ENSO predictions from the Preliminary UFS Seasonal Forecast System

Recently, United Forecast System (UFS) based coupled Global Ensemble Forecast System (GEFS) has been developed and tested at subseasonal time scales. This new system has been adapted from a fully coupled prototype (version 5; P5) with updated stochastic schemes to present a probabilistic forecast out to 35 days. A 2-year experiment (Oct. 2017 - Sep. 2019) has been made to compare to current operational un-coupled GEFS. An evaluation demonstrates that the UFS based fully coupled GEFS (CGEFS) has a comparable or better forecast skill than current operational GEFS in terms of extrapical extended weather and subseasonal prediction, tropical sub-seasonal prediction, such as MJO. The same system has been carried out, but at reduced horizontal resolution for both atmosphere and ocean, out to 9 months and 41 ensemble members, for seasonal prediction to demonstrate the ENSO prediction capability, systematic model errors and forecast uncertainty of seasonal time-scale.

These experiments were initialized on June 1st of 2012, 2015 and 2017 to simulate the neutral ENSO case in 2012, the strong El Nino event in 2015 and weak La Nina event in 2017. The investigations mainly focus on the capability of seasonal prediction from the CGEFS (or precursor to the initial version of the future operational seasonal forecast system (SFS)) with the following characteristics: 1). Impact of initial perturbations from atmosphere and ocean; 2) Impact of model perturbations (or stochastic perturbations) for atmosphere, land and ocean, and interactions between them; 3). Systematic model errors across three different Nino 3.4 SST anomalies. The signal-to-noise paradox will be used to identify the source of predictability for seasonal prediction. Nino 3.4 anomalies will be presented and discussed along with the evaluation and predictability of MJO and ENSO. Forecast uncertainties after errors are saturated will be quantified for SST and other variables. An investigation of the energy budget over the atmosphere and ocean surface will also be evaluated from different configurations and ensemble size.