Guest post by Erl Happ
This post is best read after viewing parts one, two and three that set the scene for what is described here.
I noted in parts 1 and 2 that variability in global temperature is greatest between November and March when the globe is coolest. This is related to high variability in southern summer when global cloud cover peaks. I suggested that this variability is likely related to variation in cloud cover. In part 3 I outlined a mechanism related to the coupled circulation of the stratosphere and the troposphere in the Arctic and the Antarctic that induces a variation in cloud cover and described the spatial expression of that variation.
The manner in which the planet warms is surprising. If we look hard enough, it tells us how and why it warms. The value of a good theory is that it makes explicable what we see. It is much safer therefore to look at the manner in which the planet warms, as I have done in parts 1 and 2, before theorizing.
What follows is big picture analysis jumping from highlight to highlight. For data I rely on Kalnay, E. and Coauthors, 1996: The NCEP/NCAR Reanalysis 40-year Project. Bull. Amer. Meteor. Soc., 77, 437-471. accessible here: http://www.esrl.noaa.gov/psd/cgi-bin/data/timeseries/timeseries1.pl.
In this article we see that:
- Equatorial sea surface temperature varies with equatorial sea level pressure.
- Equatorial sea surface pressure varies with the solar wind.
- In respect of anomalous behavior that is superimposed on the seasonal cycle, the hemispheres heat alternately.
- The northern hemisphere suffers the widest swings in temperature but largely in winter.
- The evolution of temperature depends in large part upon what is happening in Antarctica.
- The planet tends to heat or cool most dramatically between November and March when cloud cover is most extensive.
- The mechanism responsible for climate variation that is described here can account for the diversity in our experience of climate change over the last sixty years and the cooling to come. It is a mechanism that allows one hemisphere to warm while the other cools.
The Sun and atmospheric pressure
The Southern Oscillation Index (SOI) tracks the ENSO phenomenon in the Pacific Ocean. It is based on the difference in sea level pressure between Darwin Australia (12°south 131°east) and Tahiti (18°south 150° west in French Polynesia).
Daily sea level pressure for Darwin and Tahiti is available at: http://www.longpaddock.qld.gov.au/seasonalclimateoutlook/southernoscillationindex/soidatafiles/DailySOI1933-1992Base.txt
Because the record is short, the average is lumpy. It is assumed that, were the record long enough, it would be smooth. To obtain a smooth line the data has been adjusted manually. That smooth line is shown against the actual 30 day moving average of Darwin sea level pressure for the period January 1999-July 2011 in figure 1.
Figure 1 Seasonal evolution of Darwin daily sea level pressure in mb.
An anomaly in sea level pressure is a departure from the average daily value for a selected period. In this case the period is January 1999 to July 2011.
Figure 2 charts the relationship between the daily anomaly in sea level pressure at Darwin and the Dst index which is an index of geomagnetic activity that relates to the strength of the ring current in the ionosphere.
Figure 2 Dst Index and SLP anomaly Darwin
Left axis: Daily Dst index in nanoteslas. Source: http://wdc.kugi.kyoto-u.ac.jp/dst_realtime/201108/index.html Note that a fall in the Dst index represents increased geomagnetic activity.
Right axis: Anomaly in daily sea level pressure at Darwin, Australia in millibars.
Note that the right axis is inverted
It is plain that Darwin sea level pressure is influenced by geomagnetic activity.
Similarly, sea level pressure in Antarctica is influenced by geomagnetic activity as we see in figure 3. There is no readily available index of daily sea level pressure in Antarctica but the Antarctic Oscillation Index (AAO) is a good substitute. It varies inversely with sea level pressure at the pole.
Figure 3 DST index and anomaly and the AAO index
Left axis: Daily Dst index in nanoteslas.
Right axis: Daily AAO index. This axis is inverted.
In figure 3 we see that as the Dst index plunges into the negative the AAO index increases in value indicating falling pressure over Antarctica.
It is apparent that under the influence of geomagnetic activity the atmosphere moves away from the Antarctica towards the equator.
The same phenomenon is demonstrated using the ap index and the AAO in figure 4
Figure 4 AP index of geomagnetic activity and the AAO
Left axis: Daily Ap index in nanoteslas
Right axis: Daily AAO index
The ap index and the AAO index increase together. An increase in the AAO index indicates falling atmospheric pressure at the pole. At times the relationship seems to be better than at other times. Other variables that will be described below influence the atmospheric response to the solar wind. In particular the level of solar irradiance is important in that it governs the plasma density within the neutral atmosphere. Plasma density determines the effect on neutrals (electrically balanced particles) as plasma responds to a change in the electromagnetic field by accelerating away and bumping the neutrals along as it goes.
The data for the year 2008 shows the relationship during a protracted solar minimum when the atmosphere is least inflated because solar irradiance is weak. It can be observed that the relationship between these variables (although still imperfect) is better at solar minimum. At solar minimum the response of the atmosphere to the solar wind is amplified. Solar irradiance and geomagnetic activity do not vary together. Under high levels of irradiance the response of the atmosphere to geomagnetic activity is much reduced and harder to discern. At solar maximum the atmosphere can return to the pole regardless of the level of geomagnetic activity. High atmospheric pressure at the southern pole is associated with a cooling planet because night jet activity varies directly with pressure at the pole. The night jet brings nitrogen oxides into the stratosphere reducing ozone formation. This weakens the coupled circulation of the stratosphere and the troposphere resulting in rising surface pressure at 60-70° south (and over Antarctica generally), increased cloud and weakened westerly winds. This is a self reinforcing process.
Aspects of ENSO
The anomaly in daily sea level pressure at Tahiti has been calculated in the same way as for Darwin. Figure 5 looks at the relationship between the margin between these two pressure anomalies on the one hand and the Southern Oscillation Index on the other.
