Welcome to American Weather
  • entries
  • comments
  • views

Entries in this blog


The Quasi-Biennial Oscillation (QBO) is a cycle of zonal wind in the equatorial stratosphere with a period that varies between 24 and 30 months. This oscillation is a product of downward propagating alternating wind regimes. The current method of monitoring this oscillation is through an index, calculated by the zonal wind anomaly at 30hPa averaged along the equator. This method excludes information on the vertical structure of winds in the stratosphere, and presents the QBO as a one-dimensional temporal oscillation.

Presented here is a new framework for monitoring the QBO. This framework incorporates both the oscillation in time and in space. The two leading principle components (PCs) of equatorial stratospheric zonal winds are calculated from NCEP/NCAR reanalysis monthly mean data. These PCs are then standardized and used as X and Y coordinates in a 2-dimensional phase space. A phase angle can then be calculated from the coordinates, forming a new index that is much more representative of the state of the stratosphere. The results show a very clear pattern of zonal wind and temperature regimes in the stratosphere.

Furthermore, from each of these phases, there are physical connections to characteristics in the troposphere on a monthly to seasonal basis. These include the distribution of tropical cyclone activity, the El Niño / Southern Oscillation (ENSO), and the extratropical pattern. This new framework for monitoring the QBO is shown to be much more applicable to seasonal forecasting.





Compare the EOF-constructed oscillation to a 52-month reanalysis segment




The persistent low index blocking pattern that has lent itself to a frigid end to winter and start of spring is finally nearing an end.

An oldie but a goodie reference: Namias, J., 1950: the index cycle and its role in the general circulation. J. Meteor., 7, 130-139.

Likely culprits to our extreme blocking ... the prolonged easterly QBO and a final "warming" that resulted in a strong wave-1 response in the stratosphere.

The AO has been down in the deepest darkest depths of the basement for a while now. There is unanimous agreement in the GEFS for a world-class reversal in the NH annular mode ... i.e. a rise of over 6SD in 10 days!!! That's pretty amazing.



Tropical forcing has been persistent near the dateline over the last month, but a new MJO wave is forecast to cycle through phases 3-4-5 over the next two weeks, triggering a wave response that will have significant implications in the mid latitude pattern.

All the energy trapped in the Pacific jet gets ejected eastward and dumps into the western US. The GEFS and ECMWF are in fairly good agreement on this pattern development with just differences in the timing. Notice on the spaghetti plot, there are just some minor phase shift differences from the Pacific across the CONUS. Pretty impressive agreement actually for day 8.




Really impressive jet structure. Notice dual jet support over the Plains with the strong Pacific jet diving into the backside of the trough over the western US and the polar jet across the Great Lakes and Northeast intensified by the tightened temperature gradient in the high index pattern.


In terms of severe weather potential, the rain storm this week also doesn't hurt ... adding moisture to the boundary layer.

By the weekend, looking at large scale moisture convergence from the Gulf to the polar front across the Midwest and Northeast.


A lot of good stuff coming together to support a rapid increase in Springtime severe weather activity over the Plains by early next week!

Source: Pattern Change and Severe Weather Opportunities


To start, here's the equatorial zonal wind anomalies so far this spring. Can see the -QBO holding on by its thumbs, with the westerlies starting to descend and near 30mb (first days with a + anomaly since July 2011 WOW)


The current profile fits the 330 degree phase of my "index" pretty well. Descending westerlies above 30mb, easterlies centered around 70mb, and neutral right around 30mb.


The phase I would be keying in on for hurricane season based on average "residence time" is 60 degrees.


Here's the tropopause pressure correlation for phase 60. It's a pretty weak signal, with on average zero correlation in the MDR and negative in the central Atlantic. This may support more disturbances in the central Atlantic more recurvatures of any TCs that do form.

(Negative = a higher tropopause height, Positive = a lower tropopause height)


Here's the same map for phase 90. Notice the equator to off-equator dipole tightens and intensifies, and the Gulf and NW Caribbean is under a very low correlation (high tropopause height) and that band extends across the Atlantic between 15N and 30N.

That dipole that I mention is tightest and most intense around phase 120.


So if the QBO reaches phase 90 during the season (essentially a full-blown +QBO), this suggests an increased TC threat in the NW Caribbean and Gulf. Of course this is one factor, and probably still relatively insignificant, but something that's interesting to discuss anyway


The big consensus among model guidance at this point is that a piece of energy from the trough over the Northwest Pacific will jump to the Gulf of Alaska by Wednesday morning, then drop southeastward into California Thursday night into Friday.

