Stated another way, it is nearly impossible for me to talk about “everything” in these postings, including offering attribution to unpredictable noise. Also, plans are moving forward to have a one-day workshop on the WB (2009) GWO concepts during February 2009 in Boulder. Stay tuned for details.


There is little overall change in the spatial pattern of global SSTs. The warmest waters extend from north of Australia to near New Guinea, with totals ~30C and anomalies roughly plus 2C. SSTs 28-29.5C covers much of the TNWP southeast of the Philippines and portions of the Indian Ocean. Negative daily mean SSTAs ~1-2C (28 Nov 2008) persist within 5 degrees of the equator from about 140E to 140W, where trades have been strongly enhanced since at least early October.


How much additional SST cooling that occurs in all Nino regions is unclear. However, there has been a decided tilt toward La-Nina for at least the last 3-4 weeks, responding to a global circulation (interacting with the global oceans and land masses) that has been exhibiting those same characteristics arguably since boreal spring 2007.


http://www.wmo.int/pages/prog/wcp/wcasp/enso_update_latest.html


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)


http://www.cdc.noaa.gov/map/images/sst/sst.long.time.gif


There is no change from the scientific matters discussed in my long 22 November 2008 “treatise”. Lack of up to date AAM/GWO and related plots will also limit this writing.


Tropical convective forcing has shifted east to ~0/130-140E essentially covering most of Indonesia into northern Australia. Strong suppression exists across both the equatorial Indian Ocean and west central Pacific regions. About a month ago I wrote that the former should be enhanced at this time, which is an example of what unpredictable noise does to an outlook (more said below). In any event, this spatial pattern of Eastern Hemisphere tropical forcing is consistent with a La-Nina situation.


Intense to severe thunderstorm activity across tropical South America has resulted in deadly flooding conditions across southern Brasil during the past week. Rainfall has also been increasing across South Africa.


While there is a MJO component to the Eastern Hemisphere tropical forcing, ~1.5sigma projection in octant 6 (weaker projection in octant 5 leaving the base state in) of WH (2004) phase space, the wind signal continues to dominate. The enhanced Western Hemisphere tropical forcing is a direct response. There is still a lack of good coupling between the wind and convective signals.


In fact, reconstructed Hovmollers utilizing the WH (2004) RMMs show generally greater than 10m/s westerly wind flow anomalies at 200mb across the Western Hemisphere within 15 degrees of the equator. These same anomalous westerlies are coming back into the Indian Ocean, contributing to the convective suppression. Further, zonal mean westerly wind anomalies at 200mb across the tropical and subtropical atmospheres are ~5-10m/s, largest at 30N and 25S. There is evidence that this anomalous westerly wind flow is starting to propagate poleward into both Hemispheres.


If updated, I suspect that global relative AAM is still a good standard deviation above the R1 data climatology, and that the AAM tendency signal will not be that strong. In other words, the WB (2009) measure of the GWO may be undergoing a fast variation (looping) in octants 7-8-1 of phase space. The extended North Pacific jet/split flow situation this past week across North America was one synoptic response (giving California heavy precipitation).


Directly linked to the Eastern Hemisphere tropical forcing per above, I believe that a meridionally oriented Rossby wave energy dispersion (RWD) is arcing across the PNA sector (animations not updated). This is likely the signal that all numerical models have been playing catch-up to for the last several days. A synoptic response will be a discontinuous retrogression of the wicked ridge of the west from Hell currently just off the USA west coast to ~140W by the middle of next week.


For the past 5-6 weeks I have offered the notion of some form of an 8-1 transition in WB (2009) GWO phase space followed by a December cold regime focused on the middle of the USA. Within the envelope of probabilistic statements, this notion appears to be on track. However, the details of getting here that I have served up have been “anything but”. I do not “conveniently forget” a synoptic evolution suggested 5-6 weeks ago that did not work out. My timing was off by 1-2 weeks, for starters. Additionally, while retrogression of the west coast ridge was expected per above, I did not think that would be a response to "lingering" Indonesian tropical convective forcing.


This only reminds us of the enormous work that is always needed involving linking weather and climate, especially when dealing with red noise dynamical processes represented by the WB (2009) measure of the GWO. These forecast "busts" also offers opportunities to learn. However, I do emphasize these are probabilistic statements, and synoptic notions are presented “to communicate” the best information possible to weather sensitive users. These blog discussions are not very efficient to optimize weather-climate “insights” that can be given by the WB (2007, 2009) dynamical framework (ex., subseasonal “maps” and plume diagrams with real-time live discussions would be far more effective).


