Analysis Of The CMONC Fiducial Stations’Coordinate Time Series Variation Characteristics

Crustal movement observation network of China is constructed with the development ofGNSS. It can be widely applied in the geological subsidence monitoring, control measurementand environmental monitoring. This paper mainly focused on various feathers and variations ofcoordinate time series of the continuous operation station of the CMONC. The main contents ofthis paper are as follows:1. To acquire the coordinate time series of the continuous operation station, a detailedtreatment and adjustment program is designed. The date processing flow by using the highprecision data processing software, GAMIT/GLOBK is researched. Other technical problemsuch as parameter setting, processing step, and data partition has been carefully handled. Thuswe got the daily results with high accuracy and the coordinate time series.2. The method to preprocess the coordinate time series was researched. To resolve theunequally sampling and data losing problems of the coordinate time series, original coordinatetime series of the continuous operation station were analyzed. Different insert methods werecompared, and proper one was chosen to make sure that the new time series after being insertedis consistent with the original time series. Finally the time series which can fulfill therequirement of spectral analysis and wavelet analysis were given.3. The coordinate time series was analyzed and researched by using spectral analysis. Datafrom different plate of china during1999to2012were used, and the periodic term and value ofperiodic power in three coordinate components of the continuous operation station werecalculated. The analysis results showed that the energy of periodic movement in up direction ishigher than that of N and E.4. Noise characteristics of the coordinate time series was analyzed. The method to determinethe parameters of power spectrum function by using the non-linear least squares fitting. Thispaper figures out that both Gaussian white noise and colored noise exist in GPS coordinate timeseries.5. Multi-resolution analysis and noise reduction processing for the coordinate time serieswere used by the method of wavelet. By using the coif3wavelet filter, the multi-resolutionanalysis is applied to the coordinate time series, and then we get the detailed analysis of theperiodical movement and non-stationary characteristics of the coordinate time series. Thewavelet noise reduction method based on the adaptive threshold is researched. The soft thresholdand the hard threshold are compared. It shows that the results of soft threshold noise reductionare smoother than that of hard threshold noise reduction. The soft threshold noise reduction canbe used to analyze the tendency of continuous operation stations, but not for overall stability analysis.6. The linear movement of the coordinate time series and the periodical and seasonalvariation in up direction were analyzed. The possible reason lead to the variation was analyzed.Among them are geophysical factors, data processing factors and so on.7. The basic theories and related models of stationary time series were discussed in detail.The methods and steps to form the right ARIMA model were given. Also, the prediction modelof coordinate time series based on ARIMA was established. It was found that the differencebetween the prediction value and the real one is at mm level, which means the prediction modelis accurate enough, and it can be used in the short and medium term prediction.