WCD-02. A Hybrid Prediction (Dynamical/Machine Learning) Approach for Probabilistic Seasonal Outlook for the Famine Early Warning Systems Network

Abstract
With over three decades of continuous development, the Famine Early Warning Systems Network (FEWS NET) provides early warning and analysis of acute food insecurity based on an evidence basis that includes agro-climatic forecasts. To improve agro-climatic forecasts, we developed an experimental probabilistic multi-model ensemble (PMME) for seasonal outlook. This PMME prediction is based on the outputs of initialized dynamical forecasts from Copernicus Climate Change Service (C3S) multi-system predictions. To construct this PMME, we propose the use of the Extreme Learning Machine (ELM), a novel machine learning (ML) approach. ELM is a state-of-the-art generalized form of single-hidden-layer feed-forward neural network. However, since the traditional ELM network only produces a deterministic outcome, we use a modified version of ELM called Probabilistic Output Extreme Learning Machine (PO-ELM). Climate Hazards Group InfraRed Precipitation with Stations (CHIRPS) is used as a reference data set to evaluate the model skill. The skill and interpretability of the proposed method over FEWS NET area will be presented in the poster.