El Niņo and Australian Drought
By Dr John W Zillman AO FTSE
The El Niņo phenomenon of the Pacific Ocean has achieved considerable notoriety over the past few decades as the main villain in the periodic onset of drought over large areas of Australia. In fact, El Niņo (The Boy Child) was originally the name given to a relatively minor warm ocean current that develops off the coast of Ecuador around Christmas each year. More recently, it is used to describe the extensive warming of the central and eastern Pacific Ocean that recurs every two to seven years, may last for a year or more and leads to major dislocation of weather patterns across the Pacific and further afield in the tropical and middle latitudes of both hemispheres. Though El Niņo is known to have been recurring for many thousands of years, it is only in recent years that it has been possible to monitor its evolution and develop physically based models capable of predicting its future development on timescales of months to a year or more. This is especially significant for Australia, because it is here that it has its greatest influence on seasonal rainfall patterns, usually leading to significantly below normal rainfall, and often severe drought, over large areas of eastern Australia. In this note I will describe the role of El Niņo in the overall mechanisms of global climate including its links with the so-called Southern Oscillation and illustrate their influence on the development and evolution of the 1997 drought.
The Mechanisms of Climate
The global patterns of climate and their fluctuations over time represent the response of the coupled atmosphere-ocean "climate system" to the equator-to-pole gradients of radiant energy input from the sun, as significantly modified by the earth's rotation and the distribution of continents and oceans and major mountain ranges. The solar heating of the tropical land and ocean forces warm moist air to rise in the equatorial belt producing clouds and rain. Were it not for the earth's rotation, we might expect the rising warm air to flow poleward at high levels and sink again in the regions of radiative cooling in the high latitudes of both hemispheres. In reality, the eastward angular momentum acquired by the rising air as a result of its contact with the land and ocean surface in the tropics causes it to accelerate rapidly eastward as it moves poleward, and thus closer to the earth's axis, until it becomes unstable in middle latitudes and breaks down into a series of waves and eddies which appear as the meandering jet stream and the surface "highs" and "lows" of the daily weather chart (Figure 1). Averaged over time, the daily weather systems determine the broad geographical distribution of climate (cf, for example, the mean surface pressure pattern for the southern hemisphere shown on the right-hand side of Figure 1).
Figure 1: The main features of the general circulation of the atmosphere (show greatly exagerated vertical scale) including typical daily synoptic patterns and the mean meridional and zonal circulations (left) and the east-west Walker Circulation around the equatorial belt (right).
The ocean also plays a key role in carrying excess heat from the tropics to the polar regions and the major ocean currents significantly influence the distribution of temperature over the earth. Of particular interest for present purposes is the ocean temperature pattern in the Pacific (Figure 2) with relatively warm water in the western Pacific and considerably colder water in the eastern Pacific. The influence of the Humboldt Current which carries cold water northward along the west coast of South America combined with wind-induced upwelling of colder water from below the surface produces the tongue of relatively cool ocean extending westward from the coast of Peru into the central Pacific.
Figure 2: The annual average pattern of sea surface temperature showing the regions of warm ocean in the western Pacific and relatively cold ocean in the eastern Pacific.
It is the east-west gradient of ocean temperature across the Pacific and the associated pattern of relatively high surface atmospheric pressure over the cold water in the east and low pressure over the warm water in the west which drives the Pacific cell of the Walker Circulation shown on the right hand side of Figure 1. Surface air flows westward across the Pacific rising over the warm ocean to the northeast of Australia, producing extensive cloud and rain. It then returns to the east at high levels before subsiding in the high pressure belt over the colder water of the Eastern Pacific.
El Niņo and the Southern Oscillation
Although not all aspects of the interaction of the atmosphere and ocean during the life cycle of an El Niņo event are fully understood, El Niņo conditions are known to develop when the normal easterly winds across the Pacific weaken or reverse, upwelling of cold water ceases and a large warm anomaly in ocean temperature propagates eastward or develops in situ in the central and eastern Pacific (Figure 3).
Figure 3: Sea surface temperature anomolies (°C) for October 1997, with the warm anomolies in excess of 4°C in the eastern Pacific, typical of a well developed El Niņo event.