Figure 5 Tahiti less Darwin SLP compared to the SOI
Left axis Southern Oscillation Index
Right axis: Pressure anomaly in Tahiti less the anomaly in Darwin.
The difference in the sea level pressure anomaly between the Tahiti and Darwin tracks the Southern Oscillation Index. A fall in the index relates to El Nino warming. This is associated with a slackening of the trade winds due to a loss of the pressure differential between Tahiti and Darwin.
A slackening of the trade winds is associated with an even greater slackening of the north westerlies in the southern hemisphere.
The slackening of the north westerlies in the southern hemisphere is due to a rise in surface pressure at 60-70° south (and over Antarctica generally). This is associated with a fall in surface pressure in the Arctic ( a simple exchange of atmosphere between the hemispheres driven by change in the coupled circulation over Antarctica).
The fall in surface pressure in the Arctic is associated with an increase in the temperature of the polar stratosphere as night jet activity falls away enabling an increase in the ozone concentration of the stratosphere. Under the influence of the coupled circulation in the Arctic this affects sea surface temperature throughout the northern hemisphere but most energetically at 50-60° north. This is a winter phenomenon.
Low surface pressure in the Arctic is the expression of the warm phase of the Northern Annular Mode (also called the Arctic Oscillation) wherein the domain of the warm humid south westerlies extends to the North Pole to the exclusion of the frigid polar easterlies. Accordingly Arctic air temperature increases and the area occupied by sea ice falls away. The dominance of warm over cool episodes marked the period 1978 through to 2007. A cool mode commenced in 2007 and the northern hemisphere is currently experiencing winter temperatures not seen since the cool mode of the 1960’s and 1970’s.
The warm mode is marked by El Nino dominance in the Pacific whereas the cool mode relates to La Nina dominance. Dominance can be assessed in terms of the length of time that the index has one sign or the other or by simply accumulating index values over time. Neither the Arctic Oscillation or ENSO is ever climate neutral.
The initiating influence in this activity is the solar wind, but the effect of the solar wind is amplified by the activity of a strengthened coupling of the stratosphere and the troposphere over Antarctica.
As will be demonstrated below, the pattern of inverse pressure relations between the hemispheres dictates how the planet warms. But first, lets look at the relationship between sea level pressure and temperature at the equator.
Figure 6 Monthly anomalies in sea level pressure at Darwin and Tahiti
Figure 6 shows that although there are times when sea level pressure anomalies in Darwin and Tahiti move in the same direction at the same time, a period of intense warming like that which occurred in early 2010 is associated with positive anomalies in sea surface pressure for Darwin and negative anomalies for Tahiti (weak trades). Conversely, the period of strong cooling that commenced in mid 2010 is associated with negative pressure anomalies in Darwin and positive anomalies in Tahiti (strong trades).
The upshot is that sea surface temperature at the equator moves directly with sea level pressure in Darwin. Since the sea surface temperature response is associated with geomagnetic activity and is a global phenomenon one would expect that Darwin pressure would move in concert with equatorial sea surface pressure around the entire globe and this is indeed the case as we see in figure 7. The range of variation in Darwin is about twice the variation in near equatorial latitudes. The Pacific is a theater of extremes. Darwin sea level pressure increases when the zone of convection moves from Indonesia to the mid Pacific during warming events.
Figure 7 Sea level pressure in Darwin compared to that at 15°north to 15° south latitude.
Left axis: Monthly anomalies in sea level pressure 15°North to 15°south latitude, mb.
Right axis: Monthly anomalies in sea level pressure at Darwin, mb
How much of the change in sea surface temperature at the equator is associated with the variation in pressure in near equatorial latitudes?
Figure 8 Anomalies in sea surface temperature (10°N-10°S) and sea level pressure (15°N-15°S) with respect to the average for the period 1948 to July 2011
Left axis: Sea surface pressure in mb. Twelve month moving average of raw data centered on seventh month.
Right axis: Sea surface temperature in °C. Twelve month moving average of raw data centered on seventh month.
The closeness of the relationship that is seen in figure 8, and the fact that the curves start and finish together suggest that phenomena responsible for warming, that is allied with the rise and fall in sea level pressure at the equator is consistent with the change in sea surface temperature between 1948 and the present time. This is not the whole story however. In the short-term volcanic influences can influential. Notice the depression of temperature following the eruption of Pinatubo in 1991.
The relationship between surface pressure and geomagnetic activity
The relationship between the Dst index (or the ap index) of geomagnetic activity and sea level atmospheric pressure is non linear. From episode to episode other influences condition the surface pressure response. These influences could include:
Two factors modify the sea level pressure from day-to-day, month to month and year to year and these work in a bottom up fashion:
- Pressure changes on a daily basis with the passage of high and low pressure cells around the globe and the wetting and drying of the air.
- In near equatorial latitudes in the Pacific sphere sea level pressure is affected by the migration of the zone of convection between Indonesia and the central Pacific.
Conditions in the stratosphere and mesosphere are the strongest influence on the evolution of surface pressure. The shift of the atmosphere from high to mid and low latitudes that is monitored as the Arctic Oscillation and the Antarctic Oscillation index depends upon:
- The plasma density where plasma interacts with neutral atmospheric molecules under the influence of the changing electromagnetic field.
- The state of ionization of the atmosphere as it depends upon the changing incidence of very short wave radiation from the sun.
- The changing electromagnetic field within the solid Earth.
- The changing spatio-temporal expression of the Northern Annular Mode and the Southern Annular Mode. The mode results from the coupling of the stratosphere and the troposphere that introduces ozone from the stratosphere into the troposphere causing the troposphere to warm, lowering surface atmospheric pressure in a ring like pattern at 60-70°south latitude and also 50-60° north latitude. But the expression of the mode changes over time, for instance, a migration of zones of ozone descent affects the relativity of sea level pressure between New Zealand and the Pacific Ocean west of Chile. This is possibly involved in the El Nino ‘Modoki’ phenomenon.
