El NioLa

The climatological distribution of mean ocean surface temperature in the Pacific(see Fig. 2.2) reveals the existence of an extensive area of warm water called the warm pool (SST 28-30°C) in the western part of the equatorial Pacific (west of about the dateline) as against its eastern part where the combined effects of the Humbolt current and intense upwelling along the coast of Peru maintain an SST which can be as low as 19° C.

However, notwithstanding this normal zonal distribution of SST, almost every year, a narrow warm ocean current known as the equatorial countercurrent flows eastward from the Western Pacific warm pool a few degrees north of the equator(5N-10N) between the westward flowing north and south equatorial currents and shifts southward closer to the equator during the months January to April and ushers in, though temporarily for a few weeks only, an influx of warm water to the coast of South America. During that short period, a branch of the warm current flowing southward along the coast of Peru overrides the usual northward-flowing cold water to a latitude of about 5S. The people of Peru especially its fishermen call this warm countercurrent 'El Nino', meaning in Spanish 'the male child', indirectly referring to 'Christ child', since it comes around Christmas time.

This usual pattern of onset of El Nino, however, is upset once every 2-7 years, by a major disturbance which appears to be related to changes in atmospheric and oceanic circulation and the flow of warm water along the coast of Peru is enhanced many times and extended further south along the coast to the latitude of Lima Callao at 12S or even beyond with disastrous effects on atmospheric and oceanic environment of the region. The surface temperatures of the coastal waters on such occasions rise at places by as much as 7°C. The gradual decrease of average water temperatures towards the south on such occasions indicates that there is considerable mixing of the warm waters with the ordinary cold coastal waters, and during this mixing the organisms in the coastal current, from plankton to anchovy fish which are used to cold temperatures, are destroyed on a very large scale. Dead fish floating in the water and thrown on the beaches decompose and befoul both the water and the air. The decomposition releases so much of hydrogen sulphide that it blackens the paint of the ships, a phenomenon known as the 'Callao painter'. The wholesale loss of fish deprives the guano birds their favorite food and they die of disease or starvation or leave their nests, so that the young ones die, triggering enormous losses to the guano industry.

Table 16.1, compiled by Schott (1931) during the disturbed El Nino period of 1925, shows how and to what extent El Nino affects the water surface temperatures off the coast of Peru.

The trend shown in Table 16.1 has been substantiated by several subsequent studies of major El Nino events in the equatorial Pacific.

Table 16.1 Surface temperatures (°C) of the water, off some coastal stations of Peru in March 1925, as compared to average March temperature (Schott, 1931)

Station

Latitude

Average temp

Temp (oC) in 1925

Lobitos

4o20'S

22.2

27.3

Puerto Chicana

7o40'S

20.3

26.9

Callao

12o 20' S

19.5

24.8

Pisco

13o40'S

19.0

22.1

The meteorological phenomena which accompany a severe El Niño event are no less disastrous. Concurrent with the southward shift of the warm ocean current, the tropical rain-belt of the eastern Pacific moves south and pours torrential rain over a wide coastal belt of Peru. For example, in March 1925, rainfall at Trujillo at 8oS amounted to 395 mm, as against an average precipitation in March of the previous eight years of only 4.4 mm. These bursts of heavy rainfall cause damaging floods and erosion of land.

A recent study(Couper-Johnston, 2000) reveals that in the 473 years since 1525, catastrophic El Ninos struck the equatorial eastern Pacific seaboard as many as 116 times, giving an average frequency of one such El Nino every four years. However, there appears to be no periodicity and the gap between occurrences of two disastrous El Nino's may vary from one to thirty years. This means that in majority of the years, the water surface off the Peruvian coast remains cold with little or no rain over the region. The periods of these cold ocean-surface temperatures with drought conditions have been called 'La Niña' or 'girl child' years.