The warm ocean temperature anomalies, once they develop, may last for a year or more before the El Niņo collapses and ocean temperatures return to near normal or move into the reverse phase of the El Niņo cycle with anomalously cold water in the central and eastern Pacific, sometimes now referred to as "La Niņa" conditions.
It has long been known that this periodic warming and cooling of the central and eastern Pacific has a major influence on the patterns of air pressure across the Pacific. Indeed, it had been found empirically, even before the El Niņo mechanism was well understood, that there is a large scale east-west see-saw of atmospheric mass across the entire South Pacific basin with its nodes in the region of Tahiti and Darwin. This quasi periodic fluctuation in atmospheric pressure is known as the Southern Oscillation (Figure 4).
Figure 4: The region of influence of the Southern Oscillation shown in terms of the correlation coefficient of annual mean sea level pressure anomolies over the globe with those at Darwin.
The periodicity of the Southern Oscillation is closely linked to that of the El Niņo-La Niņa cycle. A convenient simple measure of the phase of the Southern Oscillation is provided by the Southern Oscillation Index (SOI) which is essentially the normalised difference in sea level pressure between Tahiti and Darwin (Figure 5).
Figure 5: The Southern Oscillation Index (SOI) is a measure of the departure from normal of the pressure difference (PT - PD) between Tahiti and Darwin. When the pressure at Tahiti is abnormally high and that at Darwin low, the SOI is positive, the surface easterly winds are stronger than normal, the Walker Cell strengthens and there is above average ascent and rain in the western Pacific.
When the ocean temperature in the central and eastern Pacific is abnormally cold (La Niņa), the surface air pressure there is likely to be abnormally high and that at Darwin low so that the pressure difference between Tahiti and Darwin will be larger than normal and the SOI will be positive. Because the east-west pressure gradient is steeper than normal, the easterly winds will be stronger than normal and the Walker Cell will run faster with enhanced ascent, increased cloud and above normal rain in eastern Australian longitudes. Conversely, when the eastern Pacific is warm (El Niņo), the SOI will be negative, the Walker Cell will weaken, reverse or contract eastward and there will be reduced ascent and below normal rainfall in the western Pacific.
Figure 6: The Southern Oscillation Index (SOI) for 1988-97 shown as monthly average (vertical bars) and five-month running means (solid line).
Figure 6 shows the monthly and five-monthly running mean SOI for the past decade. The notable feature of Figure 6 is the protracted period of almost continuously negative SOI from 1990 to 1995 associated with an unusually long-lasting (or twice reintensifying) El Niņo warming in the ocean.
Although every El Niņo event is different in detail, it is possible to characterise an idealised El Niņo-La Niņa cycle in terms of the typical patterns of sea temperature anomaly, the phase of the SOI and the associated shifts in the Walker Circulation as shown in Figure 7.
Figure 7: Schematic representation of an El Niņo - La Niņa cycle and the associated changes in the Southern Oscillation Index (SOI) and the Walker Circulation.
Impacts on Rainfall
Since every El Niņo event involves a different pattern and temporal evolution of ocean temperature, each has a different impact on the atmosphere and each produces a different distribution of seasonal rainfall anomalies. It is, however, possible to identify some general patterns of impact on rainfall on a global scale. Figure 8 shows rainfall anomalies expected with El Niņo (warm central and eastern Pacific Ocean) conditions persisting during the June-August and December-February periods. The patterns are broadly reversed during La Niņa (cold central and eastern Pacific) conditions.
Figure 8: Generalised pattern of rainfall anomolies during El Niņo conditions during June-August (top) and December-February (botom).