- The rate of introduction of nitrogen oxides from the mesosphere into the stratosphere over the poles affects the population of free oxygen atoms capable of forming ozone, and therefore the ozone content of the polar stratosphere. This in turn bears upon the concentration of ozone in the air that descends within the coupled circulation and the strength of the surface pressure and temperature response.
The relationship between the NAM and the SAM and sea surface temperature
The northern and southern annular modes of inter-annual climate variability influence sea surface temperature. The flow of ozone towards the equator via the high altitude counter westerlies (see part 3) warms and dries the air reducing cloud cover. Accordingly a pattern of positive sea surface temperature anomalies is generated that stretches from high southern hemisphere latitudes towards the equator in a north-westerly direction and from high northern latitudes towards the equator in a south-westerly direction. This pattern of sea surface temperature anomalies can be seen to originate from zones of increased geopotential heights at 200hPa that identify the locations of ozone descent in the coupled circulation of the stratosphere and the troposphere. This is the fingerprint of climate change as it is written in sea surface temperature.
The seasonal evolution of ENSO
Figure 9 shows the evolution of sea level pressure in Darwin and Tahiti over a year.
Figure 9 The seasonal evolution of the pressure relativity between Tahiti and Darwin
Left axis: Sea level pressure, mb. Right axis: Difference between blue and red curves, mb
The green curve represents the difference between the red and the blue curves. It shows the pressure differential driving the trade winds between Tahiti and Darwin as it evolves in an ‘average year’. It is positive in all months, builds strongly from July onwards and peaks just after the turn of the year. The Trades are weakest in mid year.
Figure 10 Variability in the raw data pressure differential between Tahiti and Darwin since 1999, mb
Figure 10 shows that in the last decade, variability in ENSO is least in mid year and greatest at the end of the year.
So the variation in cloud cover is greatest in the midst of southern summer when the globe is coolest. It is at this time that global cloud cover peaks with three percent more cloud than in July-August. In mid year cloud cover is reduced due to the direct heating of the atmosphere by the land masses of the northern hemisphere. But at the turn of the year the northern continents are least illuminated and this cloud degrading influence, a product of the distribution of land and sea, is minimal.
The influence of the coupled circulation of the stratosphere and the troposphere in the Arctic between November and March explains the strong variation in cloud cover and sea surface temperature between November and March. It is at this time that the Earth is closest to the sun, irradiance is most intense, global cloud cover is greatest and most susceptible to alteration.
Surface temperature is determined not by variations in solar irradiance (very small) but by variation in cloud cover (very large). Cloud cover relates directly to the influence of the coupled circulation between the stratosphere and the troposphere over the poles. The main driver of long term change is the coupled circulation over Antarctica but in terms of the short term jerks the Arctic circulation is important and by and large it is a mirror image of that in the south. It is the rise and fall in pressure in Antarctica that determines surface pressure in the Arctic. The Arctic is more influential in determining the evolution of cloud cover in part because cloud cover is maximal at the time that the coupled circulation in the Arctic is most active.
But, the influence of the Arctic is also supercharged due to the relatively high concentration of ozone in the northern stratosphere. Ozone levels are high precisely because the coupled circulation is intermittent and the night jet less active than it is over Antarctica. In fact when Arctic pressure is weak, a situation that has persisted for thirty year intervals (e.g. 1978-1997) , ozone depletion via night jet activity is rarely seen. The temperature of the northern stratosphere is then anomalously high.
When cloud cover is curtailed the surface begins to warm. Then the land masses of both hemispheres provide a feedback by swiftly warming the atmosphere enhancing the loss of cloud cover. Add to this the fact that wind speed is generally much lower in the northern hemisphere and we can see why gyrations in sea surface temperature that are experienced in the north Pacific and north Atlantic are about twice the amplitude of those in the southern hemisphere. Increased evaporation due to high wind speed mutes the response of surface temperature in the southern hemisphere.
Southern waters do warm as ozone is introduced to the troposphere lowering surface pressure and speeding the flow of the westerlies. But the coupled circulation is perennial in the south and stratospheric ozone levels are consequently much less than in the northern hemisphere.
When sea surface pressure is depressed in the southern hemisphere high pressure in the Arctic enhances the flow of the polar easterlies that sweep across the northern continents towards tropical latitudes. But this is largely a winter phenomenon. It is high variability in winter that marks climate in the northern hemisphere. This is most evident in the Arctic as seen here: http://ocean.dmi.dk/arctic/meant80n.uk.php
The evolution of sea surface temperature by latitude
Figure 11 The evolution of sea surface temperature at 40-55°north and 40-55° south. Anomalies with respect to the 1948-2011 average, °C.
So far as the mid latitudes are concerned, we see the sea cooling in the southern hemisphere as it warms in the northern hemisphere. Don’t be confused by the apparently consistent pattern of warming in the southern hemisphere in summer. It’s not consistent at all. Look at 2001. Similarly one notes marked warming of northern seas in winter in 2002 and 2003. The hemispheres warm and cool alternately, a pattern that is inconsistent with the notion that a greenhouse effect is responsible for temperature change. This pattern of anomalies is an expression of atmospheric circumstances post the climate shift of 1976-8. It represents the current expression of atmospheric balances that are always changing. There is not one climate system but many. If you don’t appreciate the change in its parameters, you can’t model the climate system. It’s the assumptions behind the models that give them away.
It’s a system that is open to external influences.
Figure 12 The evolution of sea surface temperature at 40-55°north and 40-55° south. Anomalies with respect to the 1948-2011 average, °C.
