The influence of intraseasonal variations of tropical convection on sea surface temperatures at the onset of the 1997-98 El Niño

John Bergman, Harry Hendon, and Klaus Weickmann

The public has become increasingly aware of climate variations particularly with variations in the El Nino- Southern Oscillation (ENSO), which exhibits year-to-year fluctuations and climate change with variations in time-scales of decades or longer. Scientists are now recognizing that atmospheric fluctuations on time-scales of days to weeks may influence the evolution of these longer time-scales processes. In particular, recent research suggested that short time-scale variations of convection in the tropical oceans (referred to as the "Madden-Julian Oscillation" or MJO) may contribute to year-to-year variations in the tropical Pacific and may help trigger the development and demise of El Niño. Interest in this possibility has increased recently because the 1997-98 El Niño, arguably the strongest of the century, followed exceptionally strong MJO activity during the winter of 1996-97.

The MJO is characterized by a large scale envelope of tropical convective activity that moves slowly eastward, from the Indian ocean into the west and central Pacific. There are two primary mechanisms through which the MJO affects sea surface temperatures. The first is the response of the local surface temperatures to enhanced winds and cloudiness that accompanies tropical convection. Both of these features act to cool surface temperatures; the winds increase surface evaporation and the clouds shade the surface from sunlight. In addition, ocean disturbances, initiated by changing wind fields during MJO activity, move eastward along the equator, altering surface temperatures along the way. MJO activity is particularly efficient at activating these disturbances, called 'Kelvin waves', because the propagating speed of the MJO closely matches the natural propagating speed of the Kelvin waves, providing a form of resonant forcing.

Fig 1 plot

Fig 1: Infrared emmisions (W/m -2) from the Earth as viewed from satellite as a function of time and longitude over the equatorial Indian and Pacific oceans. Values have been averaged over 10S to 10N. Thick lines follow the propagation of two strong MJO events

The MJO is easily identified in satellite observations because infrared (IR) emissions from the Earth are sensitive to the presence of high clouds that accompany tropical convection. For example, figure 1 shows IR observations, averaged within 10 degrees of the equator, contoured as a function of time (increasing downward) and longitude over Indian and Pacific oceans. Low values (blues in Fig. 1), where high clouds are prevalent, are concentrated in diagonally-oriented contours, indicating eastward propagation. During the winter of 1996-97 there were two strong MJO events that may have contributed to the evolution of the 1997-98 El Niño. These events are highlighted on Fig. 1 with diagonal lines that approximately follow and extend the path of propagation.

Fig 2 plot

Fig 2: Surface temperature anomalies (differences from the climatological annual cycle in Degrees Kelvin) contoured as a function of time and longitude over the equatorial Indian and Pacific oceans. Values have been averaged over 10S to 10N. Thick lines follow the propagation of two strong MJO events.<

Fig. 2 shows evidence for the impact of MJO on long-term surface temperatures. in Fig. 2. The first MJO (solid line) occurred during December 1996 and was accompanied by substantial cooling in the Indian ocean and the west Pacific (90 E to 150 E), initiating a long-term cold anomaly there. The second (dashed line) occurred during February-March 1997 and was accompanied both by cooling in the west Pacific and warming in the central Pacific (near the dateline). The warming in the central Pacific was particularly important. After that warming, convective activity, which develops preferentially over warmer waters, migrated to the central Pacific. That reduced the strength of the trade winds across the central and east Pacific, promoting further warming. This event signaled the onset of a very strong and rapidly developing El Niiño.

Fig 3 plot

Fig 3: Surface temperature (SST) and upper ocean quantities averaged over the top 50m. On the left are shown values from the Pacific (10S to 10N; 130E to 150E) for December 1996 through March 1997 of: (a) surface temperature(K), (b) net surface flux into the ocean (K per week), and temperature advection by ocean currents (K per week). On the right are shown values from the central Pacific (10SS to 10N, 160W to 180W) for February through May 1997 od:(d) surface temperature, (e) temperature advection, and (f) zonal wind stress (dyne per cm). Shaded regions indicate times of rapid surface temperature changes.

