Technical ENSO Update

18 August 2005

Current Conditions

Since mid-May, there have been no coherent anomalies observed in the ocean's subsurface structure. A negative subsurface anomaly (shoaling of the thermocline) has developed near 130-150W in late July and early August, currently centered at 140W and 100 m in depth, with maximum anomaly of 4 to 5 degrees C. However, it is has only small to moderate spatial extent and is unlikely to lead to a basin-wide development. Overall, the equatorial heat content is near average. There also is no anomalous large-scale ocean-atmosphere coupling occurring over the tropical Pacific Ocean. The zonal winds have remained near their climatological average throughout of the tropical Pacific over the last two months. Although the SST anomalies reached nominal El Nino conditions in June, NINO3.4 SST anomalies have since declined to near their average and there are no substantial or large-scale SST anomalies developing in the tropical Pacific as of mid-August.

Expected Conditions

Presently in mid-August the potential for maintaining neutral SST conditions through the Sep-Oct-Nov period carries a probability of 85 to 90%, which represents a significant enhancement of the odds relative to the historical climatological likelihood of 50%. There is a mere 2 to 3% probability for La Niņa SST conditions, leaving an approximately 10% probability for El Niņo SST conditions. 

The above assessment was made in part on the basis of an examination of the current forecasts of ENSO prediction models. For purposes of this discussion, El Niņo SST conditions are defined as SSTs in the NINO3.4 region being in the warmest 25% of their climatological distribution for the 3-month period in question over the 1950-present timeframe. The corresponding cutoff in terms of degrees C of SST anomaly varies seasonally, being close to 0.45 degrees C in northern spring season as high as 0.75 degrees C in late northern autumn. . La Niņa conditions are defined as NINO3.4 region SSTs being in the coolest 25% of the climatological distribution. Neutral conditions occupy the remaining 50% of the distribution. These definitions were developed such that the most commonly accepted El Niņo and La Niņa episodes are reproduced.

There is considerable variation among ENSO model forecasts for the coming several seasons. The number of models that are forecasting El Niņo conditions to be occurring in the Sep-Oct-Nov period of 2005 is 2 out of a total of 19, or 11%. The number of models that predict La Niņa conditions is 0 (0%). At lead times of more than 4 months into the future, statistical and dynamical models that incorporate information about the ocean's observed sub-surface thermal structure generally exhibit higher predictive skill than those that do not. Among models that do use sub-surface temperature information, 2 out of 14 (14%) call for SSTs of at least a minimum El Niņo level for the Sep-Oct-Nov period, while 0 (0%) predict SSTs cold enough to be considered a La Niņa. (Note 1). The majority of the models indicate that the SST conditions in the NINO3.4 region will continue to be near-normal over the next couple seasons. For the longer lead forecast for Dec-Jan-Feb 2005/06, 5 of the 17 models (29%) forecast El Niņo conditions, 12 models (71%) forecast neutral conditions, and no models (0%) forecast La Niņa conditions. If only the 12 models that use subsurface ocean temperature data are included, these figures become 3 (25%), 9 (75%), and 0 (0%).  Caution is advised in interpreting the distribution of model forecasts as the actual probabilities for the coming several months. The expected skill of one model versus another has not been established using uniform validation procedures, which may cause a difference in the above probability estimate from the true probability. 

An alternative way to assess the probabilities of the three possible ENSO conditions is to use the mean of the forecasts of all models, and to construct a standard error function centered on that mean. The standard error would be Gaussian in shape, and would have its width determined by an estimate of overall expected model skill for the season of the year and the lead time. Higher skill would result in a relatively narrower error distribution, while low skill would result in an error distribution with width approaching that of the historical observed distribution. When this method is applied to the current model forecasts, results indicate a probability of El Niņo that is lower than the climatological value (25%)--beginning in the 10-15% range for Aug-Sep-Oct and Sep-Oct-Nov, rising to near 40% from Feb-Mar-Apr through Apr-May-Jun of 2006. For all periods the probability for La Niņa is well below its climatological value of 25%. The same cautions mentioned above for the distribution of model forecasts apply to this alternative method of inferring probabilities, due to differing model biases and skills.

The IRI's probabilistic ENSO forecast takes into account the indications of this set of models, the outcome of the standard error approach described above, and additional factors such as the very latest observations that may have developed after the initialization times of some of the models. It indicates much higher than average probabilities for neutral conditions, and lower than average probabilities for La Niņa, throughout all of the outlook periods. Throughout most of the forecast period, chances for El Niņo are somewhat below their climatological average of 25%, but increase to that level in northern spring 2006. Neutral conditions are clearly indicated as being most likely over the remainder of 2005 and into 2006.
See also: 

• Summary of model forecasts