Left axis: Northern hemisphere
Right axis: Southern hemisphere. The right axis inverted.
In figure 12 (a restatement of the data in figure 11) we see that the cooling of the southern mid latitudes, (inverted and re-scaled) has a lot of symmetry with the warming of the northern mid latitudes. Make no mistake, sea surface temperature responds to a global stimulus with mirror image effects between the northern and southern hemisphere. This must be so, because the pattern of pressure variation at all latitudes is dictated by the evolution of surface pressure over Antarctica. If pressure is falling in Antarctica it will be rising in the Arctic and vice-versa. The variation in surface pressure is directly related to the influx of ozone into the troposphere on the margins of the Arctic and the Antarctic via the coupling of the circulation of the stratosphere and troposphere that occurs at high latitudes. The strength of the coupling varies through the year. However if one takes notice of geopotential heights at 200hPa the circulation is active to some extent in influencing surface pressure and cloud cover in both hemispheres all year round.
Figure 13 Evolution of sea surface temperature between 25 and 40° of latitude, °C.
Between 25° and 40° of latitude we see the same mirror image effect of alternate advance in sea surface temperature anomalies.
Figure 14 Evolution of sea surface temperature in near equatorial waters at 10-25° latitude. °C
In subtropical latitudes the tendency for the hemispheres to warm alternately is still apparent even though these latitudes are blessed with less cloud than higher latitudes. These latitudes are a long way away from the latitudes where the coupled circulation brings ozone into the troposphere.
Figure 15 Influence of high northern latitudes on the evolution of sea surface temperature. °C
In figure 15 we see the influence of the mid latitudes of the northern hemisphere in providing the spikes in sea surface temperature that can be seen in the evolution of sea surface temperature between 50°north and 50°south latitudes. It is not just the tropics or indeed the Pacific Ocean that is responsible for the evolution of temperature where the sun shines brightest.
Summary for policy makers
The Earth system, under the influence of solar emanations, modulates the reception of solar radiation at the surface by varying the extent of reflective cloud. The solar wind initiates this process via its influence on the distribution of the atmosphere between high and low latitudes. The effect of the coupled circulation of stratosphere and troposphere over Antarctica is to amplify these variations.
The day-to-day and year to year gyrations in cloud cover are associated with what we observe as ENSO. ENSO is a complex phenomenon that arises in part from dynamics in the Pacific including a shift in the main zone of convective activity. But the evolution of ENSO is also driven by change in surface pressure that affects deep ocean upwelling. It depends upon change in pressure at high latitudes where the stratosphere can behave like an extension of the troposphere. It does so because in winter, temperature falls away with altitude in the polar atmosphere from the surface all the way to 5hPa, encompassing both the troposphere and the stratosphere. When a convectional circulation is established the coolest parts of the stratosphere descend to elevations that we think of as the domain of the troposphere. This results in what has come to be known as the Annular Modes of inter-annual climate variation, zones of lower pressure that, as they establish reinforce the coupled circulation. These ‘annular modes’ are also involved in the evolution of climate on decadal and centennial time scales via their association with change in cloud cover. It can be shown that change in sea surface temperature and sea surface pressure in higher latitudes heralds change in the tropics.
If we were more observant we would note that gyrations in the climate are closely associated with a strong variation in the temperature in winter in the northern hemisphere. These variations are monitored as the Arctic Oscillation. This phenomenon is part of the rich texture of climate change of equal importance to ENSO. Both are dependent on Antarctic processes.
The role of trace amounts of ozone in the troposphere is critical to an understanding of cloud dynamics. It is the change in cloud cover that results in changing surface temperature.
Current understanding of what determines the ozone content and the temperature of the stratosphere is deficient. We need to understand the role of the night jet and the coupled circulations in modulating ozone concentration and therefore stratospheric temperature.
Geomagnetic activity and surface pressure variations evolve over long periods of time according to plasma dynamics that is seldom observed and little appreciated.
The dynamic described here provides a plausible explanation for the change in surface temperature that is observed. The pattern of temperature change is complex, varying by latitude and hemisphere. The fingerprint of change is inconsistent with the notion that the increase in so-called greenhouse gases in the troposphere is responsible for change.
The important thing to note is that the change is reversible and there is nothing that man can do but adapt. The temperature of the southern stratosphere has been gradually declining since 1978. A less active sun will see further falls in the temperature of the Antarctic stratosphere. This will gradually reverse the erosion of atmospheric pressure in high southern latitudes that has been influential in the warming process.
When we are dealing with complex systems like climate the idea that we can project an outcome and then qualify that projection with a statement about our degree of certainty in relation to the likelihood that we are correct, is inappropriate. Time and again we discover that our assumptions do not reflect the real world.
Those who refuse to acknowledge that their projections are inaccurate, and in any case variable from one soothsayer to the next, are not practicing science at all. They should be able to explain the variations that we see from day to day and year to year, and that includes ENSO and the Arctic Oscillation. They are in fact doing something other than ‘science’. On no account should they be suggesting that they understand the system or that their models are a source of truth.
We can not pretend that we understand the climate system unless we can explain ENSO, the Arctic Oscillation, put the Antarctic Oscillation in its context of evolving pressure relations as the Southern Annular Mode and explain the PDO and the NAO. When that is accomplished we might ask around as to whether people think the science is settled.
When we understand what determines the emanations from the sun we might hazard a forecast as to the weather to be expected in six months time.
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
Tomas speaks lucidly of temperospatial chaos, and Robert Ellison, AKA the Chief Hydrologist @ur momisugly Judy’s speaks of Dragon Kings and other associated horrors, but also understands water circulation, wherever it goes, more explainably than anyone else I’ve read.
