The Northern Hemisphere is having an old fashioned winter. I thought it might be of interest to look back at history for any lessons that might be there.
The Arctic Oscillation Index compares mid latitude sea level pressure with sea level pressure in the Arctic. There are other useful ways to compute the index based upon atmospheric phenomena that vary with the index but this is the simplest way to think of it.
The Arctic Oscillation Index and the Antarctic Oscillation Index change primarily with polar atmospheric pressure. The flux of pressure at the poles is large. In the mid latitudes the flux of pressure is small. If one observes these indexes (or polar pressure) over time it is plain that they frequently move together. Indeed, over the best part of the last sixty years there has been a loss of atmospheric mass and pressure at both poles. That process is now reversing, confounding a prediction that the AO would increase with the proportion of so called greenhouse gases in the atmosphere. The implications for the climate of the northern hemisphere are discussed in this post.
Figure 1 Sea Level pressure at the poles in summer and winter
Let us imagine that a ( mysterious) force that is capable of shifting atmospheric mass away from the poles and towards the equator increases over a period of sixty years. Then it begins to relax. First to respond is the place that exhibits the largest inter-annual fluctuation, the Arctic in winter. It is like the canary in the coal mine. Since about 1990 there is a just a small gain in the Antarctic in southern summer. By contrast the Arctic, at precisely the same time of the year is, ‘up, up and away’.
Last to respond will likely be the Antarctic in winter, where pressure is higher than anywhere else on the globe. The Antarctic is a large block of ice about as big as the USA and it doubles its surface area in winter. The southern hemisphere at any latitude is colder than the northern hemisphere at the same latitude. The pressure that the atmospheric column exerts at the surface depends upon its temperature and density. There is a natural swing of atmospheric mass towards the winter hemisphere. Cold air is denser, warm air is less dense. So, depending upon the relative temperature of the hemsipheres the swing is modulated from season to season. It follows that Arctic pressure depends upon Antarctic pressure. The Antarctic is far and away the bigger sink.
So, Arctic sea level pressure has been increasing fast since 1990, both in summer and winter as we see in figures 1 and 2.
Figure 2 SLP is sea level pressure. 12 MMAv. is 12 month moving average. AO is the Arctic Oscillation Index.
If we tip the AO on its head (as in figure 2) we see that the Arctic Oscillation Index is actually a very good measure of sea level pressure in the Arctic. Notice that the right hand axis of figure 2 is reversed with negative values on top.
Just in passing, do you notice anything in particular about figure 2? It’s like petals on a flower, or the decoration on the rim of a swirling skirt. Pressure is moving in accord with a natural process, yet to be discovered. There is no suggestion that a natural driver is being distorted by a new, third force that is giving polar pressure a downward tilt. Arctic pressure and the AO index are as high today as they have been at any time in the recent past.
Figure 3 The Monthly Arctic Oscillation Index
In figure 3 we have the AO again but this time, the conventional way up, and presented not as a twelve month moving average of monthly data but as raw monthly data, a thirteen month moving average and a fitted second order polynomial trend line. The interesting patterns in the de-seasonalised data of figure 2 have disappeared. If the polynomial curve is to be trusted as a summary of trend, we might say that the AO index has gradually risen over time but, the rate of increase has become less and less. The index is now just a smidgen below the high point that it reached about 1998. And, just to emphasize the point that things are changing, it has recently fallen into a hole. That hole is the period from December 2009 to May 2010. The data does not reflect a similar plunge in December 2010 because it only goes as far as November.
What is going to happen if we are now entering a period where sea level pressure in the Arctic is no longer subject to the insistent pull of Antarctica.
Figure 4 The monthly AO and the anomaly in sea surface temperature at latitude 30-50 north
From 1948, sea level pressure was not only high in the Arctic but for thirty years it kept rising.
As figure 4 shows the AO just happens to correlate fairly closely with sea surface temperature in the mid latitudes.
Here is a closer look at that period. Sea surface temperature is now on the right axis that is scaled so that the relationship is easier to assess.
