Note: The ESRL/PSD NCEP/NCAR Reanalysis-1 data (R1) AAM plots now present the “complete” budget (updated through 24 August as of this writing), and will be referred to.
Per NOAA/CPC and NOAA/PMEL/TAO buoy data; the overall spatial distribution of global tropical SSTs has not changed much for at least several weeks. The equatorial Indian Ocean and particularly the west central and northwest Pacific Ocean basins (with the horseshoe into the subtropics/midlatitudes) remain well above climatology. Anomaly magnitudes are ~1-1.5 deg C with the former and up to 2 deg C for the latter, with totals in excess of 30 deg C along the equator near 150E. The west Pacific Ocean anomalies are deep extending to at least 400m, suggesting an anomalously warm west Pacific Ocean warm pool.
The negative SST anomalies across the equatorial East Pacific cold tongue have weakened slightly during the past week, with recent 5-day averaged values varying from ~.5C just east of the date line to around 2C near 120W (~Nino 3). The cool anomalies still extend to around 150-200m deep in the region of 140-160W, but with values ~minus 2-2.5C at 150m/150W. It will be interesting to monitor the persistence of these anomalies going into boreal fall since the seasonal cycle favors cooling. The equatorial Atlantic Ocean and the Caribbean Sea have slightly positive SST anomalies with cool pockets inter-dispersed. Totals range from 27-28C across the former to in excess of 29C for the latter.
Again, while the global circulation is strongly La-Nina like (GSDM Stage 1), the SSTs only weakly support that response. To me, the seasonal transition to boreal fall may give us some insight whether or not the SSTs become more La-Nina like, or if a different evolution occurs. For example, the South Asian monsoon system will shut down and the Eastern Hemisphere tropical convective forcing will shift south. How will a possibly coupled system respond should, for example, MJO activity increase?
The strongest tropical convective forcing is still across the Eastern Hemisphere, centered ~10-15N/90-100E per full disk satellite imagery. From time-latitude sections of OLR/A and daily monitoring, this activity has been shifting northeast off the equator during the past week. Currently enhanced rainfall extends from northern India east-southeast into the far west Pacific Ocean. This is the third (but weak) event of this nature since about June, and is typical during boreal summer. As discussed below, the region of tropical thunderstorm clusters actually represents a weak consolidation. Other enhanced activity is occurring around the Americas and north equatorial Africa. The Wheeler and Hendon (2004) multivariate RMM phase space plots from several international weather centers generally suggest a ~1 standard deviation MJO projection across the Indian Ocean (more said below).
Since at least mid-August, relatively “well-orchestrated” Rossby wave energy dispersions (RWDs) linked to the Indian Ocean/South Asian monsoon system tropical forcing have been propagating into the Tropical Northwest Pacific Ocean (TNWP). A tropical convective flare-up over the TNWP occurred early last week, moving west and linking with the clusters farther west. However, to be brief, monitoring tells me this chain of events allowed circulation anomalies to impinge upon global north-south mountain massifs driving the mountain torque to ~plus 20 Hadleys on about 18 August. Along with a positive global frictional torque, the global relative atmospheric angular momentum (AAM ) tendency spiked to slightly greater than 20 Hadleys. Since then, the global mountain torque and relative AAM tendency have collapsed to less than minus 20 Hadleys.
The point is I think a weak mountain-frictional torque index cycle variation has occurred during the last 7-10 days within our low AAM base state. When updated, the ESRL/PSD Global Wind Oscillation (GWO) quasi-phase space plot should present a weak circuit in the direction GSDM Stage 2, before heading “down”. This notion was offered as a possibility in my posting a week ago. Some westerly wind flow has been added to both the northern subtropical and midlatitude atmospheres, and will contribute to a Stage-2 like response across the PNA sector this upcoming week.
Global total and relative AAM remain extremely low, less than 3 atmospheric momentum units (AMUs) below the R1 data climatology (at least 2.5 standard deviations) through 24 August. Deep zonal mean easterly wind anomalies still dominate much of the tropical atmosphere, with the largest magnitudes now just south of the equator near 5S (~10 m/s at 200mb). Worth repeating, the cause of this extremely low AAM, GSDM Stage 1 regime is unclear. I can speculate this all goes back to dynamical forcing-response feedback processes linked to AAM transports and large GWO circuits that started ~ 1 December 2006, when El-Nino was “stopped in its tracks”. In any event, the SSTs themselves do not support this global circulation. The intense zonal mean easterly wind anomalies have been broadening in latitude into the subtropical atmospheres of both hemispheres during the last several weeks. There is still little evidence for these zonal mean easterly wind anomalies to propagate off the equator. Daily monitoring of if/when these zonal mean easterlies do propagate poleward off the equator is critical to determining the most probable future global circulation state. I make a plea to having a diagnostic tool such as the baroclinic CHI problem to see what is going on vertically.
Animations of upper tropospheric daily mean vector wind total/anomaly fields give a loose signal of twin anticyclones ~90E with downstream very weak cyclones just east of the date line. Cross equatorial flow from the Southern Hemisphere remains robust in the region of the date line. In fact, I speculate that the intense RWDs across the Southern Hemisphere leading to this cross equatorial flow may be forcing the zonal mean anomalous easterlies closer to the equator as a feedback.
In any case, there has also been episodic (~every 3-5 days) meridional-zonal-meriodional northeast-southwest (northwest-southeast for the Southern Hemisphere) vacillations of the midlatitude synoptic eddies over the past couple of weeks. It is not unusual to observe this type of behavior during strong GSDM Stage 1 situations. Most importantly, for the first time in at least a couple of months a relatively strong inter-hemispheric symmetric AAM transport signal has appeared. The signal is most robust ~50N and 40S. Hence there is the possibility of a stronger positive mountain-frictional torque index cycle than recently observed over the next couple of weeks.
If there is any truth to the RMM predictions of a weak MJO also during the next couple of weeks, perhaps a larger circuit in GWO quasi-phase space toward Stage 2 will occur. However, this is all speculation, and I do expect an overall GSDM Stage 1 regime to persist until further notice. Only those who do rigorous daily weather-climate monitoring utilizing the GSDM will be the first to catch any kind of a rapid change. It is unlikely any multi-model ensemble prediction scheme will catch an abrupt subseasonal change “until it happens (like ~1 December 2006)”. We also need to remember that predictability with any technique/forecast process is generally quite low during seasonal transitions, which we are about to experience.
There is some model agreement for progressive troughs and ridges across the USA during the next few weeks as seasonal strengthening of the westerlies occur. After this upcoming week’s central/eastern USA trough, I offer a low probability of a stronger event having a Rockies-Plains trough later week 2 into week 3 (~ 7-16 September?). Weather ramifications should be apparent, including the first possible snowfall across the Northern Rockies and an increased likelihood of storminess across the Plains.
In the longer term, should this low AAM regime persist through boreal winter, much of the USA focusing on the western USA into the Plains may have the odds tilted toward a cold and stormy pattern (western trough/southeast states ridge), if there is an Arctic cold air source. Per above, stay tuned.
It is obviously unclear to say where any possible land falling tropical cyclone will go. However, please see the latest statements from the NOAA/NWS/Tropical Prediction Center. At least for week 1, once again relatively stable air is penetrating the deep Tropical North Atlantic Ocean (TNA) basin. Thus tropical cyclone activity for at least the TNA may be suppressed this upcoming week (understanding hybrids and what can occur across the Caribbean).