I suspect during the next 1-2 weeks the Western Hemisphere wind signal, explained by the WB (2009) measure of the GWO, will come back into the Eastern Hemisphere and excite intense tropical convection across the Indian Ocean. This is likely to initially appear across South Africa, and then shift east. A portion of the Indonesian enhanced rainfall is probable to shift into the warm southwest Pacific Ocean along a westward shifted SPCZ. By roughly weeks 3-4 much of the Indian Ocean into Indonesia ~80-120E should be strongly convectively enhanced. Perhaps there will be better coupling of both the wind and convective signals, within our quasi-stationary La-Nina base state.


An evolution shown by phases 8-1 of the GWO 250mb snr streamfunction composite anomaly plots is likely week-1. That translates to the discontinuous retrogression discussed above of the west coast ridge leading to trough amplification across the Rockies and Plains. Understanding tropical moisture transport through the Gulf of Mexico may be limited; a significant winter storm focusing on the Mississippi Valley must be a concern for the middle of next week (developing event currently across central and eastern USA understood). The coldest airmass to penetrate the lower 48 states so far this new winter season is also likely.


As the tropical forcing returns to the Indian Ocean, AAM will likely decrease to (well?) below the R1 data climatology and may be accompanied by a large negative AAM tendency. The zonal asymmetry of tropical circulation anomalies having Indian (west Pacific) Ocean twin upper tropospheric anticyclones (cyclones) will become enhanced. A strong zonal mean signal of anomalous subtropical easterly wind flow may also return, increasing the odds of another positive phase of the Branstator (2002) circumglobal teleconnection of anomalous midlatitude ridges and anticyclonic wave breaking troughs.


Regardless of the details, I do think it is probable to have a global circulation best represented by ~phases 2-4 of the GWO and MJO 250mb snr composite psi anomaly plots by the latter half of December. Synoptically across the PNA sector, this suggests a continued westward shift of extratropical circulation anomalies with the ridge perhaps to ~150W and troughs initially impacting the USA west coast. The latter should bring welcome precipitation to hopefully most of the west coast (be patient out there – good things come to people who wait!). Impacts from fast GWO variations are unclear. What is also unclear are the circulation anomalies for JFM; however, a rendition of what was observed during boreal winter 2007-08 with added subtropical westerly wind flow may be an option.


In general, USA weather ramifications were discussed in my last posting. Numerical models are going to continue to struggle in this kind of regime having tropical forcing returning to the Indian Ocean (on the list of work for WB to objectively show). I think there is an increased risk of high impact weather of many types during most of next month, and models may not catch them until “the last minute”. With variations, the emphasis may be on the western 2/3rds of the country especially weeks 3-4 (not suggesting the East Coast will be “quiet”).


Internationally, intense to severe rainfall is likely to continue across much of tropical South America including Brasil week-1 shifting into South Africa by week-2. Enhancement of thunderstorm activity is also likely at least week-1 across the southwest Pacific Ocean islands. Per above, moist convection may become less enhanced across Indonesia and northern Australia week-1 in favor of intensification back to the west across the Indian Ocean by week-3. There may be a focus of tropical convection ~100-120E during the latter half of December into January 2009 (while shifting into the Southern Hemisphere). Again, the warm southwest Pacific Ocean may become the “wild card” this boreal cold season.


Please see the latest official tropical cyclone forecasts for all basins. Currently that risk is unclear. I trust the expertise of the appropriate meteorological centers to alert the public of additional weather hazards worldwide. For instance, more strong troughs are digging into western Europe.


Appendix


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. We hope that an opportunity will arise for us (soon) to allow our dedicated web page effort to mature, expediting objectively and accountability. This web page effort will hopefully include an objective predictive scheme for the GWO with hindcasts.


The WB (2007) paper on the GSDM has been published in the February issue of MWR. The first of a two-part paper, where WB formally introduce the GWO (WB (2009)), has been accepted for publication MWR. A pdf of the in press version can be downloaded from the following link:


http://www.cdc.noaa.gov/map/clim/wb08_revised_final.pdf


In addition to the subseasonal snr composite anomaly plots, we hope near real-time discussions with “weather maps” will become a routine part of the ESRL/PSD GSDM web site sometime soon. Part-2 of our GWO paper will discuss the latter. We want to emphasize notions such as global-zonal mean-regional scale linkages as well as forcing-response-feedback (with subsequent interactions) relationships. An important purpose is to provide a dynamical weather-climate linkage framework to evaluate the numerical models in a sophisticated manner as part of a subseasonal (and any time scale) forecast process, in addition to a climate service for all users. Relying on the numerical models alone is a cookbook! I plan on posting a discussion the weekend of 6-7 December.


