[p2] Observational issues


[p2-42]

Implementing the ensemble forecast sensitivity to observations (EFSO) method with the AFES-LETKF data assimilation system

A. Yamazaki (JAMSTEC), T. Miyoshi (RIKEN), and Nobumasa Komori (JAMSTEC)

 
Abstract

A diagnostic technique called Ensemble Forecast Sensitivity to Observations (EFSO) has been implemented into the AFES-LETKF data assimilation system (ALEDAS) which produces a global ensemble reanalysis called AFES-LETKF experimental ensemble reanalysis. EFSO can quantify how much each observation has improved or degraded the forecast without data denial experiments. The EFSO diagnosis was incorporated into every forecast-analysis cycle of ALEDAS. The impacts of the assimilated observations in ALEDAS on the 6-hour forecasts were estimated. To test the implementation, ALEDAS data assimilation experiments with EFSO were conducted for the period from January to February 2016. Preliminary results of the experiments will be introduced in our presentation.