Locations from the central and eastern equatorial Indian Ocean into China and Southeast Asia are likely to continue with the poundings of intense monsoon rainfall for at least this week into possibly week 2. Perhaps the recent northward shift discussed above may be the last for boreal summer. The likelihood of severe thunderstorms including tropical cyclone activity across the TNWP including the Philippines may increase by week 2. Equatorial North Africa may remain active for at least week 1.
Appendix
An experimental quasi-phase space plot of the GSDM utilizing a time series of normalized relative AAM tendency anomaly (Y-axis) and normalized relative AAM anomaly time series (X-axis) can be found at
http://www.cdc.noaa.gov/map/images/gcm/gsdm_95d.jpg
http://www.cdc.noaa.gov/map/images/gcm/gsdm_40d.jpg
http://www.cdc.noaa.gov/map/clim/gsdm.shtml
We call the behavior of this plot the Global Wind Oscillation (GWO). While the intent of the GSDM is to extend current thinking beyond the MJO, the purpose of the GWO is to illustrate the non-oscillatory component of the GSDM.
These are probabilistic statements, and work is ongoing to quantify in future posts (for example, risk assessment maps and signal to noise ratio plots). We hope that an opportunity will arise for us 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. I will try to post an update on about Friday, 31 August.
Ed Berry
Sunday, August 26, 2007
Monday, August 20, 2007
When Will The Atmosphere Go "Up"?
Note: The ESRL/PSD AAM plots now present the “complete” budget (updated through 17 August as of this writing), and will be referred to.
Per NOAA/CPC and NOAA/PMEL/TAO buoy data; there is not much overall change to the spatial distribution of global tropical SSTs. The equatorial Indian Ocean and particularly the west central and northwest Pacific Ocean basins (with the horseshoe into the subtropics/midlatitudes) remain above climatology. Anomaly magnitudes are ~1-1.5C with the former and up to 2c for the latter, with totals in excess of 31C along the equator near 140E. The west Pacific Ocean anomalies are deep extending to at least 400m, suggesting an anomalously warm west Pacific Ocean warm pool. Will some of this warmth strongly shift east “sometime during the future?”
Anomalies across the equatorial East Pacific cold tongue remain below normal with recent 5-day averaged values varying from ~.5C just east of the date line to greater than 2C near 110W (~Nino 3). The cool anomalies still extend to around 150-200m deep in the region of 140-160W, with values less than minus 3C at 150m/140W. However, these cool anomalies are relatively localized. It will be interesting to monitor the persistence of these anomalies going into boreal fall since the seasonal cycle favors cooling. The tropical Atlantic Ocean basin remains near average with totals ~27-28C with plus 1-2C anomalies in the areas of the Caribbean and Gulf of Mexico. Overall, again, while the global circulation is strongly La-Nina like (GSDM Stage 1), the SSTs only weakly support that response.
Enhanced tropical convective forcing continues across the Eastern Hemisphere, extending from the central Indian Ocean into portions of the Tropical Northwest Pacific Ocean (TNWP). During the past few days some consolidation of enhanced rainfall has been occurring near 5N/80E, within this larger envelop. Real-time Multivariate MJO Index (RMM) phase space plots from various operational global weather centers suggest a weak MJO signal may be developing in the region of the Indian Ocean. There is also a hint of this possibility from Hovmoller plots designed to isolate coherent modes of tropical variability. Animations of upper tropospheric daily mean vector wind anomalies indicate fairly well defined twin tropical/subtropical anticyclones ~60-90E with downstream twin cyclones west of the date line. Whether or not this evolves into any real coherent MJO dynamical signal or even a northward propagating mode is unclear. Predictions from the RMMs are inconclusive.
Global total and relative atmospheric angular momentum (AAM) are at the lowest magnitudes in at least a year, with the latter near 4 AMUs (at least 3 standard deviations) below the R1 data climatology through 17 August. In fact, the quasi-phase space plot depicting the Global Wind Oscillation (GWO) is “off the scale”. Deep zonal mean easterly wind anomalies dominate much of the tropical atmosphere, with the largest magnitudes across the Southern Hemisphere near 15S (~10 m/s at 200mb). The cause of this extremely low AAM, GSDM Stage 1 regime is unclear. I can speculate this all goes back to dynamical forcing-response feedback processes linked to AAM transports and large GWO circuits that started ~ 1 December 2006, when El-Nino was “stopped in its tracks”. In any event, the SSTs themselves do not support this global circulation. The intense zonal mean easterly wind anomalies have been broadening in latitude during the last couple of weeks. Daily monitoring of if/when these zonal mean easterlies propagate poleward off the equator are critical to determining the most probable future global circulation state.
In recent days both global surface torques has become positive, with the frictional and mountain torques ~10-15 Hadleys. In addition, global relative AAM tendency is near zero and the earth AAM is slightly positive (please see ESRL/PSD R1 data plots for zonal mean details). I think this may be linked to the Eastern Hemisphere tropical convective forcing discussed above. In other words, we should see increasing trades as a response to the Indian Ocean convection. The latter is starting to occur particularly just south of the equator hence driving up the frictional torque. Additionally, I offer speculation that there may also be downward transport of the easterly wind flow anomalies to the earth’s surface coming both from the synoptic eddies (lots of Rossby wave energy dispersion recently going on linking the tropics to the extratropics) and Hadley circulations. However, a diagnostic tool such as the baroclinic CHI problem is needed in real time to quantify this diagnosis.
My bottom line is that at least some zonal mean subtropical/midlatitude westerly flow should be added to the atmosphere during the next 1-3 weeks. However, at best this will only be reflected as a weak oscillation depicted the GWO within an overall steady state GSDM Stage 1 circulation regime. Interestingly, the seasonal transition into boreal fall will impact the outcome of regional-scale circulation anomalies tied to this global weather-climate situation. For example, a reverse phase of the Pacific-North American teleconnection is most probable during GSDM Stage 1 during the cold season. Should this low AAM regime persist through boreal winter, much of the USA focusing on the western USA into the Plains may have the odds tilted toward a cold and stormy pattern (western trough/southeast states ridge), if there is an Arctic cold air source. Careful daily weather-climate monitoring will be needed this fall on just how steady state our Stage 1 regime will be. For example, should there be 2 regions of Eastern Hemisphere tropical convective forcing with the west Pacific Ocean becoming dominate, a “completely different” and sudden evolution may occur.
Most ensemble prediction schemes from operational global weather centers are coming into general agreement for the western USA trough-southeast states ridge pattern weeks 1-2. Improving model agreement makes sense given our general stationary regime having regular Rossby wave energy dispersions linking the tropics and extratropics. However, there is still large uncertainty per issues discussed above.
There is little change from the outlook I issued nearly a week ago. Cooler and wetter is the most probable trend for particularly the Rockies into the Plains of the USA during the next 1-3 weeks. In fact, portions of the northern Rockies may see their first snowfall. The center part of the country may become increasingly stormy with severe thunderstorms and flooding rainfall (which has already occurred in some areas during the last few days). Some of this activity is likely to extend into the Ohio Valley and eastern states. The southeast is likely to continue with above normal heat and humidity. It is obviously unclear to say where any possible land falling tropical cyclone will go. However, the need to monitor this situation is here, and please see the latest statements from the NOAA/NWS/Tropical Prediction Center.
Locations from the central and eastern equatorial Indian Ocean into China and Southeast Asia are likely to continue with the poundings of intense monsoon rainfall for possibly a few more weeks. This region of intense convection may shift northward week 2. The likelihood of thunderstorm activity across the TNWP including the Philippines during weeks 1-3 is unclear. However, frontal activity coming off of East Asia interacting with the warm SSTs could quickly lead to very active weather across those regions. Equatorial North Africa into the Atlantic may have climatologically active thunderstorm activity including tropical cyclone development.
