Research On Ocean Tide Loading Displacement Using GNSS

In coastal areas,the deformation caused by ocean tide loading can reach several centimeters.When GNSS,satellite laser ranging(SLR),very long baseline interference(VLBI),satellite-based Doppler orbit and radio positioning combined system(DORIS)are used for geodesy,geodynamics and geophysical research,the influence of ocean tide loading displacement must be eliminated.At present,the precision of the ocean tide loading displacement model mainly depends on the precision of the ocean tide model.However,due to the existence of shallow water effect and the lack of tide data,the precision of the ocean tide model in shallow water area is not enough,therefore,the precision of the ocean tide loading displacement model in coastal areas cannot meet the precision required by IERS specification.With the rapid development of GNSS technology and data processing methods,the deformation of coastal stations caused by ocean tide loading can be accurately obtained through GNSS,and the ocean tide loading displacement parameters can be extracted,providing a new method for studying ocean tide.The latest research shows that the precision of GNSS in estimating ocean tide loading displacement parameters can be compared with VLBI.Moreover,GNSS has the advantages of convenience,rapidity,economy and precision compared with gravity measurement and VLBI.When using GNSS to study the ocean tide loading displacement,the harmonic analysis method adopted is required to priorly specify the periods of the eight main tidal constituent to be estimated,which is likely to cause noise with similar periods to be absorbed by the tidal constituent(K1/K2 tidal constituent)corresponding to the same period,resulting in large estimation errors.Moreover,the reference model used in evaluating the ocean tide loading displacement parameters estimated by GNSS in China's coastal areas is mainly based on the ocean tide loading displacement obtained by refining the global ocean tide model or regional ocean tide model published abroad,and its precision is not enough.In addition,GPS observation data are mainly used in the estimation process.While GPS satellite orbit period and constellation repetition period are very similar to K1/K2 tidal constituent period respectively.GPS satellite orbit error is easily absorbed by these two tidal constituent periods,resulting in large K1/K2 tidal constituent error.Therefore,how to solve the large error of K1/K2 tidal constituent estimation has always been a difficult and hot issue for GNSS to study ocean tide loading displacement.Based on the regional multi-system GNSS(GPS/GLONASS)observation data,this paper analyzes the existing problems of GNSS tidal load displacement estimation method,introduces a new method of tidal load displacement,combines high-precision Chinese offshore coastal load model(CST1)to refine the global tide model and serves as a reference model.Based on the new method,the distribution uses GNSS static PPP mode and dynamic PPP mode to estimate the pressure displacement of the tide to verify the effectiveness of the new method to solve the large error of K1/K2 tide estimation.In addition,the effect of combined GPS/GLONASS observation data on static PPP mode and dynamic PPP pattern estimation results was evaluated.The accuracy of static PPP mode and dynamic PPP mode estimation is compared.The specific work and key results are as follows:1)The basic theory of tide load,including the basic theory of tidal force,tide level and equilibrium tide,as well as the estimation method of tide load displacement,points out the existing estimation method-to-analyze the problems in estimating the tidal load displacement,and then proposes a new estimation method-response analysis method for estimating the tide load displacement,followed by GNSS static PPP mode and dynamic PPP model to verify the validity of the method.2)7 global ocean tide models are refined in conjunction with China's high-precision offshore ocean tide model CST1 released by Chinese scholars and their effectiveness are evaluated.The refined global model can obviously improve the periodic effects of diurnal and semi-diurnal constituent on the coordinate time sequences of ocean tide loading.It is also possible to improve the periodic influence caused by synodic periods of O1 and K1,M2 and K2,and that of P1 and K1,S2 and K2.It is used as a reference model to evaluate the precision of the ocean tide loading displacement parameters estimated by GNSS technology.3)Response analysis method is introduced for the first time to estimate the ocean tide loading displacement.By comparing with the results of harmonic analysis,it is found that the response analysis can successfully separate GPS orbit error from K2 tidal constituent.The long-standing problem of large error in estimating K2 tidal constituent by GPS has been solved.4)Focusing on the problem that the large sampling interval is easy to lead to the large estimation error of ocean tide loading displacement parameters when GNSS static PPP obtains the coordinate change sequence of stations,the harmonic analysis method and the response analysis method are improved,and the feasibility of the improved algorithm is verified by an example.In addition,by comparing the RMS values of the ocean tide loading displacement parameters extracted from different time windows(2 hours,3 hours,4 hours and 4.8 hours),it is found that the 4-hour time window calculated by response-design algorithm has the highest precision of ocean tide loading displacement.The RMS values of K1 and K2 tidal constituents are significantly smaller than those calculated by Yuan Linguo and others.5)The ocean tide loading displacement in Hong Kong area is estimated by using GPS dynamic PPP mode,and compared with the result of 4-hour window of GPS static PPP mode.The result shows that the result of GPS dynamic PPP mode has the same precision level as that of GPS static PPP mode.When combining GPS/GLONASS observation data,the result of dynamic PPP mode is better than that of static PPP mode in 4-hour window to some extent.6)By combining GPS/GLONASS observation data,the estimation of K1 tidal constituent under static PPP model and dynamic PPP model can be improved to some extent,but the improvement is limited.