A Data Set for Intercomparing the Transient Behavior of Dynamical Model‐Based Subseasonal to Decadal Climate Predictions
Climate predictions using coupled models for different time scales, from subseasonal to decadal, are affected by initial shocks, drifts, and biases, which reduce the prediction skill. We aim to provide a multi-model data set and further insight into the mechanisms responsible for initial shocks, drifts, and biases
We present a novel data set developed within the Long Range Forecast Transient Intercomparison Project, LRFTIP. This data set has been constructed by averaging hindcasts over available prediction years and ensemble members from multiple models to form a hindcast climatology, which is a function of spatial variables and lead time, and thus results in a useful tool for characterizing and assessing the evolution of errors as well as the physical mechanisms responsible for them
Initialized predictions show zonal propagation of model errors in the tropical Pacific from the coasts of South America westward in the first three years following initialization. After the third year, the differences in most models largely subside, suggesting a role for anomalous thermocline displacements resembling equatorial upwelling Kelvin waves, which propagate eastward across the Pacific, for producing model drift. Understanding the sources of these differences improves our understanding of such model errors in initialized predictions