Appendix
An experimental quasi-phase space plot of the GSDM utilizing a time series of normalized relative AAM tendency anomaly (Y-axis) and normalized relative AAM anomaly time series (X-axis) can be found at
http://www.cdc.noaa.gov/map/images/gcm/gsdm_95d.jpg
http://www.cdc.noaa.gov/map/images/gcm/gsdm_40d.jpg
http://www.cdc.noaa.gov/map/clim/gsdm.shtml
We call the behavior of this plot the Global Wind Oscillation (GWO). While the intent of the GSDM is to extend current thinking beyond the MJO, the purpose of the GWO is to illustrate the non-oscillatory component of the GSDM.
These are probabilistic statements, and work is ongoing to quantify in future posts. We hope that an opportunity will arise for us 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. My next update will again be early next week due to travel.
Ed Berry
Per NOAA/CPC and NOAA/PMEL/TAO buoy data; there is not much overall change to the spatial distribution of global tropical SSTs. The equatorial Indian Ocean and particularly the west central and northwest Pacific Ocean basins (with the horseshoe into the subtropics/midlatitudes) remain above climatology. Anomaly magnitudes are ~1-1.5C with the former and up to 2c for the latter, with totals in excess of 31C along the equator near 140E. The west Pacific Ocean anomalies are deep extending to at least 400m, suggesting an anomalously warm west Pacific Ocean warm pool. Will some of this warmth strongly shift east “sometime during the future?”
Anomalies across the equatorial East Pacific cold tongue remain below normal with recent 5-day averaged values varying from ~.5C just east of the date line to greater than 2C near 110W (~Nino 3). The cool anomalies still extend to around 150-200m deep in the region of 140-160W, with values less than minus 3C at 150m/140W. However, these cool anomalies are relatively localized. It will be interesting to monitor the persistence of these anomalies going into boreal fall since the seasonal cycle favors cooling. The tropical Atlantic Ocean basin remains near average with totals ~27-28C with plus 1-2C anomalies in the areas of the Caribbean and Gulf of Mexico. Overall, again, while the global circulation is strongly La-Nina like (GSDM Stage 1), the SSTs only weakly support that response.
Enhanced tropical convective forcing continues across the Eastern Hemisphere, extending from the central Indian Ocean into portions of the Tropical Northwest Pacific Ocean (TNWP). During the past few days some consolidation of enhanced rainfall has been occurring near 5N/80E, within this larger envelop. Real-time Multivariate MJO Index (RMM) phase space plots from various operational global weather centers suggest a weak MJO signal may be developing in the region of the Indian Ocean. There is also a hint of this possibility from Hovmoller plots designed to isolate coherent modes of tropical variability. Animations of upper tropospheric daily mean vector wind anomalies indicate fairly well defined twin tropical/subtropical anticyclones ~60-90E with downstream twin cyclones west of the date line. Whether or not this evolves into any real coherent MJO dynamical signal or even a northward propagating mode is unclear. Predictions from the RMMs are inconclusive.
Global total and relative atmospheric angular momentum (AAM) are at the lowest magnitudes in at least a year, with the latter near 4 AMUs (at least 3 standard deviations) below the R1 data climatology through 17 August. In fact, the quasi-phase space plot depicting the Global Wind Oscillation (GWO) is “off the scale”. Deep zonal mean easterly wind anomalies dominate much of the tropical atmosphere, with the largest magnitudes across the Southern Hemisphere near 15S (~10 m/s at 200mb). The cause of this extremely low AAM, GSDM Stage 1 regime is unclear. I can speculate this all goes back to dynamical forcing-response feedback processes linked to AAM transports and large GWO circuits that started ~ 1 December 2006, when El-Nino was “stopped in its tracks”. In any event, the SSTs themselves do not support this global circulation. The intense zonal mean easterly wind anomalies have been broadening in latitude during the last couple of weeks. Daily monitoring of if/when these zonal mean easterlies propagate poleward off the equator are critical to determining the most probable future global circulation state.
In recent days both global surface torques has become positive, with the frictional and mountain torques ~10-15 Hadleys. In addition, global relative AAM tendency is near zero and the earth AAM is slightly positive (please see ESRL/PSD R1 data plots for zonal mean details). I think this may be linked to the Eastern Hemisphere tropical convective forcing discussed above. In other words, we should see increasing trades as a response to the Indian Ocean convection. The latter is starting to occur particularly just south of the equator hence driving up the frictional torque. Additionally, I offer speculation that there may also be downward transport of the easterly wind flow anomalies to the earth’s surface coming both from the synoptic eddies (lots of Rossby wave energy dispersion recently going on linking the tropics to the extratropics) and Hadley circulations. However, a diagnostic tool such as the baroclinic CHI problem is needed in real time to quantify this diagnosis.
My bottom line is that at least some zonal mean subtropical/midlatitude westerly flow should be added to the atmosphere during the next 1-3 weeks. However, at best this will only be reflected as a weak oscillation depicted the GWO within an overall steady state GSDM Stage 1 circulation regime. Interestingly, the seasonal transition into boreal fall will impact the outcome of regional-scale circulation anomalies tied to this global weather-climate situation. For example, a reverse phase of the Pacific-North American teleconnection is most probable during GSDM Stage 1 during the cold season. Should this low AAM regime persist through boreal winter, much of the USA focusing on the western USA into the Plains may have the odds tilted toward a cold and stormy pattern (western trough/southeast states ridge), if there is an Arctic cold air source. Careful daily weather-climate monitoring will be needed this fall on just how steady state our Stage 1 regime will be. For example, should there be 2 regions of Eastern Hemisphere tropical convective forcing with the west Pacific Ocean becoming dominate, a “completely different” and sudden evolution may occur.
Most ensemble prediction schemes from operational global weather centers are coming into general agreement for the western USA trough-southeast states ridge pattern weeks 1-2. Improving model agreement makes sense given our general stationary regime having regular Rossby wave energy dispersions linking the tropics and extratropics. However, there is still large uncertainty per issues discussed above.
There is little change from the outlook I issued nearly a week ago. Cooler and wetter is the most probable trend for particularly the Rockies into the Plains of the USA during the next 1-3 weeks. In fact, portions of the northern Rockies may see their first snowfall. The center part of the country may become increasingly stormy with severe thunderstorms and flooding rainfall (which has already occurred in some areas during the last few days). Some of this activity is likely to extend into the Ohio Valley and eastern states. The southeast is likely to continue with above normal heat and humidity. It is obviously unclear to say where any possible land falling tropical cyclone will go. However, the need to monitor this situation is here, and please see the latest statements from the NOAA/NWS/Tropical Prediction Center.
Locations from the central and eastern equatorial Indian Ocean into China and Southeast Asia are likely to continue with the poundings of intense monsoon rainfall for possibly a few more weeks. This region of intense convection may shift northward week 2. The likelihood of thunderstorm activity across the TNWP including the Philippines during weeks 1-3 is unclear. However, frontal activity coming off of East Asia interacting with the warm SSTs could quickly lead to very active weather across those regions. Equatorial North Africa into the Atlantic may have climatologically active thunderstorm activity including tropical cyclone development.
