Study Of Improving The Precision Of Coast Satellite Altimetry And Application To South China Sea Tide Model

A radar altimeter can measure the shape of sea surface globally and frequenlty which have a wide applications to earth and ocean science. The satellite altimeter data can be used to provide for us the undulation of the geoid, global sea surface topography, the wave on sea surface, sea surface roughness, to detect the boundary and trend of ocean current, and to analyze the thickness of ice band on polar regions. So, it is very important for people to know earth and ocean since satellite altimetry is studied in a deepgoing way.The precision of satellite altimetry can comet to 5 cm after 30 years development. However, due to the altimeter data is disturbed by land topography, automatic gain control and ocean tide model, the precision of altimetry is lower which is a bottleneck problem for application the technology.Due to the expension is greater when the microwave pulse reach sea surface, and there are some islands near the sea, the return waveform may be contaminated by the land near sea and inlands. The satellite atimeter data near HongKong is studied and analyzed to looking into the influnence in the paper. The research shows that the waveforms are contaminated by land and islands. When the land is smooth, its reflect ability is stronger, conversely, it is weaker. When the sample points is more near island and the water is shallow, noise of waveform has a obvious increasing trend.The return waveform energy can be control by automatic gain control system so that it is retained within limits. However,in coastal zones, the complex land surface may make the return wave drastic change which will lead to the difficulties of the retracking. So the change characters of automatic gain control and its influnence on waveform are studied in the paper. The results show that the values of automatic gain control are bigger, the waveforms more complex, conversely, the ocean waveform characters are not showed in the waveforms.Ocean tide model is other mian factor which influnce the precision of satllite altimeter data in coast zones. As we known, the precision near the sea of constructed global ocean tide model is lower, therefore, improving the precison of offshore ocean tide model is a main measurement to increasing the precision of altimeter data. The water level data from 15 tide stations are used to analyze and study the precision of two tide model:CSR4.0 and NAO99 in the paper. The studied results show the precisions of the two models with different zones: the precision of CSR4.0 is higher than NAO99 near the Chinese mainland's sea, however, near the Taiwan's sea, the result is opposite, by the way, the precision of the two models is basically similar near Japanese sea.Environment correction should be done before using the satellite altimeter data. The precision of wet troposphere delay correction is lowest in the environment corrections whose influnce on altimeter data come to 2-3 cm. Therefore, it is necessary to introduce the methods about the wet troposphere delay correction and an improved method is raised in the paper.In the open sea areas, the return waveform s of altimeter basically belong to Brown model. However, in coast zones, the altimeter data need be retracked due to the waveforms more complicate. The methods of retracking has been introduced in the paper. A new multi-wave optimized retracking method based on cluster analysis is raised in the paper. According to the qulity of similarity of near waveforms, making using of cluster method, a new method on waveform class is raised base on K-Means in the paper. The class method is used to classify the return waves in coast zones and retrack the waveforms. The improving ratio is about 30% for the retracked altimeter data comparing the tide station data.A method of adjusting observation weights on different depth of water is advanced in this paper according to the difference of precision of coast satellite altimeter data. Then residual orthogonal weight is computed of altimeter data. The tide model of South China Sea(SCS1.0) is contructed using average of region based on initial orthogonal weight from CSR4.0 tide model and residual orthogonal weights from the paper. Comparing the precision of partial tide of three tide model and water data from tidal stations, the result shows that the precision of M2 and O1 is better than CSR4.0 and NAO99b. The analysis results of characters of tide of South China Sea from SCS1.0 tide model is consistent to the conclusions from other studies.