It is exceptional for a meteorologist to be featured on the cover of a news magazine. But Rossby was exceptional in many ways. The Time cover of December 17, 1956 is reproduced on the front of this Newsletter to mark the Management Group visit to Norrköping in August. A report of the visit begins on page 3 and some papers from the Rossby Centre, and also from MISU and SMHI, which were presented at the meeting, appear in this issue.

A very successful HIRLAM training course for meteorologists from Estonia, Latvia and Lithuania was held in Riga in October. A report of the course, by Nils Gustafsson and Ulf Andræ, can be found on page 8. A more comprehensive report, with the full programme, a list of participants and a description of the training results, can be requested. Headmaster Gustafsson and his staff are depicted on the back cover.

Recently, there was an excellent HIRLAM Workshop in Madrid. This workshop was very well supported, with some 38 participants, from the HIRLAM countries, from DWD, from UKMO, from ECMWF and from MIT. Following the scientific presentations, the participants split into three Working Groups - on Clouds and Convection, Turbulence Schemes and Surface parameterisation - to review progress and draw up plans for future work. There is a brief report of the Workshop on page 12. The programme of scientific presentations is also reproduced. The full proceedings of the Workshop, with recommendations of the Working Groups, will be published shortly.

The HIRLAM Advisory Committee met in Madrid on 10th November. The key item on the agenda was the continuation of the HIRLAM cooperation after 1999. I am delighted to say that there is unanimous support for the establishment of a follow-up phase of the project, HIRLAM-V. This positive support clearly demonstrates that our work is appreciated and is of benefit to others. A series of recommendations for the scope and structure of the new project has been formulated by the Committee and will be presented to the HIRLAM Council at their meeting in Reading in December.

Progress in developing a variational assimilation scheme (3DVAR) is on target; a technical report on 3DVAR is nearing completion and should appear shortly. Prospects for a satisfactory version of the lagrangian advection are now good: preliminary tests with the modified version appear to solve the major problems which were reported earlier. An extensive parallel evaluation is now under way in SMHI. Results should be available within about a month. Tests are also planned at the Rossby Centre.

There has been good progress with implementation of the Rasch-Kristjánsson Cloud Scheme. Significant advances in the development of the Mass Flux Convection Scheme with CAPE closure have also been reported. A series of turbulence evaluation tests has been carried out. Both the CBR scheme and the K-e scheme were shown, in separate tests, to be superior to the reference system (H4.2); however, the tests have yet to be completed. Good progress was reported for the ISBA surface scheme, although it is still some distance from readiness for implementation in the reference system.

The consolidated parallel code of the eulerian version of the model is now ready and tests have yielded satisfactory results. The parallelisation of the semi-lagrangian version will be done next. The `millennium-proof' version of HIRLAM is now ready, but must be thoroughly tested. The date for introduction as the reference system (H4.3) is 23 November. Following a request to ECMWF, the surface geopotential is now available in the operational dissemination of the Centre's products. This provides a key part of what is required for local HIRLAM implementations to automatically adjust to changes in the orography at the Centre. The Centre has also agreed to provide a facility for extraction of data from MARS on a rotated latitude/longitude grid, following a request from the Project Leader.

Plans for 1999 are firming up. Next year's All-Staff Meeting will be held in De Bilt on 15-17 March. I look forward to seeing many of you there. A workshop on high-resolution simulations and non-hydrostatic aspects will take place at SMHI, provisionally on 10-12 May. A calendar of recent and forthcoming events appears below.

Finally, the HIRLAM Management Group would like to thank all scientists associated with the project for your work during the year, and to wish you all a