To help elucidate the potential role of MJO activity for the onset of El Niño, we examine dynamical interactions in the upper ocean that led to the important SST changes in the winter of 1996-97. Fluctuations of surface temperature that occurred in the west Pacific (Fig. 3a) during convective episodes in December 1996 and February 1997 (shaded periods) were, for the most part, forced by surface flux variations (Fig. 3b). Surface cooling was initiated by the reduction of shortwave surface fluxes due to enhanced cloud cover. Later, evaporative cooling during eastward wind anomalies reinforced that cooling. In February 1997, cool surface temperatures were transient, lasting only a few weeks. However, that cooling is interesting because vertical temperature advection (Fig. 3c) by ocean currents played an important role; off-equatorial upwelling, through an anomalously large vertical temperature gradient, contributed substantially to west Pacific cooling.

During late March and early April 1997, surface temperatures in the central Pacific (Fig. 3d) warmed in response to an oceanic Kelvin wave that was forced during the February MJO. That warming was primarily caused by temperature advection in the ocean (Fig. 3e); eastward currents acting in concert with the temperature gradient (warm waters in the west, cooler temperatures in the east) that is typically prevalent in the equatorial Pacific. After the passage of the Kelvin wave, zonal currents, which are typically westward, remained slightly eastward because trade winds (Fig. 3f) did not resume their pre-Kelvin wave strength. In addition, the east-west SST gradient was reduced after the Kelvin wave. As a result, temperature advection was negligible after the Kelvin wave, and surface temperature continued to increase due to the positive surface heat flux that is typical for the region. It therefore appears that MJO-related forcing of surface temperature fluctuations initiated the coupled interactions of the ocean and atmosphere responsible for the onset and rapid development of the 1997-98 El Niño.

For a more technical discussion, detailed analysis, see the following reference:

  • Bergman, J. W., H. H. Hendon, K. M. Weickmann, 2001: Intraseasonal air-sea interactions at the onset of El Nino. J. Climate, 14, 1702-1719

Glossary of Terms
El Niño
Spanish for "The boy". The phase of ENSO which is associated with warmer than normal SST's in the eastern Pacific and warm er than normal SST's in the west. Convection in the western Pacific tends to be further west than the climatological average. It is opposite to La Nina.
Originally, it referred to El Niño/ Southern Oscillation or the combined atmosphere/ocean system during an warm event. Currently, it generally refers to both the La Niña and El Niño phases of the coupled atmosphere/ocean system though sometimes it's still used as originally defined.
Generally, transport of heat and moisture by the movement of a fluid. In meteorology, the term is used specifically to describe vertical transport of heat and moisture, especially by updrafts and downdrafts in an unstable atmosphere. The terms "convection" and "thunderstorms" often are used interchangeably, although thunderstorms are only one form of convection. Cbs, towering cumulus clouds are visible forms of convection. However, convection is not always made visible by clouds. Convection which occurs without cloud formation is called dry convection, while the visible convection processes referred to above are forms of moist convection.
Heat flux
The movement of heat out through an imaginary box around a small parcel of fluid. Surface heat flux is the net amount of heat absorbed (or released) at the surface and will cause warming or cooling of the surface.
Kelvin Waves
In the tropics, Kelvin Waves are associated with the boundary of warm waters near the surface of the ocean and cooler waters below. Waves can form on the boundary between the cold and warm water. These waves remain trapped about the equator and move eastward at speeds of about 3 m/s.
Propagating Speed
Speed that a wave moves, or, "propagates".
Temperature advection
The movement of the ocean (air) brings warmer or cooler water (air) to another location. Temperature of the water (air) can change by means other than advection.
Trade winds
Winds that move from east to west along the surface of the tropics (30N to 30S).
Temperature gradient
Change of temperature over a distance.
Ocean water moving upward towards the surface. Since the ocean tends to cool with depth, upwelling is associated with colder waters moving towards the surface and hence cools the ocean surface. Upwelling can be forced by winds blowing along the surface of the water.
Wind stress
The force on a unit area of a surface by the wind.
Zonal currents
Ocean currents that move from the east or west (as opposed to north or south).