Study On The Theory And Method For Adjusting The Sounding Data Acquired In A Grid Pattern

Funded by the National Natural Science Foundation of China, the basic theory and method have been presented in this dissertation to detect and adjust the systematic error in sounding data acquired in a grid pattern, in order to answer a scientific question in hydrographic surveys: can the disagreements/differences at the crossing points be used to improve the precision of sounding data and assess its quality as land surveys do? The results from this dissertation show that those disagreements/differences can be used to detect effectively systematic error in sounding data and improve its quality by the theory of rank-defect adjustment when the disagreements/differences at crossing points are separated into two parts: systematic errors along sounding lines and random errors. And the main works and creations are as the following:1. The characteristics of the mode of hydrographic sounding and the spatial scope of the effects of error elements in sounding are analyzed to demonstrate that adjustment of sounding data in a grid pattern is an essential means to detect systematic error and improve the quality of sounding data. The error model of the disagreements/differences at crossing points and its rank-defect characteristics are brought forward, which are the base of this dissertation.2. The vital theory of rank-defect adjustment and its mathematic model are brought forward to detect and adjust the systematic errors in sounding data, in which the key problem is how to select datum sounding lines. The theoretical analyses show that the sounding lines, on which there are sure no systematic errors by some reason before calculation, can be selected as datum lines; otherwise, the datum lines should be selected by optimization calculation.3. It is a key problem to select datum lines for the rank-defect model before calculation. In order to solve it scientifically and reasonably, a graph comprised of circle of differences is deliberately designed to display the depth disagreements/differences at the crossing points in a visual way. The graph can help select the datum lines in a visual way before calculation. And the graph can also be used to judge directly the systematic errors and outliers in the soundings in a visual way in the field work, this is to say, the graph can help assess the quality of sounding data. Its ability to display the quality of sounding data is better than that of the forms of numeric values of depths or depth disagreements.4. An optimization method to select datum lines is put forward when the datum lines can not be determined before calculation, which is a key problem in the solution of the rank-defect adjustment model. It is founded by research that there are two effects: transferring effects and covered effects, which make a part of the systematic error along a datum sounding line transfer to the other estimates and may cause the true systematic error along a sounding line covered by the transferred systematic errors from the datum sounding lines, when the sounding lines on which there are systematic errors are wrongly selected as datum lines. On the basis of hypothesis testing theory, a t-test method is brought forward to select the datum lines step by step to minimize the two effects after the theoretical analyses.5. The effects from the weight of depth differences on the results of adjustment are analyzed detailed to present the three classifications of variances of depth differences: by sounding line, by surveying date and by depths. And then the method based on the Helmert method is tried to determine the variances of depth differences. The research shows that the Helmert method can work well to estimate the variances of the disagreements/differences at the crossing points and help improve the resolution of systematic error along a sounding line in calculation and the quality of sounding data.6. The effects of the structure of a sounding grid on the results of adjustments are analyzed and the relationships between the structure and the resolution of systematic error along a sounding line and between the structure and the reliability of the adjustment model are obtained. Based on the relationships, the advice is brought forward to optimize the sounding grid so as to improve the systematic error resolution and reliability of a sounding grid.7. Considering the complication of a sounding surrounding, a method to adjust the sounding data from an irregular grid is brought forward, in which, an irregular grid can be divided into the two parts: regular and irregular. The data in the regular part can be used to detect and adjust the systematic errors according to the method presented in this dissertation. And the data in the irregular part can be used to assess its quality in a statistical way. The example from the observations shows the method works well to process the sounding data from an irregular sounding grid.8. The software for simulation analyses and for applications of the theory and method in the dissertation are developed. The functions in the software includes: 1) visual display of sounding data, 2) the determination and visual expression of the disagreements/differences at the crossing points, 3) the selection of regular part in an irregular grid for adjustment, 3) automatic forming adjustment equations, 4) automatic selection of datum lines, 5) calculation of adjustment, 6) correction of the sounding data, 7) simulation calculation for theoretical analyses. The software provides the technical support for the theory and method in the dissertation and proves they can be used in practice.