16.8.3 Southern Oscillation (SO)

In 1877, there was a severe drought in many countries of the world, especially those bordering the Pacific and the Indian oceans. The failure of the monsoon rains in India that year caused untold suffering to the people of the country which led the then Government meteorologist, Henry Blanford, to take up a study of the problem with the ultimate objective of issuing, if possible, advance warning of the occurrence of such calamities. As a first step, Blanford wrote to meteorologists of several countries giving details of past droughts in India and enquiring if similar droughts had occurred in their countries and was surprised when he received a response from Charles Todd, South Australian Government astronomer and meteorologist, to the effect that severe droughts had occurred in Australia exactly in the same years as in India.

Encouraged by Todd's reply, Blanford and later his successor Gilbert Walker and their co-workers in India launched a world-wide search for atmospheric parameters which might show promising relationship with the Indian monsoon rainfall and which they could statistically use for prediction.

In the course of his extensive investigations, Walker (1923. 1924, 1928) found quite a few such parameters around the global tropics. The most promising of these was a relationship which he called the southern oscillation (SO) which he describes as follows (Walker,1924): "By the southern oscillation is implied the tendency of (surface) pressure at stations in the Pacific(San Francisco, Tokyo, Honolulu, Samoa, and South America), and of rainfall in India and Java ... to increase, while pressure in the region of the Indian ocean (Cairo, N.W. India, Port Darwin, Mauritius, S.E. Australia, and the Cape) decreases ... ". He wrote later (Walker, 1928): "We can, perhaps, best sum up the situation by saying that there is a swaying of pressure on a big scale backwards and forwards between the Pacific and the Indian oceans...". A schematic (Fig. 16.8, due to Berlage, 1966) shows regions of the globe encompassed by the SO.

Fig. 16.8 Map showing isopleths of correlation of monthly mean station pressure with that of Djakarta (Dj), Indonesia. Other stations shown are Cocos Island(CO), Port Darwin(D), Nauru(N), Ocean island(O), Palmyra(P), Christmas Island(X), Fanning(F), Malden Island(M), Apia(A) in Samoa, Tahiti(T), Easter Island(E), Puerto Chicama(PC), Lima(L), and Santiago(S) (Reproduced from Berlage, 1966, with kind permission of Royal Netherlands Meteorological Institute)

Fig. 16.8 Map showing isopleths of correlation of monthly mean station pressure with that of Djakarta (Dj), Indonesia. Other stations shown are Cocos Island(CO), Port Darwin(D), Nauru(N), Ocean island(O), Palmyra(P), Christmas Island(X), Fanning(F), Malden Island(M), Apia(A) in Samoa, Tahiti(T), Easter Island(E), Puerto Chicama(PC), Lima(L), and Santiago(S) (Reproduced from Berlage, 1966, with kind permission of Royal Netherlands Meteorological Institute)

According to Fig. 16.8, the isopleth of zero correlation with Djakarta pressure runs along about 170E with positive correlation values to the west and negative to the east. However, it may be noted that the isopleth of maximum positive correlation (+0.8) is not zonally oriented along the equator but inclined to it in an approximately NW-SE direction extending from Australia to central India across the Maritime continent. A similar deviation from the equatorial orientation is noticeable in the case of the isopleth of negative correlation (say, between 0 and -0.4) across the Americas. It would seem that this deviation from equatorial orientation occurs under the influence of the regional monsoons which causes a seasonal shift of the equatorial trough of low pressure and associated intertropical convergence zone (ITCZ) into the summer hemisphere across the respective connecting landmass. The orientation appears to represent the resultant of the zonal and meridional components of the correlation.

Julian and Chervin (1978) who computed the coherence-square statistic between station pressure at Port Darwin and Santiago emphasized two aspects of the maximum in the coherence-square. First, the phase angle of the cross spectrum is almost exactly ±n indicating that the pressures at the two stations are out of phase. Secondly, the bandwidth of the phenomena is rather large, giving an estimate of the period range of 87-27 months (7.2-2.2 years).

