«Medium-Range Weather Prediction Austin Woods Medium-Range Weather Prediction The European Approach The story of the European Centre for Medium-Range ...»
Through today’s eyes, the reference to influencing weather may seem a bit strange. In the mid-1960s, however, meteorologists were hopeful that soon rain and snowfall could be encouraged or reduced by artificial means, hail made less harmful, fog dissipated, hurricanes steered away from populated areas and more. Early optimism has since given way to cold realism. It remains true that “you can’t fool with mother nature”; but at the time, there was no indication that weather modification would by-and-large wither on the vine, while application of computers would become widespread in almost all areas of meteorological science.
The Working Group on Policy in the Field of Scientific and Technical Research was asked to present a report to the meeting of the Council of Ministers in October 1967. This contained a great many suggestions about The beginnings — the political background 21
areas of science and technology where there could be useful European cooperation. In June 1967, seven areas were emphasised:
• documentation research,
• language translation,
• computer installations for scientific purposes,
• materials research,
• annoyance caused by noise, and
• refuse disposal.
The Working Group decided to concentrate on the most important areas,
and in July it gave its opinion that, for the time being, four areas were worthy of promotion:
• information processing,
• traffic and telecommunications,
• oceanography, and
In October 1967, the Council of Ministers recognised that political cooperation of the six Members of the EEC had come to something of a deadlock. They adopted a resolution at a meeting in Luxembourg, which asserted that the Member States of the European Economic Community — Belgium, France, Germany, Italy, Luxembourg and the Netherlands — were willing to extend their co-operation in fields outside economics, and specifically to implement an energetic programme to promote scientific and
technical research. The Council was of the opinion:
that progress in scientific and technical matters was a fundamental factor affecting the economic growth and general development of the Member States of the Communities and in particular their competitive capability;
and that the achievements of European countries in the field of scientific and technical matters and their industrial application had not been as rapid during the previous few years as those recorded outside Europe in a certain number of branches essential to the development of modern industrial economies, and that Europe is far behind in this field creates a serious risk to its medium and long term economic and social development.
At its session on 31 October 1967, the Council of Ministers agreed to the proposal of the Working Group on Policy in the Field of Scientific and Technical Research, with minor modifications. The Council required the
Working Group to examine the opportunities for co-operation in six fields:
22 Chapter 2
• information science and telecommunications,
• development of new means of transport,
• nuisances, and
Expert Groups for each of these areas were set up. The Council requested Reports before 1 March 1968, allowing only four months for their preparation, so that it could submit conclusions before 1 June 1968. It required that the Reports should “take into consideration the co-operation existing at the present time in other international organizations, and should seek means to enable other European States to participate in such projects”.
This was particularly relevant for meteorologists, who were already well accustomed to working internationally.
The meteorologists of the Member States of the European Communities were thus presented with a unique opportunity: to study, by official order, the fields in which joint actions were possible.
It was a frustrating fact in the world of European meteorology that meteorologists from Western Europe wishing to work with other European meteorologists found it easiest to do this by going to the United States, and in some cases to the Soviet Union. The USA had a number of university departments in the field with lecturers, researchers and professors from several European countries.
Novosibirsk had a strong school in meteorology, with good expertise in numerical techniques. Guri Marchuk in 1962 had set up a computational centre of the Siberian Department of the USSR Academy of Sciences in Novosibirsk. Extensive research on atmospheric and oceanic physics, along with studies on computing technology and software, were conducted under his guidance. Marchuk later became President of the Academy of Sciences of the USSR, and served as Vice Chairman of the Council of Ministers of the Soviet Union. Lev Gandin, author of more than 200 journal articles and 14 books, was there as well. Several scientists from Météorologie Nationale, France, spent periods in Novosibirsk.
Western Europe lacked co-operative opportunities in other scientific fields as well as in meteorology. Many European scientists of several disciplines had emigrated to take advantage of the better research opportunities elsewhere. It was becoming accepted that individual states would find it difficult to resolve the problem; a common initiative was required.
Work started immediately.
Meteorological developments 1967 to 1971
A visionary concept in 1967 became ECMWF four years later.
November 1967: Longer range weather forecasting and research using a very large European computer installation April 1969: European Meteorological Computer Centre for Research and Operations May 1970: European Meteorological Computing Centre (EMCC) August 1971: European Centre for Medium-Term Weather Forecasting (ECMW) November 1971: European Centre for Medium-Range Weather Forecasts (ECMWF) It is worth looking at some of the detail of this evolution.
In November 1967, the important “Expert Group for Meteorology” was established under the Chairmanship of Dr E. Süssenberger, who had been President of the German Weather Service, Deutscher Wetterdienst (DWD), since August 1966. Within a short time, Prof E. Lingelbach, who would become President of DWD in 1977, visited the National Meteorological Services of the six countries in the EEC to elicit opinions on European cooperation in the field of weather forecasting. A physicist Mr van der Kolk from the European Communities accompanied him.
Most countries agreed that co-operation should extend beyond the
six nations of the European Communities. Many topics met with general approval:
• meteorological measurements by satellites and by EOLE constantvolume pressurised balloons,
• turbulence, 23
24 Chapter 3
• experts’ meetings,
• tropical meteorology,
• general circulation,
• influencing weather on the small scale (for example dissipation of fog),
• synoptic meteorology,
• dynamical climatology in the Mediterranean area,
• agricultural meteorology,
• infra-red measurements,
• longer range weather forecasting and research using a very large European computer installation,
• atmospheric optics,
• international research programmes,
• visual range measuring instruments,
• instruments for measuring cloud altitude,
• European manufacture of balloons,
• wind shear,
• air pollution,
• three-dimensional wind measurements,
• research stations in the Antarctic,
• exchange of research results, and
• documentation and ozone measurements.