============================
Erl Happ:
At August 23, 2011 at 5:50 pm you write:
“Perhaps those who want robust correlations could take note. Movement within the Earths crust is inferred from GA at the surface and is well correlated with the long term temperature record at the surface of the Earth once it is de-trended. LOD is a result in all this, not a cause.
Better have a close look Richard and Leif.”
Say what!?
From the very first post in this thread I have been asking you to justify your assertion of correlation where I see no indication of any kind that such correlation exists. I have repeatedly made this request because your entire argument is constructed on an assumption of the correlation although no evidence for the correlation has been presented. And your responses to my requests for evidence have failed to provide any evidence of any kind that the correlation exists.
Now you comment on an analysis by Dickey, et al which relies on correlations between other parameters than those you claim are correlated. Dickey, et al do provide evidence for the existence of the correlations they consider. And you suggest that I “Better have a close look”!
Your suggestion is disingenuous in the extreme.
Richard
Richard,
I am not being insincere. The paper that Leif put forward found a correlation between GA and global temperature. That was the what you were after because temperature and pressure are very closely related. Did you notice that? Figure 8.
Here I point to the pressure dynamic and its relation to GA and I put forward a plausible mechanism to account for the phenomenon and describe a pattern of variation in temperature that is consistent with that dependency. So many papers do not even provide a plausible mechanism.You don’t seem to be at all interested in the mechanism, or the distinctive pattern of temperature variability that arises from this mechanism that in itself confirms that the mechanism is operating, just the correlation. Naturally, I am unimpressed. Who is being disingenuous?
If you value the correlation above all else you will be interested in these papers that establish that there is a correlation between GA and pressure or temperature or geopotential heights, NAM, SAM the AO or some aspect of climate that varies with pressure or temperature. Very few offer a plausible mechanism.None describe the pattern of temperature variability that is the fingerprint of the mechanism as I describe it.
So, I see myself as being short changed.
Like this:
Geomagnetic activity forcing of the Northern Annular Mode via the stratosphere
D. R. Palamara 1, E. A. Bryant 1
(19/03/2004)
We consider various aspects of the link between solar-modulated geomagnetic activity and the Northern Annular Mode (NAM). Our results indicate that the geomagnetic forcing of atmospheric circulation in the Northern Hemisphere is temporally and seasonally restricted, modulated by the Quasi-Biennial Oscillation (QBO), and reliant on stratosphere-troposphere coupling. When the data are restricted to January values after 1965, for years in which the January QBO is eastwards, the correlation coefficient between the geomagnetic AA index and the NAM is 0.85. These results can account for many of the enigmatic features of Northern Hemisphere circulation.
OR
http://www.physics.otago.ac.nz/space/2008JA014029-pip.pdf
http://web.dmi.dk/fsweb/solar-terrestrial/staff/hgl/Pubs/TCG_GRL03.pdf
http://www.lund.irf.se/HeliosHome/NAO_article.pdf
http://www.sciencedirect.com/science/article/pii/S1364682604001750
http://www.sciencedirect.com/science/article/pii/S1364682607001368
http://www.springerlink.com/content/k5559l7046405324/
but you can use Google as well as I can.
Looking at it in the round I would say that the insistence on a correlation coefficient was a tactic to stall the discussion and cast doubt into the mind of all participants. I went to great lengths to describe the number of variables involved and the reason why a strong correlation could not be expected.
So, I am unimpressed by your demand for a correlation co-efficient. In my view, in this circumstance, its inappropriate. It reflects a simplistic, narrow minded, aggressive and unreasonable attitude.
Erl Happ:
I take severe umbrage at your message to me that says;
“Looking at it in the round I would say that the insistence on a correlation coefficient was a tactic to stall the discussion and cast doubt into the mind of all participants. I went to great lengths to describe the number of variables involved and the reason why a strong correlation could not be expected.
So, I am unimpressed by your demand for a correlation co-efficient. In my view, in this circumstance, its inappropriate. It reflects a simplistic, narrow minded, aggressive and unreasonable attitude.”
My “tactic” was to ask you to justify an assertion on which your entire argument relied.
You have failed to doi that.
My “attitude” was polite and helpful: it was not “simplistic, narrow minded, aggressive and unreasonable” as you wrongly assert.
Apologise and provide the missing data or withdraw your article.
Richard
“
Erl Happ (August 24, 2011 at 6:06 am)
“Looking at it in the round I would say that the insistence on a correlation coefficient was a tactic to stall the discussion and cast doubt into the mind of all participants. I went to great lengths to describe the number of variables involved and the reason why a strong correlation could not be expected.”
The sense in which it’s multivariate:
Changes in the state of water reverse relationships.
More importantly (the part Erl appears to overlook):
It’s spatiotemporal. Not “spatial”, nor “temporal”, nor “spatial & temporal”, but rather “spatiotemporal”, which differs fundamentally from all preceding in this list – analogous to the fundamental difference between a collection of marginal distributions and a joint distribution. This is absolute logic. Zero controversy. There’s no basis for discussion without agreement on fundamentals like black vs. white & 1+1=2.
It’s not a mere matter of clouds, but of circulation – and readers like Vukcevic could lend a courteous ear to oceanographer/commenter “sky” who keeps trying to make readers realize that winds, driven by pressure differentials, drive ocean surface currents that are orders of magnitude above effects from other types of ocean currents (such as THC).
It’s not constant laminar flow in a tube (which would yield simple linear correlation). The pump has variable speed and mass-misinterpretation of stats stems from assumptions of stationarity and misconception of relationship-reversing effects of changing size, position, & orientation of flow features.
The correlations are complex (as in complex numbers, not as in complicated). The functional numeracy of the audience here is insufficient for sensible discussion of the preceding at present, but this can change …and for the discussion to advance, this HAS to change …and in the meantime those with an interest in keeping wool over innocent eyes will continue having a comfortable, easy train ride.
