A Prototype Tool to Generate Meteorological Ensembles from Climate Scenarios

Oral Presentation 

Businesses are now planning adaptation to climate change. While output from the Climate Model Intercomparison Projects are a valuable resource in this context, they do not resolve the weather extremes which can damage assets or disrupt operations at the spatial or temporal resolution required. The short memory of the atmosphere makes it impossible to provide deterministic forecasts of future weather, but there is a growing body of evidence that the state of the ocean and land hydrological reservoirs constrains the phase space of extreme weather events. 

If so, were a weather model with high skill up to the atmospheric predictability limits coupled to sufficiently useful models of longer memory components of the Earth system, a useful ensemble of probabilistic predictions of weather extremes from any initial climate scenario could be produced. Such an ensemble would have variety of applications, most immediately to the insurance underwriting process. 

To explore this idea, we have developed a low-resolution prototype of such a modelling system by coupling the atmospheric core of the Model of Processes Across Scales (MPAS-A) to the NEMO-SI3 system for modelling ocean and sea ice components together with the WAVEWATCH III wave prediction model, and the Max Planck Institute hydrological discharge model for simulating river transport. To date this set up has been run at a global ~1/4° resolution, and an uncoupled variable resolution case up to a convection-permitting resolution over parts of Europe has been tested. We will describe the current state of our prototype alongside performance indicators benchmarked against relevant weather and climate model validation criteria. We will also report on the prospects of the prototype for delivering greater computational efficiency.

Speaker/s