Ed Berry

Saturday, November 22, 2008

No Dancing – SSTs Responding to La-Nina; Increasing Risk for USA Extreme Winter Weather Weeks 2-5

The views expressed are those of the author and do not necessarily represent those of the National Weather Service.”


Please keep in mind the ESRL/PSD GSDM web link, below, while reading this discussion.


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


The 91-day signal to noise ratio (snr) anomaly composites are currently centered on 10 November due to technical issues. This should be resolved soon meaning resumption of daily updates, and will have minimal impact on these assessments. Please see product descriptions. WB (2008), part-1 of a 2 part paper where the GWO is formally introduced, is in press for MWR publication. There is a link in the Appendix to download the manuscript.


Remember that these discussions are part of an experimental effort linking weather and climate. Until this work is formalized at the national level, many important scientific issues we are easily well aware of cannot be addressed. Stated another way, it is nearly impossible for me to talk about “everything” in these postings. Also, there are tentative plans to have a one-day workshop on the WB (2008) GWO concepts ~February 2009 in Boulder. Stay tuned for details.


The warmest SSTs globally extend from southern Indonesia to the north coast of Australia having totals at least 29-30C and positive anomalies in excess of 2C ~10-15S. The Indian Ocean has cooled as has the TNWP. The former is from severe tropical convection and latter East Asia cold surges. Significant negative anomalies extend from ~140W to at least 160E within 10 degrees of the equator. Daily anomaly magnitudes on 21 November were ~minus 1-2C with totals well under 29C. In fact, responding to continued enhanced trades since mid September, latest 5-day averaged TAO buoy data shows negative SST anomalies 0.5C and lower from ~140W to 160E within 2 degrees of the equator (please look up the relevant Nino regions).


Yes, we understand the subsurface issues and the concerns raised by the PDO crowd, latter measure projecting on an extreme negative phase. Regardless, there is always constant interaction between the global atmosphere and SSTs. The global tropical SSTs contributed to forcing the atmosphere into the low AAM side of GWO phase space (La-Nina characteristic) perhaps as early as December 2006. With variations, this has not changed. Currently, the equatorial central and eastern Pacific Ocean SSTs are responding to a global circulation state favoring coolness in those regions. This is why measures such as Nino 3.4 to define “ENSO neutral” have limited scientific defensibility, let alone basing seasonal forecasts on them.


Referring to my title, just like the 1974 Ali-Foreman fight when the former did not dance after round 1, I am not clowning around with phrases such as “El-Viejo”, “witch”, “Rottweiler statements”, etc. for this discussion. The earth is not flat, center of the universe, etc., meaning care should be exercised thinking that any weather-climate phenomena such as ENSO has to follow a "classic cycle" (ex., Rasmussen and Carpenter (1982)). While encore events tend to be weaker (there are very important sample size issues), everything I and others understand tell us, from the global ocean-land-atmosphere perspective monitoring subseasonal variability, La-Nina is back.


http://www.wmo.int/pages/prog/wcp/wcasp/enso_update_latest.html


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).


http://www.cdc.noaa.gov/map/images/sst/sst.long.time.gif


The subseasonal global circulation variability leading to the above appears that it will have significant implications for world weather including the USA well into at least December. Again, I do want to keep the rest of this “treatise” relatively concise.


Tropical convective forcing is currently centered ~5N/100E extending from the southeast Arabian Sea to north of New Guinea. Enhancement is also present from southern Indonesia to east of Australia, along a westward displaced SPCZ. The tropical forcing does have a MJO component to it, projecting ~1.5sigma in octant 5 of WH (2004) phase space, ~2sigma in octant 4 with the interannual component left in. There has been eastward propagation of ~5-6m/s along the equator for the past couple of weeks. However, just as the last MJO was truncated, I think this one will also be. During the next 1-2 weeks, a large portion of this severe rainfall is probable to shift east-southeast along the SPCZ impacting northern and eastern Australia. Impacts are also likely from southern India to the Philippines.


The large variation of the WB (2008) measure of the GWO is currently the most robust weather-climate signal. Global relative AAM (plots updated through 20 November) is ~1 standard deviation (1 AMU) above the R1 data climatology, working with ~ plus 40 Hadley positive global tendency. The former may be the highest since about a year ago.