Similar to the last event, there is a weak disturbance in the southern stream that gets ejected out ahead of the trough dropping into the Southwest ... and able to lift northeastward as the polar vortex over southeast Canada retreats poleward. So the result is some moisture return ahead of the disturbance Wednesday and Thursday ... maybe some rain too, but let's not get carried away.

The real differences between the GFS and CMC and ECMWF is with the handling of the individual impulses within the western trough. CMC and ECMWF like swinging the lead impulse into eastern Colorado Saturday afternoon. Amplitude, tilt, and timing of this particular piece of energy will be crucial. At this point, the projected evolution brings a warm front through Oklahoma on Saturday with round 1 of thunderstorm potential, and a surface low developing in eastern CO.

The ECMWF and CMC deepen this surface low into western Kansas Saturday night ... really pretty awesome placement for OK. Euro increases CAPE substantially Saturday night. The cold front catches the dryline sometime Saturday night into Sunday morning for round 2 ... so timing is an issue, but verbatim this would no doubt be a significant overnight event.

The trough continues eastward with that primary upper level impulse sharpening over the southern Plains, and the surface low intensifying across the central Plains to Midwest, introducing severe threats for Sunday and Monday for the Dixie Alley area.





The Quasi-Biennial Oscillation (QBO) as defined by the AMS:

An oscillation in the zonal winds of the equatorial stratosphere having a period that fluctuates between about 24 and 30 months. This oscillation is a manifestation of a downward propagation of winds with alternating sign.

The QBO provides a good first-order look at the background state of the stratosphere, which can be significant for seasonal prediction. It's reliable periodicity (probably the most consistent atmospheric oscillation known today) also makes it a useful tool in forecasting.

The QBO is measured by the anomaly of the winds at 30mb averaged along the equator. Here we run into the problem which is the joy with which people embrace one-dimensional indices. It is very easy to fall into the trap of calculating all sorts of statistics on these indices without any regard to the physical meaning behind them.

For as long as the QBO has been known and calculated monthly, people have been trying to determine linear correlations to all sorts of things from regional temperature anomalies to tropical cyclone activity. So first of all, many do this with complete lack of respect for the 24,000ft of vertical that divides the QBO from their backyard weather station, and no recognition of the physical relationships that can in fact draw a bridge that far.

So yes, there can be drawn physical relationships from the QBO down to sensible weather at the surface. However, not in the way everyone assumes. In fact, the relationships are probably much better than anyone realizes! This is because people only look at the number, and do not use the QBO to its full potential.

The key is in the second sentence of the AMS definition ... that the QBO is a product of the downward propagation of wind anomalies ... which can be visualized by this well-recognized time-height section:


Then remember exactly what the QBO is measuring: wind. That means a gradient by height is equivalent to wind shear. Westerly and easterly wind regimes propagating downward define the QBO, but between these regimes exists a shear zone. When one regime is descending below 30mb and the other regime is starting to develop above 30mb the QBO index is around neutral ... indicating the shear zone is around 30mb ... However there can be westerly shear and there can be easterly shear and reading a QBO of zero tells you nothing about the true state of the stratosphere.

So I am proposing that we try to expand our use of the QBO ... at the very least we can recognize it as a two dimensional index that oscillates in time and space.

I am working to develop what I call the Multivariate QBO Index (MQI). To start things off, I have simply looked at 30mb and 50mb wind data ... which as you can see both oscillate with time with similar periodicity, with a phase shift between the two indicating the descending nature of the oscillation.


I have taken this data and computed the principle components of variance to develop two orthogonal indices. I standardized the two indices, and then plotted them in 2D phase space. You can see how consistent the path is through phase space for every period of the MQI. The red line is a Barnes-like-interpolated mean at a one-degree interval. From this method, we can analyze both amplitude and phase (TRUE phase -- not just positive or negative anymore).

PC1 plotted on the horizontal axis is similar to the shear near 30mb ... or dQBO/dt, and PC2 plotted on the vertical axis is similar to the QBO.


In order to calculate the mean, I used a cosine weighting function at one-degree intervals. Below I have plotted the resulting amplitude of the mean as a function of phase angle (in red). In blue, I have plotted the total weight by phase angle ... this gives an indication of the "residence time" of the MQI around that particular phase.


Future work...

First of all, I hope to refine this index a little by using more levels within the stratosphere.

Secondly, using this index, I plan on running statistics on various motions in the troposphere ... meaningful statistics, given information on the full QBO now.

Most of all, I anticipate a better relationship to tropical cyclone activity than has been shown in the past.