Figure 5 AO and anomaly in SST 30-50°N
It looks like sea surface temperature declined at a faster rate than the AO.
In figure 6 dWN refers to the differential pressure driving the Westerly winds in the Northern hemisphere. That is gauged by subtracting the sea level pressure at latitude 50-60° north (the sink) from the sea level pressure at latitude 30-40° north (the source). The letters dEN refer to the differential pressure driving the Polar Easterlies southwards in the Northern hemisphere. That is calculated by subtracting the atmospheric pressure at 50-60° north (the sink) from the atmospheric pressure at 80-90° north (the source).
Figure 6 The differential pressure driving the Polar Easterlies and the South Westerlies in the northern hemisphere
Figure 6 is not easy to read. The polar easterlies (dEN is colored blue for cold) actually fluctuate much more than the south westerlies (dWN is colored orange for warmth). The right hand axis where dEN is plotted has a spread of 30mb and the left only 12 mb.
If the differential is negative that wind can not exist. It doesn’t even get out of bed. Its probably easier to conceptualize reality if you cover up the all that portion below zero on the vertical axis.
The warm south westerlies did not get out of bed for many short intervals up to 1969 but the cold polar easterlies were active and influential. This is what you can expect when the AO is low. After a period of low flux between 1969 and 1980 the relationship flipped. The Westerlies came into their own and the Easterlies experienced repeated intervals of slumber. The Easterlies sunk to abysmal depths of invisibility when Arctic pressure fell into a deep hole in the early 1990’s as Antarctic pressure bottomed. But Antarctic pressure is now increasing and the Easterlies are making their presence felt.
And that has a lot to do with why the ice in the Arctic comes and goes, the polar bears have been swimming further between meals and you are now shoveling snow as the price of orange juice is rising.
If you want to reproduce my sea surface temperature data please use the same database, calculate the average for the entire period and work out the anomaly from that.
Postscript: I have been wondering why Southern Greenland and Hudsons Bay have experienced unseasonable warmth at the same time as entire continents froze in December. Here is my best shot:
There are four factors contributing
1. Warm air descends from the stratosphere. As soon as the AO goes negative we see a geopotential height anomaly appear at latitudes higher than 60°north representing ozone rich air that absorbs long wave radiation from the Earth. There is no UV to be had in the polar night but plenty of outgoing long wave radiation. So the source air in the descending polar circulation is warmer. It represents air slipping down through the ozone profile with more above and less below.
2 The warmed source air is further warmed as it descends via compression. In the polar regions, during the polar night, the air at the surface is warmer that the surface itself.
3 As Tom Rude points out in the last Arctic Ice thread at WUWT: ” at the front of the high pressure anticyclones that brought cold and froze lemon trees in Florida the strong temperature gradient will force the advection of a huge amount of warm and moist air northward, that very same air that dumped snowmaggedon on the NE USA recently.”
4. All forms of precipitation release latent heat.
The Arctic and Antarctic circulations are different beasts to a regular subtropical high pressure cell of descending air. For a start the air is loaded with ozone, it starts off warm and it warming further as it it descends. The warm air wants to climb out of the funnel where the centre is colder and denser because it lacks ozone due to the influence of NOx from the mesosphere. The long and the short of it is that the coolest part of the stratosphere sinks into the troposphere at latitude 50-70. Bear in mind that the troposphere contains 75% of the atmsophere. If the troposphere is warmed kinetic energy forces the molecules apart creating a low pressure zone at that latitude. This low pressure zone markedly accelerates the circulation with an increase in the speed of the polar easterlies at the surface. Unlike the southern hemsiphere that ring of low pressure does no manifest all the way round the globe. The low pressure zones in the northern hemsiphere lie over the North Pacific and the North Atlantic leaving the continents exposed.