The astute reader may now ask how this can be if I am arguing La-Nina is back. Of course, given that the GWO is red noise (stochastic), with the large signal "something else" can happen meaning the point of this whole discussion (La-Nina) may be a poor assessment on my part. However, since I think we are seeing a repeatable event (more said below) within an emerging quasi-stationary La-Nina base state, odds are with me. Time will tell. What the large GWO signal means is global westerly wind flow has increased, much of that occurring in the subtropical atmospheres where 200mb zonal mean zonal wind flow anomalies are ~2-5m/s. The dynamical processes contributing to this have come from the global surface torques, with the mountain ~plus 30 Hadleys and frictional ~plus 20 Hadleys.


This is the point where a real-time live discussion with “maps and plots” is needed. In a sense, a repeatable variation of the 3sigma October GWO orbit in phase space is in progress. Much of the positive frictional torque is coming from the extratropics linked to poleward and downward propagation of anomalous upper tropospheric zonal mean easterly wind flow. I can also link the components of the current positive clustered mountain torque to this. All of this has involved coupled complicated interactions with tropical-extratropical Rossby wave energy dispersions (RWDs) interacting with the baroclinic eddies. At times, during the last several weeks, there have been well defined projections onto a positive phase of the Branstator (2002) circumglobal teleconnection (anomalous midlatitude ridges). Bottom line, the global wind signal is dominating the tropical convective forcing.


Since at least mid September there has been a strong zonal asymmetry of tropical upper tropospheric circulation anomalies consisting of twin Indian Ocean (west central Pacific) anticyclones (cyclones), typical of a quasi-stationary La-Nina base state. Per animations there was an attempt of subtropical transition to an opposite phase earlier this month, and there may be a feeble attempt as I type, latter across the Indian Ocean. However, going along with the sudden crash in AAM tendency and AAM I am expecting during the next 1-2 weeks, the Indian Ocean upper tropospheric anticyclones, etc., are likely to return. I do expect Indian Ocean tropical forcing to become expansive during December, then perhaps focusing ~100-120E. Another bottom line is what we may be seeing is an “AAM rally in a bear atmosphere”.


Regionally, a synoptic response across the Asia-North American sector is the extending East Asian jet as I type. Phases 4-5 of the GWO 250mb snr streamfunction composite anomaly plots best represent the extratropics. Phases 3-4 of the MJO 250mb snr streamfunction composites generally depict the zonally asymmetric tropical circulation anomalies discussed above. I have not been accurate on frankly white noise details during the last few weeks. Nevertheless, all of the above writing is on track with my notion since about mid October of another large GWO 4-5 to 8-1 evolution within a La-Nina base state during the end of this month leading to a December central USA cold regime.


Consistent with a complete forecast process, models such as the ESRL/PSD ensemble and recently the NCEP/GEFS into early week-3 are starting to catch on. I am concerned the largest negative surface air temperature anomalies may be shifted farther west (linked to Indian Ocean forcing) toward the Rockies and Plains rather than ~95W (per past discussions), particularly during the latter half of next month. Even portions of the USA Pacific Northwest may experience Arctic air during December. Implied is an active southwest flow storm track (with variations) on the Plains.


Loosely, December renditions of phase 3 (not available on-line, yet) of both the GWO and MJO 250mb snr streamfunction composite anomaly plots may represent next month. Of course, unpredictable rapid variations in GWO phase space will occur. In any event, I respectably offer that the official USA forecasts for at least December 2008 and DJF 2008-09 “may be in trouble” (seasonal mean issues understood). The anomalous warmth suggested across portions of the Deep South for DJF may be okay. There should be more of a “La-Nina flavor”.


Weather ramifications have already been suggested. The extended North Pacific Ocean jet/split flow pattern across the USA week-1 per models giving an active southern storm track is likely. This will cause holiday travel problems starting around Thanksgiving, while also giving much needed rainfall to fire-ravaged California. Weeks 2-5 is per above. Any concerns sensitive to the real possibility of extreme cold air temperatures perhaps focused from the Northern Rockies into mid/upper Mississippi Valley and possibly extending into the western Ohio Valley may want to “plan”. The above mentioned storm track may get better defined going through December. Depending on tropical moisture transport through the Gulf of Mexico, severe winter weather including blizzard conditions and intense thundersnow may impact the northern and central Plains. Other weather responses should be understood, including impacts on Christmas travel.


