High pressure cells are areas of descending air while ascending air is found in low pressure cells. Air travels from high to low pressure in a circuitous fashion, crossing isobars (lines of equal pressure). When isobars are close together, the wind velocity is greater. Speculatively, the speed and volume of flow depends upon the pressure differential and also the size of the cells involved.
The Trade Winds originate in high pressure cells centred at about 30° of latitude in winter and 50° of latitude in summer. Air flows from these high pressure cells towards low pressure cells at the inter-tropical convergence near the equator. There is a wind with a westerly component that flows towards the poles from these same high pressure cells. High pressure cells are largely cloud free. High pressure cells establish and endure most strongly over cold waters that are free of the diurnal flux in temperature evident over the land. However, a large high also establishes north of the Himalayas, on land, dominating the northern circulation in winter.
The intensity of the wind in the trade wind zone drives wave action that determines the surface area of the ocean and thereby evaporation. Under high and relatively invariable levels of sunlight, the rate of evaporation from tropical waters is the prime factor determining surface temperature. But, the trade wind also drives the flow of the equatorial currents and determines the degree of upwelling of cold waters from below. This cools the eastern margins of the oceans. Cool waters are driven in a westerly direction by the trades.
It is plain therefore that warm tropical waters are associated with slackness in the trade winds. In the Pacific this is the ‘El Nino’ situation. The reverse, ‘La Nina’ is characterized by vigorous trade winds, enhanced surface cooling by evaporation and strongly upwelling cold waters. These phenomena are seen in tropical latitudes in all oceans.
One notes that the flux in global temperature closely follows that in the tropics. It is common parlance that ‘teleconnections’ link change in temperature of the Pacific Ocean to change in the weather around the globe. The El Nino Southern oscillation (ENSO) is seen as the major mode of inter-annual variation in the climate of the earth.
Since the intensity of the Trade winds depends upon the pressure differential between subtropical high pressure cells and lows located at the equatorial convergence, we can infer the strength of the trades via the flux in surface atmospheric pressure.
The flux in surface atmospheric pressure in the subtropical latitudes between January 1948 and July 2010
The data in Figures 1 through 11 is presented as 12 month moving averages centered on the 5th month. The data refers to the entire latitude band. All data in figures 1-14 is sourced from:
This data is cited as
Kalnay, E. and Coauthors, 1996: The NCEP/NCAR Reanalysis 40-year Project. Bull. Amer. Meteor. Soc., 77, 437-471.
The pressure differential between 10-30°south and 10° north to 10°south fell between 1948 and 1978, and remained low and relatively invariable until about 1996, subsequently increasing again.
The pressure differential in the northern hemisphere is seen to be little better than half of that in the southern hemisphere. In the north the differential increased from 1948 till 1995 and fell away thereafter.
The change in differentials at this latitude is slight. The lower pressure differential in the north is probably due to the fact that the warmest equatorial waters are located between the equator and 10°north rather than at the equator itself. This means that the 10-30°north zone is warmer than in the 10-30°south zone. The northern zone does not support the generation of high pressure cells like the cooler zone in the southern hemisphere.
The inter-annual change in pressure differential between 30 to 50° and the equator (figure 2) is plainly more vigorous than it is at 10-30° latitude (Figure 1). The ENSO dynamic is plainly driven by change in the atmosphere outside the tropical latitudes.
Reversing the relationship apparent in figure 1, the pressure differential between 30-50° latitude and the equator is greater in the northern hemisphere.
The progressive collapse in pressure differential in the southern hemisphere is impressive. This collapse and recovery (in part), that is evident in the trend curves, suggest that natural, reversible cycles of multi-decadal length drive the climate system.
A decadal collapse in the differential pressure occurred in both hemispheres between 1988 and 1998. This collapse is most obvious in the northern hemisphere. In the south, the differential fell strongly, bottoming out in 1992 then rose again, but to a figure that remained below the trend line. This produced the ‘El Nino of the century in 1997-8. The recovery of this differential in 1999, to produce the La Nina of 2000, was equally spectacular.
It is the collapse in the differential in the northern hemisphere that seems to be associated with the strong El Nino of late 2010-2011.
Collapsing differentials indicate that the climate system acquires an El Nino bias. The base state then promotes continuous warming.
Change in atmospheric pressure at high latitudes
The pressure differential between 50° and 70° north and 10°north to 10°south is positive in the northern hemisphere but strongly negative (and increasingly so over time) in the southern hemisphere. Even though it occurs at high latitudes well away from the trade wind zone this development would tend to weaken the south-east trades by diverting air to the north-westerly flows tending south. The strengthening sink is in the south and it robs the tropical sink. Secondly, it is probable that there has been a diversion of air supply from the southern to the northern Hadley circulation. This is consistent with the slowly increasing differential in atmospheric pressure at 30-50°north in relation to 10°north to 10°south. This would further weaken the southern trades while strengthening the northern hemisphere circulations. Again we see a proclivity for a base state of El Nino dominance in the southern hemisphere where the expanse of heat absorbing ocean is greater than in the northern hemisphere. This must be seen to be an adequate explanation for the increase in tropical and global temperature over the period 1978 to 1998. The lack of further warming after 1998, in a regime of an unchanging differential, is wholly consistent with that explanation.
