Introduction to the cyclone tracking with the ERS Scatterometer
Figure 1: Normalized radar backscattering and wind field from the Anacelle cyclone (Indian Ocean close to Madagascar) as seen by the ERS-2 Scatterometer.
The image in Figure 1 shows a representation of the Anacelle cyclone which occurred in the Indian Ocean, close to Madagascar from 8th to 14th February 1998 based on measurements obtained by the Wind Scatterometer instrument on-board the ERS-2 satellite operated by the European Space Agency (ESA). During the lifetime of the cyclone, wind speeds greater than 90 Kts were recorded, making high quality measurements problematical and giving rise to considerable inaccuracies in the meteorological forecasting models.
The image shows data from the Scatterometer instrument acquired on 10th February 1998 and coloured red, green or blue according to which of the three antennas within the instrument provided the measurements. Each antenna provides a measure of the sea surface roughness from a particular viewing angle and this can be related to the speed and direction characteristics of the local wind field. In general, the measured sea surface roughness from each antenna are combined into a standard wind field product which provides accurate wind speed and direction measurements over an area of approximately 500 Km by 500 Km with a grid spacing of 25 Km. However, additional processing is also possible such as the analysis of measurements from different viewing angles, as shown here, which enables a greater insight to be built up regarding the dynamics associated with the evolution and measurement of intense cyclonic systems. The standard data products from the scatterometer provide not only an unprecedented volume of homogeneous wind measurements to meteorologists, but also detailed structure on intense structures such as tropical cyclones where it is normally very difficult to obtain measurements.
Due to the position of the satellite swath with respect to the cyclone, the entree structure is not visible. However, considerable levels of information regarding the dynamic and structure of the system can still be derived. In particular the direction of the wind is well represented, even at the high speeds occurring within the central areas.
Figure 2: A cross section of wind speed across the centre of the cyclone system.
As shown in Figure 2, a decrease of approximately 30 Knots can be seen between the centre and the edge of the system. Obtaining such a profile from conventional measurement techniques such as ships and buoys would be prone to significant levels of inaccuracy.
Figure 3: Meteorological warnings and cyclone structure (as sensed by the ERS scatterometer) for Anacelle.
Figure 4: Anacelle position and wind speed structure as sensed by the ERS-2 Scatterometer.
As can be seen in Figures 3 and Figure 4, there are significant differences between the scatterometer derived wind field and the meteorological prediction. In particular, the representation of the wind speed contours and a more accurate location of the centre of the cyclone system, approximately 100 Km from the position predicted by the meteorological forecast (shown as red star).
Since the launch of ERS-1 on 1991, scatterometer derived wind fields, together with wave height measurements from the ERS radar altimeter and wave spectra data, have been continuously provided within 3 hours of measurement time to national and international meteorological and oceanographic organisations via the Global Telecommunications Service of the World Meteorological Organization to improve regional and global meteorological and met-ocean forecasting and nowcasting services. Information on the atmospheric structure derived from the wind field observed by the scatterometer allows relocation of the low pressure regions by as much as several hundred kilometres in some cases, generating considerable impact on the accuracy of short and medium term forecast. Commercial companies are also exploiting these measurements to provide local wind measurements to offshore operators and other customers requiring tailored forecast and nowcast information services.
A first study within the ESRIN/PCS
This set of images was obtained from the ERS-2 Scatterometer Fast Delivery data.
The data were distributed less than three hours after sensing to European and North American meteorological centres for assimilation into their weather forecast model.
The three colour components of the images (R, G, B) have been modulated with the signals of the three Scatterometer antennas after a spatial re-sampling and a normalization with an empirical model of sea backscattering to avoid the incident angle effect.
The scatterometer provides not only a large amount of wind measurements of homogeneous quality to the meteorologists, but also detailed information on very intense structures such as cyclones.
The detailed structure of the cyclone Lisette is clearly shown; in particular one can note the sensitivity of each Scatterometer's antenna to the wind direction. The famous cyclone's eye where the wind speed falls dramatically is clearly visible at the near range of the swath. Two plots show the mid antenna sigma naught and the wind speed evolution across the cyclone's eye.
For the cyclone Justin the Scatterometer had covered only half of the structure of the tropical cyclone; due to the position of the cyclone's eye in the far range of the swath.
In this case the meteorological institutes have estimated for the sustained wind speed increases up to 90 knots (165 km/h). It is interesting to note that even at this wind speed the direction is very well captured by the Scatterometer. This is a clear indication that the physical phenomena allowing the wind to be measured (basically the sea surface roughness) are still there and that there is no saturation of the sea surface under the wind stress. The under-estimation of the wind speed retrieved by the scatterometer is due to the model used into the ground processing (CMOD-4) that is not well calibrated at such wind speed.