Abstract/Summary

Ensemble forecasting is widely recognised as a vital method to improve the skill and utility of weather forecasts. This research aims to unlock the potential of dynamic ensemble forecasts to improve operational tropical cyclone (TC) forecasting and early flood warnings. Firstly, a survey of operational TC forecast centres reveals the huge potential to improve the pull-through of ensemble forecasts in operational TC forecasts and warnings, identifying hurdles that need to be overcome to allow their full utilisation. A global evaluation of multi-model ensemble TC track probability forecasts highlights the benefit of using ensembles over a deterministic consensus forecast and demonstrates the added value gained from using multi-model ensembles. Focus then shifts to the potential use of ensembles in TC fluvial flood forecasting via the Global Flood Awareness System (GloFAS). The severity of flooding in 280 TC cases in terms of flood area, duration and magnitude is calculated using GloFAS-ERA5 reanalyses, and the key factors influencing this severity are assessed. Slow-moving, large, and intense cyclones, affecting areas with wet antecedent conditions, have the highest likelihood of causing widespread flooding, although flooding can be severe even in weaker storms. Finally, novel ensemble-based methods are developed to analyse predictability links along the GloFAS TC flood forecasting chain, demonstrated using the case of Hurricane Iota. This provides an in-depth look at the ability of GloFAS to forecast river flooding in a TC case, and reveals how the strength of the predictability links between TC track, intensity, rainfall and river discharge varies depending on catchment location. As numerical weather prediction (NWP) centres move towards purely ensemble-based forecast systems, the importance of unlocking their potential in TC forecasts and warnings grows. This project leaves a legacy of increased focus on this important topic within international forecasting and research communities via its links to World Meteorological Organisation (WMO) policy and wider collaborative projects.