Appendix
An experimental quasi-phase space plot of the GSDM utilizing a time series of normalized relative AAM tendency anomaly (Y-axis) and normalized relative AAM anomaly time series (X-axis) can be found at
http://www.cdc.noaa.gov/map/images/gcm/gsdm_95d.jpg
http://www.cdc.noaa.gov/map/images/gcm/gsdm_40d.jpg
http://www.cdc.noaa.gov/map/clim/gsdm.shtml
We call the behavior of this plot the Global Wind Oscillation (GWO). While the intent of the GSDM is to extend current thinking beyond the MJO, the purpose of the GWO is to illustrate the non-oscillatory component of the GSDM.
These are probabilistic statements, and work is ongoing to quantify in future posts. We hope that an opportunity will arise for us 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. My next update will again be early next week due to travel.
Ed Berry
Tuesday, August 14, 2007
End of Days
Note: The R1 data ESRL/PSD AAM plots now present the “complete” budget (updated through 11 August as of this writing), and will be referred to.
This posting will be relatively short. The spatial distribution of global tropical and subtropical SSTs is similar to what was discussed on 9 August. There has been cooling across the equatorial East Pacific cold tongue loosely in the Nino 3 region. Recent 5-day averaged anomalies (per TAO buoy data) are ~minus 2.5C at the surface extending to depths ~150m at 140-160W with values ~minus 4C. While the global circulation is strongly La-Nina like, perhaps the equatorial East Pacific may be starting to respond.
The strongest tropical convective forcing remains well entrenched across the Eastern Hemisphere. One anomalous region has been festering across the equatorial Indian Ocean centered ~5S/90E with another focused across the Tropical Northwest Pacific Ocean (TNWP) just east of the Philippines. The South Asian monsoon system has also intensified during the last week. There is also minor enhancement of tropical rainfall around the Americas, and easterly wave activity across northern equatorial Africa has become quite intense.
Some evidence indicates another ~40-day mode of tropical rainfall variability maybe organizing across the equatorial Indian Ocean. However, recent full disk satellite imagery already suggests a quickening of eastward movement of the thunderstorm clusters. We will see if this possible episode organizes into a MJO particularly if it propagates northeast. In any case, we are still stuck in a generally stationary GSDM Stage 1 (more said below) base state having generally weak subseasonal variations. A difference is that our nemesis of 2 regions of Eastern Hemisphere tropical convective forcing may be starting to appear as seasonal transition to the boreal cold season approaches.
Total and relative angular momentum is probably the lowest in at least a year. Both are ~minus 2.5-3 AMUs below the R1 data climatology, and much of the contribution is coming from deep zonal mean easterly wind flow anomalies throughout the tropical and subtropical atmospheres. The largest zonal mean magnitudes are in the Southern Hemisphere (SH), having 200mb easterly wind anomalies ~10m/s. There is also a strong poleward transport signal from the SH tropics into the lower midlatitudes, with a weaker counterpart developing north of the equator. Global relative AAM tendency is ~minus 25 Hadleys as of 11 August and the Global Wind Oscillation (GWO) is as far to left as I have seen it, in the GSDM Stage 4-1 quasi-phase space. The global signal from the surface torques is still weak (for now).
There are currently strong twin upper tropospheric anticyclones with extensive cross-equatorial flow centered ~80E tied to the Indian Ocean (IO) tropical forcing. A weak compensating set of cyclones is present just west of the date line. With the exception of this date line cyclone pair, the entire tropics is dominated by upper tropospheric easterly wind flow, with anomalies ~15-25m/s from Africa into the Atlantic Ocean (favorable for convectively active easterly waves). Rossby wave energy dispersions (RWDs) across the Northern Hemisphere (NH) extratropics interacting with both the IO and TNWP forcing are already impacting North America. Combined with what I expect to be strengthening zonal mean westerly flow across the NH midlatitudes, my notion offered 9 August of the trough-ridge-trough pattern across the USA shifting southeast looks probable. In fact, the western USA may see a return to a barrage of troughs similar to last spring going into September, possibly interacting with a cold air source from Canada.
Week-2 ensemble means from several international operational weather centers are offering varying solutions for North America; however, there has been a recent trend toward a western USA trough with a southeast states ridge. As before, when I see the numerical models trend toward what diagnostic reasoning suggests (per GSDM), my confidence increases. The latter should be the case since I am utilizing a proposed more complete forecast process for subseasonal prediction. Nevertheless, uncertainty is still huge given issues such as the future evolution of the tropical forcing and seasonal transition. The general synoptic pattern across the PNA sector this fall will have some dependency on the GWO. For example, will there be a return to large circuits to GSDM Stage 2, etc. (especially if the warm TNWP stays active)?
Cooler and wetter weather is the most probable trend for particularly the northwest half of the USA during the next 1-3 weeks. The southeast is likely to continue with above normal heat and humidity. It is obviously unclear to say where any possible land falling tropical cyclone will go. However, the need to monitor this situation is likely to increase (see latest statements from the NOAA/NWS/Tropical Prediction Center). The center part of the country is also likely to become increasingly stormy, including severe thunderstorms and flooding rainfall.
Locations from India into China and Southeast Asia are likely to continue with the poundings of intense monsoon rainfall possibly for several more weeks. The TNWP including the Philippines are also probable to stay active with heavy rain and tropical cyclone activity week 1, with perhaps a respite across the west central Pacific for at least week 2. Northern equatorial Africa is likely to stay active through at least week 1, and I think the Atlantic Hurricane Season “is about to get underway”.
Appendix
An experimental quasi-phase space plot of the GSDM utilizing a time series of normalized relative AAM tendency anomaly (Y-axis) and normalized relative AAM anomaly time series (X-axis) can be found at
http://www.cdc.noaa.gov/map/images/gcm/gsdm_95d.jpg
http://www.cdc.noaa.gov/map/images/gcm/gsdm_40d.jpg
http://www.cdc.noaa.gov/map/clim/gsdm.shtml
We call the behavior of this plot the Global Wind Oscillation (GWO). While the intent of the GSDM is to extend current thinking beyond the MJO, the purpose of the GWO is to illustrate the non-oscillatory component of the GSDM.
These are probabilistic statements, and work is ongoing to quantify in future posts. We hope that an opportunity will arise for us 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. My next update should be early next week. In general, due to covering shifts and travel, my postings will be irregular through at least the rest of this month.
Ed Berry
This posting will be relatively short. The spatial distribution of global tropical and subtropical SSTs is similar to what was discussed on 9 August. There has been cooling across the equatorial East Pacific cold tongue loosely in the Nino 3 region. Recent 5-day averaged anomalies (per TAO buoy data) are ~minus 2.5C at the surface extending to depths ~150m at 140-160W with values ~minus 4C. While the global circulation is strongly La-Nina like, perhaps the equatorial East Pacific may be starting to respond.
The strongest tropical convective forcing remains well entrenched across the Eastern Hemisphere. One anomalous region has been festering across the equatorial Indian Ocean centered ~5S/90E with another focused across the Tropical Northwest Pacific Ocean (TNWP) just east of the Philippines. The South Asian monsoon system has also intensified during the last week. There is also minor enhancement of tropical rainfall around the Americas, and easterly wave activity across northern equatorial Africa has become quite intense.
Some evidence indicates another ~40-day mode of tropical rainfall variability maybe organizing across the equatorial Indian Ocean. However, recent full disk satellite imagery already suggests a quickening of eastward movement of the thunderstorm clusters. We will see if this possible episode organizes into a MJO particularly if it propagates northeast. In any case, we are still stuck in a generally stationary GSDM Stage 1 (more said below) base state having generally weak subseasonal variations. A difference is that our nemesis of 2 regions of Eastern Hemisphere tropical convective forcing may be starting to appear as seasonal transition to the boreal cold season approaches.
