An Assessment of Wind Impact from Several Observing Systems in the ECMWF Analysis
This presentation characterises the impact on wind analyses arising from several observing systems in ECMWF’s 4D-Var assimilation system. This is done using so-called re-initialisation experiments, in which every assimilation cycle starts from a background obtained with the full observing system, but then assimilates only one selected observation type (e.g., AMVs, scatterometer data, MW or IR radiances, etc). This allows a detailed characterisation of the impact of each observing system on its own, and we will here focus primarily on the influence of each observation type on the wind analysis.
The study shows that in the ECMWF system a wide range of observations contributes to the wind analysis, with strengths of different observing systems in certain areas complementing each other. Radiance observations presently provide overall the largest impact, due to balance effects and implicit tracing of atmospheric structures in 4D-Var. Other observing systems complement this impact, particularly AMVs in the tropics or scatterometer data over ocean surfaces. The complementarity of information provided by Aeolus to the present operational global observing system will also be addressed.