Abstract 247

Abstract ID: 247

When on-ground decision making shapes S2S prediction research pathways

Lead Author: Neil Hart
University of Oxford, United Kingdom

Keywords: rainfall onset, convective-scale, Africa

Abstract: Remarkable advances in S2S prediction science have been made by the extensive efforts of international modelling centres to produce always improving ensemble prediction systems. However, taking these advances forward into applications is rarely straightforward and, even with the best efforts, there often remains a mismatch between what a decision maker requires and what prediction product can be provided. A growing body of literature is describing the challenges and successes of co-producing forecast products with decision makers in order to reduce these mismatches. Typically, these co-production efforts are part of developing applications for an already-achieved prediction advance. It is less common for on-ground decision requirements to directly inform more basic research and drive the prediction science advances that a given sector requires. First Rains is a new rainfall onset project trialing a decision-first approach to set basic science research pathways. The project includes development of kilometre-scale S2S model ensembles and new onset metrics for southern Africa. This contribution will discuss this decision-first approach in the context of a prediction science project, with examples of early challenges and successes.