El Ni ˜
no and Ocean Observations
81
Southern Oscillation, a see-saw in atmospheric pressure between the eastern and
western hemispheres first described by Sir Gilbert Walker in the early twentieth
century (Bjerknes, 1966, 1969a). He also realized that El Ni˜ no involved the entire tropical Pacific basin, not just the coast of South America as was previously
believed.
Bjerknes recognized that positive feedbacks between the ocean and atmosphere,
mediated by variations in surface winds and sea surface temperature (SST), were
critical to the generation of El Ni˜ no events. However, he did not understand what
caused the trade winds to weaken at the onset of El Ni˜ no, or what processes shut
down El Ni˜ no once underway. Likewise, his ideas about the physical oceanographic
processes involved in El Ni˜ no were vague. He correctly ascribed unusually high
SSTs in the eastern Pacific during El Ni˜ no primarily to a reduction in the intensity of
equatorial upwelling; however, he believed incorrectly that the cause of the reduced
upwelling was a weakening of the local trade winds.
Bjerknes was the first to link El Ni˜ no to patterns of weather variability over
North America via atmospheric teleconnections emanating from the tropics. He also
suggested that El Ni˜ no might be predictable and understood the need for systematic
ocean observations in support of climate prediction. He expressed these ideas in
particularly visionary remarks during a speech in 1969 (Bjerknes, 1969b):
In the still farther future we can visualize the creation of a worldwide service of
synoptic oceanography having as one of its most important duties to maintain
monitoring buoys reporting by way of communication satellites such data which
enter into the construction of transequatorial profiles at several selected geographical longitudes. That would usher in the era when attempts can be made to give
electronic computers the right input for global long-range dynamical predictions
of the fluctuations of the coupled circulations of the atmosphere and ocean.
El Ni˜ no research gained momentum with the 1972–73 El Ni˜ no when the Peruvian anchovy fishery, the largest fishery in world at the time, collapsed under the
combined weight of overfishing and El Ni˜ no-induced environmental stress. Anchovies
were widely used as feed supplement for poultry and livestock, so that collapse of
the fishery (which accounted for nearly one-third of Peru’s foreign exchange) rippled through the global economy (Glantz, 2001). In addressing the question of what
physical processes controlled Peruvian coastal upwelling and therefore biological
productivity during El Ni˜ no, Klaus Wyrtki identified the role of downwelling Kelvin
waves
2 generated by relaxation of the trade winds in the central Pacific, rather than
local wind variations in the eastern Pacific (Wyrtki, 1975). Shortly thereafter, the first
dynamical theories for El Ni˜ no based on this concept began to appear (Hurlburt et al.,
1976; McCreary, 1976). El Ni˜ no research in the United States was further accelerated
2 Kelvin waves are named after Lord Kelvin (Sir William Thompson), who first described these waves
mathematically in the nineteenth century (Gill, 1982).
no and Ocean Observations
81
Southern Oscillation, a see-saw in atmospheric pressure between the eastern and
western hemispheres first described by Sir Gilbert Walker in the early twentieth
century (Bjerknes, 1966, 1969a). He also realized that El Ni˜ no involved the entire tropical Pacific basin, not just the coast of South America as was previously
believed.
Bjerknes recognized that positive feedbacks between the ocean and atmosphere,
mediated by variations in surface winds and sea surface temperature (SST), were
critical to the generation of El Ni˜ no events. However, he did not understand what
caused the trade winds to weaken at the onset of El Ni˜ no, or what processes shut
down El Ni˜ no once underway. Likewise, his ideas about the physical oceanographic
processes involved in El Ni˜ no were vague. He correctly ascribed unusually high
SSTs in the eastern Pacific during El Ni˜ no primarily to a reduction in the intensity of
equatorial upwelling; however, he believed incorrectly that the cause of the reduced
upwelling was a weakening of the local trade winds.
Bjerknes was the first to link El Ni˜ no to patterns of weather variability over
North America via atmospheric teleconnections emanating from the tropics. He also
suggested that El Ni˜ no might be predictable and understood the need for systematic
ocean observations in support of climate prediction. He expressed these ideas in
particularly visionary remarks during a speech in 1969 (Bjerknes, 1969b):
In the still farther future we can visualize the creation of a worldwide service of
synoptic oceanography having as one of its most important duties to maintain
monitoring buoys reporting by way of communication satellites such data which
enter into the construction of transequatorial profiles at several selected geographical longitudes. That would usher in the era when attempts can be made to give
electronic computers the right input for global long-range dynamical predictions
of the fluctuations of the coupled circulations of the atmosphere and ocean.
El Ni˜ no research gained momentum with the 1972–73 El Ni˜ no when the Peruvian anchovy fishery, the largest fishery in world at the time, collapsed under the
combined weight of overfishing and El Ni˜ no-induced environmental stress. Anchovies
were widely used as feed supplement for poultry and livestock, so that collapse of
the fishery (which accounted for nearly one-third of Peru’s foreign exchange) rippled through the global economy (Glantz, 2001). In addressing the question of what
physical processes controlled Peruvian coastal upwelling and therefore biological
productivity during El Ni˜ no, Klaus Wyrtki identified the role of downwelling Kelvin
waves
2 generated by relaxation of the trade winds in the central Pacific, rather than
local wind variations in the eastern Pacific (Wyrtki, 1975). Shortly thereafter, the first
dynamical theories for El Ni˜ no based on this concept began to appear (Hurlburt et al.,
1976; McCreary, 1976). El Ni˜ no research in the United States was further accelerated
2 Kelvin waves are named after Lord Kelvin (Sir William Thompson), who first described these waves
mathematically in the nineteenth century (Gill, 1982).