Weeks 1-2 for the tropics are also per above. There is a strong westerly wind event from the central Indian Ocean into Australia, and that could increase the risk of tropical cyclones in that region. Diurnal thunderstorm activity tied to the GWO may intensify across tropical South America weeks 1-2. Tropical South Africa into the South Indian Ocean may see a rapid increase in rainfall by weeks 2-3, only to expand in coverage and extend eastward weeks 3-5, perhaps centered ~0/100-120E. Finally, the warm southwest Pacific Ocean may become the “wild card” this boreal cold season.


Please see the latest official tropical cyclone forecasts for all basins. I trust the expertise of the appropriate meteorological centers to alert the public of additional weather hazards worldwide. Active regions have been occurring during the last week, including the cold and stormy situation around Europe.


Appendix


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. We hope that an opportunity will arise for us (soon) to allow our dedicated web page effort to mature, expediting objectively and accountability. This web page effort will hopefully include an objective predictive scheme for the GWO with hindcasts.


The WB (2007) paper on the GSDM has been published in the February issue of MWR. The first of a two-part paper, where WB formally introduce the GWO (WB (2008)), has been accepted for publication MWR. A pdf of the in press version can be downloaded from the following link:


http://www.cdc.noaa.gov/map/clim/wb08_revised_final.pdf


In addition to the subseasonal snr composite anomaly plots, we hope near real-time discussions with “weather maps” will become a routine part of the ESRL/PSD GSDM web site sometime soon. Part-2 of our GWO paper will discuss the latter. We want to emphasize notions such as global-zonal mean-regional scale linkages as well as forcing-response-feedback (with subsequent interactions) relationships. An important purpose is to provide a dynamical weather-climate linkage framework to evaluate the numerical models in a sophisticated manner as part of a subseasonal (and any time scale) forecast process, in addition to a climate service for all users. Relying on the numerical models alone is a cookbook! I plan on posting a discussion the weekend of 29-30 November.


Ed Berry

Friday, November 14, 2008

Wrong Turn in the REAL Atmosphere???

The views expressed are those of the author and do not necessarily represent those of the National Weather Service.”


Please keep in mind the ESRL/PSD GSDM web link, below, while reading this discussion.


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


The 91-day signal to noise ratio (snr) anomaly composites are now updated daily, centered on the date shown. Please see product descriptions. WB (2008), part-1 of a 2 part paper where the GWO is formally introduced, is in press for MWR publication. There is a link in the Appendix to download the manuscript.


The warmest SSTs globally persist ~140E within 20 degrees of the equator having totals ~29-30C but with weak anomalies. Larger positive anomalies ~1-2C are in the region of the equatorial Indian Ocean with totals near 29C. A strong equatorial trade wind surge (~10m/s anomalies) continues from ~140E to east of the Dateline. There is upwelling of the oceanic thermocline east of the Dateline, leading to roughly negative 2-3C subsurface anomalies ~100-200m depth along the equator per latest 5-day averaged TAO buoy data. Remember that these anomalies are in the presence of annual cycle cooling. SST anomalies ~minus 1-2C are also present from about 160E-140W within 10 degrees of the equator.


Bottom line, I think we may be waiting for the tropical Pacific Ocean “SSTs to really kick in” responding to the global circulation already having La-Nina characteristics. In fact, speculating, are we starting an evolution of another SST cold event which may mature as late as 2009-10 that will even make the Nino 3.4 people happy? See links below.


http://www.wmo.int/pages/prog/wcp/wcasp/enso_update_latest.html


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).


http://www.cdc.noaa.gov/map/images/sst/sst.long.time.gif


I would like to keep this discussion relatively brief. The subseasonal behaviors of several components involving the earth-atmosphere-land dynamical system have not played out as I expected during the last few weeks. However, what has occurred does not surprise me. Regardless, this is an example of why rigorous daily monitoring within the WB (2007) GSDM framework is critical, and relying on the numerical models (multi model ensembles) alone is scientifically unrealistic and indefensible. Furthermore, empirical techniques employed by some synopticians to track baroclinic signals on Hovmoller diagrams as a predictive tool many weeks in advance also has little scientific merit.


Eastern Hemisphere tropical convective forcing has been shifting west during the last week or so, and is currently focused around 5N/100E. Intense to severe tropical thunderstorm activity extends from near 60E to western Indonesia and covers most of the Bay of Bengal, including a tropical cyclone about to impact India. The latter is approximately 35 days after the last period of enhanced rainfall in that region, and it is possible a MJO component may evolve. Leaving the interannual signal in, there is about a 2 sigma projection in octants 3-4 of WH (2004) phase space. A faster signal tied to the WB (2008) measure of the GWO has led to some convective enhancement across the Western Hemisphere. The recent flooding rainfall across the USA Pacific Northwest is attributable to an “atmospheric river” of moisture transport from the Eastern Hemisphere tropical convection.