Total and relative angular momentum is probably the lowest in at least a year. Both are ~minus 2.5-3 AMUs below the R1 data climatology, and much of the contribution is coming from deep zonal mean easterly wind flow anomalies throughout the tropical and subtropical atmospheres. The largest zonal mean magnitudes are in the Southern Hemisphere (SH), having 200mb easterly wind anomalies ~10m/s. There is also a strong poleward transport signal from the SH tropics into the lower midlatitudes, with a weaker counterpart developing north of the equator. Global relative AAM tendency is ~minus 25 Hadleys as of 11 August and the Global Wind Oscillation (GWO) is as far to left as I have seen it, in the GSDM Stage 4-1 quasi-phase space. The global signal from the surface torques is still weak (for now).
There are currently strong twin upper tropospheric anticyclones with extensive cross-equatorial flow centered ~80E tied to the Indian Ocean (IO) tropical forcing. A weak compensating set of cyclones is present just west of the date line. With the exception of this date line cyclone pair, the entire tropics is dominated by upper tropospheric easterly wind flow, with anomalies ~15-25m/s from Africa into the Atlantic Ocean (favorable for convectively active easterly waves). Rossby wave energy dispersions (RWDs) across the Northern Hemisphere (NH) extratropics interacting with both the IO and TNWP forcing are already impacting North America. Combined with what I expect to be strengthening zonal mean westerly flow across the NH midlatitudes, my notion offered 9 August of the trough-ridge-trough pattern across the USA shifting southeast looks probable. In fact, the western USA may see a return to a barrage of troughs similar to last spring going into September, possibly interacting with a cold air source from Canada.
Week-2 ensemble means from several international operational weather centers are offering varying solutions for North America; however, there has been a recent trend toward a western USA trough with a southeast states ridge. As before, when I see the numerical models trend toward what diagnostic reasoning suggests (per GSDM), my confidence increases. The latter should be the case since I am utilizing a proposed more complete forecast process for subseasonal prediction. Nevertheless, uncertainty is still huge given issues such as the future evolution of the tropical forcing and seasonal transition. The general synoptic pattern across the PNA sector this fall will have some dependency on the GWO. For example, will there be a return to large circuits to GSDM Stage 2, etc. (especially if the warm TNWP stays active)?
Cooler and wetter weather is the most probable trend for particularly the northwest half of the USA during the next 1-3 weeks. The southeast is likely to continue with above normal heat and humidity. It is obviously unclear to say where any possible land falling tropical cyclone will go. However, the need to monitor this situation is likely to increase (see latest statements from the NOAA/NWS/Tropical Prediction Center). The center part of the country is also likely to become increasingly stormy, including severe thunderstorms and flooding rainfall.
Locations from India into China and Southeast Asia are likely to continue with the poundings of intense monsoon rainfall possibly for several more weeks. The TNWP including the Philippines are also probable to stay active with heavy rain and tropical cyclone activity week 1, with perhaps a respite across the west central Pacific for at least week 2. Northern equatorial Africa is likely to stay active through at least week 1, and I think the Atlantic Hurricane Season “is about to get underway”.
Appendix
An experimental quasi-phase space plot of the GSDM utilizing a time series of normalized relative AAM tendency anomaly (Y-axis) and normalized relative AAM anomaly time series (X-axis) can be found at
http://www.cdc.noaa.gov/map/images/gcm/gsdm_95d.jpg
http://www.cdc.noaa.gov/map/images/gcm/gsdm_40d.jpg
http://www.cdc.noaa.gov/map/clim/gsdm.shtml
We call the behavior of this plot the Global Wind Oscillation (GWO). While the intent of the GSDM is to extend current thinking beyond the MJO, the purpose of the GWO is to illustrate the non-oscillatory component of the GSDM.
These are probabilistic statements, and work is ongoing to quantify in future posts. We hope that an opportunity will arise for us 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. My next update should be early next week. In general, due to covering shifts and travel, my postings will be irregular through at least the rest of this month.
Ed Berry
Thursday, August 09, 2007
Still Slaying -- Update
Note: The ESRL/PSD AAM plots now present the “complete” budget (updated through 6 August as of this writing), and will be referred to.
Per CPC and other data sets, global tropical SSTs are generally above average across the Eastern Hemisphere while cooler than normal across the equatorial central and east Pacific and about climatology across the Atlantic Ocean. Anomaly magnitudes vary from ~0.5-2C with the greatest across portions of the equatorial Indian Ocean and west of South America ~110-120W (even ~minus 3 on 8 August). The latter is in the Nino 3 region, and recent enhanced trades have contributed to some cooling during the last week. In fact, subsurface anomalies are as low as roughly minus 3C at 160W/150m per latest 5-day averaged TAO buoy data suggestive of the thermocline being raised. Warmest SST totals remain across the west central-northwest Pacific Ocean and the Caribbean-Gulf of Mexico regions with values ~29-30C. While the global circulation is very much La-Nina like, the SST pattern is only weakly representative of a cold event.
Weather-climate linkage signals continue to be extremely difficult to interpret. The MJO has virtually collapsed. All Wheeler-Hendon type phase space plots from international weather centers have projections less than 1 standard deviation. There is some “tilt” back toward the Indian Ocean, and attention needs to be paid to that for predicting upcoming subseasonal evolution.
As part of a consolidation with an equatorial Rossby mode, about a week ago a northeastward propagating component weakly projecting onto a MJO came out into the Tropical Northwest Pacific Ocean (TNWP). Per coherent modes Hovmollers the signal made it as far east as ~160E. A couple of tropical cyclones have developed in that region since. At this time the most significant tropical forcing remains well back in the Eastern Hemisphere centered at about 15N/120E. There is loose evidence of a tropical convective component moving into the Western Hemisphere. Recent 3-day averaged OLRA present ~minus 70 W/m**2 and lower around the Philippines. There are also negative anomalies north of the equator at the date line (a response to the weak signal moving into the Western Hemisphere) and back across the equatorial Indian Ocean. Notable positive anomalies of OLR are centered on the region of Indonesia.
Full disk satellite imagery nicely shows the convection around the Philippines while extending west and southwest into the Bay of Bengal. This is a change in the orientation discussed late week. Equatorial Africa into the equatorial Indian Ocean is getting more active. I have speculation that in the wake of our recent northward propagating event from the equator into the Southeast Asia, another one may be getting organized. It is possible this may be a form of 20-30 day variability, linked to the extratropics, and within what has been a stationary GSDM Stage 1 base state given boreal summer (discussed below).
Typical of a cold event, global total and relative atmospheric angular momentum (AAM) continues very low, roughly minus 2 Atmospheric Momentum Units (AMUs). The mass term has increased somewhat during the last week. The last MJO event did not add as much zonal mean westerly wind flow to the atmosphere as I previously thought. In fact, the Global Wind Oscillation (GWO) updated through 5 August (5-day average) objectively shows the global circulation solidly in GSDM Stage 1. There was only a pathetic circuit when the MJO came out.
Zonal mean easterly wind flow anomalies have been rapidly increasing throughout the equatorial and subtropical atmospheres over the last few days. At 200mb magnitudes are ~5m/s, flanked by zonal mean westerly wind flow anomalies at the mid and higher latitudes. A meridional distribution of zonal mean zonal wind anomalies such as these supports anomalously strong subtropical ridges from the Devil including the one impacting the USA. Hence global relative AAM tendency is negative (~15 Hadleys) and the surface torques is also trending toward less than zero. Finally, the coriolis torque is also ~minus 10-20 Hadleys when loosely it should be positive at this time.