Stuck in northwest flow for the next week, any current weather discussion here at OU has not been able to escape the use of the word "boring". However, a change is gonna come.

One of the distinct characteristics that I'm seeing of the pattern over the next week is that within this wave system, the sub-polar vortex over eastern Canada and the Northeast overwhelms the ridging over western North America. I compared it to RKW theory and the balance of vorticity in the system ... In this case the stronger trough leads to the eastern edge of the ridge slowly getting broken down ... can watch packets of anticyclonic vorticity erode away and get sucked underneath the trough

This is also why I have been very cautious about getting excited about any storm prospects for the East. Essentially we have a smooth long wave trough with periodic impulses of stronger flow consequence of s/w ridging swinging south of the trough and increasing mass gradients ... NOT conducive to amplifying disturbances for the East.

Anyway, back to the point -- this is NOT a stable pattern. Eventually, all the anticyclonic vorticity with the western ridging will get eroded southeastward ... probably by the end of next week as indicated by the long range models.

The immediate result will be a warm up for most of the Plains for a day or two.

The second result will be opening up the CONUS to Pacific disturbances ... with the blocking ridge now out of the picture.

IMPORTANT -- All the while, we still have the sub-polar vortex sitting off the northeast, over Newfoundland and the North Atlantic ... the effect being persistent background upper level confluence over the Great Lakes, Northeast, and eastern Canada.

I could see the first in the series of disturbances off the Pacific being a significant storm for the Ohio Valley to Northeast next weekend as some ridging still stands in the West.

However, as the pattern shifts to more of an RNA configuration by the following week, the next s/w (around January 28) could have a chance to dig earlier into the central Plains ... then remember the confluence to the northeast to keep the short wave from breaking north and maintain upper level dynamics ... could be a significant winter storm for someone to the north and perhaps a thunderstorm/rain threat for someone to the south.


Surprised I haven't seen any discussion about this bomb on the board.

Check out the DT pressure map I attached below. You'll never see a better signature of a tropopause fold. And obviously all sorts of upper level support with massive divergence aloft.

From Ryan Maue:

I used the full-resolution (T574) surface pressure grids from NCEP GFS for the data points:

1000 mb ... (13/12z)

994 mb ... (13/18z)

989 mb ... (14/00z)

978 mb ... (14/06z)

960 mb ... (14/12z)

946 mb ... (14/18z)

940 mb ... (15/00z)






Prolonged negative QBO

This latest easterly QBO period is on track to become the "strongest" episode on record. Below I have plotted the QBO along with an integrated QBO index which very simply sums the monthly QBO across each full easterly period.

So far this current -QBO period has an integrated magnitude of -280.62 ... which falls just short of the 2001 -QBO period with a magnitude of -280.79.

The mean is -191.6 and standard deviation is 48.6.

Data so far for January 2013 suggests the monthly QBO should come in negative once again. It's possible the recent SSW has helped stunt the descent of the westerlies, and is prolonging the easterly QBO.

This will put the -QBO period at 18 months long ... behind only 2001 which was 19 months.






For posterity's sake, here is my final call for snowfall totals for the New England area.

Expect maximum accumulations in the Monadnocks region of southwest NH and in a swath from eastern NH through western and central Maine ... where up to 18" could fall.



Using 00z NCEP (including 20 ensemble members), let's begin by defining the EOFs describing the greatest amount of variance in the ensemble forecasts:


Alright what is this telling us?

First of all, note the asymmetry in the mean/spread ... there is a maximum in the spread over E PA and NJ ... to the west-northwest of the mean low.

EOF1 -- explains 46.6% of the variance -- describes the depth of the low for the cluster of solutions to the NW of the mean ... in general the average SLP across the domain of the low ... so a positive EOF1 is related to lower SLP.

EOF2 -- explains 21.5% of the variance -- describes the west-east position of the low ... positive EOF2 is west of the mean, negative is east of the mean.

EOF3 -- explains 18.2% of the variance -- describes the north-south position of the low, but also the amplitude of the system ... notice the maximum is colocated with the mean low ... a negative EOF3 is associated with a stronger low off the delmarva peninsula with higher pressures to the north and northwest.

Perhaps as you can tell, a negative EOF3 pattern is actually quite similar to the SLP pattern we might look for in an East Coast winter storm threat. So although it explains less of the variance in the ensemble forecast, I'm personally most interested in this EOF.

So now let's use this forecast diagnostic and find the correlation to a variable in the ensemble forecast, such as 500mb height.


