Multi-diagnostic multi-model ensemble forecasts of aviation turbulence

Abstract

Turbulence is one of the major weather hazards to aviation. Studies have shown that clear-air turbulence may well occur more frequently with future climate change. Currently the two World Area Forecast Centres use deterministic models to generate forecasts of turbulence. It has been shown that the use of multi-model ensembles can lead to more skilful turbulence forecasts. It has also been shown that the combination of turbulence diagnostics can also produce more skilful forecasts using deterministic models. This study puts the two approaches together to expand the range of diagnostics to include predictors of both convective and mountain wave turbulence, in addition to clear-air turbulence, using two ensemble model systems. Results from a 12 month global trial from September 2016 to August 2017 show the increased skill and economic value of including a wider range of diagnostics in a multi-diagnostic multi-model ensemble.