Title: High-resolution time lagged ensembles: walking the resolution-
predictability tightrope

Marion Mittermaier, Mesoscale Model Development & Diagnostics, UK Met
Office 

Abstract

High-resolution precipitation forecasts can vary considerably from run
to run, especially for convective events where the model flirts with
predictability limits that kick in rather smartly at the smaller scales.
Excessive inconsistencies are undesirable, especially for the
forecaster, who seeks similarities between successive runs to gain
confidence in model guidance. 

The Met Office 4-km Unified Model (UK4) runs every 6 hours to t+36h.
Five member time-lag ensembles are created for the most recent 6-h
period to investigate run-to-run inconsistencies. Ensemble products are
compared with a verifying radar accumulation. Each of the individual
forecasts (ensemble members) are also evaluated and their contribution
to the ensemble assessed. Objective Probability of Precipitation (PoP)
charts for FLASH warning criteria (15 mm/3h) are also produced and
assessed.

Results show that the high-resolution time-lag ensemble provides a more
continuous integrated view of model output which also performs better. 
The ensemble approach ensures that the missed events (which are
potentially more serious for high-impact weather) are reduced. 

More recently the methodology has been expanded to create a multi-
resolution "super" lagged ensembles, combining operational 12 km and UK4
forecasts to extend the forecast lead time to 30h. A number of case
studies will be presented to show the diagnostic capabilities that the
technique provides.