The collapse in the pressure differential at 50-70° south is much greater than in the northern hemsiphere. It is more exaggerated on both inter-annual (ENSO) and multi decadal time scales. The negative pressure differential at 50-70°S increased by 8 bars between 1948 and 1978. This was due to a strong fall of pressure at 50-70° south and a slight gain at the equator. The collapse in the differential was most extreme in the period 1994 to 1998.
There is a positive pressure differential between 70-80°north and 10°north to 10°south in the northern hemisphere but the differential has diminished over time. The differential ran to the negative in the 1990’s and from 2007 it has recovered.
The inter-annual (ENSO) fluctuation is most extreme in the southern hemisphere. Lowest pressure relativity between 70-80° and the equator was seen in 1994.
Again, we observe that the origin and extent of the El Nino of 1997-98 lies in the low differentials established in mid and high latitudes of both hemispheres during the period 1990-94.
The pressure differential between the high polar latitudes and the equator is normally positive in both hemispheres. But notice the negative pressure differential in the Arctic over the period 1990-1996 leading up to the El Nino of the century in 1997.
The Antarctic has suffered a marked reduction in surface pressure over the period. This is of the order of 11 bars between 70 and 90°south.
The pressure differential in the northern polar region recovered strongly from 1990. The differential at 80-90°north recovered to a high point (in terms of the record as a whole) during at the start of 2010. The resulting flow of cold air from the Arctic into continental Europe and North America prompted many observers to remember similar winters in the 1970’s.
During Southern hemisphere winter in mid 2010 Antarctic differential pressure has also been high with cold polar air affecting Australia, and South America.
Globally, the most severe fluctuations in surface pressure are seen at 70-80°south. It is interesting therefore to compare the flux in atmospheric pressure at 70-80°south to that at 10N-10S. This data is presented in figure 6.
Atmospheric pressure at 10°north to 10°south is plotted on the right axis that has a restricted scale by comparison with the left hand axis. It must be remembered that the equatorial surface area is much greater than the at the pole. It takes a large drop of pressure at the pole to produce a small increase in atmospheric pressure at the equator.
Generally, the relationship between atmospheric pressure at 70-80° south and pressure at 10°north to 10°south is inverse, confirming that it is the loss of pressure in high southern latitudes that allows the gain in pressure at low latitudes and in the northern hemisphere generally. The gain in pressure in the low and mid latitudes of the northern hemisphere is apparent in figures 1 and 2 above.
Peak pressure at 10°north to 10°south occurred in 1998 and pressure has been in decline since that time. This will markedly assist the recovery of the trade winds and the end of the El Nino base state. In fact, data for the SOI (Southern Oscillation Index) (not shown) suggests that the base state has been La Nina dominant since 2007.
Figure 7 relates changing pressure differentials to sea surface temperature between 20°north and 20°south latitude. However, in this figure the pressure differential between 30-50° south latitude and 10°northto 10°south is inverted by reversing its sign. In this graph a rise in the pressure data indicates a fall in the differential. A fall in the differential causes sea surface warming. The blue line confirms the point.
It is apparent that change in the pressure differential ordinarily precedes the change in sea surface temperature. The pressure differential is a predictor of sea surface temperature. But, is it reliable?
Sea surface temperature at 20°north and 20°south depends upon the activity of both the north-east and the south-east trades, and as we have seen, pressure changes rather differently in each hemisphere. Let us now look at the northern hemisphere.
Notice the strong relationship between the northern pressure differential and sea surface temperature in 1998 and 2011. It appears that the northern differential has been the prime driver of tropical sea surface temperature since 1996. Notice also that sea surface temperature leads the northern pressure differential between 1972 and 1996. In this period the southern differential seems to havebeen more influential. However, in the 1963 El Nino it was the north that drove the change.
This paper therefore presents a new ‘Southern Oscillation Index ‘ and for the first time, an equally valid ‘Northern Oscillation Index’. Both indices rely upon the pressure differential across the entire latitude band rather than that between spot locations. We should be interested in what drives the trade winds in all oceans, not just the Pacific. Figure 9 presents these two indices on a common axis.
It is apparent that the indices are as different as the hemispheres that give rise to them. It is also apparent that an El Nino bias is currently reversing in both hemispheres.
Let us compare the new (pressure band) ENSO Oscillation Index with the old SOI? Figure 10 relates to the southern hemisphere. Remember, a fall in the SOI indicates warming in the tropics.
The southern pressure band index is more volatile than the SOI. The SOI is not a good guide to global pressure relations at 30-50°south. The pressure band index frequently leads the SOI revealing that change begins outside the confines of the Pacific Ocean.
And in figure 11 there is the comparison of the Northern (pressure band) Oscillation Index with the traditional SOI.
Plainly the SOI is not a good guide to the pressure relations driving the trade winds in the northern hemisphere.
