The recent All-Staff Meeting in KNMI, De Bilt, reviewed on page 3, was the largest of , with 42 participants. It was regarded by many as the most successful so far. The cover illustration, chosen in celebration of this, is not the new anemometer at KNMI, but a beautiful windmill in Dordrecht, drawn by Ad Bakker.
The visited in early February to listen to presentations of work in Dublin and to discuss plans with the staff there. A report of this visit appears on page 9. The development of the variational assimilation systems continues to progress well. A report of a recent meeting, by , can be found on page 12. In view of the importance of verification, and the urgent need to develop more refined and comprehensive methods for validation and verification of models, a workshop on verification was held in De Bilt in April. A report by appears on page 15. Ben invites comments, suggestions and criticisms, and I urge all of you involved in this area to consider the report carefully and respond to him.
This is the `operational' edition of the Newsletter, and accounts of the operational implementations of the system in the member countries are presented, together with details of verification results in each centre. There follows a selection of scientific contributions on various aspects of the system, starting with an overview by of the new parallel code for the forecast model.
In the final section there are several articles dealing with verification. First, presents some thought-provoking results on the factors contributing to differences between and ECMWF forecast quality. This is followed by a comparison of verification results undertaken at . This article illustrates the pitfalls of verifying forecasts near the domain boundary, and reinforces the importance of acquiring more timely boundary data for the operational forecasts.
I have been invited to deliver a keynote address at the European Conference on Applications of Meteorology (ECAM 99) in Norrköping in September, entitled `State-of-the-art of Short Range Forecasting. Activities'. An abstract of this talk, prepared for the conference proceedings, is reproduced on page 128 below.
There have been major developments to the Reference System over recent months. A `millennium-proof' version of the system (H4.3) was introduced late last year. The MPP (massively parallel) version of the model (H4.4) was introduced in April. Several other important changes were made and an outline table of modifications appears on page 132. As usual, fuller details are available on .
Preparations are under way for the Workshop on High-Resolution Integrations and Non-hydrostatic Modelling. This is being organised by and , and takes place in mid-May in Norrköping. The proceedings of the Madrid Workshop on Physical Parameterisation have now been published. Technical Reports 39 and 40 have also appeared since the last Newsletter. Two further Technical Reports are at an advanced stage of preparation and sould be distributed soon.