Phase 4 of the WH (2004) MJO snr 250mb composite psi anomaly plots reasonably depicts the current tropical circulation. In fact, per animations of upper tropospheric daily mean vector wind anomalies, the spatial pattern of Indian Ocean anticyclones and west Pacific cyclones is similar to a couple of weeks ago. The notion offered that the opposite sense to this phase would not be as robust was correct.


So, what happened? Obviously not the coherent eastward shift of tropical forcing/circulation behavior from the Eastern into the Western Hemispheres I discussed was likely in my last several postings. Nevertheless, I do feel there is a scientifically defensible attribution, and this will be one of the many foci of WB during the next several weeks. I think dynamical processes explained by the WB (2008) measure of the GWO contributed, starting with the ~3 sigma orbit in octant 4 of phase space during mid October. Poleward AAM transports and surface torques created a large tendency working to increase zonal mean westerly wind flow in the subtropical atmospheres.


Cutting to the chase, it is possible Rossby wave energy dispersions (RWDs) linking the tropics and extratropics (fast GWO) worked to oppose this tendency, wanting to retain the global circulation in “La-Nina”. Specifically, and beyond the scope of this discussion, the latter is in reference to complex dynamics working to reduce global westerly wind flow, best represented synoptically by ~phase 3 of the WB (2008) GWO 250mb snr composite psi anomaly plots. That is, a positive phase of the Branstator (2002) circumglobal teleconnection of anomalous midlatitude ridges, observed during the boreal winter of 2007-08. Please remember that I am talking about forcing-response-feedback relationships and the not the unrealistic notions of equilibrium states (ex., Charney and DeVore 1979).


Currently, and loosely similar to about a month ago, there is an interhemispheric signal of poleward AAM flux, ~45N and 50S. The global surface torques is positive including a ~20 Hadley East Asian mountain torque (see AAM plots for details). A synoptic response has been for a cold outbreak off the coast of China and an expanded North Pacific Ocean jet. This is an example of a situation when the jet “outruns” the Eastern Hemisphere tropical convective forcing. Phases 4-5 of the WB (2008) GWO snr 250mb composite psi anomaly plots best depict the extratropics, and should through at least week-1.


What road the atmosphere takes from here is extremely uncertain, particularly for timing and details. Statements such as forecaster confidence of something like 4 on a scale 1-5 for a USA week-2 prediction of temperature and precipitation anomalies are unrealistic. The numerical models have been inconsistent, to say the least, for the last 1-2 weeks. The tropics and extratropics are still out of sync.


That said, what appears most probable is for a GWO 8-1 transition ~weeks 2-3 per above, then perhaps coupling of the tropics and extratropics across the Eastern Hemisphere weeks 3-4.That is, possibly a superposition of the circulation states shown by the GWO/MJO phases 3-5 of the snr 250mb composite psi anomaly plots during the first half of December. Given base state issues briefly discussed above, I do feel confident that the La-Nina side of phase space is the most probable “until further notice”.


Weather ramifications include above (below) average temperatures returning to the eastern (western) USA during roughly week-2/3 along with an active southwest flow storm track on the Plains. Much of the west coast could receive generous precipitation. Anomalous cold may then focus on the central states afterward. Interestingly, what may actually occur is ~1-2 week delay of the outlook statements offered in past postings. Monitoring is critical, including the future of the Eastern Hemisphere tropical forcing, which will impact the USA. The risk of extreme weather events (ex., blizzard conditions and severe thunderstorms) affecting large portions of the country, including having adverse impacts on Thanksgiving and even Christmas travel, is possible to be elevated the next several weeks.


It is unclear if and when the above mentioned tropical forcing will coherently shift east. However, locations from the west central into the Southwest Pacific Ocean are probable to be at risk from severe rainfall and cyclones ~weeks 2-4 (week-1 understood). Generally climatological precipitation should continue for tropical South America and Africa the next several weeks. It is not out of the realm of possibilities to see multiple regions of enhanced tropical convection during the next several weeks, but with a focus shifted toward the Indian Ocean and Indonesia ~100-120E (as part of what may be an emerging quasi-stationary component of a La-Nina base state).


Please see the latest official tropical cyclone forecasts for all basins. I trust the expertise of the appropriate meteorological centers to alert the public of additional weather hazards worldwide. There has been an increase during the last week.