Now let us focus on the last couple of decades using monthly data rather than a 12 month moving average. Figure 12 is compiled using a 5 month moving average of monthly data. The green and brown bars are a means of locating each pressure collapse within the annual cycle.
The leap in SST in early 2010 represents the first El Nino of solar cycle 24 just as the leap in temperature in 1997 was the first El Nino of solar cycle 23. These events, of almost equal stature, were plainly driven in the main by a collapse of differential pressure in the northern hemisphere. The La Nina that prevailed through 2001 and 2002 persisted through solar maximum. A similar experience might be expected in 2011 and 2012. The southern index is currently plummeting and the northern index may well fall as far as it did in 2008.
It is apparent that the northern hemisphere pressure differential frequently collapses in mid winter (1998, 2001, 2002, 2003, 2004, 2010) or in spring (1999, 2000, 2005, 2007, 2008).
The southern pressure differential collapses most frequently in late winter/southern spring. However, there is one instance of autumn collapse in 2004 (slight) and one in midsummer 2010 (again a slight collapse).
It is noted that the SST response to a collapse in the differential in atmospheric pressure appears without delay. See for instance the events of 1997 and 2010 when the collapse in pressure occurred in both hemispheres at the same time.
Figure 13 is compiled from a 12 month moving average of monthly data. It compares the changing pressure differential at mid latitudes with air temperature at 10hPa over Antarctica.
An enhanced pressure differential at 30-50°south seems to be associated with a plunge in 10hpa temperature at 80-90°south, the latter implying a vigorous polar vortex. A vigorous vortex brings nitrogen oxides from the mesosphere, eroding ozone. Ozone is a primary driver of temperature in the stratosphere because ozone is a greenhouse gas par-excellence, reacting strongly to long wave radiation from the earth.
Figure 14 presents monthly data. The calculated anomaly represents the departure from the average for the entire period January 1948 through to July 2010.
The dramatic fall in sea level pressure at 70-90°south in July 2010 is associated with a sudden warming of the stratosphere between Antarctica and Australia in that month. The course of 10hPa temperature at 65-90° south is shown in figure 15.
Figure 15 shows the range of data over the years 1979-2010. In the period 1948-1978 winter temperature were much cooler than after 1978.
Returning to figure 14, relative to the period as a whole, 10hpa temperature has been anomalously high in winter when the high altitude atmosphere reaches its lowest temperature (as seen in figure 15). Figure 14 shows that this anomalous warmth at 10hPa in midwinter is associated with a collapse of surface atmospheric pressure at 80-90°south. This represents a change in the base state of the climate system that weakens the differential pressure driving the trade winds. The change in the base state happened in 1976-1978. It is well documented in my post ‘The climate Engine’.
Tropical sea surface temperatures respond to the change in surface pressure across the globe and in particular to the differential between mid latitudes and the near equatorial zone.
The southern hemisphere and high latitudes in particular experience marked flux in surface pressure. This leads directly to a change in the trade winds and tropical sea surface temperature.
There is an asymmetry between the hemispheres with loss of pressure in the southern hemisphere compensated to some extent by a gain in pressure in the northern hemisphere.
If we wish to understand the ENSO phenomenon we must look beyond the tropics for causal factors. ENSO in the Pacific is just one facet of change in the tropics. Change is driven by air pressure variations at mid and especially high latitudes. This determines the strength of the trade winds and the temperature of the tropical ocean (where solar insolation is greatest and cloud cover is least). There are knock on effects for heat transfer from the tropics to mid and high latitudes, rainfall, flood, drought and tropical cyclone activity worldwide. The tropical oceans are the Earths solar array.
The flux in surface pressure appears to be cyclical. However, the cycle is longer than the sixty years of available data. We cannot say for sure what the cycle length may be or how it varies over time. However, there is good evidence that the warming cycle that began in 1978 peaked in 1998. Cooling is underway.
We must acknowledge that the ENSO cycle is not temperature neutral. There are short ENSO cycles of just a few years and long ENSO cycles that are longer than 60 years.
Is there evidence that the activity of man (adding CO2 to the atmosphere) is tending to produce more severe El Nino events. The answer is no. The flux in surface pressure is responsible for ENSO and for the swing from El Nino to La Nina dominance. In spite the activities of man, the globe is currently entering a La Nina cooling cycle testifying to the strength of natural cycles and the relative unimportance atmospheric composition in determining the issue (if the much touted greenhouse effect exists at all) .
Is there evidence that the ENSO phenomenon is in fact ‘climate change in action’, driven by factors other than the increase of atmospheric CO2? Yes, it appears that whatever drives the flux in surface atmospheric pressure drives ENSO and with it, climate change.
Is recent ‘Climate Change’ driven by greenhouse gas activity? No, it appears that the cause of recent warming and cooling relates to long-term swings in atmospheric pressure that changes the relations between mid and low latitudes thereby affecting the trade winds that in turn determine the temperature of the Earth’s solar array, its tropical ocean, and ultimately the globe as a whole.
Bill Illis on the connection between the Trade winds and ENSO at
And the conventional viewpoint on ENSO can be found here: