Large Scale Observational and Model Analysis of the Unusual 2015-2016 El Niño Event: Implications for California Precipitation

Forecast models generally predicted above average winter precipitation for California during the 2015-16 El Niño event, similar to what occurred during the 1997-98 and 1982-83 events. The large scale pattern that actually occurred during the winter of 2015-16 showed differences from the earlier El Niño events, including a weaker enhancement of the Aleutian Low, and an Eastern Pacific jet shifted further to the south. The result was low observed precipitation over California. Comparison of the observations to the ensemble seasonal forecasts made with the Climate Forecast System Version 2 (CFSv2) model for the winters of 2015-16 and 2016-17 showed that the model failed to fully capture the observed structure of the tropical vertically integrated diabatic heating as well as the jet stream and the height field at 200 hPa during the 2015-16 winter. The circulation predicted by the model caused a shift in precipitation comparable to a typical El Niño in the model.

This study investigates the degree to which the model’s failure to predict the dynamic circulation was due to the misrepresentation of model heating by performing intervention experiments. The estimated diabatic heating boreal winter mean and seasonal cycle were estimated from ERA-Interim reanalyses and from the CFSv2 forecasts as a residual in the thermodynamic equation. Subsequently the ensemble mean error in the seasonal cycle of heating was subtracted from the model’s internally generated heating during a new ensemble of integrations. The procedure is iterated until the ensemble mean diabatic heating is realistic. Similar experiments are carried out with the Community Earth System Model (CESM). It was shown that the added heating runs created a canonical ENSO response resulting in more storms tracking through California, different than what is shown in observations. This experiment failed to resolve the unusually low precipitation anomaly and therefore was not a response to the tropical heating.