What causes one year to be warmer or cooler than other years?

  Why has 1998 been such a warm year (as measured by the satellites)?

  Now that there is nearly 20 years of satellite temperature data available, it is now possible to determine the primary cause of year-to-year changes in global temperature, and to show how and why 1998 has been the warmest year since 1979 (the year the satellite monitoring began).

  We will deal with two variables only, the first being the temperature of the lower troposphere as measured by satellites, and the second being the Southern Oscillation Index.

  This is an index number derived by comparing air pressure at sea level between Darwin and Tahiti. During an El Niño episode, the index becomes a negative number, and is characterised by warming of the central and eastern Pacific Ocean, disturbing many of the world’s weather systems in the process. During a La Niña episode, the reverse happens and a cooling takes place in these same waters. The cycling between these two episodes is known as the El Niño Southern Oscillation, or ENSO for short. (The cycle between El Niño and La Niña and the significance of the Southern Oscillation Index is discussed more fully on this website here).

  After 20 years of satellite-measured global temperature data, it is now clear that the Southern Oscillation is the primary driver of year-to-year global temperature changes.   This is best illustrated in this chart -

  The chart shows global temperature in red, and the S.O.I. in blue. The SOI has been inverted to more easily reveal the relationship between the index and temperature. That is, actual negative index values (characteristic of an El Niño episode) are inverted to be displayed as positive values, while actual positive index values (a La Niña episode) are shown as negative values. Both graphs represent a 3-month running mean of monthly values of SOI and temperature.

  From the chart, it becomes immediately apparent that during periods where there is no significant volcanic activity, global temperature lags the SOI by between 6 and 9 months. This is very easily apparent when looking at 1987-1989 where an El Niño, quickly followed by a La Niña, resulted in a temperature profile closely mirroring the SOI, but after a 9 month time lag. The same pattern is even more dramatically shown in 1997-98. At most points of the chart, movements in temperature can be traced back to an earlier matching movement in SOI many months earlier.

  This leaves the interesting question. Why did the 1997-98 El Niño result in such a large surge in global temperature when the even stronger 1982-83 El Niño only resulted in a small rise in temperature? Was this a greenhouse warming finally taking hold?

  The grey shaded areas of the chart give the answer. The El Chichon eruption of 6th April 1982 was powerful enough to eject aerosols and dust particles into the stratosphere, resulting in greater scattering and reflection of incoming sunlight, thus dampening the warming effect of El Niño. Had El Chichon not intervened, global temperature would have been free to soar to the values we have seen in 1998 where no volcanic activity was present to inhibit the warming effect. Even so, the 1982-83 El Niño did manage a small warming in spite of El Chichon’s countervailing cooling effect.

  This volcanic cooling effect is revealed very clearly by the Mount Pinatubo eruption, considered to be the biggest this century, with a cooling effect lasting even longer than El Chichon, and with a much more dramatic effect on global temperature, cooling the earth in spite of mostly El Niño conditions during the early 1990s. It will be noticed that the only times in the 20 years when global temperature did not show the full 9-month delayed response was during those periods shown in grey shade when volcanic activity intervened to produce a modified outcome. In 1982-83, we get the predicted warming response, but at a much weakened level, while the Pinatubo eruption effectively cancelled (and even reversed) the effect of the Southern Oscillation

  The two data streams are well correlated, but because of the 6 to 9 month time lag, it is clear that the Southern Oscillation is the causative agent. An `effect’ can only follow a `cause’, it cannot precede it, and so there can be no dispute here about what the chain of cause and effect must be.

  A more interesting question then arises as to what `causes’ the cycles in the Southern Oscillation. We know it results from gargantuan changes in ocean currents and in deep water upwelling in the eastern Pacific, but as to what triggers this response is still an unknown. The answer may well lie in long-term solar changes. The Greenhouse industry readily blames greenhouse gases, but the idea that a few parts per million of CO2 can cause the overturning of trillions of megatonnes of sea water is fanciful to say the least, a reasoning based more on ideology than on science. Those who point to greenhouse gases as the `cause’ of El Niño fail to describe exactly what mechanism they imagine the gases to be performing to achieve such a feat.

  It will be noticed from the chart that the two volcanic eruptions had no effect whatever on the course of El Niño or La Niña. The SOI cycled back and forth, quite oblivious to the cooling being imposed by the eruptions. Thus the Southern Oscillation is not dependent on, or influenced by, changes in atmospheric temperature, whether these changes are caused by volcanic eruptions or, by alleged changes in temperature due to greenhouse gases. In fact, it is quite the opposite - the Southern Oscillation forces the temperature to change, not the other way around.

  Based on the the assumption that the Southern Oscillation is the primary driver of year-to-year global temperature, with a 6 to 9 month lag time, we can now predict that since the SOI has now gone sharply into La Niña mode in the last 6 months, global temperature will follow (with the predicted time lag) and fall to below the zero line (the long term average of temperature) in the next few months. The latest monthly value for temperature was +0.33°C in October 1998, after reaching a peak of +0.72°C in April. Since the SOI moved into La Niña mode in June, we can expect global temperature to fall below the zero line by March 1999.

John L. Daly (22 Nov 1998)

Update: 20th March 2000

Since writing the above article over 15 months ago, global temperature did indeed fall below the long-term average by March 1999. Since then, the earth has continued in La Niña mode with consequent below-average temperatures as measured by the satellites. Temperature continues to closely track the SOI with a time lag.

For the latest version of the comparison chart, click here  - John L. Daly