COST is the abbreviation for 'European co-operation in the field of scientific and technical research'. The COST framework is considered appropriate for technical and scientific co-operation in research projects and it is also open for European countries not belonging to the European Communities.
The COST-76 action entitled 'Development of VHF/UHF wind-profiler radars and other vertical sounders for use in European observing systems' was created in early 1994 in order to encourage the development and implementation of wind profiler networks in Europe. COST-76 is an follow-up project of COST-74 'Utilization of UHF/VHF radar wind profiler networks for improving weather forecasting in Europe', which started in 1987 to co-ordinate the efforts of researchers, potential users and manufacturers interested in wind profiler radars. For this kind of European actions the COST framework is appropriate in the pre-competitive stage, in other words before equipment is readily available from manufacturers.
Scientists from 13 countries participate in COST-76. Also many representatives from industry attend the meetings regularly.
During the last 20 years, so-called Doppler radars were systematically developed to probe the atmosphere and derive the wind profile (i.e. the speed and direction as function of height) from echoes of the transmitted radio waves produced by turbulence in the clear air. A wind profiler is the operational application of a radar originally developed by scientists for measuring the echo intensity and the wind profile up to about 30 km with height resolutions from 100 to 1,500 m. Sequences of high power pulses are radiated in the vertical and in oblique directions. By analysing the received echoes, the radial velocity and the turbulence intensity can be computed. Observations from at least three directions are necessary to determine direction and speed of the wind. A related development, the Radio Acoustic Sounding System (RASS) provides profiles of temperature. This is achieved by transmitting a strong but short acoustic 'beep' vertically upwards. This tone burst travels as a compression wave with the speed of sound upwards in the atmosphere. The wind-profiling radar is now used to measure the speed of propagation of this sound burst, which also produces an echo of the radar signal. Since the speed of sound depends on the air temperature, this latter value can then be computed.
The maximum reachable height of a wind profiler radar depends, among other parameters, on the operating frequency. To monitor atmospheric processes up to 30 km, 16 km and 5 km, wind profiler systems with operating frequencies at about 50 MHz (above the shortwave radio range), 400 MHz (in the television range) and 1,000 MHz (above the cellular telephone range) are used.
What are the advantages of a wind profiler?
At present, meteorological organisations use balloon-borne systems to measure profiles of wind, temperature and humidity from the ground to high up in the atmosphere. While current wind profiler radars do not operationally measure all these parameters, they do have several advantages in comparison to the balloon based systems.
Since wind-profiler radars can be adapted to measure temperature profiles up to about 5 km when they are used in conjunction with a Radio-Acoustic Sounding System (RASS), the possibility to obtain temperatures profiles much more frequently than when using balloon tracking. No other measurement technique will present comparable advantages within the foreseeable future, including satellite borne sensors.
What are the uses of wind-profiler data?
Wind-profiler radar provide wind measurements and turbulence information as a function of altitude in most weather conditions. The region of observations of these radars ranges from the ground up to 30 km, the altitude resolution is from 30 m up to 1,500 m, and the time delay between profiles from about ten minutes to an hour. The existence of several types of wind profilers allows the user to choose the instrument best suited to his needs. The present development of weather forecasting requires frequent, closely spaced, and high quality wind data with improved accuracy from near the Earth's surface to high in the atmosphere.
Wind data based principally on balloon-borne instruments, satellite measurements, and automated aircraft reporting systems are insufficient to satisfy the needs of the increasingly high-resolution atmospheric computer models as well as those on man-machine interactive forecasting systems. Without substantial increases in high-resolution wind data, the capacity of these new models and interactive systems being deployed later in this decade to improve weather forecasting and severe weather warnings, will be greatly limited.
Due to their capabilities, the development of wind profilers is increasing, and many networks are under study, construction or in operation for research purposes and meteorological prediction, both in the United States and in Europe. Wind profiler radar can be used for several applications.
What are the most important problems?
Because of the physical principle used, wind-profiler radars require a large amount of bandwidth in the electromagnetic spectrum already crowded by television and radiostations as well as by two-way radio and wireless telephones. Further, the choice of the operating frequency is not entirely free, because the processes producing the echoes occur only at certain frequencies. From the very beginning of the activities in COST concerning wind-profiler radar, one of the main activities was related to obtaining frequency allocations. Recognising the absolute necessity for such allocations on one hand, and the difficulties in accommodating profiler operation in the already crowded frequency bands on the other, an activity was initiated which consisted in using all available administrative channels for placing a request.
In the field of frequency allocations, the work was conducted in close relation with the World Meteorological Organization, the International Telecommunication Union, and the numerous national and international telecommunications authorities. This process is complex and lengthy, and, although considerable progress was made, frequency allocations on a European basis may not be expected before the World Radicommunication Conference 1997 (WRC-97).
Many potential users are now waiting for a decision about frequency allocation before ordering wind profiler radars. Many manufacturers also wait for this decisions in order to develop instruments adapted to the market. It is hoped that after a positive decision at the WRC-97, national weather services will be able to accelerate work on the practical implementation of the European Windprofiler Network.
What other observing systems can complement wind profilers?
The restriction of wind-profiler radars to measure wind only (and not, as in the case with conventional radiosondes, to include other atmospheric characteristics like temperature, humidity and pressure) caused some hesitation in applying wind-profiler radars in operational services. Future work will be aimed at integrating other ground-based remote sensing techniques with wind profilers. In the last few years, new equipment like Differential Absorption Lidar, Microwave Radiometer, GPS-Receiver, Fourier Transform Infrared Radiometer, and Optic Spectrometers have been developed to monitor, e.g. the structure of the atmospheric water vapour field.
Radio Acoustic Sounding Systems (RASS) are remote-sensing systems for the measurement of the temperature profile in the (lower) atmosphere. RASS is deployed routinely in experiments and at monitoring sites as a simple addition to a wind profiler. The principle of operation is to track the propagation speed of the transmitted acoustic pulses with a Doppler wind-profiling radar. The acoustic wave speed depends on the virtual temperature*. Therefore, measuring the acoustic wave speed with the wind profiler yields in a straightforward way the virtual air temperature Tv. The conversion is written as
Tv = (M/yR) Vac2
where Vac is the acoustic speed, M is the molecular weight, y is the ratio of specific heat and R is the universal gas constant.
The most important source of error in the measurement is caused by (vertical) air motion. The measured acoustic speed Vrass is namely the true acoustic speed Vac added to the background vertical air motion w, i.e. Vrass = (Vac + w). To compensate for this error most RASS systems measure the acoustic speed and the vertical air speed w simultaneously. However, whether this correction is applied in real time depends on the data processing of each specific system. For the RASS systems operated in the CWINDE97 project the correction is applied in real time to the data from Lindenberg, but not applied to the data from Cabauw.
In addition some smaller error sources exists, but their magnitude is in generally small. These error sources are described in 'Errors in Radio Acoustic Sounding of Temperature', W.M. Angevine and W.L. Ecklund, 1994, Journal of Atmospheric and Oceanic Technology, Vol. 11, 837-842.
*The virtual temperature is defined as the temperature the air would have if it were completely dry and is at the same pressure.