Regional NWP: why bother?

In 2012 the Met Office's IBM supercomputer will be upgraded with a ~3x increase in computing power. It is anticipated that this may permit the retirement of all regional (European, S. W. Asia, etc) NWP configurations, with the convective-scale (1.5km stretched grid) UK domain being driven directly from a ~16km global model. The talk will begin with a brief review of the origin of NWP improvements over past decades, as justification for continued investment of some of the enhanced computing resources in further increases to global model resolution and sophistication.

Initial preparations for the 2012 upgrade have included studies to assess the value of regional NWP in current generation models: 25km global, and 12km regional. Results will be presented that indicate the value of current deterministic regional NWP is limited to particular surface and cloud parameters. Given these fields are key components of the Met Office's UK skill index, the current regional deterministic NWP configuration is therefore justified. However, it is clear that at some point in the near future it should be possible to retire regional NWP and refocus those computing/human resources on enhanced global/convective-scale and probabilistic NWP capabilities.

It is anticipated that enhancements to data assimilation capabilities will continue to make a significant contribution to improving NWP skill in future. The talk will end with a brief review of current data assimilation developments at the Met Office and an assessment of their likely impact in near-future operations.