Regards.
Richard S Courtney says:
August 24, 2011 at 6:32 am
My “tactic” was to ask you to justify an assertion on which your entire argument relied.
You have failed to do that.
Indeed, Erl has failed to show that there is a correlation, but as he admits, he wouldn’t expect any, so no wonder that there is none.
Paul Vaughan says: August 24, 2011 at 7:20 am
Eloquently put.
I wonder if we are to get a response from Richard and Leif on the matters that you raise. While they are at it perhaps they might address the question as to whether there are other ways to assess a relationship than via a correlation co-efficient.
Here is one: “Ye shall know them by their fruits.”
And the fruits being:
The hemsipheres warm alternately on monthly, annual and longer time scales.
The SST variation increases with latitude in the northern hemisphere and falls away with latitude in the southern hemisphere.
SST varies with surface pressure in high latitudes.
Variation in the strength of the winds occurs with the change in high latitude surface pressure.
In the southern hemisphere the sea warms as the wind blows harder.
The north westerlies in the southern hemisphere fell away for sixty years prior to 1920 and have strengthened since.
You will also be informed via the use of logic.
Is temperature in a single hemisphere reacting to a unitary forcing or twin forcings?
What are the candidates for twin forcings?
Is sea surface temperature at high latitudes responding to the coupled circulation of the stratosphere and the troposphere at high latitudes?
Is the activity of the coupled circulation tied to time of year?
Why does the coupled circulation vary as it does? Sometimes both poles together and at other times alternately?
What is the role of the night jet? How does in vary over time and why? Is this and the coupled circulation the primary control of ozone levels in the polar stratosphere?
Is it the variation in ozone that drives the coupled circulation?
Is this system externally forced? By what mechanism might it be forced?
What could explain the fall in surface pressure at 60-70° south over the last sixty years.
What is required to reverse the fall in pressure at 60-70° south.
Is it conceivable that all this is externally driven?
Does the atmosphere move when the electromagnetic field changes? What direction does it move in. Where is the electromagnetic force greatest?
Is there a mechanism within the Earth system that could amplify surface pressure variations arising from electromagnetic effects?
What happens to the temperature of the stratosphere over the pole when surface pressure falls? Why does it vary on daily, weekly, monthly, annual, inter-annual, decadal and longer time scales? Might this variation be tied to variations in surface temperature?
erl happ says:
August 24, 2011 at 4:36 pm
I wonder if we are to get a response from Richard and Leif on the matters that you raise. While they are at it perhaps they might address the question as to whether there are other ways to assess a relationship than via a correlation co-efficient.
If there is a good relationship the extra machinery Paul likes is not needed at all. If the relationship is spurious, Paul’s methods will find all kinds of cycles, equally spurious.
The hemsipheres warm alternately on monthly, annual and longer time scales.
The SST variation increases with latitude in the northern hemisphere and falls away with latitude in the southern hemisphere. […] etc
All of these things are incorporated in the current atmospheric models. And are not so mysterious as you want to believe.
Leif Svalgaard says: August 24, 2011 at 4:45 pm
All of these things are incorporated in the current atmospheric models. And are not so mysterious as you want to believe.
OK I am mightily reassured. But hang on, which ones of these are the current atmospheric models that I can hold in high esteem?
http://iri.columbia.edu/climate/ENSO/currentinfo/SST_table.html
Pull the other leg.OR Address the question with some attempt at integrity.
erl happ says:
August 24, 2011 at 7:15 pm
OK I am mightily reassured.
I don’t think you mean this.
But hang on, which ones of these are the current atmospheric models that I can hold in high esteem?
You can hold them all in high esteem for the effort and science going into them. That they don’t have all the answers is just the nature of good science, rather than the pseudo-science practiced by self-assured amateurs.
with some attempt at integrity.
Still in the insulting business I see.
As an example of what I’m talking about, refer to this paper:
“Growth increment series in Eocene fossils from the Antarctic Peninsula exhibit spectral peaks consistent with modern ENSO, the region is strongly teleconnected to the equatorial Pacific today, and ENSO variation and similar teleconnection during the Eocene are predicted by a coupled
climate model.” http://www.leif.org/EOS/2011GL048635.pdf
Leif Svalgaard wrote (August 24, 2011 at 4:45 pm)
“If there is a good relationship the extra machinery Paul likes is not needed at all. If the relationship is spurious, Paul’s methods will find all kinds of cycles, equally spurious.”
Correct on the 1st point. Absolutely NOT on the 2nd.
On the 1st point: What’s up for discussion is interpretation.
For example, AGF linked to this [ http://www.nasa.gov/topics/earth/features/earth20110309.html ] which includes (for a taste) this:
“Other possibilities are that […] an external (e.g. solar) process affects the core and climate simultaneously. “
I’m not convinced of arguments about the core (which are based on model assumptions). I’m also not convinced that the researchers understand LeMouel, Blanter, Shnirman, & Courtillot (2010) and insights from optimized extension of comparable methods to other indices (e.g. aa index).
I disagree with most of the wild speculation of some WUWT commenters.
The discussion of interpretation will have to wait for another week or month, as I’m working 7 days per week, but if I can find time tomorrow I’ll dig out links to Milanovic comments for Erl.
Regards.
Leif Svalgaard says: August 24, 2011 at 7:58 pm
The presence of ENSO variation during this markedly warmer interval argues for the persistence of robust interannual variability in our future greenhouse world.
It’s been with us through warmer times and cooler times. Nobody yet knows what causes it and I know of a whole group of people who don’t want to know including the authors of this paper and Leif Svalgaard.
What is a ‘fully coupled Eocene climate model’? Will it produce an ENSO variation? Do any of the models that I point to above manage to predict it? If they do, tell me which one so that I can check which way it is pointing? Then we can check if it’s any good or not.
