As this is the final Newsletter of the HIRLAM-4 Project, I should like to take the opportunity to briefly review progress over the past three years. There have been very substantial developments in all major areas of endeavour. The 3D Variational Assimilation is now at an advanced stage of development. It should be ready for operational implementation around mid-2000. Indeed, several centres are planning early implementation of 3DVAR. A feasibility demonstration for 4DVAR is planned before the end of this year. The development of this system will be a major focus of the new Project.
Substantial model developments have taken place over the course of the Project. The straco cloud system has proved very beneficial. Other highlights include the new CBR turbulence scheme and Digital Filtering Initialization, both of which have been shown to lead to significant reductions in model error. Substantial improvements have been made to the semi-Lagrangian advection scheme. The priorities for the remaining weeks of the Project are:
This Newsletter contains reports of several HIRLAM meetings which took place over the past few months. The Management Group Visit to DMI is summarised (page 3). The cover illustration, chosen in harmony with this `Danish' opening, has some curious meteorological evocations (see page 82). Satellite meetings in Noordwijk an zee are presented next. The ECMWF Optional Project Boundary Conditions for Limited Area Modelling will start next Spring. The proposal for this project presented to the ECMWF Council in June is reproduced on Page 9, and contains information about the additional model runs for boundary conditions. We should see significant operational benefits from the enhanced boundary data as soon as we start to use them. On page 12 there is a status report on verification. We may note that HIRLAM has recently proposed that it act as the SRNWP Lead Centre for verification, with KNMI as the responsible organisation, and the SRNWP coordinator, Jean Quiby, has warmly welcomed this proposal.
There is a variety of scientific articles relating to HIRLAM. Per Undén describes several significant modifications to the OI analysis system. Simo Järvenoja presents the results of some pre-operational tests at FMI. Bent Hansen Sass discusses the implicit treatment of turbulent kinetic energy dissipation in the CBR turbulence scheme. Björn Bringfelt describes the land surface treatment used for the second regional climate simulation at the Rossby Centre. Ray McGrath and I consider the impact of the lateral boundaries on the digital filter initialization scheme in HIRLAM. Toon Moene has an article describing HIRLAM on Linux systems. Finally, Clive Wilson writes about Unified Model experience with higher resolution (this article follows from the Norrköping Workshop).
In an article starting on page 76, Jim Hamilton and Tom Keane summarise annual verification scores of the operational implementation of HIRLAM at Met Éireann in Dublin, and show how there has been a clear trend towards better forecasts over the period of the HIRLAM-4 Project. At the end of this Newsletter I have written a short note of thanks to you all, and a wish for the ongoing success of the HIRLAM Co-operation.
Forthcoming Meetings: