Reforecasting the ENSO Events in the Past Fifty-Seven Years (1958-2014)

A set of ensemble seasonal reforecasts for 1958-2014 is conducted using the National Centers for Environmental Prediction (NCEP) Climate Forecast System, version 2 (CFSv2), initialized with observation-based ocean, atmosphere, land and sea ice reanalyses, including the Eu­ropean Centre for Medium-Range Weather Forecasts (ECMWF) global ocean reanalysis version 4, the ERA-40 atmospheric reanalysis, the NCEP CFS Reanalysis for atmosphere, land and sea ice, and the NASA Global Land Data Assimilation System reanalysis version 2.0 for land. The purpose is to examine a long and continuous seasonal reforecast dataset from a modern seasonal forecast system to be used by the research community. In comparison with other current reforecasts, this dataset allows us to evaluate the degree to which El Niño and Southern Oscillation (ENSO) events can be predicted, using a larger sample of events. Furthermore, we can directly compare the predictability of the ENSO events in 1960s-70s with the more widely studied ENSO events occurring since the 1980s to examine the state-of-the-art seasonal forecast system's capability at different phases of global climate change and multidecadal variability. A major concern is whether the seasonal reforecasts before 1979 have useful skill when there were fewer ocean observations. Our preliminary examination of the reforecasts shows that, although the reforecasts have lower skill in predicting the SST anomalies in the North Pacific and North Atlantic before 1979, the prediction skill of the ENSO onset and development for 1958-1978 is comparable to that for 1979-2014. The skill of the earlier predictions declines faster in the ENSO decaying phase because the reforecasts initialized after the summer season persistently predict lingering wind and SST anomalies in the eastern equatorial Pacific during the decaying phase of several major ENSO events in the 1960s-70s. Since the 1980s, the reforecasts initialized in fall overestimate the peak SST anomalies in strong El Niño events. Both facts imply that the model air-sea feedback is overly active in the eastern Pacific before ENSO termination, likely induced by the model warm bias in the eastern Pacific during boreal winter and spring.