Prediction of El Niño
Despite uncertainty about the causes, the system's inclination towards an event can be predicted with increasing success. Prediction is generally achieved by monitoring the events described above under "effects". Small fluctuations in these can show the b
eginnings of an ENSO event before it can be recognised in other ways. There is considerable doubt about the cause and effect sequences of these "teleconnections". For some reason, the precise timing of an event is harder to predict than its form or effect
s. Nevertheless, it is difficult to predict factors such as whether California will be wet or dry, since very small changes in SST can be magnified.
Some prediction attempts are centred around the use of computer models of the atmosphere (global circulation models, or GCMs) which discretize the entire atmosphere and solve equations for various parameters in each cell. As computer power has increased o
ceanic interactions have been included and their resolution is now sufficient to reproduce the effects of ENSO, especially those at the beginning of an event. The 1990 El Niño was successfully predicted months in advance by a model at Lamont-Doherty Geological Observatory of Columbia University, though it also predicted the same event to end in 1992: it is still going. Other models have shown that the predicted doubling of CO2 levels by 2050 will probably increase the effects of ENSO. Details of another model are given here.
Prediction will be aided by improvements in measurement technology. Sea temperature may now be measured to a depth of 500 metres by a buoy network and satellites, while several major surveys have recently been done. These things are all coming together in advisory reports such as this or this. Committees are being formed out of the organisations trying to predict it. There seems to be considerable hope for prediction in the future.
Previous: research on El Niño
Next: conclusions