Impact of Hyperspectral Sounders

Trial on Raijin

• u-bq615 - control
• u-bq977 - experiment that excludes all hyperspectral IR sounders

The ​GES output shows,

• A general degradation of forecasts when hyperspectral sounder observations were removed. However, there are some improvements for longer forecast ranges in the Northern Hemisphere and European verification regions
• The degradation in the Southern Hemisphere forecasts when hyperspectral sounder data are removed is marked.

Comparing with OSE's conducted in UKMO

Stuart Newman (SA, UKMO) provided some results of a number of OSE's conducted in UKMO using PS37 baseline global suite. Here's a score card from one such trial where all IR radiances were removed (blue means the denial run was degraded),

Here the verifying truth was the ECMWF analyses and the overall score is -1.28%. However when the trial was verified against observations the overall score is -0.3%. This number is comparable to my trial where all hyperspectral sounder data were removed (overall score of -0.26%). Here is Stuart's email detailing the result of the UKMO trial,