Appendix


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. We hope that an opportunity will arise for us (soon) to allow our dedicated web page effort to mature, expediting objectively and accountability. This web page effort will hopefully include an objective predictive scheme for the GWO with hindcasts.


The WB (2007) paper on the GSDM has been published in the February issue of MWR. The first of a two-part paper, where WB formally introduce the GWO (WB (2008)), has been accepted for publication MWR. A pdf of the in press version can be downloaded from the following link:


http://www.cdc.noaa.gov/map/clim/wb08_revised_final.pdf


In addition to the subseasonal snr composite anomaly plots, we hope near real-time discussions with “weather maps” will become a routine part of the ESRL/PSD GSDM web site sometime soon. Part-2 of our GWO paper will discuss the latter. We want to emphasize notions such as global-zonal mean-regional scale linkages as well as forcing-response-feedback (with subsequent interactions) relationships. An important purpose is to provide a dynamical weather-climate linkage framework to evaluate the numerical models in a sophisticated manner as part of a subseasonal (and any time scale) forecast process, in addition to a climate service for all users. Relying on the numerical models alone is a cookbook! I plan on posting a discussion the weekend of 22-23 November.


Ed Berry

Saturday, November 08, 2008

"Season of the Witch Part Deux", for Real?

The views expressed are those of the author and do not necessarily represent those of the National Weather Service.”


Please keep in mind the ESRL/PSD GSDM web link, below, while reading this discussion.


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


The 91-day signal to noise ratio (snr) anomaly composites are now updated daily, centered on the date shown. Please see product descriptions. WB (2008), part-1 of a 2 part paper, where the GWO is formally introduced is in press for MWR publication. There is a link in the Appendix to download the manuscript.


As expected from any slow boundary forcing on the atmosphere, global tropical and extratropical SSTs have changed little from a week ago. The warmest ocean waters have shifted south of the equator to the region of Australia and west of New Guinea, ~140E. Totals are ~30C, with 29C SSTs still covering much of the TNWP. The anomalous warmth across the Indian Ocean has increased during the past week, with totals roughly 29C and anomalies ~plus 1C. The equatorial Dateline back to 150E has ~minus 2C SST anomalies, responding to trade surges.


An equatorial trade wind surge from the Dateline to Indonesia has resumed, with anomalies ~10m/s. Further, per latest 5-day averaged TAO equatorial depth sections, negative subsurface anomalies 3C and larger centered ~200m/160W have evolved, presumably due to an upwelling oceanic Kelvin wave. The bottom line is I think the “Nino SSTs” are now responding to the global La-Nina base state. This is further evidence that “ENSO neutral” is an inaccurate description of the global weather-climate situation. It is not a question whether “La-Nina” has “resumed”, but simply how intense “round two” will be. In fact, many well posed numerical and statistical models support the latter, as do more complete indices than “Nino 3.4”. Please see links below.


http://www.wmo.int/pages/prog/wcp/wcasp/enso_update_latest.html


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).


http://www.cdc.noaa.gov/map/images/sst/sst.long.time.gif


I expanded a bit on the SSTs in an attempt to keep the rest of this discussion shorter. While in the minds of many the current weather-climate situation may appear confusing with (understandably) a significant noise factor, I do think non-trivial signals are coming back. My attribution comfort level remains okay. While the details during the last couple of weeks have not played out as “expected”, I do think the previous probabilistic predictive statements offered are loosely on track. However, my confidence in any outlooks discussed below has decreased, and also serves as an example that any good numerical model agreement does not imply high forecast confidence (yes, insert the Rottweiler!).


The MJO component of the Eastern Hemisphere tropical convective forcing came out to about 160E, then “fell apart”. Full disk satellite imagery and other tools show the tropical forcing consolidating around 0/120E while extending from the South Indian Ocean to the west central Pacific. Having my reasons, I think we are seeing a response to our La-Nina base state, including the trade surge per above. In fact, the latest WH (2004) measure of the MJO retaining ENSO shows ~1.5 sigma projection in octant 3 of phase space. Animations of upper tropospheric daily mean vector wind anomalies show tropical circulation anomalies that are reasonably consistent with ~phase 4 of the 250mb snr composite psi plots.


Phases 8-1 of the 250mb snr composite psi anomaly plots best describe the global extratropics. During ~mid October there was a large positive global AAM tendency forced by eddy poleward transports and surface torques. A response was to add weak zonal mean westerly wind flow anomalies to the subtropical atmospheres and decouple the WB (2008) measure of the GWO from the WH (2004) MJO measure. Tropical forcing then intensified across the Western Hemisphere, and recently left severe tropical cyclone Paloma in the wake.


