Precipitation Biases along the Warm Pool Edge as a Cause of Poor El Niño Forecasts During Boreal Spring

El Niño forecasts in seasonal forecast models struggle with El Niño false alarms and over-confident forecasts under conditions of recent westerly wind bursts or an extended western Pacific warm pool. There are two major contributors to El Niño prediction: ocean subsurface heat content and westerly wind bursts. Westerly wind bursts, which can more generally be described as weather events, play a major role in the spread of the El Niño forecasts. They interact with the upper ocean at the edge of the warm pool to extend the warm pool eastward through momentum, heat, and freshwater fluxes. We hypothesize that forecast models struggle to correctly capture the interaction between weather events and the warm pool edge, which leads to over-confident El Niño forecasts and false alarms. Specifically, we hypothesize that forecast models do not correctly represent the upper ocean response to the strong precipitation associated with weather events at the warm pool edge. We therefore propose to study the role of weather events in producing over-confident El Niño forecasts. We will use forecasts from the North American Multi-Model Ensemble (NMME) and Subseasonal Experiment (SubX) projects to assess the role of weather events in producing El Niño, comparing to ocean reanalysis and satellite data. Details of the weather events will be quantified through mixed-layer heat and salt budgets computed from reanalysis data; comparison to forecast data will enable us to identify the processes that forecast models may be getting wrong. A major focus will be the role of precipitation, which generates upper ocean stratification anomalies that may not be well captured by models. We will assess the impact of both the precipitation fields in the forecast models as well as their representation of upper ocean processes in forecast skill. We will also look at how the ocean response influences the creation of future weather events.