These findings emphasize the fact that the Southern Oscillation is not periodic, but certainly oscillatory, as found by the Peruvian fishermen in the oceanic case. Various combinations of stations have been used to compute an index for the SO from differences in station pressure. Most commonly used have been Djakarta (Dj), Port Darwin (D), Santiago (S), Apia (A) Samoa, Tahiti (T), and Easter Island (E), the locations of which are shown in Fig. 16.8. In some recent studies, the normalized pressure difference between Tahiti and Darwin has been used to compute a SO Index (SOI); a large negative value of the index indicates El Nino and positive value La Nina.

16.8.4 The Walker Circulation - ENSO

Examining monthly mean sea surface temperature (SST) and air temperature and precipitation data at Canton island (2°48'S, 171°43'W) during a period of 5 (1963-1967) years, Bjerknes (1969) found that whenever SST was higher (lower) than air temperature at the station, there was a large increase (decrease) in rainfall, the two anomalies being positively correlated. Canton-type rainfall regime is known to prevail along the equatorial Pacific from about 165°E eastward to the coast of South America. He also found that temperature and pressure were negatively correlated, which means that a warm (cold) anomaly was associated with a fall (rise) of pressure. The zero isallobar was found to lie close to the dateline. One of his most important findings was that warm and cold water regimes and pressure and rainfall patterns associated with them alternated between the western and the eastern parts of the equatorial Pacific with the period of about two years(1964 and 1966 were El Niño years; 1963, 1965 and 1967 were La Nina years).

Bjerknes (1969) interpreted the findings of his study at Canton Island in terms of a vertical circulation the rising (sinking) branch of which would be located near the island whenever it had the warm (cold) SST anomaly. A warm anomaly epoch at the island would be associated with lower pressure to which air from the region of higher pressure would converge, rise in convection carrying moisture evaporated from the warm surface to higher levels where it would condense and produce cloud and precipitation. On cooling at higher levels, the rising currents would diverge and after long travel sink over region of higher surface pressure from where they would flow back again towards the island to complete a vertical circulation. On the other hand, during the period of a cold anomaly, the surface pressure at the island would be higher with cold air diverging, so it would come under the sinking branch of the vertical circulation and there would be little or no cloud or precipitation. Bjerknes called this vertical circulation the "Walker Circulation", as its east-west movement was consistent with Walker's southern oscillation in the Pacific.

Thus, Southern oscillation was shown to be directly coupled to El Nino/La Niña events through ocean-atmosphere interaction. Hence the combined oscillation is termed ENSO.

16.8.5 Evidence of Walker Circulation in Global Data

The Walker circulation is clearly evidenced by the field of upper-air divergence computed from the annual mean wind fields which have little or no monsoonal effects

(e.g., Krishnamurti, 1975, 1979; Murakami, 1987). The study by Murakami who computed the annual mean wind vectors and corresponding velocity potential fields at 200 hPa during the FGGE year (December 1978-November 1979) highlights three prominent centers of divergence (D +) along the equator, one over Brazil (60°W), the second over Africa (20°E) and the third over the Maritime continent (170°E). Of these, the divergence center over the Maritime continent which reflects the dominance of the Southwest Pacific Convergence Zone (SPCZ) over the area appears to be the most prominent and strongest of the three to have within its domain a fourth divergence center near Malayasia (100°E). The annual mean winds are nearly symmetric about the equator as they become predominantly zonal on approaching the equator.

The near-symmetry of the equatorial winds between the two hemispheres is not surprising, as it is strongly suggested by several theoretical studies (e.g., Matsuno, 1966; Gill, 1980; Lim and Chang, 1983). Using a linear diagnostic model with equatorial heat sources and sinks prescribed as external forcing, the studies showed that the atmospheric response is generally confined to the equatorial latitudes, where the Rossby radius of deformation is approximately 1000 km, with its structure resembling a Walker circulation. Thus, the Walker circulation is of Kelvin wave type, but stationary. Relative to the divergence center near New Guinea (170°E), equatorial winds are westerlies to the east and easterlies to the west. These equatorial easterlies blow between the zonally-oriented subtropical ridge axes of the two hemispheres along about 10°N and 10°S. Poleward of these ridge axes, winds are westerly with two jetstreams, one near Japan in the northern hemisphere and the other over Australia in the southern hemisphere. There is little doubt that these northern and southern hemispheric W'ly jetstreams are strengthened by the local Hadley circulations between the equator and the higher latitudes.

