For long-range predictions (e.g., seasonal), it is a common practice for retrospective forecasts (also referred to as the hindcasts) to accompany real-time predictions. The necessity for the hindcasts stems from the fact that real-time predictions need to be calibrated in an attempt to remove the influence of model biases on the predicted anomalies. A fundamental assumption behind forecast calibration is the long-term stationarity of forecast bias that is derived based on hindcasts.Hindcasts require specification of initial conditions for various components of the prediction system (e.g., ocean, atmosphere) that are generally taken from a long reanalysis. Trends and discontinuities in the reanalysis that are either real or spurious can arise due to several reasons, for example, the changing observing system. If changes in initial conditions were to persist during the forecast, there is a potential for forecast bias to depend over the period it is computed, making calibration even more of a challenging task. In this study such a case is discussed for the recently implemented seasonal prediction system at the National Centers for Environmental Prediction (NCEP), the Climate Forecast System version 2 (CFS.v2).Based on the analysis of the CFS.v2 for 1981-2009, it is demonstrated that the characteristics of the forecast bias for sea surface temperature (SST) in the equatorial Pacific had a dramatic change around 1999. Furthermore, change in the SST forecast bias, and its relationship to changes in the ocean reanalysis from which the ocean initial conditions for hindcasts are taken is described. Implications for seasonal and other longrange predictions are discussed.
The characteristics of El Niño-Southern Oscillation (ENSO) variability have experienced notable changes since the late 1990s, including a breakdown of the zonal mean upper-ocean heat content as a precursor for ENSO. These changes also initiated a debate on the role of thermocline variations on the development of ENSO events since the beginning of the twenty-first century. In this study, the connection between thermocline variations and El Niño and La Niña events is examined separately for the 1980-98 and 1999-2012 periods. The analysis highlights the important role of thermocline variations in modulating ENSO evolutions in both periods. It is found that thermocline variation averaged in the central tropical Pacific, including both equatorial and off-equatorial regions, is a good precursor for ENSO evolutions before and after 1999, while the traditional basinwide mean of equatorial thermocline variation is a good precursor only before 1999. The new precursor, including both high-frequency variability in equatorial regions and low-frequency variability in off-equatorial regions, is found to be indicative of multiyear persistent warm and cold conditions in the tropical Pacific. Further, it is found that the strength of the subtropical cells (STCs) interior mass transport in both hemispheres increased rapidly around the late 1990s. It is proposed that the strengthened STC interior transports provide a pathway for the enhanced influence of off-equatorial thermocline variations on the development of ENSO events after 1999.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.