Animations of upper tropospheric daily mean vector wind anomaly fields suggest twin tropical/subtropical anticyclones redeveloping across the Indian Ocean while the pair across the west central Pacific weakens. There are weak twin cyclones along and just east of the date line. The upper tropospheric Western Hemisphere Pacific Ocean equatorial westerly wind flow anomalies also discussed last week have weakened significantly and are going into the Southern Hemisphere. It was Rossby wave energy dispersions (RWDs) arcing across Asia interacting with the west Pacific anticyclones that contributed to eastward shift of what was the “wicked ridge of the west” (western USA for international readers).I can already see the RWDs coming out of the Indian Ocean that will lead to some retrogression of the Deep South ridge back toward the Plains. This will not be as far west as some numerical ensemble solutions were indicating a week-10 days ago.
Uncertainty remains huge for making any subseasonal forecast, particularly as we start to head into fall. During the last couple of weeks I think have had somewhat better success with what I have believed to be the most probable options. I am going to be bold here and speculate that the MJO dynamical signal, having interactions with the extratropics (forcing-response-feedbacks, etc.), is going to get its act back together across the Indian Ocean during the next week or so and then propagate east-northeast into the TNWP by roughly weeks 3-4. How intense this event may be and its impact on the GWO is unclear. I would like to see the Wheeler-Hendon plots come out of their “1 sigma” circles “soon” across the region of the Indian Ocean. Should this occur, any model, numerical, statistical, etc. is unlikely to predict this possibility right now. It would be interesting to see what “brute force stochastics” might give us right now about MJO and anything else dealing with making predictions in our current lazy GSDM Stage 1 regime. For weeks 1-4, let’s stick with Stage 1 through week 2 possibly getting perturbed a bit to Stage 2 during weeks 3-4.
I think the ramifications for the USA are obvious by now. For much of week 1 deadly intense heat is a good bet centered from the Plains into the Deep South. Severe MCS activity will occur on the northern fringes of this ridge from Hell, likely from the Northern Rockies to the Great Lakes into portions of the eastern USA. This general trough-ridge-trough pattern across the USA should shift southeast weeks 2-4 if the scenario discussed above is realized.
Some respite for portions of India and Bangladesh from the severe storms and flooding may occur during a portion of week 1. However, locations from India into China and Southeast Asia are likely to continue with the poundings of intense monsoon rainfall possibly for several more weeks. As I type a tropical cyclone is trying to develop across the far northern Arabian Sea, which should weaken as it gets inland (but not before causing “problems”). While northern equatorial Africa is getting active, dry stable air from the North Atlantic Ocean has been inhibiting tropical cyclone development of the waves as they move off the West African coast. While I can offer some attribution to this, suffice to say this situation needs to change if there is to be an active Atlantic tropical cyclone season (understanding the Caribbean, etc.). Finally, more strong troughs appear on the horizon for Europe for at least the next couple of weeks.
Stay tuned if we start to observe 2 regions of Eastern Hemisphere tropical convective forcing this fall, the Indian and west central Pacific Oceans. If the former dominates, GSDM Stage 4-1 would be most probable.
Appendix
An experimental quasi-phase space plot of the GSDM utilizing a time series of normalized relative AAM tendency anomaly (Y-axis) and normalized relative AAM anomaly time series (X-axis) can be found at
http://www.cdc.noaa.gov/map/images/gcm/gsdm_95d.jpg
http://www.cdc.noaa.gov/map/images/gcm/gsdm_40d.jpg
http://www.cdc.noaa.gov/map/clim/gsdm.shtml (general link)
We call the behavior of this plot the Global Wind Oscillation (GWO), and one of its purposes is to extend thinking beyond the MJO.
These are probabilistic statements, and work is ongoing to quantify in future posts. We hope that an opportunity will arise for us 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. I will try to do an update early next week. In general, due to covering shifts and travel, my postings will be irregular through at least the rest of this month.
Ed Berry
Per CPC and other data sets, global tropical SSTs are generally above average across the Eastern Hemisphere while cooler than normal across the equatorial central and east Pacific and about climatology across the Atlantic Ocean. Anomaly magnitudes vary from ~0.5-2C with the greatest across portions of the equatorial Indian Ocean and west of South America ~110-120W (even ~minus 3 on 8 August). The latter is in the Nino 3 region, and recent enhanced trades have contributed to some cooling during the last week. In fact, subsurface anomalies are as low as roughly minus 3C at 160W/150m per latest 5-day averaged TAO buoy data suggestive of the thermocline being raised. Warmest SST totals remain across the west central-northwest Pacific Ocean and the Caribbean-Gulf of Mexico regions with values ~29-30C. While the global circulation is very much La-Nina like, the SST pattern is only weakly representative of a cold event.
Weather-climate linkage signals continue to be extremely difficult to interpret. The MJO has virtually collapsed. All Wheeler-Hendon type phase space plots from international weather centers have projections less than 1 standard deviation. There is some “tilt” back toward the Indian Ocean, and attention needs to be paid to that for predicting upcoming subseasonal evolution.
As part of a consolidation with an equatorial Rossby mode, about a week ago a northeastward propagating component weakly projecting onto a MJO came out into the Tropical Northwest Pacific Ocean (TNWP). Per coherent modes Hovmollers the signal made it as far east as ~160E. A couple of tropical cyclones have developed in that region since. At this time the most significant tropical forcing remains well back in the Eastern Hemisphere centered at about 15N/120E. There is loose evidence of a tropical convective component moving into the Western Hemisphere. Recent 3-day averaged OLRA present ~minus 70 W/m**2 and lower around the Philippines. There are also negative anomalies north of the equator at the date line (a response to the weak signal moving into the Western Hemisphere) and back across the equatorial Indian Ocean. Notable positive anomalies of OLR are centered on the region of Indonesia.
Full disk satellite imagery nicely shows the convection around the Philippines while extending west and southwest into the Bay of Bengal. This is a change in the orientation discussed late week. Equatorial Africa into the equatorial Indian Ocean is getting more active. I have speculation that in the wake of our recent northward propagating event from the equator into the Southeast Asia, another one may be getting organized. It is possible this may be a form of 20-30 day variability, linked to the extratropics, and within what has been a stationary GSDM Stage 1 base state given boreal summer (discussed below).
Typical of a cold event, global total and relative atmospheric angular momentum (AAM) continues very low, roughly minus 2 Atmospheric Momentum Units (AMUs). The mass term has increased somewhat during the last week. The last MJO event did not add as much zonal mean westerly wind flow to the atmosphere as I previously thought. In fact, the Global Wind Oscillation (GWO) updated through 5 August (5-day average) objectively shows the global circulation solidly in GSDM Stage 1. There was only a pathetic circuit when the MJO came out.
Zonal mean easterly wind flow anomalies have been rapidly increasing throughout the equatorial and subtropical atmospheres over the last few days. At 200mb magnitudes are ~5m/s, flanked by zonal mean westerly wind flow anomalies at the mid and higher latitudes. A meridional distribution of zonal mean zonal wind anomalies such as these supports anomalously strong subtropical ridges from the Devil including the one impacting the USA. Hence global relative AAM tendency is negative (~15 Hadleys) and the surface torques is also trending toward less than zero. Finally, the coriolis torque is also ~minus 10-20 Hadleys when loosely it should be positive at this time.