Subseasonal dynamical processes explained by the WB (2008) measure of the GWO are not only currently working to decrease global relative AAM, but also push the global atmosphere back toward the La-Nina attractor. For instance, updated through 6 November the computed global AAM tendency was ~minus 20 Hadleys having zonal mean contributions from negative East Asian, North American and Andes mountain torques. The strong positive Coriolis torque and negative earth AAM tendency, working along with the above, have forced the GWO to orbit back to octants 8-1 of phase space. The latter terms, part of the earth-atmosphere AAM budget, suggests an intensifying Hadley cell leading to mass removal from the tropics.


Regardless of how confusing the above may appear to the reader, my point is to give some evidence that the tropics and extratropics may be getting back in sync shifted toward La-Nina. I think it is probable to for both the WB (2008) GWO and WH (2004) MJO to orbit in similar octants of phase space during the next few weeks. That is, initially octant 3 then perhaps 4-5. The corresponding snr composite anomaly plots for both the GWO and MJO should represent North American weather roughly weeks 2-4.


Sources of uncertainty include the likelihood of more western Pacific Ocean and Western Hemisphere tropical forcing, the seasonal cycle, and the inherent red noise of the GWO. In fact, some models of the WH (2004) MJO (La-Nina removed) suggest a “loop” into the Western Hemisphere weeks 1-2. In any case, I think our quasi-stationary global circulation component per above may dominate going into DJFM, resulting in at least some weather similarities to boreal winter 2007-08 including the USA.


It is possible that the “8-1 evolution” of the global circulation discussed in past postings may be occurring as I type. Regionally for North America, the Pacific Ocean jet is extended leading to a split flow storm track. As tropical-extratropical coupling occurs, this jet should retract leading to a spatial pattern similar to phase 3 of the WB (2008) GWO 250mb snr composite psi and 2m air temperature anomaly plots by ~weeks 2-3. That is suggestive of troughs focused on the Plains, after initially impacting the western states. On average, this situation shifts west heading into boreal winter. Whatever the details, there may be weather concerns for much of the country during Thanksgiving travel. Additionally, there is no change to the “December cold scenario” discussed in the last few postings.


Internationally, intense to severe thunderstorm activity is likely to remain focused on Indonesia (~120-140E) weeks 1-2 while locations including Southeast Asia and portions of the TNWP also have weather impacts. The on-going trade surge should limit tropical cyclone activity; however, hybrids could affect locations such as the Philippines. While currently convectively suppressed and would be expected to remain so, flare-ups particularly across the South Indian Ocean are possible “anytime”, including a “rogue” tropical cyclone for locations such as Madagascar. No more than the typical diurnal thunderstorm activity appears probable for regions such as Brasil. Rainfall may also be quite intense at times for equatorial Africa (shifting south).


Broadly, going into boreal winter enhanced rainfall appears probable to have a quasi stationary component ~120E while shifting slowly south. The warm west central and southwest Pacific Ocean will remain the “wild card”, as there will likely be severe convective flare-ups including cyclones in those regions. The role of the MJO is unclear, and there will be tropical impacts from at least rapid ~10-20 day GWO variations that can only be captured from daily monitoring.


Please see the latest official tropical cyclone forecasts for all basins. I trust the expertise of the appropriate meteorological centers to alert the public of additional weather hazards worldwide.


Appendix


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. We hope that an opportunity will arise for us (soon) to allow our dedicated web page effort to mature, expediting objectively and accountability. This web page effort will hopefully include an objective predictive scheme for the GWO with hindcasts.


The WB (2007) paper on the GSDM has been published in the February issue of MWR. The first of a two-part paper, where WB formally introduce the GWO (WB (2008)), has been accepted for publication MWR. A pdf of the in press version can be downloaded from the following link:


http://www.cdc.noaa.gov/map/clim/wb08_revised_final.pdf


In addition to the subseasonal snr composite anomaly plots, we hope near real-time discussions with “weather maps” will become a routine part of the ESRL/PSD GSDM web site sometime soon. Part-2 of our paper will discuss the latter. We want to emphasize notions such as global-zonal mean-regional scale linkages as well as forcing-response-feedback (with subsequent interactions) relationships. An important purpose is to provide a dynamical weather-climate linkage framework to evaluate the numerical models in a sophisticated manner as part of a subseasonal (and any time scale) forecast process, in addition to a climate service for all users. Relying on the numerical models alone is a cookbook! I plan on posting a discussion the weekend of 14-16 November.


Ed Berry