The equatorial westerlies over the Pacific appear to be the strongest near 140°W. This may be partly due to the equatorward extension of two mid-Pacific upper-air troughs in the westerlies, one to the north and the other south of the equator. Thus, the annual mean wind fields are symmetric not only over the equatorial latitudes but also in higher latitudes, especially in regard to the locations of the subtropical ridge axes and troughs and Jetstreams in westerlies.

16.8.6 Mechanism of ENSO?

No one knows for sure what causes the ENSO and how it is caused. From time to time, scientists have pointed at a variety of possible influences from far-off places, such as excessive snowfall over Eurasia, changes in Antarctic pack ice, seismic activity following volcanic eruptions, etc., but it is believed that these could only have some marginal effects, if any, and that the main cause may be lurking closer to home.

Current thinking is that whatever the real cause a self-perpetuating loop between the ocean and the atmosphere is initiated by it. For example, a perturbation in the form of an anomalous burst of westerly winds along the equator will move the warm pool water a little to the east. This will cause further relaxation of the trade winds allowing more warm water to move towards the east. The movement of warm water flattens the thermocline and less cold water is upwelled in the east along the entire equator. This decreases the temperature difference and hence reduces the pressure gradient between the east and the west. With further weakening of the trades, more warm water flows to the east, preparing the east to have a full-fledged El Nino.

Thus, at the peak of an El Niño, the broad picture of the Pacific is very different from the norm and is described by Couper-Johnston (2000), as follows: "The thermohaline has flattened out considerably. So the deep, cold water normally close to the surface off South America is up to 30 m deeper than usual. The sea levels on both sides of the ocean are comparable. And the pressure difference between east and west has disappeared and, at times, even reversed. This is the flipside of the atmospheric seesaw that Gilbert Walker first noticed when he coined it the words "southern oscillation". Because the pressure difference drives the winds, the trade winds disappear and are replaced by westerlies that can blow nearly all the way to the Americas. The zone of cloudiness and heavy rain that characterizes the western Pacific is now located across much of the equatorial Pacific, and in its place across Australasia the abnormally high pressures bring warmer drier conditions. The clear skies associated with the usual zone of high pressures off the western coast of South America are replaced by rainclouds, as the warmer water off the coast heats up the air above it."

The above-mentioned physical processes at the ocean surface are reinforced by two other very important and related processes a few tens of metres below the ocean surface. The first is the excitation of a series of massive eastward-propagating Kelvin waves. Starting at the western boundary of the ocean, these waves as they move have the effect of lowering the thermocline in the east and thereby moving warm water eastward which may surface later. The second, which is sparked off by the first, is the generation of a series of westward-propagating Rossby waves which have the effect of raising the thermocline in the west and bringing it closer to the surface. The process goes on till the Kelvin waves, traveling at about 100 km (day)-1 and depending upon their point of origin, reach the coast of South America in about two to three months to start a full-fledged El Nino phase. There, they get partially reflected up and down the coast to advect heat to higher latitudes and partially reflected back across equatorial Pacific as Rossby waves. Now, Rossby waves travel at one-third the speed of the Kelvin waves, so they take any time between six and twelve months to reach the western boundary of the ocean. From there, they are reflected eastward as Kelvin waves, but this time as upwelling waves, i.e., raising the thermocline in the east and lowering it in the west. The process goes on till the reflected Kelvin waves reach the South American coast to end the El Niño and replace it by a La Niña phase. Thus, as Kelvin and Rossby waves keep shuttling across the equatorial Pacific Ocean from one side to the other, they would appear to provide the necessary mechanism to start or end an El Niño or a La Niña.

In 1997-1998, there was a pronounced El Niño not only in the equatorial eastern Pacific but also in the other global oceans, which lasted about a year and a half. Fig. 16.9 presents equatorial depth-longitude sections of ocean temperature

Was this article helpful?

0 0

Post a comment