Animations of upper tropospheric daily mean vector wind anomaly fields suggest twin tropical/subtropical anticyclones redeveloping across the Indian Ocean while the pair across the west central Pacific weakens. There are weak twin cyclones along and just east of the date line. The upper tropospheric Western Hemisphere Pacific Ocean equatorial westerly wind flow anomalies also discussed last week have weakened significantly and are going into the Southern Hemisphere. It was Rossby wave energy dispersions (RWDs) arcing across Asia interacting with the west Pacific anticyclones that contributed to eastward shift of what was the “wicked ridge of the west” (western USA for international readers).I can already see the RWDs coming out of the Indian Ocean that will lead to some retrogression of the Deep South ridge back toward the Plains. This will not be as far west as some numerical ensemble solutions were indicating a week-10 days ago.
Uncertainty remains huge for making any subseasonal forecast, particularly as we start to head into fall. During the last couple of weeks I think have had somewhat better success with what I have believed to be the most probable options. I am going to be bold here and speculate that the MJO dynamical signal, having interactions with the extratropics (forcing-response-feedbacks, etc.), is going to get its act back together across the Indian Ocean during the next week or so and then propagate east-northeast into the TNWP by roughly weeks 3-4. How intense this event may be and its impact on the GWO is unclear. I would like to see the Wheeler-Hendon plots come out of their “1 sigma” circles “soon” across the region of the Indian Ocean. Should this occur, any model, numerical, statistical, etc. is unlikely to predict this possibility right now. It would be interesting to see what “brute force stochastics” might give us right now about MJO and anything else dealing with making predictions in our current lazy GSDM Stage 1 regime. For weeks 1-4, let’s stick with Stage 1 through week 2 possibly getting perturbed a bit to Stage 2 during weeks 3-4.
I think the ramifications for the USA are obvious by now. For much of week 1 deadly intense heat is a good bet centered from the Plains into the Deep South. Severe MCS activity will occur on the northern fringes of this ridge from Hell, likely from the Northern Rockies to the Great Lakes into portions of the eastern USA. This general trough-ridge-trough pattern across the USA should shift southeast weeks 2-4 if the scenario discussed above is realized.
Some respite for portions of India and Bangladesh from the severe storms and flooding may occur during a portion of week 1. However, locations from India into China and Southeast Asia are likely to continue with the poundings of intense monsoon rainfall possibly for several more weeks. As I type a tropical cyclone is trying to develop across the far northern Arabian Sea, which should weaken as it gets inland (but not before causing “problems”). While northern equatorial Africa is getting active, dry stable air from the North Atlantic Ocean has been inhibiting tropical cyclone development of the waves as they move off the West African coast. While I can offer some attribution to this, suffice to say this situation needs to change if there is to be an active Atlantic tropical cyclone season (understanding the Caribbean, etc.). Finally, more strong troughs appear on the horizon for Europe for at least the next couple of weeks.
Stay tuned if we start to observe 2 regions of Eastern Hemisphere tropical convective forcing this fall, the Indian and west central Pacific Oceans. If the former dominates, GSDM Stage 4-1 would be most probable.
Appendix
An experimental quasi-phase space plot of the GSDM utilizing a time series of normalized relative AAM tendency anomaly (Y-axis) and normalized relative AAM anomaly time series (X-axis) can be found at
http://www.cdc.noaa.gov/map/images/gcm/gsdm_95d.jpg
http://www.cdc.noaa.gov/map/images/gcm/gsdm_40d.jpg
http://www.cdc.noaa.gov/map/clim/gsdm.shtml (general link)
We call the behavior of this plot the Global Wind Oscillation (GWO), and one of its purposes is to extend thinking beyond the MJO.
These are probabilistic statements, and work is ongoing to quantify in future posts. We hope that an opportunity will arise for us 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. I will try to do an update early next week. In general, due to covering shifts and travel, my postings will be irregular through at least the rest of this month.
Ed Berry
Friday, August 03, 2007
Update from the Slayer
The spatial distribution of global tropical/subtropical SSTs is similar to what was discussed on 31 July. There remains warmth across the Indian and west central Pacific Oceans with cool anomalies across portions of equatorial Indonesia. Impressive warmth remains around the South China Sea and northwest Pacific Ocean with SST totals in excess of 30C. The equatorial East Pacific cool anomalies are ~minus 1-2C while the subsurface continues to moderate.
Tropical convective forcing has become better organized centered ~10-15N/120E while extending in a band from India east-southeast into the west central and northwest Pacific Ocean. Three-day averaged OLRA are roughly minus 50-90W/m**2 with this enhanced rainfall, and there has been a slow east-northeastward shift of this activity toward the Tropical Northwest Pacific Ocean (TNWP). Other somewhat enhanced rainfall continues along the East Pacific ITCZ and Africa north of the equator.
Animations of upper tropospheric daily mean vector wind anomalies do give a nice signal of slow eastward movement across the tropics/subtropics. Since 31 July robust twin anticyclones have appeared ~140E having 150mb anomalies around 15-30 m/s while weaker anticyclones persist across Africa. There are downstream anomalous troughs just west of the date line with equatorial westerlies (15-25 m/s anomalies) across the Western Hemisphere to about South America. Rossby wave energy dispersions from both hemispheres are interacting with these features. Included is a Northern Hemispheric component across the Pacific Ocean to the first in a series of troughs about to dig along the west coast of the USA.
This circulation response working with other dynamical processes such as global frictional and mountain torques is adding westerly flow to the atmosphere. In fact, calculated relative AAM tendency (per R1 data) through 28 July which includes budget terms such as Coriolis torque and transports was ~plus 20 Hadleys. However, total AAM remains very low, roughly 3 AMUs (Atmospheric Momentum Units) below normal, and the Global Wind Oscillation (GWO) updated through 29 July is well to the left. The global circulation continues to behave La-Nina like, solidly in GSDM Stage 1. As the Eastern Hemisphere tropical forcing continues into the TNWP, interactions with the extratropics do include the possibility of a mountain-frictional torque index cycle. The latter would increase zonal mean westerly flow across the subtropical atmospheres of both hemispheres (ideally).
I remain unchanged in my feeling that this GSDM Stage 1 base state should be perturbed enough for the GWO to orbit to GSDM Stage 2. The magnitude and amount of persistence of this perturbation are unclear, particularly considering seasonal cycle issues. Overall, we may see GSDM Stage 1-2 for the next few weeks, then collapse to Stage 4-1 per reasons discussed last posting. Uncertainty remains huge about this specific subseasonal issue. However, at least synoptically for the USA the major circulation pattern reversal with an anomalous western USA trough and Deep South ridge is probable. Furthermore, Rossby wave energy dispersions arcing from the Indian Ocean tropical forcing into the higher latitudes will contribute to an anomalous deep cyclone covering much of western and central Canada. Hence a full latitude trough extending into the western USA is a good forecast meaning the polar westerlies should shift southward. Most numerical ensemble prediction schemes have now captured this change.
There has been some thought from at least one official forecast center based on model output the ridge may return to the Rockies during week 2. While this is an option, I could easily argue a “cold/wet regime” from the east slopes of the Rockies into the Plains weeks 2-3 should the TNWP become extremely active with several typhoons. Again, this emphasizes the uncertainty given the stochastic nature of the atmosphere. Right now there is little if any support from ensembles for a central USA cold/wet pattern weeks 2-3 (for example, like August 2004 as discussed in one of our weather-climate discussions on the ESRL/PSD MJO web site). Should this less probable scenario happen, it is currently outside the envelope range of uncertainty sampled any of the operational global models (multi-model ensemble spread). Again, this emphasizes the need for a weather-climate linkage diagnostic component (the GSDM) including daily monitoring to correctly making subseasonal predictions as part of a complete forecast process. Multi-model ensembles alone are not enough, particularly considering their biases, data assimilation issues, not predicting tropical convection correctly after about day 5, etc…………..