Here we can see that the greatest correlation (sensitivity of the EOF3 to the 500mb height) is associated with the trough moving through the Northeast and Quebec over the next 48 hours. Lower heights over the northwest Atlantic over the next two days correlates with a negative EOF3 ... and a SLP pattern characteristic of Northeast snowstorms.

More later.


Let's begin with the state of winter right now. Where is it? Recent reports say Russia has found it


In other news, the planetary reconfiguration that is occurring right now. Major ridging burgeoning from the Aleutians to the Bering Strait and right up over the North Pole by Saturday. The effect? Well at 22/00z that's 125kts at the DT crossing the NP toward the western hemisphere.


Clear at this stage the magnitude of nonlinear processes leading to anticyclonic breaking of the ridge over the NP toward Greenland this weekend. Quite a signal for developing negative AO and negative NAO regime.

At the same time, we have this blizzard ongoing over the Plains associated with an impressive trough, continuing to amplify tomorrow (around -2.5 to -3SD anomaly at 500mb). The trough breaks over the Northeast into Quebec this weekend, amplifying downstream ridging from the northern Atlantic into the Davis Strait. By Christmas Eve, this ridge is starting to bridge with the NP ridge. By Christmas Day, we're actually on our way to achieving a quasi-stable anticyclone from the Hudson Bay to Baffin Island ... i.e. a west-based negative NAO.


This pattern is clearly characterized by a tumble in both the AO and NAO domains as seen on the forecast maps I'm presenting here.

Notice also in the above map the disturbance (lowered DT / PV anomaly) over the Davis Strait. Tracking it back in time reveals that it originates over eastern Russia currently and is brought around the Pacific ridge as it is bursting poleward. From the Davis Strait, this disturbance continues westward (south of the anticyclone) into eastern Quebec and Newfoundland and Labrador by the 27th.

In the end this disturbance in fact plays a crucial role in the 50N/50W slot, with its miriad relationships in Northeast snowstorms.

So I see you perked up at the mention of the word "snowstorm." Yes, I suppose I should get to the actual storm itself.

The storm of interest originates as a shortwave from the GOA vortex. The shortwave is ejected into the Pacific Northwest on the 24th. At this time, attention turns back to the North Pacific where a strong shortwave is plowing under the recently broken-off ridging from the Aleutians to the NP. The shortwave continues eastward, amplifies and breaks just off the west coast. This results in the Pacific jet briefly splitting as ridging amplifies downstream over the western U.S. ...can also note a relate spike in the PNA at this time ... so certainly convergence of teleconnections supporting this event.


Coinciding with this PNA spike, the shortwave associated with our storm amplifies into the southern Plains by late on the 25th-26th.

The general progression of events that I outlined above is supported in the model guidance. But some minor discrepancies leads to large disagreements in the end result for the eastern third of the nation.

Some of these disagreements I think can be related to the span of time between the ejection of the shortwave and the PNA spike ... the greater separation there is between these two events, the further east the shortwave will amplify.

...Can be thought of simply as timing of the group energy catching up to the shortwave.

By the evening of the 26th, we're watching the development of beautiful dual-jet structure, with excellent upper level mass evacuation over an intensifying low.

The ECMWF has tended to want to cut this low inland ... actually track it along the spine of the Appalachians, meanwhile attempting but failing to produce a secondary low along the coastal baroclinic zone. The GFS and CMC both achieve this secondary development with the coastal low taking over.

Below I've included some maps from the 20/12z run of the GFS that presents the features I discussed above. It's worth noting on the DT theta map signs of convection wrapping into the system from the central Atlantic -->

Meteorology ... helps amplify downstream ridging and the overall system.

Modelology... can introduce potential convective feedback errors in the model forecast.




Some things to think about for the coming week.


A significant cyclogenesis event in the North Pacific will be the primary catalyst of a pattern change over the next few days.

Let's take a look at the evolution over the Pacific into western North America.

Notice our N Pac storm gets dual jet support and really goes to town over the Aleutians. Wait isn't that where a massive ridge was last week? ... or for that matter the last three months. Well that's one thing that changes. The ridging once over the Aleutians and Bering Straight retrogrades west, while the N Pac storm breaks cyclonically northward. It's really a beautiful event. Minor flat ridging over the eastern Pacific gets pumped up into the Gulf of Alaska, with an intensifying polar jet streak pushing north into the Yukon.



What happens after that? Fantastic classic anticyclonic wave breaking into British Columbia! The result is a 170kt+ northerly jet diving into the Pacific Northwest. So you know the trough and all that energy bottled up over Alaska and the Gulf? Well a lot of that comes pouring southeast on the wings of this jet into the northern Rockies



So that's the atmosphere's plans for Saturday.