Data Assimilation Meetings at Reading
| Date | Meeting type | Speakers |
| 15 August 2012 | DARC seminar |
Peter Lean (UoR) Quantifying errors in Atmospheric Motion Vector wind retrievals in a perfect model framework Atmospheric Motion Vectors (AMVs) are wind retrievals derived by tracking coherent features (e.g. clouds or clear sky water vapour structures) in consecutive satellite images. The assimilation of AMVs into numerical weather prediction (NWP) models has provided benefits to operational forecasts for many years. However, the observation operators used in the assimilation process are still relatively crude and potential remains for improvement. While sonde wind observations are routinely used to monitor the accuracy of AMVs, the sparse coverage of the sonde network in both space and time means that only a small fraction of the AMVs can be verified. Simulation studies, whereby AMVs are derived from model cloud fields, provide a useful framework to characterise and understand the sources of errors in AMV wind estimates. The latest generation of high resolution NWP models with grid lengths of order 1km are able to resolve most convective cloud features explicitly. These models have been shown to realistically simulate the life cycle and motion of many cloud features. In this talk, I will present the latest results from a simulation study where the Met Office 'UKV' mesoscale model is used to generate synthetic AMVs. Comparisons will be presented between the synthetic AMVs and the 'true' model winds. Further preliminary results will be presented of an experiment to understand the extent to which sources and sinks of clouds degrade the quality of AMVs. Finally, there will be a discussion of how this work will lead to improved quality control and ultimately an improved observation operator for AMVs. |