As the atmospheric shift occurs, increasing the general level of polar pressure, the circulation intensifies and the whole caboodle descends to the surface (its usual habitat in the Arctic is in the stratosphere) and starts to mimic what happens in the Antarctic all year round. This is really a case of the dynamics that drive the stratospheric circulation becoming apparent at the surface. What happens in the stratosphere is determined in the stratosphere, it is a function of the stratospheric circulation and the night jet that starts in the mesosphere and the dynamics alter with the shift of atmospheric mass to the Arctic. Those who would suggest that it is all driven from the troposphere are deluding themselves. It is their want to consider the atmosphere as a closed system so that their AGW fantasies are cosseted and conserved. I would hope that there is a bit of soul searching going on at the moment amongst those who can glimpse the bigger picture.
The climate of the northern hemisphere depends upon the general level of atmospheric pressure in the Arctic. If pressure is low the hemsiphere warms and if presssure is high the hemsiphere cools. It is a winter time dynamic.
This explanation of climate dynamics focuses on shifts in atmospheric mass over time as a function of external stimuli. I can conceive of no internal process that would shift the atmosphere towards the equator from the poles. In a warming globe those parts that are warming most should see a loss of atmospheric pressure. Since the climate shift of 1978 the entire northern hemisphere has gained atmospheric mass as it has warmed and the surface air temperature has increased at a faster rate than in the southern hemisphere. If the north had cooled relative to the south I could understand the increase in mass. But no, something is slewing the atmsophere northwards.
I am aware that sea surface temperature is modulated by the activity of the adjacent pole as it determines the flux of ozone into the troposphere. That is the ENSO driver. But that is a story for another day.
I can not imagine any internal dynamic that would produce a swing in Antarctic sea level pressure over 100 years. But I am aware of an external influence that does vary on that time scale.
Corroboration that the Stratosphere/Mesophere is the driver of the AO:
Stratospheric Harbingers of Anomalous Weather Regimes Mark P. Baldwin Timothy J. Dunkerton
Observations show that large variations in the strength of the stratospheric circulation, appearing first above ∼50 kilometers, descend to the lowermost stratosphere and are followed by anomalous tropospheric weather regimes. During the 60 days after the onset of these events, average surface pressure maps resemble closely the Arctic Oscillation pattern. These stratospheric events also precede shifts in the probability distributions of extreme values of the Arctic and North Atlantic Oscillations, the location of storm tracks, and the local likelihood of mid-latitude storms. Our observations suggest that these stratospheric harbingers may be used as a predictor of tropospheric weather regimes.
JOURNAL OF GEOPHYSICAL RESEARCH, VOL. 104, NO. D24, PP. 30,937-30,946, 1999
Propagation of the Arctic Oscillation from the stratosphere to the troposphere Mark P. Baldwin Timothy J. Dunkerton
Geopotential anomalies ranging from the Earth’s surface to the middle stratosphere in the northern hemisphere are dominated by a mode of variability known as the Arctic Oscillation (AO). The AO is represented herein by the leading mode (the first empirical orthogonal function) of low-frequency variability of wintertime geopotential between 1000 and 10 hPa. In the middle stratosphere the signature of the AO is a nearly zonally symmetric pattern representing a strong or weak polar vortex. At 1000 hPa the AO is similar to the North Atlantic Oscillation, but with more zonal symmetry, especially at high latitudes. In zonal-mean zonal wind the AO is seen as a north-south dipole centered on 40°–45°N; in zonal-mean temperature it is seen as a deep warm or cold polar anomaly from the upper troposphere to ∼10 hPa. The association of the AO pattern in the troposphere with modulation of the strength of the stratospheric polar vortex provides perhaps the best measure of coupling between the stratosphere and the troposphere. By examining separately time series of AO signatures at tropospheric and stratospheric levels, it is shown that AO anomalies typically appear first in the stratosphere and propagate downward. The midwinter correlation between the 90-day low-pass-filtered 10-hPa anomaly and the 1000-hPa anomaly exceeds 0.65 when the surface anomaly time series is lagged by about three weeks. The tropospheric signature of the AO anomaly is characterized by substantial changes to the storm tracks and strength of the midtropospheric flow, especially over the North Atlantic and Europe. The implications of large stratospheric anomalies as precursors to changes in tropospheric weather patterns are discussed.