There is little change in my outlooks for weeks 1-3 (through ~24 August). The “wicked ridge of the west” will expand into the Deep South and eastern USA as the westerlies come southward into the Rockies and Plains downstream from the western USA trough. A reversal of the temperature and precipitation anomalies observed across much of the country for the last 4-6 weeks may occur. Severe MCS activity with flooding rainfall is likely from the Northern Plains into the Ohio Valley periodically through at least the middle of this month. This “storm track” should slowly shift south. Intense heat may focus on the south central into the mid-Atlantic states, along with high dewpoints.
Internationally, severe thunderstorms and heavy rainfall are probable from southern China into the Philippines at least weeks 1-2. Additional typhoon activity is probable for the TNWP weeks 1-3, possibly impacting locations such as China and Japan. Finally, tropical waves moving west from Africa have been recently intensifying. However, for numerous reasons I am now unclear about the risk of tropical cyclone development across the North Atlantic Ocean at least through week 2. The latter is yet another example why there is not a linear relationship between La-Nina basic states and an increased risk of North Atlantic tropical cyclone activity, at least subseasonally.
Appendix
An experimental quasi-phase space plot of the GSDM utilizing a time series of normalized relative AAM tendency anomaly (Y-axis) and normalized relative AAM anomaly time series (X-axis) can be found at
http://www.cdc.noaa.gov/map/images/gcm/gsdm_95d.jpg
http://www.cdc.noaa.gov/map/images/gcm/gsdm_40d.jpg
We call the behavior exhibited by this plot the Global Wind Oscillation (GWO), and one of its purposes is to extend thinking beyond the MJO.These are probabilistic statements, and work is ongoing to quantify in future posts. We hope that an opportunity will arise for us 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.
My next posting will not be until late next week since I have to cover a set of overnight shifts. In general, due to covering shifts and travel, my discussions will be irregular through at least the end of this month.
Ed Berry
Tropical convective forcing has become better organized centered ~10-15N/120E while extending in a band from India east-southeast into the west central and northwest Pacific Ocean. Three-day averaged OLRA are roughly minus 50-90W/m**2 with this enhanced rainfall, and there has been a slow east-northeastward shift of this activity toward the Tropical Northwest Pacific Ocean (TNWP). Other somewhat enhanced rainfall continues along the East Pacific ITCZ and Africa north of the equator.
Animations of upper tropospheric daily mean vector wind anomalies do give a nice signal of slow eastward movement across the tropics/subtropics. Since 31 July robust twin anticyclones have appeared ~140E having 150mb anomalies around 15-30 m/s while weaker anticyclones persist across Africa. There are downstream anomalous troughs just west of the date line with equatorial westerlies (15-25 m/s anomalies) across the Western Hemisphere to about South America. Rossby wave energy dispersions from both hemispheres are interacting with these features. Included is a Northern Hemispheric component across the Pacific Ocean to the first in a series of troughs about to dig along the west coast of the USA.
This circulation response working with other dynamical processes such as global frictional and mountain torques is adding westerly flow to the atmosphere. In fact, calculated relative AAM tendency (per R1 data) through 28 July which includes budget terms such as Coriolis torque and transports was ~plus 20 Hadleys. However, total AAM remains very low, roughly 3 AMUs (Atmospheric Momentum Units) below normal, and the Global Wind Oscillation (GWO) updated through 29 July is well to the left. The global circulation continues to behave La-Nina like, solidly in GSDM Stage 1. As the Eastern Hemisphere tropical forcing continues into the TNWP, interactions with the extratropics do include the possibility of a mountain-frictional torque index cycle. The latter would increase zonal mean westerly flow across the subtropical atmospheres of both hemispheres (ideally).
I remain unchanged in my feeling that this GSDM Stage 1 base state should be perturbed enough for the GWO to orbit to GSDM Stage 2. The magnitude and amount of persistence of this perturbation are unclear, particularly considering seasonal cycle issues. Overall, we may see GSDM Stage 1-2 for the next few weeks, then collapse to Stage 4-1 per reasons discussed last posting. Uncertainty remains huge about this specific subseasonal issue. However, at least synoptically for the USA the major circulation pattern reversal with an anomalous western USA trough and Deep South ridge is probable. Furthermore, Rossby wave energy dispersions arcing from the Indian Ocean tropical forcing into the higher latitudes will contribute to an anomalous deep cyclone covering much of western and central Canada. Hence a full latitude trough extending into the western USA is a good forecast meaning the polar westerlies should shift southward. Most numerical ensemble prediction schemes have now captured this change.
There has been some thought from at least one official forecast center based on model output the ridge may return to the Rockies during week 2. While this is an option, I could easily argue a “cold/wet regime” from the east slopes of the Rockies into the Plains weeks 2-3 should the TNWP become extremely active with several typhoons. Again, this emphasizes the uncertainty given the stochastic nature of the atmosphere. Right now there is little if any support from ensembles for a central USA cold/wet pattern weeks 2-3 (for example, like August 2004 as discussed in one of our weather-climate discussions on the ESRL/PSD MJO web site). Should this less probable scenario happen, it is currently outside the envelope range of uncertainty sampled any of the operational global models (multi-model ensemble spread). Again, this emphasizes the need for a weather-climate linkage diagnostic component (the GSDM) including daily monitoring to correctly making subseasonal predictions as part of a complete forecast process. Multi-model ensembles alone are not enough, particularly considering their biases, data assimilation issues, not predicting tropical convection correctly after about day 5, etc…………..
There is little change in my outlooks for weeks 1-3 (through ~24 August). The “wicked ridge of the west” will expand into the Deep South and eastern USA as the westerlies come southward into the Rockies and Plains downstream from the western USA trough. A reversal of the temperature and precipitation anomalies observed across much of the country for the last 4-6 weeks may occur. Severe MCS activity with flooding rainfall is likely from the Northern Plains into the Ohio Valley periodically through at least the middle of this month. This “storm track” should slowly shift south. Intense heat may focus on the south central into the mid-Atlantic states, along with high dewpoints.
Internationally, severe thunderstorms and heavy rainfall are probable from southern China into the Philippines at least weeks 1-2. Additional typhoon activity is probable for the TNWP weeks 1-3, possibly impacting locations such as China and Japan. Finally, tropical waves moving west from Africa have been recently intensifying. However, for numerous reasons I am now unclear about the risk of tropical cyclone development across the North Atlantic Ocean at least through week 2. The latter is yet another example why there is not a linear relationship between La-Nina basic states and an increased risk of North Atlantic tropical cyclone activity, at least subseasonally.
Appendix
An experimental quasi-phase space plot of the GSDM utilizing a time series of normalized relative AAM tendency anomaly (Y-axis) and normalized relative AAM anomaly time series (X-axis) can be found at
http://www.cdc.noaa.gov/map/images/gcm/gsdm_95d.jpg
http://www.cdc.noaa.gov/map/images/gcm/gsdm_40d.jpg
We call the behavior exhibited by this plot the Global Wind Oscillation (GWO), and one of its purposes is to extend thinking beyond the MJO.These are probabilistic statements, and work is ongoing to quantify in future posts. We hope that an opportunity will arise for us 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.
My next posting will not be until late next week since I have to cover a set of overnight shifts. In general, due to covering shifts and travel, my discussions will be irregular through at least the end of this month.
Ed Berry
Subscribe to:
Posts (Atom)