There is some science that you think is convincing and other science that you seem to think is deficient but when you cite papers like this its hard to tell the difference.
Still in the insulting business I see.
Which are you, the pot or the kettle?
I don’t like your methods.Too much reliance on authority, no attempt to address the question or deal with the gist of the argument. In sport they call it sledging. Ultimately it erodes authority rather than enhances it.
That they don’t have all the answers is just the nature of good science
For myself I like to see problems solved and understanding enhanced.
erl happ says:
August 24, 2011 at 9:03 pm
What is a ‘fully coupled Eocene climate model’?
How about you actually reading the paper, study the models, etc
There is some science that you think is convincing and other science that you seem to think is deficient but when you cite papers like this its hard to tell the difference.
I look like you didn’t even read the paper
Which are you, the pot or the kettle?
So you admit it.
I don’t like your methods.Too much reliance on authority,
The main authority I rely is myself. But there are times when it pays to bow to expert knowledge. Ever had on operation? Wasn’t the doctor an authority you relied on?
no attempt to address the question or deal with the gist of the argument.
I have gone to the core of your argument, all the was back ozone being soluble in water and folloed each step, and, frankly, they don’t hang together.
For myself I like to see problems solved and understanding enhanced.
Science is a collaborative effort and your one-man show does not cut it.
Leif Svalgaard says: August 23, 2011 at 6:45 pm
Leif, I have read enough of the paper at http://iri.columbia.edu/climate/ENSO/currentinfo/SST_table.html to have formed an opinion.
I suggest you read the Character of climate change parts one and two as posted here…………………and we can then discuss the paper, and the issues that I raise above at August 23, 2011 at 6:45 pm on the basis of a shared understanding that the Greenhouse World that is mooted for the Eocene is not something that we are currently experiencing.
Erl,
By referencing Tomas Milanovic notes from Dr. Judith Curry’s blog Climate Etc. I’m trying to help readers see [ http://tallbloke.wordpress.com/2011/08/21/delve-into-halcrut-at-the-poles/#comment-8355 ] that Tim Channon’s summary [ http://tallbloke.files.wordpress.com/2011/08/had-uahlat-corr-1.png ] is a mere (single) cell of an infinite multiscale array.
At best, narrowly-focused autocorrelation & cross-correlation summaries PATENTLY canNOT handle the job; at worst, they’ll hopelessly mislead [spatiotemporal version of Simpson’s Paradox] the unwary.
The Milanovic quotes may serve as energizing food for thought in your efforts to interpret complex correlations, which are sometimes driven turbulently in parallel by different processes. (Not saying that’s the case, but a data explorer’s wise to always remain vigilantly alert to the possibility of confounding.)
Important:
It’s not just turbulence that flips relationship signs, it’s also spatiotemporal aliasing. (Many in these discussions forget that the “data” aren’t actually data; they’re stats. [Worse: Probably most here don’t even know the difference.])
You may be pleased with comments slipped into a WUWT thread late (in early July after everyone left the June discussion) by Marcia Wyatt:
“[…] recent research suggests AMO is connected to multidecadal variability in frequency of sudden-stratospheric-warmings, which are related to both tropical convective processes and to the integrity of the polar vortex – both features wielding hemispheric influence on the climate. In addition, longitudinal and latitudinal placements of the atmospheric centers-of-action shift with multidecadal variations in AMO, as does the meridional placement of the mean intertropical convergence zone (ITCZ), along with associated changes in Atlantic hurricane activity and in frequency of occurrence of Atlantic-NINOs. Likewise, with multidecadal variability in PDO come changes in placement and strength of atmospheric centers-of-action and in placement and strength of associated oceanic gyres and in meridional mean location of the Pacific ITCZ. […]”
http://wattsupwiththat.com/2011/06/08/on-the-amopdo-dataset/#comment-695279
I disagree with Wyatt, Milanovic, Tsonis, Swanson, & Kravtsov fundamentally on many points, but these folks bring something stimulating to the discussion.
I’ll have a few more Milanovic notes to add in the Channon thread over the next few days, including a quote where Milanovic reveals his primary reason for entering the climate discussion.
Regards.
Hi Paul,
Thanks for your helpful observations, some of which I understand and the rest I appreciate are probably indecipherable due to my lack of background in using the tools that you have at your command.
Atmospheric phenomena respond to multiple influences. A good correlation for a period of a month or two on two variables that swing on a daily time scale should alert people to the presence of a causative relationship. And decay in the correlation or a simple displacement either vertically or horizontally tells you that something else is involved.
But one doesn’t go looking for those things without a theory that suggests that two things should be related and that theory must be based on a thorough exploration of how the atmosphere has behaved in the past. You must have a clear idea of what you are trying to explain. You have to do the hard yards of data exploration first.
So, you develop a theory and you go looking for relationships that confirm that the posited cause and effect are operable. And one doesn’t keep looking unless the strike rate from these data explorations is exciting. End of the day it is a whole array of observations and a chain of logic about cause and effect that convince you that you have a handle on phenomena.
It strikes me that this correlation http://tallbloke.files.wordpress.com/2011/08/had-uahlat-corr-1.png tells us that equatorial processes do not drive climate change.
Just as the diversity of thermal experience across the hemispheres, by latitude and season should tell you that the greenhouse hypothesis is inoperable.
So, when Wyatt says : Sudden-stratospheric-warmings, which are related to both tropical convective processes and to the integrity of the polar vortex – both features wielding hemispheric influence on the climate.
She betrays confusion about causation that should not be there. The equatorial forces push in one direction and should provoke a wave of reactions spreading towards the poles like a ripple in a pond in both hemispheres, albeit biased according to season. What I see is altogether different. Antarctica is the driver and the Arctic a bit player.
