Font Size: a A A

Study Of Sea Surface Wind Remote Sensing By Satellite Multi-sensor Data

Posted on:2005-01-14Degree:DoctorType:Dissertation
Country:ChinaCandidate:Q FengFull Text:PDF
GTID:1100360125965676Subject:Physical Oceanography
Abstract/Summary:PDF Full Text Request
Microwave scatterometer, radiometer and synthetic aperture radar (SAR) -have1 the capability of monitoring sea surface winds (SSW) under both clear-sky and: cloudy conditions. In this thesis, the general principles and methods of SSW retrievals using satellite scatterometer, radiometer and SAR data are reviewed. The fundamental for SSW retrievals is to establish the relationship between the normalized radar cross sections (NRCS) and wind vectors or relationship between sea surface brightness temperatures and wind speeds.Because empirical model functions for SSW retrievals are related to other environmental parameters of ocean, e.g. sea surface temperature and ocean wave etc, the establishment of model functions for SSW retrievals in the China Seas will contribute significantly to China's satellite scatterometer projects. Using NSCAT NRCS, Special Sensor Microwave/Imager (SSM/1) wind speeds and ECMWF model wind directions in the China Seas, the first empirical model function for the China Seas is established and the coefficients of this model function are obtained by Quasi-Newton iteration. This model function is applied to NSCAT data to retrieve wind vectors, with the Maximum-Likelihood Estimator to get a set of ambiguous wind vectors and a vector filter technique to remove ambiguities. The results are compared with NSCAT wind vector products. The mean and standard deviation of wind speeds difference are 0.22 and l.l0m/s, respectively. The mean and standard deviation of wind directions differences are 1.4?and 6.4?, respectively. The results are also compared with JMA buoy data. The mean and standard deviation of wind speeds difference are 0.28 and 1.40m/s, respectively. The mean and standard deviation of wind directions differences are 2.9?and 16.2?, respectively. It indicates the validity of this model function, which provides preliminary experience for the establishment of the model function for China's future satellite scatterometer.Ambiguity removal of satellite scatterometer depends on its antenna design and scanning mode. Using ECMWF model winds, the relationships of ambiguity selection accuracy with azimuth angle, incidence angle and wind speeds are analysed. It indicates that ambiguity selection accuracy of conical scanning scatterometer is better than that of fan-beam scatterometer, especially at low winds. The results can be useful for the design of antennas of satellite scatterometer to be launched by China.Though scatterometer is designed to acquire measurements of sea-surface winds under all weather conditions, the atmospheric attenuation still has influence on the measured NRCS. More accurate NRCS can be obtained after eliminating the effect caused by water vapor and cloud liquid water. The study of the atmospheric correction of NRCS is carried out based on radiative transfer model and simultaneous microwave radiometer data. The result shows the obvious improvement on the performance of wind retrievals after NRCS correction.The advantage of the SAR measurements is the capability of high-resolution wind fields retrieval. In this thesis, high-resolution wind fields are derived from C-band RADARSAT SAR data using ERS/SCAT C-band CMOD series model functions. C-band polarization ratios are needed since CMOD series model functions are applied to the C-band vertical polarization ERS scatterometer data and can not be used to retrieve wind fields from the C-band horizontal polarization SAR data. The comparison of four polarization ratio models is carried out. The wind speeds from horizontal polarization RADARSAT SAR data using different polarization ratio models are compared with collocated QuikSCAT wind speeds. The results show that Kirchhoff polarization ratio model is suitable for wind fields retrieval using horizontal polarization RADARSAT SAR data. The mean and standard deviation of wind speeds difference is 0.37m/s and 1.54m/s, respectively.Some preliminary results about the application of satellite sea surface winds in the track and stucture of typhoon, atmospheric front and forcing of oce...
Keywords/Search Tags:Scatterometer, Synthetic Aperture Radar, Radiometer, Sea surface wind, Retrieval, Model function
PDF Full Text Request
Related items