|
|
|
Disasters
|
Tropical Cyclone Monitoring Using INSAT, ERS-1, NOAA and DMSP Satellites
B. M. Rao, M. S. Narayanan, C. M. Kishtawal, P. K. Pal
Meterology and Oceanography Division,
Space Applications Centre (SIRO).
Ahmedabad - 380 053, India
V. S. Prasad, A. N. Nath, K. H. Rao, M. V. Rao
National Remote Sensing Agency, Hyderabad
S. K. Subramanyam
India Meteorological Department, Madras
Abstract
The 80-100 topical cyclones that occur each year affects many counties around the globe and exacts tremendous annual losses in life and property. Seatellites are uniquely suited to the task of tropical cyclone surveillnace. Recent observational studies uing extensive satellite data to supplement conventional surface and upper air data have revealed the structure of tropical cyclones. An attempt has been made here to demonstrate the multi-sensor approach based on VHRR data from INSAT, Microwave sounder data from NOAA, ERS-1 scatterometer derived surface wind vectors and rain rate measurements from SSM/I in the diagnosis of tropical cyclones in the Indian region.
1.0 Introduction
Tropical cyclone ranks with earthquakes as the major geophysical causes of loss of life and property. A tropical cyclone is a warm-core low-pressue system that develops over the warm tropical oceans in association with an upper-level anticyclone, with intense winds moving anticlockwise (clockwise) in the northern hemisphere (southern hemisphere) around its clear, calm eye region. After crossing the land they cause destruction in the coastal areas because of the storm surge (rise in sea level inundating coastal areas), strong winds and heavy and widespread rainfall causing floods. All over the world increasing attention is being paid to the understanding of the mechanism of formation, development, structure and movement of these cyclones.
During the last few years a number of polar and geostationary satellites carrying a sophisticated array of instruments making measurements in the various regions of electromagnetic spectrum have provided valuable quantitative information about the cyclones. Geostationary satellite INSAT with a Very High Resolution Radiometer (VHRR) onboard provides visible and infrared pictures once in 30 minutes These are operationally used for locating the centre of the cyclone as well as to estimate the current intensity of the storm. NOAA series of satellites with a sounder facilitates the retrieval of atmospheric temperature profile even in the cloudy conditions. Unpprecedeted views of surface wind fields in tropical cyclone are now provided by the 5.3 GHz scatterometer on the ERS-1 satellite. Similarly SSM/I observations from DMSP satellite provide information on the convection and instantaneous rain rate in the cyclones.
An attempt has been made here to demonstrate the usefulness of this multi-satellite multi-sensor approach in the diagnosis of the tropical cyclones in the India region. One of the major drawbacks of tropical cyclone research with extensive satellite data has been the lack of high quality data sets to verity the satellite diagnosis. A concerted effort is required to organize aircraft reconnaissance flights over the Indian region to provide high quality data sets which would go a long way in improving the understanding of these recurring devastating atmospheric phenomena.
2.0 Surface wind observations using ERS-1 Satellite
European Remote Sensing Satellite (ERS-1) launched in 1991, with an onboard 5.3 GHz Scatterometer, provides surface wind fields, which is one of the most important parameters along with surface heat fluexes in the feed back mechanism driving tropical cyclones. A number of cyclones in the Indian region sicne 1991 with ERS-1 coverage have been analysed to study different aspects of cyclones particularly the evolution of rotating wins. Here we present two typical cases of scatteroomter-derived surface wind fields beneath the cyclones.
Fig. 1 (a) shows the surfae winds (fast deliver product from ESA) asociated with the cyclne in the Bay of Bengal on November 19,1992. On this day the cyclonic storm had attained the hurricane intensity and was located at 11.5° N and 880 E. The ERS-1 sub satelite track was passing almost over the centre of the cyclone.
Figure 1 ERS-1 Scatterometer derived Surface wind field over the Bay of Bengal cyclone on November 19, 1992.
Although the spiraling flow is quite evident from fig.1 (a) , a close inspection reveals that from most of the vectors the ambguity in the seen to be out of phase by 1800 Attema (1993) has also reported a similar directional ambiguity in the case of typhoon Elsie south of Japan. The wind vectors have been corrected under the assumption of the cyclonic flow around the known centre of the cyclone. The corrected wind field (fig. 1 b) depicts the typical surface gradients in the wind speed with low wind speeds in the calm eye region of the cyclone and highest wind speeds just 10-20 km away from the centre are clearly see. In general the wind speeds ranged bewteen 5 and 8 ms-1 in the outhe region, whereas the maximum wind speeds in the range o 22-25 ms-1 in the eyewall region are seen. On this day (November 19,1992) the intensity of the cyclone as inferred from the INSAT-VHRR imagery was T-5.5 on the Dovrak's T-number scale (Dvorak, 1975). Using the empirical relation for Bay of Bengal, this T- number corresponds to a maximum wind speed of around 50 ms-1, whereas the ERS-1 Scatterometer estimates are around 25 ms-1. Obviously, the high wind speeds associated with the tropical cyclones are highly smoothed in ERS observations. Similar features were noticed in the case of Arabian Sea cyclone on November 14, 1993 (fig.2)
Figure 2 Same as fig 1 except for the Arabian Sea cyclone on November 14, 1993.
Quilfen et al (1993) have analysed the scatterometer surface wind fields of a number of hurricanes and have listed the main reasons for the shortcomings as (I) the inadequac of the C-band model function for high wind speed conditions (ii) coarser spatial resolution (50 m) of the scatterometer and (iii) the attenuation of the radar signals by the intervening atmosphere.
|
|
|
|
|
|
|