In this rather long list we can find the first mention of what became the European Centre for Medium-Range Weather Forecasts.
The Expert Group for Meteorology first met on 4 January 1968 in Brussels. Of the six countries of the European Communities, only Luxembourg was not represented. Detailed discussions were held on the projects. Three sub-committees were established to define the individual projects more precisely and to assess their merits.
• Sub-committee I: Structural problems, secretariat, and integration into the EC, general questions of standards, standardisation and industrial questions, basic questions of common programmes.
• Sub-committee II: Development of new instruments, standardisation of existing instruments, new measuring procedures, documentation.
• Sub-committee III: Scientific programmes in general, homogenising ozone research, seminars, floating balloons, buoy systems, computer centre, satellites in meteorological research and operations.
Meteorological developments 1967 to 1971 25 The members of Sub-committee III, responsible inter alia for the “computer centre”, were Prof Rosini (Italy), Prof Schmidt (the Netherlands), Prof Lebeau (France), Dr van Isacker (Belgium), Dr Lingelbach (Germany) and Mr van der Kolk (European Communities).
One important question considered by Sub-committee III was whether it was justified at that time to invest considerable sums of money in a centre for Numerical Weather Prediction (NWP) using a powerful computer — was the science sufficiently developed to consider such a project?
In April 1967, Dr Süssenberger had informed the WMO Congress that in
weather prediction for the general public, aviation and shipping, posed a crucial problem. A frontier had been reached which could not be crossed by conventional methods; beyond a relatively short-range prediction period of 36-48 hours, the accuracy of forecasts left much to be desired.
With the benefit of hindsight, one can see that the Centre developed and grew during the 1970s and early 1980s in intimate association with the vast scientific and technical work of the Global Atmospheric Research Programme (GARP).
Following the establishment of the World Weather Watch in 1963, GARP was perhaps the most ambitious scientific undertaking in the history of meteorology, indeed perhaps in the entire field of geophysical science.
GARP aimed to reveal nothing less than the details of the dynamics of the atmosphere of the planet. Launched in 1967 by WMO, with the collaboration of the International Council for Science (ICSU), GARP lasted 15 years.
Its field experiments led to dramatic progress in weather forecasting. One of these, the GARP Atlantic Tropical Experiment (GATE), which took place from June to September 1974, was unprecedented in its scale and success.
Some 70 countries participated in GATE. A huge observational system, including over 40 ocean research vessels and a number of meteorological aircraft, as well as balloons and meteorological satellites, was deployed. The unique results were fundamental to our understanding of the large-scale weather systems of the tropics. Some thirty years later, the Centre as part of its Re-Analysis Project would use the unique and valuable collection of GATE observations again to prepare analyses of the global atmosphere at that time — we will return to this in Chapter 13.
The crowning achievement of GARP was undoubtedly the First GARP Global Experiment (FGGE), planned first for 1977, then 1978, and brought to fruition in 1979. [Since there was no “SGGE”, the alternative official name “Global Weather Experiment” should perhaps be used. However, the
26 Chapter 3expression “Figgy” has been embedded in the hearts of many at the Centre over the years, so we continue with FGGE here.] The National Meteorological Services of 170 nations, as well as space agencies and research institutes, participated in FGGE. It is due to the work of FGGE that we have the vast observational network of the World Weather Watch that constantly measures and probes the atmosphere, the sea and the land today. FGGE laid the foundation of the global system of geostationary and polar orbiting satellites, which now form the space-based observing system of the World Weather Watch. New methods of analysis in operational weather forecasting were developed — in fact from necessity. Major NWP centres around the world found, somewhat to their dismay, that their systems then in use were quite unable to produce good analyses of the tropical atmosphere! Major improvements were made in the forecasting models.
Forecasts improved steadily during the years from 1980, as a result of improvements in the global observing system, more powerful computers power, and advances in the science: in the Centre’s data assimilation system and forecast model. Seven-day forecasts in the Northern Hemisphere became more accurate than five-day forecasts of 1980, and five-day forecast accuracy reached that of the three-day forecasts made 25 years earlier.
In the Southern Hemisphere, the improvement was even more marked. In the early 1980s Southern Hemisphere three and five day predictions were not much better than those of the Northern Hemisphere for five and seven days respectively. Two decades later, forecasts for both Hemispheres were of similar accuracy - a gain of about four days in the accuracy of Southern Hemisphere predictions.
The shaded area shows the differences in forecast accuracy between the Hemispheres. Score: Anomaly correlation, 500 hPa height. See Simmons AJ and Hollingsworth A (2002) Some aspects of the improvement in skill of numerical weather prediction. Quart J Roy Meteor Soc 128: 647–678.
Meteorological developments 1967 to 1971 27 Other GARP field experiments included the Alpine Experiment (ALPEX) in 1982, which led to greater understanding of cyclogenesis and the mechanisms driving local mountain winds, and the Monsoon Experiments of 1978–1979, which improved forecasting of regional monsoon circulation.
Such historic experiments have contributed to the remarkable headway that has been made in moving the time-scale of skilful weather forecasts in midlatitudes using NWP from two or three days ahead, the best achievable in the 1960s, to seven to ten days ahead today.