Quasi-geoid Refinement Based On EGM2008 In Jilin Province

Quasi-geoid is the datum of normal height system.The traditional method to obtain the normal height of ground points is leveling.In order to overcome the disadvantage of large workload and low efficiency of leveling,the method of obtaining normal height by combining GPS surveying with quasi-geoid model is widely used.GNSS positioning technology can quickly obtain high-precision geodetic height,so it is important to obtain high-precision quasi-geoid model in GPS height measurement.Based on the global gravity field model of EGM2008 and taking Jilin Province as an example,this paper studied the principle and algorithm of geoid refinement,and compiled the code of geoid refinement based on the platform of Matlab.The main contents include ground gravity data preprocessing,free air correction based on twodimensional Fast Fourier Transform,plate correction,terrain correction,isostatic correction,data gross error detection and gravity data gridding method.The gravity geoid of Jilin Province was calculated by using SRTM elevation data and ground measured gravity data,and according to Stokes theory and removal recovery technology.The quadratic surface fitting was used to eliminate the systematic error of quasi-geoid,and the high-precision quasi-geoid model in Jilin Province was obtained.Using 63 B-level first-class GPS leveling points in Jilin Province,the overall accuracy of the refined quasi-geoid model was analyzed according to the three indexes of error extreme value,error interval and stardard error.According to the difference of terrain complexity and altitude in Jilin Province,the accuracy of geoid model in plain,mountainous and high mountain areas was evaluated,and the systematic error characteristics of each region were analyzed.After refinement,the error interval of the geoid model in Jilin Province is 0.182 m,the standard deviation is ±0.043 m,the original EGM2008 model is 0.329 m and the standard deviation is ±0.063 m.The refined model significantly improves the accuracy of the EGM2008 model.Because of the difference of terrain and the uneven distribution of gravity data in plains,mountainous areas and alpine areas,there are obvious differences in the effect of refinement in the three regions.Quasi-geoid refinement can be carried out according to the difference of system errors in each region to achieve better results.The results of quasi-geoid refinement in Jilin Province proved the validity of the removal and recovery technology.The refined geoid model combined with JLCORS system can widely replace the conventional leveling method.The effect of removecompute-restore method to refine quasi-geoid is closely related to GPS elevation and ground gravity data.Improving the accuracy and resolution of the two data can effectively improve the accuracy of quasi-geoid model.The main purpose of Refining Geoid based on EGM2008 model is to improve the precision and resolution of the model.The gravity field model selected in the refining process directly affects the refining effect.Therefore,the use of higher precision models such as EGM2020 will further improve the refining effect.Topographic complexity is the main factor affecting gravity reduction.Selecting the appropriate integral radius is conducive to improving the accuracy and efficiency of calculation.For the small area quasi-geoid refinement,the model accuracy and resolution can be improved by improving the resolution of GPS leveling data and residual fitting method.