Summary of ENSO Model Forecasts

Summary of ENSO Model Forecasts

17 July, 2002

A note on interpreting the model forecasts:

The following graph and table show forecasts made by dynamical and statistical models for SST in the Nino 3.4 region for eight overlapping 3-month periods. Note that the expected skills of the models, based on historical performance, are not equal to one another. The skills also generally decrease as the lead time increases. Thirdly, forecasts made at some times of the year generally have higher skill than forecasts made at other times of the year--namely, they are better when made between June and October than when they are made between January and April. Differences among the forecasts of the models reflect both differences in model design, and actual uncertainty in the forecast of the possible future SST scenario.

The set of dynamical and statistical model forecasts issued during late June and early July shows a range of possible sea surface temperature conditions for the coming 3 to 8 months (August-September-October through March-April-May 2003). Most models are indicating a continuation of warm conditions. The remaining models call for just slightly above average--but neutral conditions. Many of the models forecast warming sufficient to be called an El Nino (e.g., warming to 0.6 degrees C or more above average in the Nino 3.4 region for the September-October-November seasonal average). A smaller proportion are also forecasting ENSO conditions in the upper half of the neutral category-between 0.0 and 0.6 degrees C above normal. The warmest forecast for the August-September-October period comes from the statistical CLIPER (multiple regression) model of Colorado State University and AOML, U.S. (1.45 degrees C above normal), and the coldest one is from the NOAA CDC's statistical Linear Inverse Model, calling for SST anomalies of 0.1 degrees C. The SST anomalies forecast by the models for August-September-October tend to be forecast also for later in the year, such as December-January-February 2003. For this later time, 9 of the 13 models that forecast to that long a lead time suggest El Nino development: the NCEP coupled, JMA, Scripps, LDEO, BMRC, CPC Markov, CPC-CCA, Colorado State CLIPER, and the UBC nonlinear CCA. The Korean SNU model and the CPC Constructed Analogue model are near the borderline (between 0.5 and 0.6 degrees C), and the other two models (NASA/NSIPP and CDC Linear Inverse Model) predict slightly above average SST that falls short of El Nino levels.

FORECAST SST ANOMALIES (deg C) IN NINO 3.4 REGION

Model ASO SON OND NDJ DJF JFM FMA MAM

==Dynamical models==

NASA/NSIPP model 0.2 0.2 0.1 0.1 0.1 0.3 0.3 0.4

NCEP Coupled model 0.6 0.6 0.6 0.8 0.8 0.8 0.8

Japanese Met. Agcy. model 0.9 1.1 1.2 1.1 0.9

Scripps Inst. model 1.0 1.1 1.1 1.2 1.3 1.1 0.9 0.7

Lamont-Doherty model 0.8 0.9 1.0 1.2 1.3 1.3

BMRC intermed. model 1.0 1.0 1.0 1.0 0.9 0.8 0.8

CSIRO model 0.3 0.2 0.1 0.1

ECMWF model 1.0 0.9 1.0

KMA SNU (Korea) model 0.5 0.5 0.6 0.5 0.5 0.5 0.5 0.4

Average, dynam. models 0.7 0.7 0.8 0.7 0.9 0.8 0.7

==Statistical models==

NCEP/CPC Markov model 0.8 0.9 1.0 1.1 1.1 1.1 1.0 1.0

NOAA/CDC Linear Inverse 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.2

Dool Constructed Analog 0.4 0.4 0.5 0.5 0.5 0.5 0.4 0.3

NCEP/CPC Can Cor Anal 0.9 1.1 1.3 1.4 1.4 1.3 1.1 1.0

Landsea/Knaff CLIPER 1.1 1.5 1.8 1.6 1.4 1.2 1.0 0.8

Univ. BC nonlinear Can Cor 1.2 1.3 1.3 1.4 1.3 1.2 1.2 1.1

Average, statistical models 0.7 0.9 1.0 1.0 1.0 0.9 0.8 0.7

AVERAGE, ALL MODELS 0.7 0.8 0.9 0.9 0.9 0.9 0.7 0.7

Some notes about the formulation of the entries in the table above:

=>Only models producing forecasts on a monthly basis are included. This means that some models whose forecasts appear in the Experimental Long-Lead Forecast Bulletin (produced by COLA) do not appear in the table.

=>The SST anomaly forecasts are for the 3-month periods shown, and are for the Nino 3.4 region (120-170W, 5N-5S). Often, the anomalies are provided directly in a graph or a table by the respective forecasting centers for the Nino 3.4 region. In some cases, however, they are given for 1-month periods, for 3-month periods that skip some of the periods in the above table, and/or only for a region (or regions) other than Nino 3.4. In these cases, the following means are used to obtain the needed anomalies for the table:

o temporal averaging,

o linear temporal interpolation,

o visual averaging of values on a contoured map, and

o regional SST anomaly adjustment using the climatological variances of one region versus that of another.

As an example of the last case, suppose only the Nino 3 anomaly is provided. The Nino 3.4 anomaly is then obtained by decreasing the Nino 3 anomaly by the factor defined by the ratio of the year-to-year variance of Nino 3.4 to the year-to-year variance of Nino 3 SST, for the 3-month season in question.

The anomalies shown are those with respect to the base period used to define the normals, which vary among the groups producing model forecasts. They have not been adjusted to anomalies with respect to a common base period. Discrepancies among the climatological SST resulting from differing base periods may be as high as a quarter of a degree C in the worst cases. Forecasters are encouraged to use the standard 1971-2000 period as the base period, or a period not very different from it.

To top | Back | Technical Discussion | ENSO Update