Basically, using the pond analogy: Where is the biggest splash occurring? Where do you see the greatest variation in the variables under observation? It’s as simple as that. Then you ask, what’s causing the splash? Why is it manifesting in this place?
When Wyatt says this:
longitudinal and latitudinal placements of the atmospheric centers-of-action shift with multidecadal variations in AMO, as does the meridional placement of the mean intertropical convergence zone (ITCZ), along with associated changes in Atlantic hurricane activity and in frequency of occurrence of Atlantic-NINOs. Likewise, with multidecadal variability in PDO come changes in placement and strength of atmospheric centers-of-action and in placement and strength of associated oceanic gyres and in meridional mean location of the Pacific ITCZ.
It would be a lot simpler of she were to look at the Southern Annular Mode and the Northern Annular Mode and how they evolve over time. These are acknowledged modes of inter-annual climate variability located close to the poles, obviously related to the coupling of the troposphere and the ‘startosphere’.
That little dyslexic typing mistake is something that Kim will appreciate. It’s accurate. The characteristics of the stratosphere are determined by a circulation wherein the stratosphere interacts with the mesosphere. The basic dynamic determining the variability in the interaction is surface pressure. And surface pressure responds in the first instance to the solar wind.And in the second instance it responds to the dynamics created by the interaction of the stratosphere and the troposphere that amplifies the pressure response due to the electromagnetic effect. The circulation is not maintained without constant stimulus.
erl happ says:
August 27, 2011 at 6:07 pm
The characteristics of the stratosphere are determined by a circulation wherein the stratosphere interacts with the mesosphere. The basic dynamic determining the variability in the interaction is surface pressure. And surface pressure responds in the first instance to the solar wind. And in the second instance it responds to the dynamics created by the interaction of the stratosphere and the troposphere that amplifies the pressure response due to the electromagnetic effect.
Not to reopen the thread, but the above is totally, utterly wrong, and incoherent to boot.
Hi Leif,
Don’t worry, I will keep working on you until it is coherent, and if not to you, then until it is sufficiently coherent to your wife or some other person close to you that you will trust implicitly.
erl happ (August 27, 2011 at 6:07 pm)
“It strikes me that this correlation http://tallbloke.files.wordpress.com/2011/08/had-uahlat-corr-1.png tells us that equatorial processes do not drive climate change.”
I have to point that this is a misinterpretation.
(Part of the burden of being a former stats instructor.)
Not challenging your idea – but the graph is about something else…
…and also misinterpreted by its author (TC).
Bob Tisdale would have an easy time giving misinterpreters a schooling on this one.
Thanks for the articles Erl.
Best Regards.
Sorry Paul, It seems that I may have misconstrued what that correlation was about. And its not plain in the original.
erl happ says:
August 28, 2011 at 1:58 am
It seems that I may have misconstrued what that correlation was about. And its not plain in the original.
It is called confirmation bias: you misconstrue to see what you wish.
Tim’s aggregation criterion: Summarize by latitude.
The problem: Line transects crossing alternating spatial patterns (e.g. sideways V-patterns one sees on global SST maps caused by opposing average trades & westerlies …& there are 2 looping basins in the north, a different pattern in the south – the equator is the only straight line transect and even so, it’s NOT without spatial modes, etc., etc.)
http://upload.wikimedia.org/wikipedia/commons/6/67/Ocean_currents_1943_%28borderless%293.png
In layman’s terms: Tim used a blender on the spatial patterns.
One can design other blending schemes …and get completely different results.
Plus: Tim looked only at a marginal distribution. He went from studying the temporal dimension blindered to the spatial one to the other extreme of studying the spatial dimension blindered to the temporal one. The sum of the 2 in isolation [even if one designs sensible aggregation] is NOT a joint distribution. Bob Tisdale would know very well that Tim would obtain completely different results if he laid down his line transects &/or quadrats differently and looked at lags.
It’s a sampling issue.
Aggregation criteria FUNDAMENTALLY affect measures of pattern (aliasing, integration across spatiotemporal harmonics, etc.). Change the size, shape, or orientation of your transect &/or quadrat and you measure a DIFFERENT pattern from the SAME field. (This is why Milanovic is so worked up.)
In landscape ecology we went through a bit of a revolution in the 90s where some folks were waking up to this and rocking the boat. There are some in climatology who know this VERY well (I took a graduate level course from one 16 years ago on exactly this), but my experience as an educator tells me this stuff isn’t always intuitive for students …and some forget right after they learn. We certainly have a challenge here for capable educators. The online medium is excellent for communication, but designing the needed messages (with good “picture worth 1000 words” visuals) takes FAR more time than most have available.
I suggest that everyone take the time to understand Milanovic’s primary concern, even if he hasn’t packaged the message the way people need it.
Sorry for the fast & loose message. I wish I had time to give the course, but since I don’t I can at least alert you to the issue. Erl, add to your list of “other factors” the following PRIMARY one: spatiotemporal aggregation criteria.
If you want to discuss this further, I suggest doing so over there [ http://tallbloke.wordpress.com/2011/08/21/delve-into-halcrut-at-the-poles/ ] (a still-active thread) since everyone has moved on from here by now.
Regards.
Leif,
It is called confirmation bias: you misconstrue to see what you wish.
It’s always a danger and sooner or later it catches up with you, and you come a nasty cropper. Good to admit that when it when you see it.
erl happ says:
August 28, 2011 at 7:24 am
Good to admit that when it when you see it.
Yes, and you caught it in time.
Erl, re the EUV impacts on the atmosphere, here’s a new report from NASA about flares.
http://science.nasa.gov/science-news/science-at-nasa/2011/19sep_secretlives/
Seems there’s often a delayed blast of EUV, about 4X the energy of the previous flaring.