The 1997 El Niņo Event
The 1997 El Niņo event first began to appear in March and developed rapidly with widespread impacts across the Pacific basin. Notable features of the event through 1997 included:
- ocean surface temperatures in excess of 28°C right across the equatorial Pacific with warm anomalies in excess of 5°C in the east;
- consistently negative values of the SOI since March 1997 with negative values in excess of -20 in several months;
- greatly weakened trade winds across the Pacific, giving way, on occasion, to periods of surface westerlies;
s a shift of the focus of tropical convection to the central and eastern Pacific with severe drought over Papua New Guinea and Indonesia and long-term rainfall deficiencies over most of eastern Australia;
- flooding and loss of life on the tropical west coast of South America;
- drought in northern China and suppressed Indian Monsoon rainfall;
- development of dry conditions in southern Africa.
Across southeast Australia, autumn and winter (March-August) rains were well below normal with extensive areas receiving totals in the lowest ten per cent of all those previously recorded. Unusually, however, for a strong El Niņo event, September produced two good rain-bearing weather systems which, in total, resulted in above average rain for the month over much of the Murray Darling Basin and beyond. October and November were again below normal in most areas though not excessively so. The accumulated rainfall deficiencies over Australia for the period March-November are shown in Figure 9. Despite the good rain over most of New South Wales in September and above average falls in northern New South Wales and southern Queensland in October, most of eastern and southern Australia is now experiencing long-term rainfall deficiencies with serious to severe deficiencies in much of Victoria and parts of New South Wales.
Figure 9: The distribution of rainfall anomolies for the period March-November 1997 corresponding to the period of influence of the 1997 El Niņo event so far. The rainfall totals are shown in terms of decile ranges. Decile target 1 means the rainfall received was in the lowest 10% of all previous falls for the same nine-month period, decile range 2 in the lowest 20% and so on.
Although the skill levels are not yet high, the Bureau of Meteorology's National Climate Centre issues regular Seasonal Climate Outlooks for Australia within the framework of the World Meteorological Organisation's internationally co-ordinated Climate Information and Prediction Services (CLIPS). Close co-operation is maintained with overseas National Meteorological Services and with State Primary Industries Departments and other major user sectors. Updated climate outlooks are provided to the community at large each month through the mass media.
On the basis of physical models of the ocean behaviour and consideration of analogues of the evolution of the SOI during previous El Niņo events over the past century, present indications are that the 1997 El Niņo will persist well into 1998 with a relatively weak monsoon and likely below normal rainfall across much of northeast Australia during the 1997-98 wet season.
El Niņo and Climate Change
There has been considerable recent speculation that the influence of global warming due to the enhanced greenhouse effect could lead to changes in the frequency and/or intensity of El Niņo events (e.g. Meehl and Washington, 1996; Houghton et al, 1996). Indeed, some scientists have suggested that the highly unusual nature of the protracted El Niņo conditions during 1990-95 (Figure 6) could, in fact, already be a product of greenhouse influences (Trenberth and Hoar, 1997).
While such influences cannot be ruled out and, in the view of some, are highly plausible, it must be stressed that the physically based atmosphere-ocean climate models being used for greenhouse studies (Zillman, 1994) do not yet capture the natural El Niņo processes with sufficient accuracy to justify confidence in model assessments of the likely effects on El Niņo of increased greenhouse gas concentrations. It may be some years yet before such links, if they exist, can be established unambiguously.
It is important to recognise, also, that, even if the frequency or intensity of El Niņo events were to increase significantly in a warmer world, it is by no means clear what the implications would be for Australian rainfall. Generalised extrapolations to scenarios of "permanent drought" and/or "more disastrous flooding" have little sound scientific basis at this stage.
References
- Houghton, J T, L G Meira Filho, B A Callander, N Harris, A Kattenberg and K Maskell, 1996. Climate Change 1995: The Science of Climate Change (Contribution of the Working Group I to the Second Assessment Report of the Intergovernmental Panel on Climate Change). Cambridge University Press, 572 pp.
- Meehl, G A and W M Washington, 1996. El Niņo-like Climate Change in a Model with Increased Atmospheric CO2 Concentrations. Nature, 382, pp 56-60.
- Trenberth, K E and T J Hoar, 1995. El Niņo and Climate Change. Geophysical Research Letters (in press).
- Zillman, J W, 1994. The Climate of the Twenty-First Century. Ian McLennan Oration 1993. ATSE Focus No. 83, July/August 1994, pp 5-9.
|
